qctreecare

U.S. Secretary of Agriculture Brooke L. Rollins Statement on President Donald J. Trump’s Support for the Nationwide Year-Round Sale of E-15

Wed, 04 Feb 2026 06:08:40 GMT

(Washington, D.C., January 27, 2026) - U.S. Secretary of Agriculture Brooke L. Rollins issued the following statement applauding President Donald J. Trump’s support for the nationwide year-round sale of E-15:

“Yet again President Trump is honoring his commitment to America’s farmers and energy producers today in Iowa by announcing his support for the nationwide year-round sale of E-15. As Congress continues to work through the details, the President has been clear - get a bill that allows nationwide E-15 to his desk, and he will sign it to unleash American homegrown row crops for biofuel use like never before. America’s national security depends on our energy security, and biofuels are a crucial asset that brings more jobs and helps farmers in rural America. This action will allow up to 2 billion more bushels of corn to be consumed domestically.

There is no greater advocate for our nation's corn, sorghum, and soybean growers than President Trump. We are seeing increased biofuel demand both at home and abroad like never before. American corn, sorghum, and soybean growers fuel America and the world, and we will continue to ensure they are able to do that, but at an even faster rate under the Trump Administration. American ethanol exports are up 11% in the last year alone and this is a major opportunity as other countries expand their energy demand for biofuels. The President has negotiated historic, unprecedented trade deals and framework agreements that expanded ethanol access with new purchase agreements including in the UK, Japan, Malaysia, and Cambodia.

The Trump Administration has proven it is the most pro-biofuels administration in our nation’s history, sending a clear market signal there is a growing need for American grown commodities for fuel use. With support for nationwide year-round E-15, the highest Renewable Volume Obligation (RVO) proposal in our nation's history, and extending the 45Z biofuel tax credit through 2029 in the One Big Beautiful Bill, the Trump Administration is unleashing new domestic and international markets for our farmers and ranchers like never before. We are just getting started.”

Keep Food Safety in Play this Super Bowl

Wed, 04 Feb 2026 06:07:58 GMT

WASHINGTON, January 28, 2026 — Super Bowl parties often feature takeout, delivery and foods that are served over several hours. To help prevent foodborne illness, the U.S. Department of Agriculture’s (USDA) Food Safety and Inspection Service (FSIS) is reminding fans to keep food safety in play on game day.

“When food is served throughout the Super Bowl, it can be easy to lose track of how long it’s been sitting out,” said USDA Under Secretary for Food Safety Dr. Mindy Brashears . “As Americans gather to enjoy the game, keeping food hot or cold and serving it promptly helps protect family, friends, and guests from foodborne illness.”

Many Super Bowl favorites including pizza, chicken wings, hamburger sliders, and chili should not be left out at room temperature for more than two hours, what USDA calls the Danger Zone (temperatures between 40 and 140 degrees F). If perishable food is left out without hot or cold sources for too long, bacteria will multiply to unsafe levels that can cause foodborne illness.

FSIS encourages hosts to follow these food safety tips:

#1 Handle Takeout and Delivery Safely

Transport takeout in insulated bags if travel time exceeds one hour.
Serve food promptly or divide into smaller portions and refrigerate until ready to reheat and serve.
Keep food hot at 140 degrees F or above using a preheated oven, warming tray, chafing dish or slow cooker.
Reheat food containing meat or poultry to an internal temperature of 165 degrees F as measured by a food thermometer.
If reheating in the microwave, spread food evenly, stir thoroughly, and check for cold spots to ensure a safe internal temperature is reached throughout.
Reheat liquid foods like soups and sauces to a boil.

#2 Keep Food out of the Danger Zone

Serve food in smaller batches. Bring out one round of food during the first half of the game and another during the second to ensure your food doesn’t stay out for more than two hours.
Discard perishable foods left out for longer than two hours. To prevent food waste and enjoy leftovers after the game, refrigerate or freeze perishable items within two hours.
If food will be out for more than two hours:
Keep cold foods at a temperature of 40 degrees F or below by nestling in ice.
Keep hot foods at a temperature of 140 degrees F or above by placing food in a preheated oven, warming trays, chafing dishes or slow cookers.

#3 Use a Food Thermometer

Ensure your food reaches a safe minimum internal temperature when cooking at home:

Meat (whole beef, pork and lamb) 145 degrees F with a 3-minute rest
Ground meats 160 degrees F
Poultry (ground and whole) 165 degrees F
Eggs 160 degrees F
Fish and shellfish 145 degrees F
Leftovers and casseroles 165 degrees F

If serving chicken wings, use a food thermometer on several wings to gauge the doneness of the entire batch. If one is under 165 degrees F, continue cooking all wings until they reach the safe internal temperature.

#4 Follow the Four Steps to Food Safety

Clean: Wash hands for 20 seconds before and after handling your takeout or delivered food, as well as any raw meat or poultry you prepare at home. Clean hands, surfaces and utensils with soap and water before and after meal prep. Sanitize any surfaces that may have come in contact with food using a commercial or homemade solution (1 tablespoon of unscented, liquid chlorine bleach per gallon of drinking water).
Separate : Use separate cutting boards, plates and utensils to avoid cross-contamination between your takeout or delivery foods and any raw meat or poultry you are preparing at home.
Cook : Confirm foods are cooked or reheated to a safe internal temperature by using a food thermometer.
Chill : Chill foods promptly if not consuming immediately after cooking.

For food safety questions, contact the USDA Meat and Poultry Hotline at 1-888-MPHotline (1-888-674-6854) or email MPHotline@usda.gov from 10 a.m. to 6 p.m. Eastern Time, Monday through Friday.

Access news releases and other information on FSIS’ website at www.fsis.usda.gov/newsroom . Follow FSIS on X at @usdafoodsafety and USDA on Instagram @usdagov and Facebook .

This is a subtitle for your new post

USDA Encourages Ag Producers, Residents to Prepare for Weekend Bomb Cyclone Winter Storm

Wed, 04 Feb 2026 06:06:03 GMT

(Washington, D.C., January 30, 2026) - The U.S. Department of Agriculture is encouraging farmers, ranchers, families, and small businesses across the Southeast, southern Virginia, and potentially parts of the mid‑Atlantic and New England to prepare for a rapidly intensifying winter storm expected to develop into a bomb cyclone this weekend. USDA staff in regional, state, and county offices are ready to assist communities before, during, and after the storm.

USDA’s Disaster Resource Center and Disaster Assistance Discovery Tool offer easy access to information on programs that support recovery from natural disasters. USDA also encourages residents and producers to contact their local USDA Service Center to determine which programs may meet their needs.

Food safety guidance:

Strong winds and heavy snow may lead to scattered power outages. USDA recommends the following steps to keep food safe:

A refrigerator will keep food below 40°F for up to 4 hours during an outage. A full freezer stays cold about 48 hours (24 hours if half full). Keep doors closed as much as possible.
Do not place perishable food out in the snow. Outside temperatures can vary and food can be exposed to unsanitary conditions and animals.
Freeze containers of water ahead of the storm or make ice in containers left outside to freeze; place them around food to help maintain cold temperatures.
Freeze refrigerated items you may not need immediately—such as leftovers, milk, and fresh meat or poultry—to extend their safe storage time.
Consider purchasing 50 pounds of dry or block ice if a long outage is expected; this can keep an 18-cubic-foot freezer cold for two days.
Group foods together in the freezer to help them stay cold longer.
Keep several days’ worth of ready-to-eat foods that do not require cooking or cooling.

For food safety questions, call the Meat and Poultry Hotline at 1-888-674-6854 (Monday–Friday, 10 a.m.–6 p.m. ET), or email MPHotline@usda.gov . Meat and poultry businesses may contact the FSIS Small Plant Help Desk online 24/7, call 1-877-FSIS-HELP (1-877-374-7435) or email InfoSource@usda.gov .

Protecting pets and livestock:

USDA’s Animal and Plant Health Inspection Service (APHIS) urges everyone in the potential path of the storm to prepare now:

Ensure animals have shelter, dry bedding, and access to unfrozen water.
If moving livestock across state lines, contact the receiving state’s State Veterinarian’s Office. APHIS Veterinary Services state offices can also assist.
Follow instructions from emergency officials, especially in areas expecting blizzard conditions or coastal flooding.

Risk management and disaster assistance for agricultural operations:

USDA offers several programs to help producers recover from winter storm impacts.

Producers with Federal Crop Insurance or Noninsured Crop Disaster Assistance Program (NAP) coverage should report losses to their crop insurance agent or local Farm Service Agency (FSA) office within 72 hours of discovering damage and follow up in writing within 15 days.

Other key programs from USDA include:

Livestock Indemnity Program and Emergency Assistance for Livestock, Honeybees, and Farm-raised Fish .
Tree Assistance Program for damaged orchards and vineyards.
Emergency Conservation Program and Emergency Forest Restoration Program for land and forest recovery.
Producers should document all losses and contact their local USDA Service Center as soon as possible.

USDA’s Natural Resources Conservation Service (NRCS) also provides financial and technical assistance through the Environmental Quality Incentives Program and can support local governments through the Emergency Watershed Protection Program , which may be especially helpful in areas expecting coastal flooding or erosion.

FSA also offers financial support to farmers and ranchers impacted by natural disasters, including:

Direct and guaranteed farm loans , including operating and emergency farm loans, to producers unable to secure commercial financing.
Low interest emergency loans for producers in designated disaster areas to help recover from production and physical losses.
Loan servicing options for borrowers unable to make scheduled payments due to circumstances beyond their control.

Loans help producers replace property, livestock, equipment, feed and seed, cover living expenses, refinance farm-related debts and more.

Producers can also use tools on farmers.gov , including the Disaster Assistance Discovery Tool and Disaster Assistance-at-a-Glance fact sheet , to identify applicable recovery programs. For crop insurance claims, contact your insurance agent . For FSA or NRCS programs, reach out to your local USDA Service Center .

USDA’s Food and Nutrition Service is also ready to work with the Federal Emergency Management Agency, and is standing by for requests for emergency nutrition assistance from states and local authorities.

Statement by Secretary of State Marco Rubio and Secretary of Agriculture Brooke Rollins

Wed, 04 Feb 2026 06:04:00 GMT

(Washington, D.C., February 3, 2026) – On January 31st, the U.S. Department of Agriculture and the U.S. Department of State welcomed a new commitment between the United States and Mexico that strengthens implementation of the 1944 Water Treaty, providing greater certainty for farmers, ranchers, and producers in South Texas who rely on consistent water deliveries from the Rio Grande.

This announcement follows a call last week between President Trump and President Sheinbaum, during which both leaders reaffirmed their commitment to resolving longstanding water management challenges and supporting communities and producers on both sides of the border.

“Water is the lifeblood of the farmers and ranchers who power South Texas’s agricultural economy,” said U.S. Secretary of Agriculture Brooke L. Rollins. “This understanding between our countries is a direct result of President Trump’s determination to secure fair, practical deals that deliver for American agriculture, and we’re grateful to President Sheinbaum and the Government of Mexico for their partnership in this effort.”

“Under President Trump’s leadership and direction,” said U.S. Secretary of State Marco Rubio, “the Department of State, Department of Agriculture, and the U.S. International Boundary and Water Commission have worked to secure Mexico’s commitment to meet its obligations under the 1944 Water Treaty, while also providing a plan to eliminate the deficit from the prior cycle, strengthening water security for Texas communities and U.S. agriculture. This is another example of how the Trump Administration continues to produce benefits for the American people on issues ranging from illegal immigration, countering cartels, and modernizing trade, as well as securing water for our farmers.”

Under the negotiated outcome, Mexico committed to deliver a minimum of 350,000 acre feet of water per year to the United States during the current five year cycle, providing stability for agricultural producers and rural communities in the Lower Rio Grande Valley. Mexico has also committed to a detailed plan to fully repay all outstanding water debt accrued during the previous cycle.

Additionally, both parties will hold monthly meetings to ensure timely, consistent deliveries and prevent future deficits. USDA, the Department of State, and other federal partners will continue to work closely as implementation moves forward.

USDA Forest Service issues revised oil and gas leasing rule

Tue, 27 Jan 2026 06:09:52 GMT

New rule speeds leasing and permitting for federal oil and gas development

(Washington, D.C., January 27, 2026) — The U.S. Department of Agriculture’s Forest Service finalized revisions to its regulation governing federal oil and gas resources on National Forest System lands. The revision modernizes and streamlines the process for managing energy development across millions of acres.

U.S. Secretary of Agriculture Brooke L. Rollins and Interior Secretary Doug Burgum announced the updated rule today, emphasizing the Trump Administration’s joint commitment to eliminating outdated and burdensome processes and advancing President Donald J. Trump’s Executive Orders on Declaring a National Energy Emergency and Unleashing American Energy .

“President Trump has made it clear that unleashing American energy requires a government that works at the speed of the American people, not one slowed by bureaucratic red tape,” said Secretary Brooke Rollins . “This rule gives energy producers the certainty they need to expand supply to make energy more affordable, create jobs, and ensure America remains the dominant force in global energy markets - all while safeguarding forests and communities. Energy security is national security. These revisions create clarity and alignment across federal agencies, allowing our teams to move swiftly on leasing and permitting so American families and businesses can rely on affordable, dependable energy, while continuing to be good stewards of our public lands.”

“We are replacing the Biden administration’s bureaucratic delays with American innovation and efficiency,” said Secretary Doug Burgum . “These new rules provide the certainty needed to boost production, slash energy costs, and guarantee our global leadership. By streamlining permitting and cutting bureaucracy, we are lowering costs for families, creating jobs, and securing our nation all while protecting our public lands."

The final rule (36 CFR 228 Subpart E), now published in the Federal Register, updates and simplifies federal oil and gas leasing procedures, allowing the Forest Service and Bureau of Land Management (BLM) to seamlessly coordinate when issuing permits. By establishing a single, clearly defined leasing decision point and reducing duplicative analysis, the rule improves response times to industry requests, helps reduce longstanding backlogs, accelerates lease issuance, and supports the timely processing of applications for permits to drill.

Under federal law, the Forest Service manages the surface estate of National Forest System lands, while the BLM manages the subsurface mineral estate. The two agencies work together to develop permitting conditions under their respective authorities.

Currently, 5,154 federal oil and gas leases cover approximately 3.8 million acres (about 2%) of National Forest System lands. Of these, roughly 2,850 leases spanning 1.8 million acres across 39 national forests and grasslands contain producing federal oil or gas wells.

USDA Launches New Online Portal for Reporting Foreign-Owned Agricultural Land Transactions

Thu, 22 Jan 2026 06:11:37 GMT

(Washington, D.C., January 22, 2026) – The U.S. Department of Agriculture (USDA) is launching a new online portal to streamline reporting of transactions involving U.S. agricultural land by foreign persons, which can include businesses and governments, under the Agricultural Foreign Investment Disclosure Act of 1978 (AFIDA). The new online portal is part of a broader effort to strengthen enforcement and protect American farmland as USDA continues its implementation of the National Farm Security Action Plan (PDF, 1.2 MB).

“President Trump is putting America First, and this includes increasing transparency and scrutiny of one of our most valuable national assets, American farmland. We are working to improve reporting of foreign owned land in the United States. This move to streamline the reporting portal will increase compliance and assist our efforts to effectively enforce accurate reporting of interests held by foreign adversaries in U.S. farmland,” said Secretary Brooke Rollins . “The online portal will allow us to obtain verifiable information about foreign interests in American agricultural land and protect the security of our farmers.”

The new online portal is available at afida.landmark.usda.gov . Users can access the portal with Login.gov, a sign in service that provides secure online access to participate in certain government programs and reporting requirements.

The new digital portal will gather the same information found on the current form FSA-153 and those subject to filing may still file using the current FSA-153 hard copy form if desired. However, filers should not duplicate filings by using both submission options.

About the National Farm Security Action Plan

One of the key tenets of USDA’s National Farm Security Action Plan (PDF, 1.2 MB) is strengthening processes around disclosure of foreign persons who have an interest in U.S. farmland. This historic plan, announced in July 2025, calls for aggressive implementation of reforms to the AFIDA process including improved verification and monitoring of collected AFIDA data. In addition to the new portal, USDA published an Advanced Notice of Proposed Rulemaking for AFIDA in December 2025.

About AFIDA

The new portal is part of USDA’s efforts to streamline its process for electronic submission and retention of AFIDA disclosures, as initially required by the Consolidated Appropriations Act, 2023. Today USDA also shared its annual AFIDA report for 2024 with Congress, which is available online. The report lists foreign holdings of U.S. agricultural land as 46 million acres, as of December 31, 2024 and includes a section on land held and acquired by China, Russia, Iran, and North Korea in recent years.

AFIDA became law in 1978, and its regulations were created to establish a nationwide system for the collection of information pertaining to foreign ownership of U.S. agricultural land. The regulations require foreign investors who acquire, transfer or hold an interest in U.S. agricultural land to report such holdings and transactions to the Secretary of Agriculture.

The data obtained from AFIDA disclosures are used in the preparation of an annual report to Congress, which is published online .

The AFIDA regulations define the term “foreign person” and specifies the information that must be included in the report. AFIDA focuses on foreign persons who hold direct or indirect interest in the agricultural land, provided those foreign persons with an indirect interest have “significant interest or substantial control” in the direct interest holder.

USDA Announces New World Screwworm Grand Challenge

Wed, 21 Jan 2026 06:13:01 GMT

Driving Innovation to Combat NWS and Prevent its Northward Spread

(Washington, D.C., January 21, 2026) – Today, U.S. Secretary of Agriculture Brooke L. Rollins announced the launch of the New World Screwworm (NWS) Grand Challenge. This funding opportunity marks a pivotal step in USDA’s comprehensive strategy to combat NWS and prevent its northward spread.

“This is a strategic investment in America’s farmers and ranchers and is an important action to ensure the safety and future success of our food supply, which is essential to our national security,” said Secretary Brooke Rollins . “These are the kinds of innovations that will help us stay ahead of this pest and protect our food supply and our economy, protecting the way of life of our ranchers and go towards rebuilding our cattle herd to lower consumer prices on grocery store shelves. We know we have tried-and-true tools and methods to defeat this pest, but we must constantly look for new and better methods and innovate our way to success. Together, through science, innovation, and collaboration, we can ensure we’re utilizing the latest tools and technology to combat NWS in Mexico and Central America and keep it out of the United States.”

As part of the Grand Challenge, USDA’s Animal and Plant Health Inspection Service (APHIS) will make up to $100 million available to support innovative projects that enhance sterile NWS fly production, strengthen preparedness and response strategies, and safeguard U.S. agriculture, animal health, and trade.

Priority Areas for Funding

APHIS invites proposals that support one or more of the following objectives:

Enhance sterile NWS fly production
Develop novel NWS traps and lures
Develop and increase understanding of NWS therapeutics/treatments (i.e. products that could treat, prevent, or control NWS) for animals
Develop other tools to bolster preparedness or response to NWS

The notice of funding opportunity, including application instructions, eligibility, and program requirements, is available on the NWS Grand Challenge webpage . Applicants can also find information on the ezFedGrants website or Grants.gov by searching USDA-APHIS-10025-OA000000-26-0001.

Eligible applicants are invited to submit proposals that align with and support these priorities by the deadline on February 23, 2026 at 11:59 PM ET.

Entities interested in submitting a proposal should ensure they are registered with the U.S. Government System for Award Management (SAM) . Learn more about the basics of the funding process and  how to get ready to apply .

For more information about NWS, visit screwworm.gov . Sign up here to have the latest NWS updates delivered to your inbox .

ICYMI: Secretary Rollins Pens Op-ed in Fox News “Trump Brings Whole Milk Back to Schools, Undoing Obama’s War on Real Food”

Tue, 20 Jan 2026 06:16:16 GMT

This is a subtitle for your new post

(Washington, D.C., January 20, 2026)  – U.S. Secretary of Agriculture Brooke L. Rollins published an opinion piece in Fox News highlighting how President Trump is delivering on his promise to put the welfare of American farmers and children first by signing the Whole Milk for Healthy Kids Act.

“The childhood health crisis currently facing our nation is nothing less than an existential threat. Over 75% of kids in America struggle with  obesity, poor physical fitness , or related health challenges. These rising rates of chronic disease are influenced by several factors, but diet plays a central role,” said Secretary Rollins. “The absence of whole milk from schools has long been overlooked by countless public officials, but President Trump noticed and has done something about it. This administration understands that the national health crisis cannot be overcome without reorienting federal nutrition policy around science and real-world outcomes.”

Read the full piece below.

In 2012, the  Obama administration  decided that America’s kids didn’t need  whole milk . As a result, our children missed out on essential nutrition and our farmers lost critical income. Obama-era economic stagnation and anti-agriculture policies, including those promoting the Green New Scam multiplied hardships on the farm and many hardworking Americans began to lose hope.

Nearly one year ago, President Donald Trump’s inauguration restored that hope, and today he renews it. In signing the  Whole Milk for Heathy Kids Act , President Trump delivers on his promise to put the welfare of American farmers and American children first.

While President Barack Obama took away market share from America’s dairy farmers to fight the war on healthy fats, President Trump is expanding markets both at home and abroad, pushing for better real food options for our kids.

The Whole Milk for Healthy Kids Act, sponsored by Sens. Roger Marshall, R-Kan., and Peter Welch, D-Vt., and championed by Rep. Glenn Thompson, R-Pa., and Sen. John Boozman, R-Ark., restores whole milk to schools across the nation, delivering real food for the next generation and standing up for the farmers who feed this country.

This issue is near and dear to my heart. Last year at my confirmation hearing, Sen. Marshall asked me if whole milk belongs in school lunches. I enthusiastically agreed. I also shared that growing up, my mom made sure that our refrigerator was always well stocked with whole milk. She instinctively knew that whole milk was a building block to a healthy future for me and my younger sisters.

So much has changed since the founding of our nation 250 years ago, but the benefits of drinking whole milk have remained the same. If anything, the nutrients that whole milk naturally provides are more in demand than ever before.

The childhood health crisis currently facing our nation is nothing less than an existential threat.

Over 75% of kids in America struggle with  obesity, poor physical fitness  or related health challenges. These rising rates of chronic disease are influenced by several factors, but diet plays a central role.

We have a responsibility to help fix this crisis, especially since it was partly driven by misguided federal nutrition policies that replaced real food with ideology.

The absence of whole milk from schools has long been overlooked by countless public officials, but President Trump noticed and has done something about it. This administration understands that the national health crisis cannot be overcome without reorienting federal nutrition policy around science and real-world outcomes.

Let’s be clear—whole milk isn’t just another drink on a school lunch tray. It’s a nutrient-dense, affordable source of protein, calcium, vitamin D, and healthy fats that growing bodies and minds need to thrive.

Bringing whole milk back to schools also builds on this month's release of the  Dietary Guidelines for Americans , 2025–2030, which recognize full-fat milk, protein and healthy fats as essential building blocks of a balanced diet.

For the first time in years, federal guidance and school meal programs will complement one another, sending a consistent message to families about what healthy eating really looks like.

The Whole Milk for Healthy Kids Act restores flexibility to schools, allowing them to offer whole, reduced-fat, low-fat, or fat-free milk. This is a win for local communities and parents, who can now make choices that best serve their kids.

And equally important, it’s also a win for American farmers—the backbone of rural America.

School meal programs create consistent demand for their products, strengthen local economies and reconnect children to the food that truly fuels them. And after Thursday's announcement, the demand for whole milk will take off like a rocket.

So, to America’s dairy farmers: get ready. Gone are the days of declining milk consumption driven by failed Obama-era policy. Your hard work is back where it belongs, front and center in feeding our nation’s children.

This isn’t about partisan politics. It’s about  practical, commonsense government policy , and it’s exactly the kind of real-world reform President Trump was elected to carry out.

Brooke L. Rollins is the U.S. Secretary of Agriculture.

On One-Year Anniversary of President Trump’s Second Term, Secretary Rollins Signs Commissions for Senate-Confirmed USDA Leadership

Tue, 20 Jan 2026 06:14:54 GMT

(Washington, D.C., January 20, 2026) – U.S. Secretary of Agriculture Brooke L. Rollins today signed the official commissions of all Senate-confirmed officials serving at the U.S. Department of Agriculture (USDA), marking one year since the start of President Trump’s historic second term in office.

“One year into President Trump’s second term, we have assembled a results-driven team that is fully committed to putting farmers and ranchers first, strengthening our national security through agriculture, restoring common sense to government, and protecting America’s food supply,” said Secretary Brooke Rollins . This commission signing represents more than a milestone; it reflects our shared mission to fight for the American people, defend rural communities, and deliver on the promises President Trump made to the country.”

The commission signing took place in the Secretary’s office and brought together USDA’s full slate of Senate-confirmed leaders, along with their families, for a single, historic moment underscoring the Administration’s commitment to service, accountability, and results. Secretary Rollins signed each commission individually and commemorated the occasion with photographs alongside each official and their loved ones.

The following Senate-confirmed USDA officials received their commissions:

Devon Westhill , Assistant Secretary for Civil Rights
Dudley Hoskins , Under Secretary for Marketing and Regulatory Programs
John Walk , Inspector General
Luke Lindberg , Under Secretary for Trade and Foreign Agricultural Affairs
Michael Boren , Under Secretary for Natural Resources and Environment
Dr. Mindy Brashears , Under Secretary for Food Safety
Richard Fordyce , Under Secretary for Farm Production and Conservation
Scott Hutchins, Ph.D., Under Secretary for Research, Education, and Economics and USDA Chief Scientist
Tyler Clarkson, General Counsel, U.S. Department of Agriculture
Yvette Herrell, Assistant Secretary for the Office of Congressional Relations

Students as Researchers: An Inquiry into University Courtyards as Diverse and Inclusive Areas for Social Connection and Wellbeing

Thu, 01 Jan 2026 11:00:00 GMT

Background

Urban forests enhance mental health by reducing loneliness, fostering connections to nature, and reducing stress and anxiety. There is growing interest in understanding how urban forests can help support mental health across the life course, including among young adults. Given the known psychological and social benefits of nature-rich environments, it is critical to evaluate the functionality and usage of urban forest spaces for specific groups, particularly those at higher risk of mental health conditions like the members of this age group.

Methods

This student-led research study at the University of British Columbia’s Vancouver campus applied a mixed methods approach to assess the role of campus courtyards in supporting student wellbeing, with the ultimate aim of informing inclusive and effective spatial planning. Eight courtyards were analyzed via surveys and participant observation to understand their restorative and social benefits. Involving students as researchers played a vital role in offering alternative perspectives that helped identify previously overlooked gaps in this field.

Results

Our findings highlight the value of nearby, convenient greenspaces for young adults. There were 46 survey participants who shared their experiences in UBC courtyards, focusing on restorative and social benefits; 139 courtyard uses were observed by student researchers. Courtyards varied in biodiversity, order, and seclusion. Biodiverse courtyards received higher ratings for restoration, while social courtyards were linked to less reported guilt due to taking breaks. Across courtyard design typologies, students valued privacy, vegetation, and a sense of inclusion, although feelings of loneliness and discontent persisted.

Conclusions

This study demonstrates the value in engaging students as researchers to understand student perceptions of a campus urban forest for supporting wellbeing, social connection, and academic achievement. Although greenspaces such as courtyards are known to have restorative potential, they are not always designed to fully support student needs, highlighting the importance of student-informed planning frameworks that address existing gaps and foster more accessible, functional, and representative greenspaces on campuses.

Introduction

As our understanding of the health benefits of exposure to urban forests has developed (Cox et al. 2017; Nieuwenhuijsen et al. 2022), it has become clear that a diversity of experiences and user perspectives must be taken into account to maximize these benefits (Bratman et al. 2021; Larson et al. 2022). In response, recent studies have sought to understand variations in people’s experiences of urban forests as they age, as well as the ways in which those experiences have a differential impact on wellbeing at specific points in human growth and development (Besser et al. 2023). A recent systematic review found that biodiversity and natural characteristics of greenspaces—a broad category that includes urban forests as well as more tended areas that feature vegetation, such as parks, green roofs and walls, laneways and transit corridors, and plazas and courtyards—support human wellbeing by enhancing health, particularly mental health, and by fostering positive social interactions (Reyes-Riveros et al. 2021).

One population that has been less studied in urban forest research is young adults, those aged 18 to 25 (Peterson et al. 2024), despite the fact that members of this group are particularly vulnerable to poor mental health. Three-quarters of lifelong mental health conditions emerge at this age (Kessler et al. 2003); self-harm is a leading cause of death among young adults aged 15 to 24 (Ward et al. 2021); and COVID-19 sharply increased levels of psychological distress within this group (Wiedemann et al. 2022). In addition to these psychological vulnerabilities, young adulthood is a critical period for establishing lifelong behaviour patterns, including a relationship with nature, but studies have shown a decreasing amount of time spent in nature (Zamora et al. 2021) and an increasing sense of disconnection (Whitten 2025) at this age in comparison to earlier in life. Finally, urban forests may provide a particularly important location for fostering social connections among young adults (Collins et al. 2024), especially those individuals residing in dense urban environments, which have been linked to higher levels of isolation and loneliness (Moore 2003).

Seeking to fill this gap, the role of university campus environments in supporting student wellbeing has grown in interest (Li et al. 2024). Emerging research emphasizes the importance of designing spaces that cater to the specific needs and preferences of diverse student populations (Ha and Kim 2021; Sun et al. 2021; Guo et al. 2023). Alves et al. (2022) point to the role of biophilic attributes in enhancing students’ sense of connection, suggesting that spaces designed specifically for social interaction can improve students’ perceptions of both the space and nature as a whole. Conversely, enclosed classrooms and densified urban campuses that lack natural elements can amplify stress and anxiety (Peters and D’Penna 2020; Alves et al. 2022), potentially increasing feelings of isolation and loneliness in the student population.

Sun et al. (2021) emphasize the varying restorative impacts of different types of campus environments, highlighting the importance of incorporating diverse landscape types, including both greenspace and bluespace (such as streams, ponds, and even fountains), to cater to a wide range of preferences and needs. Here, we use ‘urban forests’ as an umbrella term to capture the diversity of nature found across university campuses. Moving beyond the broader design elements captured above, emerging evidence indicates that this specific aspect of campus design can be critically important for student success, health, and social connections. Studies have shown that urban forests on campus play an important role in student learning and engagement, increasing attention (Lee et al. 2015; Alves et al. 2022), improving mental wellbeing (Ha and Kim 2021; Guo et al. 2023; Wen et al. 2025), and supporting a sense of belonging (Menatti et al. 2019). These greenspaces can also increase a sense of connectedness to nature, or CTN (Alves et al. 2022), offsetting the so-called “teenage dip” (Hughes et al. 2019).

Urban forests grow across our campus communities in a myriad of conditions and range of sizes, making it vital to identify the restorative effects of specific characteristics, including size, plant configurations, relationship to buildings, and location of outdoor furniture (Li et al. 2024). Additionally, many studies have overlooked the diverse cultural, social, and personal factors that influence how individual students interact with campus spaces, limiting the design of inclusive environments that support the mental health of all students, particularly members of underrepresented groups such as international students or those with disabilities.

Our current research effort seeks to fill these gaps in the literature while also responding to calls to integrate young adults’ voices into greenspace research and design processes (Alnusairat et al. 2022; Barron and Rugel 2023) via the participation of young adult researchers. Student researchers contributed as equals in the research design, data collection, data analysis, and writing of this project. In addition, student researchers selected the unique setting for this study—campus courtyards—highlighting their potential as informal meeting places connected by a network of pathways and trails that provide green routes. Taking their lead, our mixed methods approach combines biodiversity audits, participant observation, and a brief survey of courtyard visitors to uncover nuances in how students experience campus courtyards as a specific form of urban forests.

Methods

Location

This study took place on the University of British Columbia (UBC) campus in Vancouver, Canada, which is set 10 km from downtown Vancouver and occupies 402 ha on the traditional, ancestral, and unceded territories of the Musqueam people. In 2024, UBC Vancouver had 48,149 undergraduate and 11,139 graduate students. The university operates on a semester system, with 2 longer terms in the winter (roughly 4 months each) and 2 short terms in the summer (approximately 1 month duration). Most students take courses during both the fall (September to December) and spring semesters (January to April). Our study took place in the fall semester of 2024.

This study adopted a ‘researchers as participants’ approach where 5 undergraduate student researchers, all young adults, led key aspects of the research with structured mentorship from a faculty member. The student researchers coproduced the survey with the assistance of the faculty member and worked independently in conducting biodiversity audits and participant observations. With support from the faculty member, the students analysed the data and assisted in the drafting and editing of the manuscript. The faculty member provided guidance throughout the study, including methodological training during the study design phase; facilitating research group discussions during site selection and data analysis; and offering feedback and collaboration on survey development and manuscript drafting. Mentorship occurred during weekly group discussions and ad-hoc consultations as needed, ensuring students had significant autonomy while benefiting from expert oversight. This balance allowed students to drive the research process while leveraging faculty expertise to refine their work.

To investigate student uses and preferences for campus spaces, the research team selected a range of campus courtyards that students might use throughout the day. Campus courtyards (see Figure 1) were selected as the study’s setting by student researchers because they are located near classrooms, provide seating, and are designed for restorative and/or social activities. The research team mapped all potential campus courtyards, identifying 22 potential sites. Two student researchers spent time at each of the identified courtyards to assess their potential for restoration, focusing on three qualitative criteria: perceived biodiversity (e.g., presence of diverse plant species); capacity for seclusion (e.g., presence of enclosed or quiet areas); and degree of order and maintenance (e.g., cleanliness, upkeep of landscaping). These assessments were based on visual observations and the researchers’ experiences as students, without formal scales, to prioritize the selection of sites aligning with young adults’ likely preferences.

Figure 1.

Locations of campus courtyards measured in this study (pink boxes).

Following these initial visits, the research team convened a group discussion to narrow the initial list of 22 to 8 specific sites for detailed study. This discussion involved ranking sites based on the 3 criteria (biodiversity, seclusion, and order) to capture a range of courtyard types. Additional considerations included proximity to academic hubs (e.g., classroom buildings, cafes) and high student traffic, as the student researchers believed these factors would increase student use, advancing the study’s feasibility. Other variables—including accessibility (e.g., wheelchair access) and comfort features (e.g., shade, benches), ecological, and spatial characteristics—were also taken into consideration. The 8 selected courtyards were spread across campus while being located near major academic hubs and main campus walkways, ensuring relevance to the young adult population.

After site selection, biodiversity audits and observational measurements (see below) provided quantitative data on biodiversity, seclusion, and order, which informed the grouping of the 8 courtyards into 4 pairs. This grouping was achieved through team consensus, aligning with Barron and Rugel’s (2023) framework for tolerant greenspaces, as shown in Figure 2.

Figure 2.

Qualities of campus courtyards. We grouped the 8 courtyards into pairs according to their qualities of biodiversity, order, and seclusion (adapted from Barron and Rugel 2023). Two courtyards scored highly in all three qualities. Four courtyards scored medium for two categories and low in one. Two courtyards scored low in diversity and order, but medium for seclusion.

Biodiversity Audits and Observational Measurements

On an initial visit in September or October, a team of two student researchers completed a robust biodiversity audit at each site. Parameters measured included plant species, size, and a range of functional plant traits such as leaf shape, size, bark texture, and presence of fruits or flowers. Because data were collected in the fall, not all traits were observable.

As a complement to these audits, observational measurements were drawn from the Survey Instrument for Plaza Stationary Activity Scans (Kim 2016), prioritizing variables that the student researchers felt resonated with their personal campus experiences. The key adaptation was to add ‘studying’ as a potential activity, as suggested by the student researchers. Observations were taken at each courtyard in October and November during 3 separate visits by a single student researcher that lasted between 1 and 3 hours apiece in a range of weather conditions. This timing was guided by the student researchers’ interest in capturing data during potentially stressful times for the young adult population of interest, including exam periods.

Grouping the courtyards into pairs occurred after the observational and biodiversity measurements, which allowed the students to better understand the qualities of each space. Following a discussion, the research team came to a consensus that the courtyards naturally grouped into 4 pairs (Figure 2). The first pair, biodiverse courtyards, had similar characteristics of a high biodiversity of plants, high amount of planted material, and less open space for social activities. The next pair, open courtyards, had less plant material and more unprogrammed space, giving an impression of extent with less seclusion. The third pair, social courtyards, had cohesive designs focused on social spaces, with lower biodiversity and less sense of seclusion. Finally, the last pair, forgotten courtyards, ranked low across all 3 qualities except seclusion, because they were tucked away and had fewer signs of maintenance.

Survey

Signs were posted in each courtyard encouraging visitors to participate in a short online survey, along with a QR code link to the survey itself. Student researchers also approached users of the space with the survey link. Because the UBC campus hosts a number of facilities that are used by members of the public—including a hospital, museums, and sports and event facilities—participation was open to undergraduate and graduate students, faculty and staff members, and visitors to campus in order to increase the total number of participants. As a mixed methods study, we did not pre-specify an adequate sample size figure but instead sought to collect enough data to reach saturation during thematic analyses.

The survey was developed iteratively by the entire research team with students offering feedback on how they would interact with such a survey and what the optimal length would be to encourage participation by their peers. Each member of the team identified potential items from other campus environment studies, testing them with the team as a whole, and adapting them for the young adult population. Adjustments were made to individual items to ensure clarity and to make them comprehensible from the perspective of both researchers and students. Importantly, two of the student researchers edited items to ensure they were understandable by individuals with dyslexia and ADHD. We strove to create surveys that felt approachable and empathetic, rather than harsh, intrusive, or dehumanizing.

The survey included basic demographic questions, including gender identity and relationship to UBC. Participants were asked questions to gauge their current psychological state, including their ability to concentrate and their feelings of loneliness. They were then asked questions about their experience of the courtyard space they were using. The psychological questions included a short version of Hartig et al.’s (1991) Perceived Restorativeness Scale (PRS)(Negrín et al. 2017), which gauges whether a space induces feelings of ‘being away’, ‘fascination’, ‘coherence’, ‘compatibility’, and ‘scope’. Our survey also included questions to learn more about how using the space made participants feel, how often they used the space, and why they used the space. All items were measured on a 5-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree’. The Appendix includes the full survey instrument.

To gauge feelings of social connectedness, we chose the 3-item UCLA Loneliness Scale because it was brief but effective (Hughes et al. 2004) and because the language resonated with the student researchers. We used the same Likert scale as the loneliness questions because the student researchers agreed that a consistent scale across measures was easier for participants to follow. As a result, this survey component was scored slightly differently than typical applications, which ask participants to respond ‘hardly ever, some of the time, often’ to each item and for which a score of 3 to 5 is categorized as ‘not lonely’ while 6 to 9 is ‘lonely’ (Hughes et al. 2004). Finally, two open-ended items asked participants to complete a sentence, offering additional insights into their other survey responses. The prompts were: “I spend time here because…” and “After spending time here, I feel…”.

Human Ethics approval for this study was obtained from the UBC Research Ethics Board prior to collecting participant data, and all analyses were conducted in Microsoft Excel for Mac, version 16.91 (Microsoft, Redmond, WA, USA).

Results

Participants

A total of 46 full survey responses were recorded. Of the participants, 25 identified as female, 19 as male, and 2 as nonbinary. Undergraduate students made up 37 participants, 4 were graduate students, 4 were faculty/staff members, and 1 was not affiliated with UBC.

Research suggests that many university students experience chronic stress and that this stress can impact their ability to succeed in the classroom environment (Foellmer et al. 2021; Guo et al. 2023; Wen et al. 2025). In order to get a sense of how participants were currently performing, we collected data on participants’ abilities to concentrate and to pay attention. Participants were generally neutral on their reported ability to pay attention and/or concentrate on long lectures. In response to the question “Typically, I can concentrate on full lectures” using a 5-point Likert scale, the average response was 3.45, between neutral and somewhat agree. Broken down, 22 participants ‘somewhat agreed’ that they could concentrate on a full lecture, while 17 ‘somewhat disagreed’, and 7 were neutral. When asked about their ability to “pay attention to a long lecture (over one hour)”, the average response was 3.17.

When asked about loneliness, on average, participants indicated that they were lonely. Responses ranged from 3 to 9 out of a total of 9 on our adapted version of the 3-item UCLA Loneliness Scale, with an average of 6.17. A score below 6 indicates that participants did not feel they lacked companionship, felt left out, or felt isolated; a score of 6 or greater indicates that participants sometimes felt these 3 dimensions of loneliness; a score of 9 indicates that participants often felt this way. In detail, 12 participants (26%) scored under 6, which we categorized as ‘not lonely’, while the remaining 34 (74%) participants were categorized as ‘lonely’.

Biodiversity and Other Courtyard Characteristic

To elucidate the survey findings, we measured the characteristics of each courtyard, including biodiversity, exploring connections between courtyard characteristics and participant responses. Select results on the measured biodiversity, order, and seclusion of the 8 courtyards are shown in Table 1. We report biodiversity in terms of number of species. Some courtyards had only one type of tree, while others had up to ten unique species. Understory diversity ranged from one to seven species. Order was categorized by researchers as ranging from ‘low’ to ‘high’.

The survey asked participants to list any features that attracted them to a space, with nearly half (20 of 46) mentioning vegetation and trees. Some participants linked these natural features to providing a sense of being away: “It doesn’t look like a school” and it “feels kind of wild and overgrown, the trees and brick building have a fun vibe”. Four participants mentioned specific tree species they connected with: Pinus coulteri, Acer palmatum (two), and Fagus sylvatica. Others noted that large trees provided privacy. One spoke of viewing the courtyard from class: “The bushes and scenery around the buildings give a sense of nature while attending classes”.

Open space was highlighted as important by 5 participants. An opportunity to sit in the sun was mentioned by 3 participants, which is likely important in the rainy autumn months during which this study took place. Shelter from rain was also noted as an attractive feature by 2 participants. Seating, including benches and picnic tables, and amenities such as electrical outlets were noted by 7 participants. Two spoke of the importance of student life and interacting with others: “best times were chilling here before class started in the morning”. In addition, multiple participants mentioned seclusion as attractive: “I like how it’s seemingly private but it has an outlet into the more open portions of Main Mall, it’s a good mixture” and that the courtyard provides “privacy away from walkways”. Other notable features mentioned included proximity to class (4), maintenance (3), and design (2). Two participants also mentioned the red brick colour of one courtyard as an attractive feature.

Courtyard Uses

Based on the student researchers’ observations, the number of users at each site varied greatly, ranging from 2.9 to 20.6 per hour, as shown in Table 2. We noticed a divergence in the number of users by courtyard type, with the social courtyards receiving the most visitors per hour, on average. Open courtyards had the second highest number of users, suggesting that university students prefer less-secluded spaces, independent of size. Observational results reveal that higher biodiversity and better maintenance of courtyards were not necessarily valuable characteristics in attracting students. With respect to individual activities, courtyard users were seen eating, talking, watching electronic devices, studying, and smoking. In the social courtyards, talking and studying were primary activities. In biodiverse courtyards, visitors primarily ate and used their phones, while forgotten courtyards were most commonly used for smoking.

Looking next at the survey responses, the most common reasons for visiting courtyards were ‘passing through’ (80%), ‘relaxing and unwinding’ (53%), to ‘enjoy nature’ (46%), to ‘eat or drink’ (37%), ‘getting away’ (35%), and to ‘chat with friends’ (35%).

The potential for courtyards to primarily serve as a transitional space was highlighted by the open-text responses, with 15 respondents completing the phrase “I spend time here because…” with the concept of passing through. These qualitative responses offer additional insights into how simply passing through can still provide a sense of restoration, however, with 8 of the 15 respondents noting that the space was relaxing and others remarking on courtyards as places that provide quiet as well as a safe space for immunocompromised individuals.

Although chatting with friends was a common motivation for visiting a courtyard, most participants reported visiting these spaces on their own (70%). Again, open-ended responses offer a unique perspective on how such solo visits might be restorative, with one participant highlighting the benefits of having a space that “helps me remain centred and enjoy quality time by myself”. Other participants noted that spending time in the courtyard “makes me happy”, “helps me remain centred”, and “releases some stress and anxiety”. One participant noted, “I spend time here because I have to[;] otherwise I’m sad”. Open-text comments also clarified social uses of the courtyards, with 3 participants mentioning using the space to eat lunch or chat with friends.

Table 3 shows the results of the survey questions asking whether courtyard visitors feel bothered by the presence of others, or, conversely, whether the presence of others creates a sense of belonging. The overall score was 2.77 for “I feel bothered by the presence of others”, between ‘somewhat disagree’ and ‘neutral’; for “the presence of others creates a sense of belonging”, the overall score was 3.05, just slightly above ‘neutral’. There was little variation in response by courtyard type, although participants felt somewhat less bothered and a greater sense of belonging from the presence of others in the open and social courtyards.

Restorativeness

A key finding was that the biodiverse courtyards that measured highest for biodiversity, order, and seclusion were also those that rated highest for restoration according to the PRS scale, with the exception of the ‘being away’ measure (Figure 3). This finding aligned with the student researchers’ pre-study intuitions that more nature would be more restorative. The forgotten courtyards were more highly rated for ‘being away’, perhaps as a result of their smaller sizes and sense of isolation. In general, the biodiverse courtyards were rated more highly across all PRS measures, while the social courtyards had the lowest ratings.

Figure 3.

Courtyard ‘restorativeness’ according to the PRS scale.

The biodiverse courtyards were also the most highly rated in the survey questions asking about health benefits of spending time in those urban forests (Table 3). The overall findings fell between ‘neutral’ and ‘somewhat agree’ for questions asking about increased energy, balance, and concentration after spending time in these courtyards.

The open courtyards scored lowest for being bothered by the presence of others and highest for feeling that the presence of others creates a sense of belonging. The qualitative responses add some depth to these findings. When participants were asked to complete the sentence “After I spend time here, I feel…”, the majority wrote ‘relaxed’ (10) and/or ‘calm’ (10). Other positive outcomes included: ‘refreshed’ (4), ‘better’ (2), ‘good’ (2), ‘happy’ (2), ‘productive’ (1), and ‘motivated’ (1), with one participant responding that “after I spend time here, I feel like I belong”. Not all participants found a benefit from spending time in courtyards, however: 6 had a neutral response, 2 felt confined, 1 felt stressed that they had to go back to campus, and 1 noted feeling ‘dead inside’.

Discussion

Courtyards Provide Restoration for Stressed University Students

Our study aligns with others that have reported that young adults are facing challenges with psychological well being (Barron and Rugel 2023; Li et al. 2024) and loneliness (Astell-Burt et al. 2022), with a number of qualitative responses offering insights into the ways in which students are struggling with these issues. Our respondents’ lack of ability to ‘concentrate’ and ‘pay attention’ is particularly troubling in light of the fact that 80% were university students who must depend upon these capacities to succeed in their education. Other studies with this age group have demonstrated similar results (Peters and D’Penna 2020; Guo et al. 2023). There is a wide range of potential causes for this inability to pay attention—including poor pedagogy (boring lectures), inconsistent sleep patterns, or overwhelming course loads—and these factors are important to explore in future studies. The higher restoration scores we found for visits to biodiverse courtyards point to a promising trend, corroborated by other studies, that biodiverse nature is especially psychologically restorative (Ha and Kim 2021). Together, these results indicate that increasing biodiversity across campus landscapes could support psychological restoration. Unfortunately, however, the results of our participant observation show that fewer students visit such sites in comparison to open or social courtyards.

We tested the potential for courtyards to serve as psychologically restorative campus spaces via the PRS, finding a connection between high biodiversity and the specific PRS component of ‘fascination’. Visits to biodiverse courtyards were rated as more fascinating, while social courtyards rated low for ‘fascination’. Other components of the PRS have less clear relationships with our biodiversity assessments, although qualitative feedback indicated that participants value courtyards with greater vegetation and few hard surfaces. Openness, captured by the component ‘scope’ in our study, was highest in the biodiverse and open courtyards. Sun et al. (2021) similarly found that ‘extent’ (a sense of space and escape) most clearly connected with restoration in campus spaces. Wen et al. (2025) reported that openness, particularly where building enclosures allowed views of vegetation instead of hardscape, positively impacted evaluations of restorativeness. Height also appears to be important, with Huang et al. (2024) noting that tree height, ground texture, and moderate building heights were all critical to the restoration offered by pedestrian spaces. Through the qualitative comments and restorativeness scores in our study, we note that ‘scope’ was rated highly across courtyard types, suggesting that views of vegetation are an important aspect of restoration in campus urban forests.

Our findings align with previous work highlighting the value of nearby, convenient greenspaces for young adults (Foellmer et al. 2021; Peterson et al. 2024; Whitten 2025) rather than other studies that have asserted that students prefer greenspaces located away from buildings and social campus environments (Windhorst and Williams 2015). The high number of participants reporting that they used courtyards for ‘passing through’ highlights the potential importance of incidental nature exposure, which would benefit from additional research. This possibility is supported by qualitative responses indicating that convenience is a factor in engaging with courtyards. Sun et al. (2021) note that longer exposure to restorative spaces does not necessarily enhance their benefits, indicating that quality and design are more influential than duration.

Diversity of Designs Supports Diverse Opportunities for Wellbeing

Our findings align with Foellmer et al.’s (2021) framework, which suggests that healthy academic spaces exist across a spectrum of experienced, symbolic, and social spaces. The framework suggests that campus greenspaces support student wellbeing in 3 ways: (1) by introducing a restorative effect; (2) by creating campus identity and a sense of belonging; and (3) by creating spaces for social encounters. As shown in Figure 4, different courtyard types can achieve each of these aims, but no single site is sufficient. The variety of spaces and the spectrum of potential wellbeing benefits could explain the neutral responses to our survey questions about health benefits. Our averaged findings fall between ‘neutral’ and ‘somewhat agree’; however, we weren’t able to parse out whether participants were in the right space for their specific needs. For example, a student seeking quiet restoration might not find health benefits from a more socially oriented courtyard design. Cultural and psychological differences may influence responses, and future studies could explore the extent to which specific demographic groups of students are bothered by the presence of others. As noted in other studies (Ha and Kim 2021; Sun et al. 2021), there is no uniform solution to support student wellbeing via the integration of campus nature.

Figure 4.

Healthy academic courtyards (adapted from Foellmer et al. 2021). Our courtyard study did not include active spaces, which would also contribute to physical wellbeing.

hat said, specific design characteristics of vegetation have been linked to greater restoration, including layered hues and textures, softening harsh edges of buildings, and increasing a sense of seclusion from high-stress environments (Huang et al. 2024). Increasing biodiversity does not always provide conditions for restoration, however (Aleves et al. 2022). Some promising trends were highlighted in our observations and survey, particularly related to the social courtyards. Visitors to the open and social courtyards in our study reported higher scores in the socially relevant domains of our survey, with participants suggesting that they ‘somewhat agree’ that the presence of others contributes to their sense of belonging. Interestingly, the social courtyards had a greater number of students who reported low guilt over taking breaks. Perhaps seeing others taking a break alleviates feelings of guilt from taking a break oneself.

Our survey expands upon other studies that have found that young adults are lonely (Astell-Burt et al. 2022) while also highlighting the difference between feeling lonely and spending time alone for restoration (Zhang et al. 2024; Rodriguez et al. 2025). As we observed, and as qualitative responses within our survey reveal, solitary restoration in urban forests is an important activity for young adults. These findings align with those of a study in Australia which reported that young adults found respite in greenspaces for solitary restoration during the COVID-19 pandemic, continuing this behaviour following the lifting of public health restrictions on group activities (Peterson et al. 2024). Increasing opportunities for social engagement appears to be an important contribution of urban forests to student wellbeing, but providing spaces for solitary restoration is similarly important, with our study indicating that biodiversity is a key quality to include in spaces designed to achieve the latter aim.

Biodiversity can take a range of forms. Although the student researchers measured a wide range of biodiversity attributes, we only report tree species diversity and understory diversity. Functional trait diversity, fauna diversity, and fractal patterns are important considerations beyond the scope of our study. Similarly, diversity of locally endemic flora has some connection to sense of belonging (Menatti et al. 2019), but the courtyards that we studied included a broad range of species, both local and exotic.

Benefits of Including Students as Researchers

Reaching young adults to understand their unique needs and preferences can be challenging (Birch et al. 2020), and our study demonstrates the benefits of bringing students into the research process itself in order to achieve this aim. Alnusairat et al. (2022) emphasize the importance of student input in designing outdoor courtyard spaces, noting that connectivity between static spaces is an important feature to consider. Such insights, combined with evidence-based strategies from environmental psychology, can inform the creation of inclusive, restorative campus environments that prioritize health and wellbeing.

An important finding from our research was learning which survey questions most resonated with students. Detailed discussions with student researchers during the survey design stage pointed to the importance of acknowledging and respecting the lived experiences of the survey audience: in our case, these were students who are often busy, stressed, and managing anxiety within a fast-paced academic environment. Many of the student researchers expressed guilt about taking breaks away from their computers, particularly breaks spent outdoors, so we included a survey question about this potential reaction to assess how this sentiment resonated with the broader student population. Overall, our survey found that participants felt relatively neutral regarding guilt for taking breaks in contrast to Foellmer et al. (2021), which found stronger consensus for students having a ‘bad conscience’ for spending time taking breaks in campus greenspaces.

The inclusion of students as researchers offered a range of other benefits in addition to identifying unique topics. The students suggested how their peers might interpret surveys and the issues they might face while responding, refining the survey as whole to feel more relatable to their peers. As student researchers, our team members gained a better understanding of how students respond to questions and what they find challenging, such as long surveys, overly descriptive language, or poorly structured questions. When surveys are designed without consideration for the participants’ context, they risk coming across as treating participants as mere data points, rather than as individuals.

Our study also created a training opportunity for students interested in further research work as graduate students, a benefit previously noted by Adebisi (2022). According to the student researchers, their university studies suffered from a gap between theoretical learning and practical application that often left them with a feeling of incomplete accomplishment regarding their research capabilities. In contrast, engaging in research offered them a unique sense of achievement and success, aiding in the training of the next generation of academics (Russell et al. 2007). In addition, having student researchers performing observations allowed them to blend in better, supporting the robustness of their findings because visitors were less likely to feel watched.

Reflecting on our experience as a team, we believe that having students involved in research can significantly increase engagement, especially by younger populations. Young adults are often one of the hardest groups to reach, but they may feel more compelled to contribute when a study is created and distributed by someone in their peer group. A connection with the researcher can make the study feel more personal, increasing both the frequency and quality of responses. Moreover, conducting research within the familiar campus environment generated a sense of personal relevance and local impact to the student researchers. The familiarity of the study setting also provided a sense of security during site visits, because student researchers were well acquainted with the campus layout and resources.

When students share an interest in the research topic and possess knowledge of the subject, collaboration and productivity are significantly enhanced (Adebisi 2022). In our case, these factors enabled effective brainstorming, open communication during meetings, and cooperative work outside of structured sessions. The shared enthusiasm and understanding created an environment conducive to problem solving and the successful production of research outcomes. Based on our experiences, several elements are necessary to effectively engage students as researchers: a clear and well-defined objective, the opportunity to gain relevant experience, alignment with their interests, and sufficient mentorship (Russel et al. 2007). When these factors are in hand, students are more likely to contribute their full effort, leading to meaningful research experiences.

Limitations

This study was limited to one type of urban forest (courtyards adjacent to campus buildings) and took place during one time of year. This timing limits the generalizability of our results, although it should be noted that the courtyards were observed for a large portion of one of two main academic terms, occurring over midterm exams through the last week of classes. This timing also posed a limitation due to poor weather, which likely discouraged time spent in courtyards. In addition, the fact that it took place during a stressful time of the academic year may have influenced people’s ability to get out and take a break. Our survey sample of 46 respondents did not represent every academic department offered at UBC, similarly limiting generalizability. This study primarily utilized quantitative methods, supplemented by a few open-ended qualitative questions at the end of the survey. To gain a deeper understanding, more comprehensive qualitative data collection, such as interviews or focus group discussions, is recommended, similar to recommendations from Foellmer et al. (2021).

Our study did not observe or test ‘active landscapes’ where students might move or participate in sport, but inclusion of active spaces would be an important addition to future studies. Guo et al. (2023) note that dynamic activities such as exercise had a far greater restorative impact than static leisure behaviours, suggesting that campuses should prioritize spaces encouraging movement and interaction.

Future Research Directions

Our student researchers suggested additional approaches to engage young adults, specifically university students, in future studies. They suggest collaborating with diverse academic departments, particularly those focused on mental health, psychology, forestry, and environmental studies. Additionally, connecting with student-led organizations—such as student wellbeing unions, mental-health societies, and other campus-affiliated groups—could broaden the reach of the survey. Leveraging platforms such as student-society blogs, university social media accounts, and online student communities would further amplify visibility. These methods could not only reach a more diverse audience but also foster a sense of campus-community involvement, encouraging greater participation and enriching data quality.

Specific aspects of urban forests beyond the scope of the current effort would benefit from additional study. For example, future studies could test the restorativeness of urban forest spaces with local flora, particularly looking at ‘coherence’ and users’ sense of belonging. The high number of participants reporting that they used courtyards for ‘passing through’ highlights the potential importance of incidental nature exposure, which would also benefit from additional research. Length of stay was not tested in our study, but future efforts should investigate length of exposure as a factor. Because guilt over taking a break generated a heated discussion amongst the student researchers, and was also mentioned by survey respondents, this area should be considered by other teams as well.

Conclusion

Young adults, including university students, represent a population group that is experiencing challenges with poor mental health and loneliness. Diverse urban forest spaces can provide a range of related benefits, including opportunities for psychological restoration, belonging, and social interaction. Our study found that students reported finding some amount of restoration in campus courtyards, and that more biodiverse courtyards were perceived as more restorative. Our model of students as researchers benefited both the study’s robustness and the participating students themselves by offering valuable insights into effective research design and giving young adults a voice. Moving forward, campus design must adopt a participatory approach, incorporating student engagement and interdisciplinary collaboration to create natural spaces that reflect the diverse needs of their specific communities.

Conflicts of Interest

The authors reported no conflicts of interest

Acknowledgements

Impacts of Periodical Cicada Egg Laying on Young Trees in the District of Columbia and Limitations of Using Remote Sensing Products to Assess Their Damage

Thu, 01 Jan 2026 11:00:00 GMT

Background

The periodical cicadas of Brood X (Hemiptera: Cicadidae) emerged in the summer of 2021, shortly after several thousand trees were planted throughout the District of Columbia, USA, during 2020 to 2021. There was concern that the millions of cicadas would negatively impact recently planted trees. This presented an opportunity to assess the impacts of periodical cicadas on an urban tree planting program.

Methods

Newly planted trees were surveyed, with field inspections and remote sensing techniques, for evidence of cicada egg laying and twig damage related to tree genus and location within trees and throughout the District.

Results

There were significant differences in egg-laying behavior among 10 tree genera, with the most damage observed in Acer and Nyssa. Egg laying occurred most often in scaffold and lateral branches. Egg-laying intensity was positively correlated with percent foliage lost. There was little evidence of cicada impacts on tree health as tree condition was unrelated to egg-laying intensity. Survival of the 2020 to 2021 tree cohort was similar to previous years without a large-scale cicada emergence. Remote sensing was unsuccessful in differentiating cicada caused damages from other damage.

Conclusions

Municipal urban forestry departments facing a large periodical cicada emergence may continue tree planting campaigns and avoid damage to new trees by choosing less preferred genera. Although useful, remote sensing products are not currently at the point where a non-remote sensing specialist can acquire and utilize these tools for identifying cicada damage. Field surveys are recommended for accurate delimitation of cicada activity in urban settings.

Introduction

During the summer of 2021, the District of Columbia, as well as adjacent states in the Mid-Atlantic, experienced the emergence of periodical cicadas from Brood X ( Oldland and Turcotte 2017 ; Kritsky 2021 ). The District of Columbia was inundated with millions of cicadas in parks and neighborhoods, but also in the news with warnings of a new “plague,” one comprised of large, screaming insects. Most District community members were put at ease to learn that cicadas are only interested in mating, do not bite, and, though they feed on trees as adults, appear to do little damage in the process ( Dybas and Lloyd 1974 ). However, their egg laying was a point of concern among arborists and residents, particularly regarding young trees. The prevailing message from extension offices and local experts was to avoid planting trees during a periodical cicada emergence year ( Day et al. 2021 ) or to cover newly planted trees with protective netting that would exclude egg-laying females ( Raupp 2021 ). Excluding cicadas from access to young trees with netting has proven effective ( Frank 2020 ). However, this was of little help to urban foresters in the Mid-Atlantic engaged in ambitious tree-planting campaigns, such as those outlined in the Keystone Ten Million Trees Partnership ( Ten Million Trees 2018 ). The District of Columbia’s Urban Forestry Division (UFD) was one such municipal department and, at the time of the 2021 Brood X emergence, had planted over 8,000 trees during the 2020 to 2021 planting season ( District Department of Transportation 2025 ). The District’s UFD primarily plants trees along the streets in right of way, with additional trees planted in public parks and schools. In this study, the majority of sample trees were planted along streets (approximately 87%) with other trees planted in public parks and schools (9%) and the remaining trees planted in other public spaces. Protecting more than 8,000 trees with netting was not realistic as some trees reached heights of 6 to 8 feet (1.8 to 2.4 m), nor was it economically feasible. Presented with this impending challenge, there was an interest in understanding the impacts of periodical cicada activity on newly planted trees and determine if any particular species or genera of tree were preferred or more susceptible to damage. Understanding such impacts can be used to guide tree plantings in other areas of the country during large emergences of the 13- or 17-year periodical cicadas.

Brood X periodical cicadas are comprised of 3 species: Magicicada septendecim (L. 1758), M. cassini (Fisher 1851), and M. septendecula (Alexander and Moore 1962), all with a 17-year life cycle. Periodical cicada nymphs spend the majority of their lives underground feeding on plant roots until they emerge en masse to molt into their adult phase to find mates and reproduce ( Williams and Simon 1995 ). Adults are active for approximately 4 to 6 weeks ( Williams and Simon 1995 ) and during the Brood X emergence in 2021, adults were observed from April through June ( Kritsky 2021 ). Magicicada spp. adults aggregate in trees for courtship and mating, where males produce loud calls attracting both males and females prompting females to click their wings in response to initiate mating ( Cooley and Marshall 2001 ). Karban (1981) found that periodical cicadas rarely fly distances greater than 50 meters, as they are weak and clumsy fliers and there is such a high incidence of potential mates in any given location that there perhaps is no need to fly far. Periodical cicadas live approximately 4 to 6 weeks as winged adults ( Williams and Simon 1995 ). During that time, 8 to 16 day old females lay eggs in pencil-sized twigs in a variety of species of host trees ( Smith and Linderman 1974 ; Williams and Simon 1995 ). Females use their ovipositor to slice open the stem and insert approximately 20 to 30 eggs ( Forsythe 1976 ; Williams and Simon 1995 ), creating a small egg nest. Egg laying creates an injury in stems, resulting in a visible wound and subsequent scar ( Williams and Simon 1995 ; Miller and Crowley 1998 ).

Periodical cicadas may negatively impact trees through their nymphal feeding on roots, females ovipositing on stems and twigs, or the introduction of pathogens associated with egg laying ( Williams and Simon 1995 ). Excessive egg-laying behavior causes twig dieback at locations of oviposition scars. Oviposition scarring can restrict movement of phloem and xylem solutions and increase the occurrence of decay in branches long after cicada emergence, based on each trees’ individual ability to compartmentalize decay. There is some evidence of reduced growth in individual tree species subsequent to a periodical cicada emergence year ( Speer et al. 2010 ), presumably due to egg-laying behaviors. However, other studies indicate minimal impacts of cicada oviposition behavior on long-term growth in trees ( Clay et al. 2009a ) or on forest dynamics ( Cook and Holt 2002 ).

Oviposition behavior is influenced by site factors such as forest patch size ( Lloyd and White 1976 ; Cook et al. 2001 ) and sunlight exposure ( Yang 2006 ) where, cicadas preferentially oviposited in trees with the greatest exposure to light. Yang (2006) suggests that such oviposition preferences may be explained by potential physiological benefits for adults in predator avoidance or offspring with respect to development, or more likely that well-lit trees are an indirect indicator of root habitat quality. Trees in canopy gaps and with increased sunlight are associated with greater root growth and root biomass, respectively ( Naidu and DeLucia 1997 ; Sevillano et al. 2016 ). Ovipositing in trees located in an open canopy environment may benefit offspring as trees with high sun exposure present greater habitat quality among roots. More broadly, distribution of cicadas may be impacted by conditions of the urban environment such as soil compaction due to difficulties related to burrowing ( Moriyama and Numata 2015 ) or urban heat island effect, demonstrated by greater cicada densities with higher temperatures ( Nguyen et al. 2018 ).

When female cicadas oviposit in small branches and twigs a common sign of damage is called “flagging” ( White 1981 ). Flagging occurs when physical damage causes the twig to break and bend but remain hanging on the tree ( Figure 1 ). The vascular cambium and phloem are disconnected in these two pieces of the branch, and the leaves will die and lose their green coloration. Branches full of dead brown leaves will often be left in the tree canopy for the remainder of the summer after a cicada emergence. This flagging can be seen in ground, drive-by, and aerial surveys.

Figure 1.

Flagging and dieback in (A) shingle oak street tree (Quercus imbricaria) and closeup (B) showing flagging in the District of Columbia.

Large emergences of adult periodical cicadas allow for an ideal survey of insect activity because the flagging damage they cause produces a unique defoliation pattern and is visible to observers on site and via remote sensing, while also occurring at predictable intervals and time of the year.

The Brood X cicada event presented an opportunity to utilize USDA Forest Service remote sensing resources for monitoring landscape level forest disturbance. This project utilized several remote sensing products including ForWarn II’s near-real-time forest disturbance monitor (Norman et al. 2013), high-resolution forest mapping (HiForm) 10 m pixel (10-m2 cell) change detection (Norman and Christie 2022), and high-spatial resolution satellite (WorldView 2) image interpretation. ForWarn II has readily available coarse resolution 232-m pixel (232-m2 cell) change detection products and was the logical first choice in trying to assess large-scale forest change, while the high-resolution forest change detection product (HiForm), which utilizes 10 m Sentinel-2, was requested from the Southern Research Station Threat Assessment Center (Asheville, NC, USA) and has a cell size that approximates the footprint of a single large canopy tree (Norman and Christie 2022). High resolution satellite imagery was requested from the Remote Sensing Application Center (RSAC)(Salt Lake, UT, USA). The imagery RSAC provided was WorldView-2 imagery which provides commercially available panchromatic imagery of 0.46-m resolution, and 8-band multispectral imagery with 1.84-m resolution, representing one of the highest available spaceborne resolutions on the market.

Both change detection products (ForWarn and HiForm) use satellite imagery to produce maps of the Normalized Difference Vegetation Index (NDVI) values that compare current to historical NDVI values to identify change (Norman and Christie 2020; Pontius et al. 2020). The NDVI is a widely used index that compares the red and near infrared spectral bands, available on most multispectral satellite systems, to measure the vigor, or “greenness,” of a pixel (cell) and thus measure the impact of environmental and biotic factors on vegetation (Norman et al. 2013). Changes in greenness are then used to determine or identify changes to tree health. High-spatial resolution satellite imagery is primarily available from private satellite companies such as Maxar and Planet. Products vary in spectral, spatial, and temporal resolutions.

This study describes a survey of Brood X periodical cicada damage on a young tree cohort of varied species in 2021 and also incorporates inspections of newly planted trees carried out by Urban Forestry Division staff. The USDA Forest Service applied remote sensing to quantify damage done to mature street trees. The objectives were to (1) assess the impact from cicada egg laying on newly planted trees through field inspections, and (2) determine the effectiveness of utilizing remote sensing to detect periodical cicada damage on urban street trees.

Materials and Methods

Site Description

The Urban Forestry Division (UFD) manages approximately 200,000 publicly owned trees. These trees are located in the public right of way and in public schools and parks. Tree species, size, date planted, and other data are stored and continuously updated in an open access geodatabase using ArcGIS Online, Pro, and Collector (Esri, Redlands, CA, USA). Information for this project was accessed on or before 2023 April 6.

Field Inspections

The UFD planted 8,149 trees beginning on 2020 October 5 until 2021 May 18. During the 2020 to 2021 planting season, planted trees were comprised of 149 species and cultivars, where no individual species or cultivar made up greater than 4% of all trees planted. “Planting season” refers to the years that trees were planted. For example, the “2020 to 2021 planting season” covers the span of the autumn of 2020 through the winter to the end of spring in 2021. Due to high species richness among newly planted trees, many species were represented by just a few individuals. To ensure large enough samples sizes in statistical analysis, trees were tallied by genus rather than species. Only the top 10 most commonly planted tree genera were chosen for the survey. These genera were Acer, Celtis, Cercis, Gymnocladus, Magnolia, Nyssa, Prunus, Quercus, Taxodium, and Ulmus. Each of these genera comprised between 4% to 11% of the total genera planted during the 2020 to 2021 planting season (Table 1). These 10 genera represented 52% of the trees planted during the 2020 to 2021. Of these 10 genera, a random sample of 30 individual trees were selected using ArcGIS Pro. Due to limitation on staff time, 175 out of the 300 randomly selected trees were surveyed over 10 days throughout a 2-month sampling period during the periodical cicada emergence. In short, during the late summer of 2021, the impact of periodical cicada was assessed on 175 newly planted trees planted during the 2020 to 2021 planting season that comprised of 10 genera (Acer = 22; Celtis = 15; Cercis = 19; Gymnocladus = 23; Magnolia = 11; Nyssa = 19; Prunus = 20; Quercus = 19; Taxodium = 15; and Ulmus = 12) within 7 families (Table 2).

Each tree surveyed was first inspected for standard tree inventory data collected by District Urban Foresters, such as tree condition, size measured as diameter at breast height (DBH), and date of inspection (Table 3). Tree condition was defined broadly based on a combination of criteria that include condition of the stem and branches, aboveground roots, and crown vigor. Cicada survey data were collected by the District Forest Health Coordinator using Arc-GIS Collector (iOS app version 21.0.4) by slowly walking around the tree, conducting a 360° visual inspection of each tree. Trees were evaluated for percent foliage loss, presence/absence of egg-laying scars, location of oviposition scars (lead stem, scaffold branches, or both), and egg-laying intensity estimated as percentage of branches with oviposition scars: 0% Null, 1% to 24% Trace, 25% to 49% Light, 50% to 74% Moderate, and 75% to 100% Heavy.

Observations of oviposition scars were described as located on lead or scaffold branches. The lead refers to the “leader,” an upright stem that dominates growth ( Lilly et al. 2019 ). Oviposition scars located on other lateral or scaffold branches were referred to as “located on scaffold branches.” Location of oviposition scars on individual trees was assessed because the impact of cicada damage on the leader could be greater compared to the impact on scaffold branches due to apical dominance growth patterns in some tree species. Data were collected on the following dates: 2021 August 3, 23, and 25, and September 2, 7, 10, 16, 22, 24, and 29. The sampling dates were selected to occur after peak egg-laying periods for the periodical cicadas so well-developed damage would be observable.

Because the trees used in this trial were located in different parts of the city with different habitats, and a complex mix of private and public ownership, we used a visual ground field assessment “line of sight” method to estimate cicada activity in the area around the individual tree. This was accomplished by estimating the percentage of visible trees showing cicada damage (flagging): 0% Null, 1% to 24% Trace, 25% to 49% Light, 50% to 74% Moderate, and 75% to 100% Heavy in a 360° arc around the tree. The Urban Forestry Division’s annual evaluation of newly planted trees was used to compare survivorship of trees the year of cicada emergence to previous years. Urban Forestry Division inspects newly planted trees immediately after 1 growing season, and 1 year later (after 2 growing seasons). The first inspection of newly planted trees confirms that trees are in Good or Excellent condition ( Table 3 ) by District Urban Foresters. Any trees identified with a condition of Fair, Poor, or Dead are replaced per the tree planting contract. Such trees require a warranty replacement; this is reflected in Table 4 . After 2 growing seasons, a sample of newly planted trees (approximately 10%) are evaluated for survival, condition, and several traits related to the urban environment. Additionally, observations of pest and pathogen presence are evaluated per the i-Tree Inventory and Pest Evaluation Detection (IPED) protocol ( USDA Forest Service 2010 ). Newly planted trees are evaluated after 2 growing seasons. For example, the tree cohort planted between October 2020 and April 2021, just prior to the emergence of Brood X, were evaluated for survival in the fall of 2022. A total of 786 trees were evaluated, out of 8,149 planted between 2020 to 2021. The 786 trees evaluated consisted of the following sample sizes for each of the 10 genera referenced above: Acer = 39; Celtis = 26; Cercis = 48; Gymnocladus = 23; Magnolia = 28; Nyssa = 44; Prunus = 68; Quercus = 78; Taxodium = 33; and Ulmus = 38. Survival and condition of these 786 trees in fall 2022, after Brood X, was compared to survival rates from previous years. Occasionally, upon inspection, trees are missing from the planting location; these trees are identified as “Absent” in the survival study ( Table 4 ).

Remote Sensing

ForWarn II: The initial assessment of landscape level cicada damage in the District was done using the ForWarn II online mapper system ( Norman and Christie 2022 ). ForWarn II currently uses European Space Agency’s Sentinel-3 imagery for tracking change in the Normalized Difference Vegetation Index (NDVI). Products are at 232-m spatial resolution (13.3 acres or 5.4 hectares). ForWarn II uses a historical NDVI database from 2000 to the present to compare baseline data to current NDVI values ( Norman and Christie 2022 ).

High-Resolution Forest Mapping (HiForm): HiForm is an online application that maps forest change utilizing Google Earth Engine and 10-m Sentinal-2 imagery and the NDVI to calculate change ( Norman and Christie 2022 ). The HiForm product used in this study was received directly from the Southern Research Station, and there was no additional processing or production of the files by the authors. The Southern Research Station used Surface Reflectance adjusted 10-m Sentinel 2 NDVI products. The 10-m high-resolution NDVI data layer from the Southern Research Station was filtered using the District of Columbia’s Department of Transportation, Urban Forestry Division’s tree canopy layer to eliminate areas that were not covered by tree canopy. The NDVI for the peak period of branch flagging (July to September 2021) was compared with a baseline of the highest NDVI for the same time from the previous 2 years. The threshold to target the level of change to the signature was adjusted to the desired threshold range of −5% to −15% lower NDVI (S. Norman, personal communication). The reasoning for this was that less than 5% is too subtle a reduction and may not mean anything and a greater than 15% reduction may represent a more drastic change like tree removal or death within a pixel. This layer was further simplified to presence/absence for NDVI change; cells falling within the −5% to −15% range were classified as NDVI change present and all others were classified as no NDVI change present.

Scan and Sketch Assessment: This portion of the study aimed to assess the ease of obtaining high spatial resolution imagery and attempted to digitize these products on GIS software. No statistical comparisons were made using this method. Satellite imagery data were downloaded for the period of peak cicada impact from various sources including Worldview-2 and Sentinel-2 databases (USDA Forest Service Remote Sensing Applications Center, Salt Lake City, UT, USA). Cloud free digital color images were processed using a “scan and sketch” approach within ArcMap ArcGIS 10.2 software, while using high-resolution imagery. This is meant to mimic an aerial detection survey ( McConnell et al. 2000 ) by allowing the user to “move” across the image and digitize areas of perceived damage on a computer screen as opposed to from a low-flying airplane. This period of peak impact occurred starting in late July and ran through late September 2021. Percent affected was used as the attribute to map in the program. The resulting polygons were then classified ( Hanavan et al. 2022 ) and attributed to mimic current USDA Forest Service national reporting standards ( USDA Forest Service 2022 ). For this project, a single surveyor did all the digitizing to minimize bias between surveyors, and a polygon was assigned a value based on the percentages of damage (percent affected).

HiForm vs Field Inspection: To compare the ground field assessment of cicada intensity around the sample trees to the HiForm detected change product, we created polygons that approximated the area of what the on-the-ground field technician saw when doing their assessment. Google Earth Engine’s (EE) Google Street View (leaf on) imagery from the vantage point of the sample tree was used to create a cicada damage intensity (DI) 3D-polygon of the visible area (line of site) around each individual survey tree as seen by the observer. Although the Google Street View imagery varied by location and time taken, we tried to use, when possible, the closest month available to the actual cicada injury month (July 2021). The resulting damage intensity polygons were then superimposed over the NDVI layer and the damage intensity for that polygon was calculated as the number of pixels designated as having cicada damage (positive for −5% to −15% reduction in NDVI) over the total number of pixels in the polygon and then categorized. Polygons were classified using the same categories as the field assessment. These data were then used to formulate an error matrix ( Johnson and Ross 2008 ; Campbell and Wynne 2011 ).

Statistical Analysis

The impact of periodical cicada emergence on newly planted trees was analyzed using Statistica 13.5.0.17 software (TIBCO Software Inc., Palo Alto, California, USA). We used ordinal regression and ordinal logit model because the data did not meet the assumptions of normality or homogeneity. Means of significant effects were then compared with Tukey’s contrast test. Ordinal regression analysis was used to calculate the relationship between egg-laying intensity (dependent variable) with DBH, planting year, foliage loss, and ground field assessment of cicada intensity around the sample trees, separately. Ordinal regression was also used to estimate the relationship between tree condition (dependent variable) with egg-laying intensity, ground field assessment of cicada intensity, tree DBH, and planting year, separately. Egg-laying intensity, egg-laying location, foliage loss, and tree condition were used to determine the impact of periodical cicada damage on tree genera using ordinal logit model. Trees without cicada damage were excluded from the egg-laying location analysis. All dependent variables had an ordinal distribution, and independent variables were categorical such as tree genus or continuous such as DBH. Multiple linear regression models were not used, eliminating issues of multi-collinearity.

Kappa statistics was used to quantify the level of agreement between egg-laying intensity and ground field assessment of cicada intensity around the sample trees in SAS (SAS®, version 9.4, SAS Institute Inc., Cary, NC, USA). The simple kappa coefficient, K, ( Cohen 1960 ) is a measure of interrater agreement, where the row and column variables of the 2-way table are viewed as 2 independent ratings. When there is perfect agreement between the 2 ratings, the K coefficient is +1. When the observed agreement exceeds chance agreement, the value of K is positive, and its magnitude reflects the strength of agreement. The minimum value of K is between −1 and 0, depending on the marginal proportions (SAS System Documentation).

The error matrix follows Johnson and Ross (2008) and contains columns and rows and diagonals to calculate the following statistics: the overall accuracy, omission error, commission error, the consumer’s accuracy, producers’ accuracy. To measure the degree of agreement between the two rating systems (field inspections and NDVI), the kappa statistic and Bowlers’ test of symmetry were calculated using Proc FREQ with the AGREE options in SAS.

Results

Field Inspections

The newly planted trees used in this study (n = 175) had an average DBH of 1.75 inches (4.45 cm)(+/− 0.04 SE). The ordinal regression analysis indicated that egg-laying intensity was not related to DBH (Wald test, X2 = 3.52, df = 1, P = 0.061) on trees that have a range of 0.5 to 2.60 inches (1.3 to 6.6 cm) of DBH, nor the year trees were planted (Wald test, X2 = 0.69, df = 1, P = 0.405). In contrast, egg-laying intensity showed significant differences among tree host genera (Wald test, X2 = 35.15, df = 9, P < 0.001)(Figure 2, Table 5). The impacts of egg laying on Gymnocladus spp. trees (Wald test, X2 = 15.28, P < 0.001) were significantly lower compared to the other genera, with a mean ranking of Null (Figure 2, Table 5). Trees in Acer spp. (Wald test, X2 = 18.02, P < 0.001) and Nyssa spp. (Wald test, X2 = 11.09, P < 0.001) experienced the highest impacts of egg-laying activity, with a mean ranking of Light. Trees in the other 7 genera (Celtis, Cercis, Magnolia, Prunus, Quercus, Taxodium, and Ulmus) showed egg-laying intensities intermediate between Gymnocladus and Nyssa/Acer genera, with the majority of observations showing impacts in the Trace to Light egg-laying intensity. This could indicate that female cicadas prefer ovipositing in some tree genera versus others.

Figure 2.

Egg-laying intensity in newly planted trees by genus. Egg-laying intensity was estimated as 0% Null, 1% to 24% Trace, 25% to 49% Light, 50% to 74% Moderate, and 75% to 100% Heavy (n = 175). The bars indicate the lower and higher confidence interval, the vertical line in the middle of the bar represents the means, and lines extending from the bar represent the minimum and the maximum.

Most sampled trees were located in areas with Trace (31%), Light (41%) and Moderate (18%) cicada damage (Figure 3). In contrast, 9% of sampled trees were in areas with no damage and only 1% in heavily damaged areas. This indicates that the damage a street tree experienced was not related to the estimated cicada damage of the surrounding public or private trees. Indeed, results of the kappa analysis showed that there is no agreement (K = 0.008) between egg-laying intensity and ground field assessment of cicada intensity around the sample trees (Bowker’s test of Symmetry of Disagreement; Chi-Square = 54.85, df = 10, P < 0.001)(Table 6). Similarly, the ordinal regression analysis between the estimated cicada activity in the area (ground field assessment of cicada intensity) around the individual trees and egg-laying intensity was not significantly different (Wald test, X2 = 4.80, df = 4, P = 0.307).

Figure 3.Estimate of damage to nearby trees within sight of each study tree (includes private and publicly owned trees) (n = 175).

This study found that a greater proportion of the trees surveyed exhibited egg-laying scars in scaffold branches than did trees with scarred lead stems. Excluding trees without cicada damage, 77% of trees had cicada damage on the scaffold branches, 1% on the leader and 22% in both scaffold branches and leader. Figure 4 shows all trees surveyed, including those where no egg-laying scars were observed. However, tree genus did not influence egg-laying location (Wald test, X 2 = 0.71, df = 9, P = 0.999).

Figure 4.

Percent of tree which had oviposition scars on lead branches, scaffold branches, both branch positions, or no oviposition damage (n = 175).

The percent of foliage lost was low but was significantly different among genera (F = 4.59; df = 9, 160; P = 0.005)(Figure 5). Nyssa had the highest reduction of foliage with an average of 4% of foliage lost followed by Acer with 2%. Foliage lost was statistically similar among the rest of the genera (Celtis 1.33%, Cercis 0.79%, Gymnocladus 0%, Magnolia 0%, Prunus 0.26%, Quercus 1.38%, Taxodium 0%, and Ulmus 0.42%). According to the ordinal regression analysis, egg-laying intensity was related to the percent foliage lost (Wald test, X 2 = 21.79, df = 1, P = 0.005) which indicated that the dieback recorded was caused by the cicada damage during the egg-laying process.

Figure 5.

Estimate of percent foliage lost by tree genus (n = 175).

Although, effect of egg-laying intensity and foliage lost were statistically significant, tree condition was not different among genera (Wald test, X 2 = 9.90, df = 9, P = 0.359). In addition, the data did not show relationships between tree condition with egg-laying intensity (Wald test, X 2 = 6.98, df = 3, P = 0.072), ground field assessment of cicada intensity (Wald test, X 2 = 1.25, df = 4, P = 0.869), DBH (Wald test, X 2 = 0.92, df = 1, P = 0.337) nor the year the tree was planted (Wald test, X 2 = 0.01, df = 1, P = 0.950). Most of the trees were ranked as good (68%) and excellent (20%) in their overall health condition. In contrast, 7% of the trees were ranked fair, 4% poor, and 1% dead. Results from Urban Forestry Division’s annual survival evaluation revealed that after 2 years of growth, trees planted in the 2020 to 2021 season had an estimated survival rate of 97%. This rate was higher than rates in previous years, which ranged from 90% to 95% and average of 92% survival (Table 4).

Remote Sensing

ForWarn II: The District of Columbia is 17,702.6 hectares with each Forwarn II cell covering 5.4 hectares, meaning there are roughly 3,200 pixels (cells) within the District. Although, Forwarn ll did capture change throughout the period of peak cicada activity (July to September 2021); with the maximum departure occurring on the 2021 July 4 to 27 product. During this period the extracted summary values showed only 4.8% of the cells were substantially “browner” than the baselines. As this did not appear to capture the full spatial extent of the cicada damage observed by ground observations and reports (Figure 6), we transitioned to focus on the high-resolution HiForm data.

Figure 6.

Screen capture from ForWarn II from the 2021 July 27 cycle for the District of Columbia, in addition to adjacent states of Maryland and Virginia. Resulting map included few larger areas of disturbance when reported cicada damage was widespread.

HiForm: The high-resolution data, with its 10-m pixel cell size, provided more damage locations and better represented field inspection reported cicada damage (Figure 7). The 2021 August 8 change product was chosen as it appeared to best represent the scale and distribution of the cicada outbreak when compared to field observations. In covering the District, this product contained more than 1,493,000 pixels (cells) in which 416,373 pixels (cells) were identified as having change in the recommended range when compared to previous years data (Figure 8). This product appeared to better represent the distribution and scale of the cicada’s activity when compared to field observation and reports.

Figure 7.

(A) Cicada damage visible from Worldview-2 satellite imagery at the National Arboretum, District of Columbia in July 2021; (B) ground photo of the trees with flagging symptoms in the National Arboretum taken in July 2021.

Figure 8.

High resolution (10 m) NDVI data derived image from 2021 August 8th that utilized the target threshold of −5% to −15% reduction in NDVI. Image cut to show 2020 canopy layer for the District of Columbia, and the white areas represent no canopy cover.

Scan and Sketch: High-resolution satellite images were limited due to the infrequency of collection and the need for cloud-free images. Although many platforms (e.g., Sentinel and others) were accessed for high-resolution images, only a single image from July to September 2021 was found to be useful (> 50% cloud free) and that image only covered approximately one third of the District. This was a Worldview-2 image from 2021 July 21. This single image provided enough resolution to view the signature of cicada damage (flagging) for the use of the “scan and sketch” method. This resulted in 144 polygons being mapped and categorized, of which 6% were Light, 31% Moderate, 63% Severe, and none found to be very severe (Figure 9).

Figure 9.

Results of the ground surveys and the scan and sketch survey methods. Ground surveys were carried out throughout the district while scan and sketch was restricted to areas for which aerial imagery was available.

HiForm vs. Field Inspection: The ground field assessment of cicada activity around the 175 individual survey trees produced 140 3-D variable sized polygons (some polygons overlapped with trees on the same street). Digitized intensity (DI) polygons ranged in size from 0.25 to 19 acres (0.1 to 7.7 hectares). The ground field assessment of DI resulted in 18 None, 36 Trace, 57 Light, 28 Moderate and 1 Heavy ground polygon being assessed. In the comparison to the high-resolution NDVI change detection product and the ground field assessment data, the high-resolution (HiForm) NDVI change detection product picked up detectable damage in all the sample areas. As the ground field assessment identified 18 areas without cicada damage, these observations were omitted from the error matrix to make for a symmetrical analysis. These results can be seen in the error matrix of Table 7. The diagonal from the upper left to lower right shows the proportion of correctly classified areas between NDVI and the field inspections. The overall classification accuracy of 24%, with only 30 of 122 areas in agreement. The agreement for the two methods was low kappa (K = 0.016) and was not significant because the 95% confidence interval (0.0625 to 0.0938) includes 0; this does not support any agreement between the two methods. The Bowker’s test of Symmetry of Disagreement (X 2= 68.08, df = 6, P < 0.001) indicates that the disagreement is asymmetrical (null hypothesis of symmetry is rejected), meaning the observations contributing to asymmetry are represented by heavier counts in the upper right hand side corner of the diagonal in the error matrix. Specifically, there were many more polygons rated as severe and very severe by NDVI, while most of them were not rated as damaged by cicadas in the field ground assessment.

Discussion

This study evaluated a sample of newly planted trees in the District of Columbia after the growing season in the summer of 2021, during the Brood X emergence. Trees were evaluated for evidence and impacts of cicada oviposition behavior. Egg-laying preference was different among tree genera and did not have a relationship with the ground field assessment of cicada intensity. This indicates that females might discriminate against certain tree genera. Acer and Nyssa trees experienced the highest levels of egg-laying activity (Light), while Gymnocladus trees exhibited the fewest incidents of egg laying. Females may exhibit preferences for particular tree genera, but acting on such preferences was possibly limited by short dispersal flights within the surrounding area from which they emerged.

Other studies have reported differences in periodical cicada egg laying according to host tree species and genus (Dybas and Lloyd 1974; Miller and Crowley 1998; Clay et al. 2009b). This study found the greatest egg-laying impacts on Acer and Nyssa, similar to an earlier study by Miller and Crowley (1998) reporting heavy egg-laying damage and flagging in Acer spp. and no egg laying in Gymnocladus. Though other studies have reported egg-laying preferences for Quercus spp. as hosts (Dybas and Lloyd 1974; Perkovich and Ward 2022). Observed differences in egg-laying intensity by genus could be explained by overlap in distribution between periodical cicadas and host trees or by chemical or morphological host tree defenses. Acer (based on the most widespread Acer spp., A. rubrum)(Peters et al. 2020) and Nyssa (Coladonato 1992) have quite broad distributions in central, eastern, and southern states that overlap with Brood X (Kritsky 2021). Gymnocladus is represented by just one species (G.dioicus) in North America and is a relatively uncommon species in its native range (Beckman et al. 2021). With the limited exposure of Brood X periodical cicadas to Gymnocladus due to its rarity and limited range, this genus may be unfamiliar to Brood X periodical cicadas as a potential host tree. In addition, it has relatively few pests or pathogen threats (Beckman et al. 2021).

Host tree defenses may limit egg laying, for example conifer species with resin are associated with less egg-laying activity (Miller and Crowley 1998; Clay et al. 2009b). This study did include one conifer genus (Taxodium) comprised mostly of T. distichum. However, Taxodium spp. was subject to intermediate egg-laying activity by cicadas. Gymnocladus with its stout stems (Row and Geyer 2007) and peeling bark may present a challenge to egg-laying cicadas, as suggested by Miller and Crowley (1998). In a study of egg-laying impacts on trees, Miller and Crowley (1998) note that the stem diameter of Gymnocladus typically were larger (closer to 10 mm) than the mean maximum diameter of twigs exhibiting egg-laying scars at 9.12 mm.

Newly planted trees in the District of Columbia experienced minimal effects from periodical cicada egg-laying behavior, as demonstrated by tree condition and the location of egg laying within the trees. Though egg-laying damage was relatively low in most trees, egg laying was related to percent foliage lost. Most importantly, the condition of newly planted trees was not affected by periodical cicada egg-laying damage. Egg-laying damage was minimal enough that trees with the highest amount of damage were still considered in good or excellent condition, as assessed by District Urban Foresters. In addition, the location of damage within trees varied, with egg-laying scars more frequently observed in scaffold and lateral branches compared to the lead stem. Consequently, the risk of damage from cicada egg laying on street tree canopy structure was somewhat minimized as it occurred infrequently in lead stems. Egg laying and related damage in the lead stems has the potential for long-term adverse impacts similar to those observed in topping, resulting in poor branch structure, internal decay, and greater maintenance needs (Gilman and Lilly 2002). Given the minimal impacts of periodical cicada egg laying on tree condition and form, current and prohibitory recommendations for postponing tree planting during periodical cicada emergence years in urban environments, such as Washington, DC, should be revised and reconsidered.

Studies have also found that large scale cicada emergences can actually aid in plant growth in the long term due to an increase in available nutrients as the cicada bodies decay (Yang and Karban 2019). These increases in growth are noticeable 1 to 5 years after a large cicada emergence (Speer et al. 2010; Yang and Karban 2019). Therefore, perhaps newly planted trees during a cicada emergence may receive an extra push towards successful establishment due to this increased release of nutrients.

In the months leading up to the Brood X emergence of 2021, concern for the fate of trees in the Eastern United States came to the forefront in popular and social media (University of Michigan 2021). There was cause for concern given that in the Mid-Atlantic states surrounding the District of Columbia (DE, MD, VA, WV), the private nursery and tree nursery industry accounted for a direct economic footprint of $103 million in direct sales and services in 2017 (Thompson et al. 2021). In addition to direct economic benefits of the tree nursery industry, tree planting is increasingly relied upon for climate adaptation benefits (Eisenman et al. 2021; Pataki et al. 2021). In 2022, the USDA Forest Service announced $1.5 billion in funding for urban and community forestry grants, the majority of which will focus on tree planting and maintenance, while also addressing climate adaptation and extreme heat (USDA Forest Service 2023). Considering the pressing need and value of tree planting in urban areas, it may not be possible for growers and municipalities to avoid planting during large-scale periodical cicada outbreaks.

A large-scale periodical cicada emergence presents an interesting challenge to urban forests and practitioners in the midst of a tree planting campaign. Periodical cicada egg-laying behavior is considered a risk for young trees, but less so for mature trees (Raupp 2021). Periodical cicadas (M. septendecim and M. cassini) preferentially oviposit in small stems of approximately 5 to 10 mm diameter (Miller and Crowley 1998). The impact of such behavior on mature trees is less of a concern compared to younger trees (Raupp 2021), presumably due to the greater overall proportion of small stems on younger trees compared to mature trees. This study, and others (Miller and Crowley 1998; Clay et al. 2009b), demonstrates that periodical cicadas do exhibit host-tree preferences in egg laying. However, the genera and species impacted by egg laying is still quite broad (Clay et al. 2009b), and not all species within a genus are preferred. For example, studies have reported high and low susceptibility to cicada egg laying for Quercus alba and Q. palustris, respectively (Clay et al. 2009b). While other studies report the highest oviposition rates in all host tree species surveyed for Quercus, such as Q. alba, Q. palustris, and Q. rubra (Perkovich and Ward 2022). These impacts differ from specialist urban forest pests, such as emerald ash borer, which has the potential to kill approximately 1.3 million street trees (Fraxinus spp.) between 2020 to 2050 (Hudgins et al. 2022) and can essentially double municipal budgets during peak years of infestation and management (Hauer and Peterson 2017). In comparison, a large-scale periodical cicada emergence is unlikely to have similar long-lasting impacts on urban forest health. This study did not find any evidence of mortality related to cicada egg laying. Overall condition of the tree cohort planted in 2020 to 2021 was similar to previous years in the absence of a periodical cicada emergence. In addition, egg laying occurred less frequently in lead stems compared to scaffold branches. Intact lead stems are important for maintaining a strong central leader in young trees, which helps avoid structural defects later in life such as codominant stems (Luley 2019).

Although Moderate Resolution Imaging Spectroradiomete (MODIS) data has been around since 2000 (Spruce et al. 2011), and ForWarn II is readily available through its website and is easy to use (Norman et al. 2013), the ForWarn II system didn’t provide high enough resolution for this region and pest of interest. The scale of the mapper, which utilizes 232-meter (5.4 hectare) MODIS appeared to be too large to notice the signature of unique foliage damage (flagging) caused by cicadas. This application has been used successfully for numerous change detection forest surveys (Olsson et al. 2016) at the site, county, state, and national level (Norman and Christie 2022) and on numerous forest pests (Eklundh et al. 2009; Spruce et al. 2011; Norman et al. 2013).

The HiForm change detection on the other hand provided detailed maps of NDVI change throughout the cicada outbreak. Although agreement was not as good as in other studies (e.g., Eklundh et al. 2009), the high-resolution change detection did offer a District-wide view of change and gave the best assessment of the potential distribution of cicada activity across the district. That said, ground confirmation would be required to create a higher confidence map of cicada damage. The greater damage intensity found by the remotely sensed product was likely due to a number of issues, such as not selecting the best threshold range for this unique signature, seasonal tree developmental differences (Lukasová et al. 2014), or unique rainfall and temperature patterns (Revadekar et al. 2012) occurring concurrently and impacting the NDVI products. Another issue could be that the high-resolution change detection product was picking up other pest damage (e.g., insect or disease) in addition to the cicada damage. Another factor working against the change detection system was the ephemeral nature of the signature of the cicada and its damage (flagging) do not tend to stay on the tree (Raupp 2021). The damaged branch tips tend to break, die, and are shed by the tree during and not long after the outbreak (Gerhard 1923; Miller 1997). Another issue impacting our remote sensing results could have been perspective; trees appear to have more flagging when viewed from above than from the side. This phenomenon could have caused the overhead perspective to appear more prominently than what appeared from the side of the tree (Figure 7).

In the end, this method (change detection analysis of high-resolution imagery) may not be universally available as it required access to satellite imagery and remote sensing expertise. Aerial imagery of sufficient quality (resolution and lack of clouds) for use in photo interpretation in a scan and sketch project was difficult to obtain, time consuming to look for, and had a time delay associated with its acquisition. Urban areas are known to influence cloud cover during the spring and summer seasons, making working with this type of data more difficult when insects like the cicada are active (Romanov 1999). When a cloud-free image is obtained, it is generally easy to download and incorporate, although images tend to have large file sizes and take long periods to download. These downloads also require a large amount of disk space to store and utilize. In addition, the scan and sketch process is subjective and time consuming but does allow the user an alternative to traditional aerial surveys (Hanavan et al. 2022), whereas forest damage aerial surveys have mostly consisted of forest health specialists observing forest canopies from small, fixed-wing aircraft flown at low elevations (McConnell 2000). Unfortunately, most remotely sensed images available ended up not producing useful information for this study due to cloud cover and had to be omitted as only one viable image was considered usable in the processing.

It was the hope that remote sensing could be a strong tool for creating an accurate map of the cicada damage that was being seen across the District. Remote sensing was not able to do this alone, however ground truthing the remotely sensed map layer was required, as false positives would be included at an unknown rate if all the damage layer positive pixels were assumed to be cicada damage. This did not make the results a failure as the high-resolution damage map layer still provided an operational guide to direct field surveys. With the broad media coverage of the Brood X emergence in 2021, citizen science and social media posts were explored as an additional source of observations of Brood X activity in the district and the surrounding metro area (Kritsky 2021). However, at the time that this research occurred, there were not enough resources to fully pursue community science as a viable resource. It could be a valuable avenue in the future to see cicada impacts on urban trees, especially with newly developed apps like Cicada Safari that are gaining popularity during these large emergence events and collecting data on brood densities and locations.

Conclusions

In this study, periodical cicada activity of one of the largest broods, Brood X, and their potential impacts on young trees in urban environments, such as Washington, DC, was investigated. A sample of street trees in the District of Columbia revealed few impacts of cicada egg laying, where defoliation (the percent foliage lost) was minimal for most trees surveyed and tree condition was not adversely affected. In general, cicadas exhibited a preference for particular genera and scaffold or lateral branches. Together, these results indicate little risk of planting trees in urban areas prior to a periodical cicada emergence. With some tree genera faring better than others, this may allow for tree planting recommendations for particular genera in the planting season preceding a large periodical cicada outbreak. The need for planting trees in urban areas is only growing as city residents depend on them to mitigate the impacts of extreme heat and reduce stormwater runoff, among many other benefits trees provide as nature-based solutions. Fortunately, this study indicates that, at least in the short term, there appears to be little risk to planting new trees in the same year as a large-scale periodical cicada emergence. Follow-up studies of these trees would be beneficial in understanding the long-term impact of egg laying on branch form and canopy structure. As for the use of remote sensing products, it is clear to the authors that although these products and methods are somewhat readily available, the unique damage signature of cicadas may not have been a good match for either of the NDVI products or for the urban environment. Although high-resolution imagery (1.28 feet [0.30 m] or greater) could be used, the difficulty in obtaining high resolution, cloud-free imagery and the need for specialized skills out-weigh what could be accomplished by a timely on-the-ground monitoring system and simple roadside driving surveys.

Conflicts of Interest

The authors reported no conflicts of interest.

Acknowledgements

This research was supported in part by the USDA Forest Service, Forest Health Protection Core Funding to the District of Columbia, District Department of Transportation (21-DG-11094200-163). The District of Columbia Forest Health Coordinator Kasey Yturralde’s salary was partially funded by USDA Forest Service Forest Health Protection Core Funding in federal fiscal years 2021 and 2022. The authors would like to thank the following individuals for their contributions as related to data access: Rowan Moody with Red Bird Consulting (aerial imagery), Chris Hayes with the USDA Forest Service Forest Health Protection (ForWarn Imagery), data analyst Steve Norman and US Forest Service Region 8 Research Team (HiForm), and John Stanovich, retired from the US Forest Service (error matrix statistics). Thank you to Ida Holaskova with West Virginia University, Davis College for her invaluable assistance with statistical analysis. Thank you to the District Department of Transportation Urban Forestry Division staff for their support in the planning and implementation of the annual tree survival study, particularly Earl Eutsler and Simoun Banua.

Street Tree Management Challenges in Small Cities, Iowa, USA

Thu, 01 Jan 2026 11:00:00 GMT

Background

Small and large cities typically both have street trees, but small cities have fewer resources to manage them. Three 1980s papers assessed that small rural cities in the state of Iowa, USA, were at a disadvantage in managing street trees as reflected in the diversity, age, and condition of their street tree populations. This paper analyzes more current street tree inventory data in conducting a similar assessment.

Methods

Street tree inventory data were obtained from small, midsize, and large Iowa cities for 2008 to 2024. Tree diversity, age, and condition were analyzed based on city size. Diversity was assessed by relative abundance percentages and diversity index statistics, age by trunk DBH distributions, and condition by ratings for tree wood and leaves.

Results

Small cities have less street tree diversity than midsize and large cities. A diversity t-test found statistically significant differences based on city size. Small and midsize cities were found to have older tree population profiles than large cities. Large and midsize cities had better condition wood ratings than small cities, but little difference was found between cities for leaf condition.

Conclusions

Inventory data suggest that Iowa’s small cities still face challenges managing their street trees. Lack of funding is the reason most often cited. Progress is being made although its extent cannot be determined due to lack of longitudinal data. There remains a need to assist Iowa’s small cities in sustainably managing their street trees and maintaining the benefits they provide.

Introduction

The devastation wrought by Dutch elm disease (DED)(Ophiostoma ulmi, O. novo-ulmi), which was introduced into the United States in the late 1920s (Copeland et al. 2023) and killed millions of Ulmus americana (American elm) planted as street trees (Campanella 2011), was a calamitous event in the history of urban forestry in North America. It prompted Jorgensen (1970) to coin “urban forestry” as a term, which he defined as the management of urban trees on an areawide, systematic basis rather than on the basis of the individual tree. Managing trees on an areawide, systematic basis required a municipality to have an urban forestry management program (Grey and Deneke 1986). Components of such a program were envisioned ideally to include a street tree inventory and management plan (Tate 1985), a municipal arborist and trained workforce to plant and care for trees (Beatty and Heckman 1981), and a street tree ordinance (Hudson 1985). These components required organization, personnel, and funding by a municipality over a number of years (Clark et al. 1997). However, Kielbaso et al. (1982) found in updating a national survey conducted by Ottman and Kielbaso (1976) of municipal tree programs in the United States that less than 50% of responding municipalities had a systematic tree management program that addressed current and long range needs and that communities with smaller populations were less likely to have such a program than more populous ones. Similarly, on the level of the individual states, Reynolds et al. (1981) found that smaller cities in Minnesota, USA, often did not have a sufficient governmental structure to operate a shade tree program, and Tate (1984) reported that New Jersey, USA, cities with populations less than 5,000 expended far less money per tree on management than the state’s more populous cities.

Subsequent research has confirmed that municipalities with larger populations are more likely to have the resources to establish or sustain an effective tree management program than municipalities with smaller populations (Schroeder et al. 2003; Ries et al. 2007; Grado et al. 2013; Hauer and Peterson 2016; Hargrave et al. 2024). Consequently, Karlovich et al. (2000) found that street trees in 4 small, rural Illinois, USA, communities were characterized by low species diversity and tree topping; Galvin and Bleil (2004) found that in many smaller communities in Maryland, USA, there was no staff person or department legally responsible for the care and management of the community’s trees; Neupane et al. (2022) found that in the Southeastern United States the presence of municipal vegetation ordinances depended upon community population size, and communities with smaller populations were more likely not to have such ordinances; and Groninger et al. (2002) questioned if the challenges faced by small, rural communities practicing urban forestry in the United States were so great that these communities could even practice urban forestry.

In the state of Iowa, United States, 3 papers written between 1981 and 1985 identified and discussed street tree management issues and concerns in the state’s smaller, more rural communities amidst the devastation wrought by DED (Wray and Jungst 1981; Wray and Prestemon 1983; Wray and Mize 1985). These papers made use of data collated from street tree inventories conducted in 1978 as part of a research project by the Department of Forestry at Iowa State University in conjunction with the Iowa Conservation Commission’s Forestry Section, a precursor of the state’s Department of Natural Resources. Inventories were conducted in 40 randomly selected, geographically dispersed municipalities with populations between 500 and 10,000. Communities of this size, deemed to be “smaller communities,” were targeted because municipalities with populations greater than 10,000, deemed to be “larger communities,” were considered more likely to have programs and departments managing their street trees due to greater resources (Wray and Prestemon 1983). In fact, these papers found that none of the selected smaller communities had an urban forestry program with continuous, regularly scheduled activities; maintenance was sporadic and often performed only in cases of severe need; street tree planting was a low priority item because of limited financial resources; and management plans were nonexistent (Wray and Jungst 1981). Analysis of the inventory data focused on 3 aspects of street tree populations: tree diversity, tree age as measured by trunk diameter at breast height (DBH), and tree condition. Results of this analysis were then used to make judgments on the efficacy of street tree management practices in the selected smaller communities.

This paper takes another look at street trees in Iowa’s smaller communities. It revisits many of the issues discussed in the papers written by Wray and Jungst (1981), Wray and Prestemon (1983), and Wray and Mize (1985). It does so by analyzing more recent street tree inventory data from small, midsize, and large Iowa municipalities. These data include a large dataset compiled by the Iowa Department of Natural Resources from small and midsize communities and inventories conducted by commercial vendors mostly from larger communities. Findings are made from these more recent data for street tree composition and structure on the basis of community size. Emphasis is given to the diversity of street tree species and genera as a crucial factor in the sustainable management of street trees given the legacy of DED, but tree age and condition are also considered. Finally, findings made from these more recent data provide a starting point to discuss some of the issues faced today by smaller Iowa communities in managing their street trees in order to maintain the environmental, social, and economic benefits these trees provide (Mullaney et al. 2015). Comparisons are made where valid with the findings contained in the papers written by Wray and Jungst (1981), Wray and Prestemon (1983), and Wray and Mize (1985).

Methods

Study Area

Iowa is the 23rd largest of the 50 states comprising the United States, with a land area of 144,659 km2 (55,853 mi2) and a population of 3.24 million (US Census 2024). The state is divided into 99 counties which in turn contain 948 cities and 86 Census Designated Places (CDPs)(Iowa State Data Center 2025). A CDP is a densely settled population center that has a name and community identity and is not part of any incorporated place (US Census 1994). The state also contains townships which are administrative subdivisions of counties that provide limited functions such as fire protection, cemetery management, fence maintenance, and boundary dispute resolution in rural areas outside the scope of counties and cities (Story County Iowa 2025). Somewhat confusingly, Iowa’s small rural incorporated communities are often referred to as towns. For example, the 3 papers written by Wray and Jungst (1981), Wray and Prestemon (1983), and Wray and Mize (1985) referred to the 40 municipalities where street tree inventories were conducted as towns, when, based on state code, they are more accurately cities (Iowa General Assembly 2025). For the purposes of this study and to avoid confusion, the state’s small rural incorporated communities will be referred to as cities.

Iowa’s climate is humid continental (Köppen Dfa and Dfb) with year-round precipitation (Bailey 2016). The majority of the state is located in the Southern Shortgrass Prairie biome (Great Plains), and the remainder of the state is located in the Central Appalachian Forest biome (Eastern Temperate Forest) (Bailey 2016). USDA Plant Hardiness Zones 4b, 5a, 5b, and 6a are contained within the state (Figure 1). Zone boundaries between the 2012 and 2023 hardiness zone maps have shifted from south to north as average annual minimum winter temperatures have increased (PRISM Group 2012, 2023). Zone 4b, the coldest zone, has decreased in area from 134,886 km2 (52,080 mi2) to 37,925 km2 (14,643 mi2), and Zone 6b, the warmest zone, has increased in area from 425 km2 (164 mi2) to 42,945 km2 (16,581 mi2).

Figure 1.

Iowa, USA, 2012 and 2023 USDA plant hardiness zone maps.

The grasslands associated with the prairie ecosystem, when tilled under, produced some of the most fertile soils for agriculture in the United States (Mutel 2008), and agriculture became Iowa’s most important industry after statehood in 1846 (Schwieder 1996). In 1900, 97% of Iowa land was occupied by farms, and the rural population totaled 2.23 million, which was 74.4% of the state’s population (US Census 1900). However, as farms became more mechanized and the state’s industrial sector grew, the rural population decreased (Mutel 2008), and in 2020 the rural population totaled 1.18 million (36.8% of the state’s population) and the urban population totaled 2.01 million (63.2% of the state’s population)(Burke 2021). Additionally, the rural population is aging due to the out-migration of youth under 25 years of age and experienced workers aged 45 to 64 to the state’s large metropolitan areas, as well as due to a decline in the number of births as compared to the number of deaths (Peters 2024). Despite these trends, Iowa’s small rural communities are not dying. The decrease and aging of the rural population has been moderated by an influx of persons of color (Peters 2024), and polling indicates that living in small rural communities still has considerable appeal (Parker et al. 2021). Moreover, Iowa’s small rural communities remain a significant part of its cultural and economic heritage (Bremer 2023), and state government and other entities have made efforts to better understand and address the challenges facing these communities, including street tree management, in order to revitalize them.

Datasets

As stated above, the Wray and Jungst (1981), Wray and Prestemon (1983), and Wray and Mize (1985) papers focused on the challenges faced by small, rural Iowa cities in managing their street trees by analyzing data from street tree inventories conducted in 1978 in 40 randomly selected Iowa cities having a population between 500 and 10,000 in the 1970 United States Census. In 1970 there were 413 cities in Iowa with a population between 500 and 10,000 (Iowa State Data Center 2025). Therefore, the 40 cities selected comprised a 9.7% percent sample of all Iowa cities in that population range. Cities were assigned to one of four groups based on population size. There were 13 cities (32.5%) with a population between 500 and 999; 13 (32.5%) with a population between 1,000 and 2,449; 8 (20.0%) with a population between 2,500 and 4,999; and 6 (15.0%) with a population between 5,000 and 9,999. For each of the 40 cities, a complete inventory was conducted that included trees along streets on public property and in publicly owned parking areas (Wray and Mize 1985). Park trees were not inventoried. Data were collected for more than 39,000 trees. The raw data from these inventories are unavailable. However, the 3 papers provide summary statistics for tree species relative abundance (minimum 100 trees per species), trunk diameter size (5 classes in inches), and tree condition (good, fair, and poor). Tree condition was reported according to these criteria: good trees were those requiring no care or maintenance or only minimal maintenance such as minor pruning; fair trees were those in need of corrective treatments if they were to function for more than 10 additional years; and poor trees were those that probably would not survive longer than 2 years without corrective action (Wray and Mize 1985). The inventories providing the data for the papers written by Wray and Jungst (1981), Wray and Prestemon (1983), and Wray and Mize (1985) are hereafter referred to as the 1978 dataset.

More recently, beginning in 2009, Iowa’s Department of Natural Resources (IDNR) initiated a program to conduct street tree inventories and create urban forestry management plans for small cities in the state; small cities were defined as those with a population of 5,000 or less (Iowa DNR 2024a). This definition is consistent with the United States Census Bureau’s definition of a small town or city (Toukabri and Medina 2020) but does not correspond with the definition of ‘small’ by Wray and Jungst (1981), Wray and Prestemon (1983), and Wray and Mize (1985). The IDNR also defined a midsize city as having a population of 5,001 to 49,999 and a large city as having a population of 50,000 or more. Based on the IDNR definitions, the 1970 United States Census found that there were 931 small cities, 56 midsize cities, and 7 large cities in Iowa, whereas the 2020 Census found 857 small cities, 74 midsize cities, and 11 large cities (Iowa State Data Center 2025). For the purposes of this paper, Iowa cities are classified according to size by their 2020 Census population. As of 2024, IDNR had conducted tree inventories in 448 small cities (52.3% of small cities statewide), 47 midsize cities (63.5% of midsize cities statewide), and 1 large city (9.1% of large cities statewide). Of those inventoried by IDNR, 128 cities (25.8%) had a population of 500 or less.

ArcGIS Collector was used by IDNR personnel to collect tree inventory data in walking surveys; data were then run through either i-Tree Streets or Eco (Nowak 2024), depending on the year in which the inventory was conducted, to calculate street tree benefits (Hannigan 2025, personal communication). Data were collected for more than 330,000 trees, including both street trees and park trees. Data attributes for each tree included trunk diameter size (9 DBH classes in inches), land use type, location type, and longitude and latitude coordinates. Each tree received one of four conditions ratings—good, fair, poor, and dead or dying—for wood and for leaves according to i-Tree protocols, which are also commonly used by private data collection software. Species and genus were identified for most but not all trees. For example, 10,518 trees (3.17%) were classified as Broadleaf Deciduous, Broadleaf Evergreen, or Conifer Evergreen, and 292 trees (0.09%) were classified as Unknown. In addition, 58,637 trees (17.65%) were classified at the genus level but not at the species level. For example, 773 trees were classified as birch (Betula spp.), 297 trees were classified as dogwood (Cornus spp.), and 1,863 trees were classified as oak (Quercus spp.), etc. The management plans based on the inventories included recommendations to improve tree diversity, to plant additional trees and reduce the maturity of the overall tree age profile, and to address maintenance issues associated with tree health. The plans also included a 6 year tree maintenance plan based on existing municipal funding and a summary of annual tree benefits calculated with i-Tree software.

Besides the data contained in the IDNR street tree inventories, data were obtained for this paper from 2 midsize cities and 7 large cities that conducted inventories between 2008 and 2024 independently of IDNR. These cities employed different commercial vendors to conduct the inventories. Data were collected for more than 198,000 trees including both street trees and park trees. Data attributes collected for each tree included species and genus, trunk diameter size (DBH in inches), location information, and longitude and latitude coordinates. Some, but not all, inventories contained ratings on tree condition which approximate the ratings made in the IDNR inventories. The data associated with the IDNR inventories and the data associated with the inventories conducted between 2008 and 2024 independently of IDNR are hereafter referred to collectively as the 2008 to 2024 datasets. Table 1 contains the sources and characteristics of the datasets discussed in this paper.

Ideally, all the data obtained for this study would have been collected utilizing the same methods and metrics so that the data are fully comparable. This was not the case, which is not surprising given the typical lack of standardization in tree inventory data collection (Roman et al. 2013). As a result, analyses of these data and the validity of any findings are conditioned by the differences in the datasets. Additionally, since only summary data from the 1978 dataset are available, the ability to make longitudinal comparisons with the 2008 to 2024 datasets is limited, and caution is exercised in making them.

Street Trees and Park Trees

The 1978 dataset contained data for trees located along streets and in publicly owned parking but not in parks (Wray and Mize 1985). The 2008 to 2024 datasets contain data for both street trees and park trees; park trees account for 33% of all trees inventoried by IDNR and for 37% of all trees inventoried independently of IDNR. Data attributes for the 2008 to 2024 datasets allow street trees to be differentiated from park trees. This is important because studies have found that growing conditions and population structures associated with street trees and park trees are significantly different, and they should be evaluated separately (Welch 1994; Nielsen et al. 2007; North et al. 2018). Accordingly, to support the validity of findings made from the 2008 to 2024 datasets and comparisons made between those datasets and the 1978 dataset, parks trees were differentiated from street trees in the 2008 to 2024 datasets and are not included in those analyses.

Tree Diversity

Wray and Prestemon (1983) believed that “diversity in street tree populations is a desirable goal because it reduces potential problems with insects, diseases, and species adaptability.” Based on the finding that the 3 most abundant species in the 1978 dataset varied from 30.2% to 64.6% by city, they judged the “mix and distribution of [tree] species” to be inadequate and “the largest potential management problem that the smaller communities face.” This appraisal was in large part a reaction to the overplanting of elms as street trees in Iowa and the subsequent impact of DED. For example, in Des Moines, Iowa’s largest city and the state capital, DED killed approximately 175,000 elms on public and private property between 1960 and 1970, and the cost to remove and dispose of dead elms was nearly 20 million USD (Wray and Jungst 1981). The concern with tree diversity was prescient because, despite the lessons ostensibly learned from overplanting elms, ash trees were planted in large numbers in many Iowa communities to replace the elms killed by DED (Thompson et al. 2021) and since 2010 have become fodder for the emerald ash borer (EAB)(Agrilus planipennis)(Iowa DNR 2016).

A commonly used metric to assess the diversity of a community’s tree population is a relative abundance distribution which quantifies the percentage of each tree species and genus relative to the population as a whole. Santamour (1990) made use of relative abundance in proposing the 10-20-30 rule whereby no tree species should comprise more than 10%, no tree genus should comprise more than 20%, and no tree family should comprise more than 30% of a municipally managed tree population. He believed these limits on abundance would render a tree population less vulnerable to a pest or disease and avoid repeating what had happened with DED. Santamour’s 10-20-30 rule has since become widely accepted in municipal tree management to benchmark diversity and is applied most often to tree species and genera. However, because some pests and diseases affect more than a single tree species or genus, it has been argued by some that the 10-20-30 rule does not go far enough in benchmarking diversity and that a 5-10-15 rule should be used instead (Bassuk et al. 2009; Watson 2017). Additionally, climate change, which is generally associated with more extreme weather such as prolonged periods of drought and more intense rainfall events (Brandt et al. 2021), and which has also increased the range of pests and diseases threatening urban trees (Tubby and Webber 2010), has further emphasized the importance of diversity in tree population stability (Morgenroth et al. 2016).

In addition to relative abundance, other metrics commonly used to assess the diversity of street tree species and genera include the following diversity statistics (Table 2): the Buzas and Gibson evenness index (Buzas and Gibson 1969), which appraises the degree to which all species and genera abundances are divided equitably within the species and genus distributions; the Simpson Diversity Index (SDI) (Simpson 1949), which evaluates the dominance of species and genera within those distributions and is sensitive to species and genera evenness; and the Shannon-Wiener Diversity Index (Shannon 1948), which is more sensitive to species and genera richness than to evenness in gauging diversity. Additionally, the Inverse of the Simpson Diversity Index (Inv SDI or 1/SDI) is sometimes preferred to the SDI because its values increase inversely to dominance and therefore equate more clearly with diversity (Sun 1992; Sreetheran et al. 2011), and effective diversity (eH where H is the Shannon-Wiener Diversity Index value) is sometimes preferred to Shannon-Wiener because, unlike Shannon-Wiener, its scale is linear and not logarithmic, and the statistic is therefore more directly comparable (Jost 2006).

The 3 papers based on the 1978 dataset summarized the relative abundance percentages of the 10 most common street tree species found in the 40 inventoried cities. For this paper, street tree species and genera for each city contained in the 2008 to 2024 datasets were grouped according to the city’s population and the IDNR definitions of small, mid-size, and large size cities. The relative abundance percentages were then calculated for those species and genera associated with each city size class. The percentages of the 5 most prevalent species and genera in each city class were identified and summed. In addition to relative abundance, this paper calculated statistics for street tree species and genera at each city size class for the diversity metrics in Table 2. Diversity statistics cannot be calculated for the 1978 dataset because they require tree data for each city and only summary data are available. Also, for the 2008 to 2024 datasets only, a diversity t-test (Hutcheson 1970) was utilized to assess the statistical significance of differences (P < 0.05) for Shannon-Wiener Index values between city size classes. The diversity statistics and t-test were calculated with PAST Paleontological Statistics software version 4.17 (Hammer et al. 2001).

Tree Age

Street tree populations are vulnerable not only to pests and disease, but also to other factors (Steenberg et al. 2019). Prominent among these factors are environmental stressors including but not limited to more extreme temperatures, de-icing salts, and soil compaction, which increase tree mortality (Carol-Aristizabal et al. 2024). Wray and Mize (1985) were cognizant that “street-tree planting environments are among the most severe encountered” and that “not all climatically adapted species are good candidates for use in street or similar environments.” Accordingly, they judged those tree species which had most frequently grown to maturity while also remaining in good condition to be species best suited as street trees in Iowa. As part of this assessment, data were taken for tree species, many of which were among the most common street tree species found in the 40 inventoried cities, based on 5 DBH size classes: 0 cm to 7.62 cm (0 in to 3 in), 7.62 cm to 15.24 cm (3 in to 6 in), 15.24 cm to 30.48 cm (6 in to 12 in), 30.48 cm to 50.80 cm (12 in to 20 in), and > 50.80 cm (> 20 in). Unfortunately, results were reported for each species as a percentage of the DBH class found in good condition without specifying the number of trees contained in each DBH class. Therefore, DBH distributions cannot be generated for prevalent street tree species from the 1978 dataset. However, Wray and Jungst (1981) did report size distribution percentages of all inventoried street trees by the same DBH size classes utilized by Wray and Mize (1985), and a DBH distribution of all trees contained in the 1978 dataset can be generated.

At about the same time that Wray and Jungst (1981) and Wray and Mize (1985) collected trunk diameter data and assessed tree species best suited as street trees in Iowa’s small cities, Richards (1983) reasoned that the stability of a street tree population depended not only on the selection of those tree species “adapted for long-term success” but also on a population with “good age diversity” with larger numbers of young trees relative to the number of mature trees in order to account for the mortality of young trees after planting in the establishment phase. Subsequent research has confirmed that street tree population growth is constrained by high mortality among young trees in the first few years after planting when environmental stressors are particularly acute and that annual mortality is lower for midsize and large trees (Roman et al. 2014; Czaja et al. 2020). This has led to wide acceptance of the construct of a descending tree size distribution based on trunk diameter (DBH) as desirable in street tree management and particularly to tree establishment; however, this construct may not be as applicable to the management of other urban trees, such as park trees and trees on private property, or if maximizing ecosystem services is a priority (Morgenroth et al. 2020).

The IDNR management plans based on the inventories in the 2008 to 2024 datasets agreed with the desirability of a descending DBH distribution to ensure enough young trees were being planted to account for the mortality of older trees as they aged and their health declined (Iowa DNR 2013). Similarly, the management plans associated with the inventories conducted independently of IDNR recommended a descending DBH tree distribution not only to account for tree attrition (City of Des Moines 2020) but also to ensure the continued flow of tree benefits and a more uniform workflow allowing managers to more accurately allocate annual maintenance funds (Iowa City 2018). Accordingly, trunk diameter data were collected in the IDNR inventories. These data were aggregated into 9 DBH size classes: 0 cm to 7.62 cm (0 in to 3 in), 7.62 cm to 15.24 cm (3 in to 6 in), 15.24 cm to 30.48 cm (6 in to 12 in), 30.48 cm to 45.72 cm (12 in to 18 in), 45.72 cm to 60.96 cm (18 in to 24 in), 60.96 cm to 76.20 cm (24 in to 30 in), 76.20 cm to 91.44 cm (30 in to 36 in), 91.44 cm to 106.68 cm (36 in to 42 in), and > 106.68 cm (> 42 in). For tree inventories conducted independently of IDNR, DBH data either were collected in inches and then aggregated into the 9 DBH classes listed above or were assigned during collection into the same 9 DBH classes. Data in the 0 cm to 7.62 cm (0 in to 3 in) DBH size class can be instructive in understanding the number and species of those trees most recently planted. However, it also creates in tandem with the 7.62 cm to 15.24 cm (3 in to 6 in) DBH size class unequal frequency intervals and complicates the analysis of DBH distributions. Therefore, for all DBH data contained in the 2008 to 2024 datasets, data in the 0 cm to 7.62 cm (0 in to 3 in) and 7.62 cm to 15.24 cm (3 in to 6 in) size classes were aggregated to create a 0 cm to 15.24 cm (0 in to 6 in) size class, resulting in 8 DBH size classes instead of 9. DBH distributions were then calculated for each city size class and for the most prevalent tree species and genera found in those size classes.

Tree Condition

There are many reasons for rating tree condition in a street tree inventory. These include an assessment of tree health, the risk of tree failure, the presence of tree pests and disease, understanding and budgeting for maintenance needs, and calculating compensatory tree replacement cost (Ma et al. 2021). As stated above, Wray and Mize (1985) rated mature tree condition in terms of survivability and the need for corrective maintenance. For example, many trees required minimal pruning and were therefore classified as being in good condition since minimal pruning was not considered a threat to tree longevity. Ratings were intended not only to inform prospective maintenance costs but also to suggest those tree species best suited as street trees in Iowa. Species where 90% or more of trees received good ratings were judged to be acceptable candidates for street trees, and species exhibiting the biggest declines in good ratings as trunk diameter increased were judged to be poor choices.

More recently, importance has been given to tree condition because trees in good health provide more ecosystem services than trees in poor health (Hintural et al. 2024). Both i-Tree Streets and Eco incorporate tree condition ratings when estimating ecosystem services and other benefits provided by urban trees. Inventories conducted by IDNR made separate ratings for wood condition and leaf condition in accordance with i-Tree protocols. Each tree received one of four ratings—good, fair, poor, and dead or dying—for wood and for leaves. For i-Tree Streets, these ratings reflect replacement factor percentage values utilized by the Council of Tree & Landscape Appraisers (USDA Forest Service 2010). For i-Tree Eco, these ratings reflect percentages of crown health and dieback (USDA Forest Service 2018). Some, but not all, of the inventories conducted independently of IDNR utilized the same rating scale to make condition ratings for both wood and leaves. Other inventories used the same scale but combined wood and leaves into a single overall rating for tree condition. Since an overall tree condition rating might not agree with separate ratings for wood and leaf condition, inventories that combined wood and leaves into an overall condition rating were excluded from assessing tree condition on the basis of city size, both for all inventoried trees and for prevalent street tree species.

Results

Tree Diversity

For the 40 small cities comprising the 1978 dataset, Acer saccharum (sugar maple, 16.9%) was found the most common street tree species, followed by Fraxinus pennsylvanica (green ash, 13.7%), A. saccharinum (silver maple, 12.0%), A. platanoides (Norway maple, 8.1%), and Celtis occidentalis (northern hackberry, 4.5%)(Table 3). Relative abundance percentages for street tree genera were not reported, but the 3 Acer (maple) spp. above account for 37.0% of all inventoried street trees. For the 2008 to 2024 datasets, A. saccharinum (14.52%) was found to be the most abundant species in small and midsize cities and Malus spp. (crabapple, 10.51%) the most abundant species in large cities (Table 4). Percentages of A. saccharinum, F. pennsylvanica, and A. platanoides slightly exceeded Santamour’s 10% benchmark for species relative abundance in small and midsize cities, but not in large cities. At the genus level (Table 4), Acer spp. was found to be the most abundant genus in small, midsize, and large cities. The percentage of Acer spp. substantially exceeded Santamour’s 20% benchmark for genus relative abundance in small and midsize cities but only slightly exceeded that benchmark in large cities. Little difference was found between small and midsize cities in the summed percentage of the 5 most prevalent street tree species. However, the summed percentages of the 5 most prevalent street tree species and genera in large cities were found to be substantially less than the summed percentages in small and midsize cities.

In addition to assessing street tree diversity based on relative abundance percentages, statistics were calculated for the 2008 to 2024 datasets at species and genus levels for SDI, the inverse of the SDI (Inverse SDI), the Shannon-Wiener Diversity Index, distribution evenness, and effective diversity (Table 5). For street tree species and genera, the Inverse SDI and effective diversity increased as city size increased, as did species and genera richness (i.e., the number of species and genera). Increases were larger between midsize and large cities than between small and mid-size cities. Results for distribution evenness were mixed. Evenness for street tree species declined from small to large cities while evenness for street tree genera evenness increased from small to large cities. Lastly, diversity index t-tests (Hutcheson 1970) assessed the statistical significance of differences in the Shannon-Wiener Diversity Index for all street tree species and genera between city sizes. Statistically significant increases (P < 0.05) were found in the Shannon-Wiener Diversity Index for both street tree species and genera between successive city size classes (i.e., diversity increased significantly from small to mid-size cities and from midsize to large cities)(Table 6).

DBH Distributions

For the 40 small cities comprising the 1978 dataset, a DBH distribution was generated for all inventoried street trees. Results showed a strong descending trend line in the 2 smallest size classes and a peak in the 30.48 cm to 50.80 cm (12 in to 20 in) size class (Figure 2). Wray and Prestemon (1983) noted the large proportion of trees (nearly 50%) in the two smallest size classes and believed it represented a surge in new tree planting in the wake of DED. They attributed the smaller numbers of large trees not only to the impact of DED but also to severe urban planting environments which resulted in slower tree growth.

Figure 2.

DBH distribution, all street trees, small cities, 1978 dataset.

For the 2008 to 2024 datasets, DBH distributions were generated by city size for all street trees and for prevalent street tree species and genera. Distributions showed peaks in the 30.48 cm to 45.72 cm (12 in to 18 in) DBH size class for small and midsize cities. These peaks suggest that older aged profiles will result due to an insufficient number of young trees in the 0 cm to 15.24 cm (0 in to 6 in) DBH size class to compensate for tree mortality. By contrast, results for large cities showed a consistent descending trend line from small to large DBH size classes, suggesting a sufficient number of young trees to compensate for tree mortality (Figure 3).

Figure 3.

DBH distributions, all street trees by city size, 2008 to 2024 datasets.

DBH distributions by city size were also generated for the 5 most prevalent street tree species in the 2008 to 2024 datasets. Distributions for A. saccharinum, F. pennsylvanica, A. platanoides, and A. saccharum showed peaks in the DBH classes between 45.72 cm and 76.20 cm (18 in to 30 in) for all city sizes, suggesting older age profiles for those species. The distribution for C. occidentalis showed peaks for small and midsize cities in the 60.96 cm to 76.20 cm (24 in to 30 in) DBH size class; the distribution for large cities was mixed, showing a peak in the 45.72 cm to 60.96 cm (18 in to 24 in) size class, but also a large percentage of trees in the 0 cm to 15.24 cm (0 in to 6 in) size class, suggesting a recent surge of planting this species in large cities. The distribution for Malus spp. showed a descending trend line from small to large DBH size classes suggesting sufficient young trees were planted to compensate for tree mortality. The lack of many Malus spp. trees in size classes greater than 60.96 cm (24 in) reflects the comparatively small stature of this species and its many cultivars. Statistics and graphs for the DBH distributions of prevalent street tree species can be found in the Appendix (Figures S1 to S6).

DBH distributions by city size for the 5 most prevalent street tree genera were more nuanced. Distributions for Acer spp., Fraxinus (ash) spp., Celtis (hackberry) spp., and Tilia (linden) spp. evidenced peaks in the DBH classes between 30.48 cm and 76.20 cm (12 in and 30 in) for all city sizes. At the same time, there were strong descending trend lines for Acer spp., Celtis spp., and Tilia spp. between 0 cm and 30.48 cm (0 in to 12 in) in large cities, suggesting a surge of new plantings in those communities. The surge in Acer spp. was due mostly to big increases in A. rubrum (red maple) and A. tataricum (Tatarian maple) in the 0 cm to 15.24 cm (0 in to 6 in) DBH size class. Conversely, there appeared to be a large decline for Fraxinus spp. in the 0 cm to 15.24 cm (0 in to 6 in) size class, although the drop-off in large cities was not as large as in small and midsize cities. The distribution for Quercus (oak) spp. showed a descending trend line for all city sizes, but the trend was especially strong in large cities. The distribution for Malus (apple) spp. was similar to the Malus (crabapple) spp. distribution, which is not surprising since street trees in that genus are comprised mostly of Malus (crabapple) spp. Statistics and graphs for the DBH distributions of prevalent street tree genera can be found in the Appendix (Figures S7 to S12).

Condition

For the 40 small cities comprising the 1978 dataset, Wray and Prestemon (1983) reported that the percentage of all street trees receiving good condition ratings improved slightly from 87.7% to 91.2% as city population increased. Wray and Mize (1985) reported results for individual street tree species and for trees in each species rated as being in good condition. Among the most prevalent street tree species (Table 3), C. occidentalis had the most trees (96%) and A. saccharinum the fewest trees (87%) rated in good condition; U. americana received the lowest ratings among all street tree species with only 54% of trees rated in good condition. Wray and Mize (1985) also reported results for tree species and the decline in the percentage of trees for each species rated in good condition as trunk diameter increased. They found A. saccharum and F. pensylvanica to be the species with the most deterioration in tree condition as trunk diameter increased and A. platanoides the species with the least.

Distributions for tree condition (wood and leaves) were generated from the 2008 to 2024 datasets for all street trees by city size (Table 7). Results for wood condition for all street trees indicated an increase in the percentage of trees with good wood ratings and a decrease in the percentage of trees with fair wood ratings as city size increased. There was a slight decrease in poor wood ratings as city size increased but an increase in dead or dying wood ratings as city size increased from midsize to large. Ratings for leaf condition for all street trees did not reflect a clear trend as city size increased. Results for midsize cities were inconsistent.

Distributions for wood condition and for leaf condition were also generated from the 2008 to 2024 datasets for prevalent street tree species by city size. Results for wood condition for all prevalent street tree species indicated an increase in the percentage of trees with good wood ratings and a decrease in the percentage of trees with fair wood ratings as city size increased (Table 8). The percentage increase in good wood ratings as city size increased was greatest for A. saccharum and least for Malus spp. Between small and large cities, there was an increase in dead or dying wood ratings for all species except for F. pennsylvanica and A. saccharum and a decrease in poor wood ratings for all species except for Malus spp. and C. occidentalis. Results for leaf condition for prevalent street tree species did not indicate a clear trend as city size either increased or decreased (Table 9). In fact, for all species except C. occidentalis, the largest percentages of trees with good leaf condition ratings were associated with midsize cities.

Finally, distributions for tree condition (wood and leaves) were generated from the 2008 to 2024 datasets for all street trees by DBH classes. Results for wood condition for all cities indicated a decrease in the percentage of trees with good ratings and an increases in the percentages of trees with fair, poor, or dead or dying ratings as DBH size increased (Figure 4). Conversely, results for leaf condition for all cities did not indicate a clear trend as DBH size either increased or decreased (Figure 5).

Figure 4.

Ratings, wood condition, all street trees by DBH class, 2008 to 2024 datasets.

Figure 5.

Ratings, leaf condition, all street trees by DBH class, 2008 to 2024 datasets.

Discussion

Wray and Jungst (1981) believed that small Iowa cities with populations of 500 to 10,000 faced significant challenges in managing their street trees due to a lack of funding, knowledge, and organization. They further believed these challenges were reflected in findings made from a statewide sample of street tree inventory data collected in the 1978 dataset. These findings focused on 3 structural features of street tree populations: diversity, age, and condition. According to Ma et al. (2021), these are the 3 types of data most commonly collected in street tree inventories for the purpose of street tree management. This paper analyzed a more recent statewide sample of street tree inventory data collected in the 2008 to 2024 datasets to also explore the challenges facing small Iowa cities with populations of 5,000 and less in managing their street trees. It made findings for tree diversity, age, and condition not only for small cities but also for midsize and large cities as well.

The 1978 and 2008 to 2024 datasets were collected in the aftermath of recent losses caused to a prevalent street tree species and genus by a fungus or pest: DED for the 1978 dataset and the EAB for the 2008 to 2024 datasets. As a result, tree diversity was a major concern for the 3 papers based on the 1978 dataset and for the IDNR management plans based on the 2008 to 2024 datasets. The 1978 and 2008 to 2024 datasets both indicated that an increase in street tree diversity was warranted, especially for tree genera in small cities due to an overabundance of trees belonging to the Acer genus. The need for increased diversity at the species level did not appear to be as pressing since A. saccharum, the most abundant tree species in the 1978 dataset, accounted for 16.9% of all trees, and A. saccharinum, the most abundant tree species in the 2008 to 2024 datasets for small and midsize cities, accounted for 14.52% of all trees in small cities (Tables 3 and 4). By contrast, Ball et al. (2007) found in a study of 34 communities in South Dakota that, for 21 communities with populations less than 5,000, F. pennsylvanica accounted for 41.77% of all street trees. However, the dominance by a particular street tree species or genus in Iowa can vary greatly by the city, and Fraxinus spp. make up over 50% of the street tree canopy in some Iowa cities (Iowa DNR 2025).

Because the 2008 to 2024 datasets contained inventory data for small, midsize, and large cities, this paper was able to assess tree diversity on the basis of city size for those datasets. Relative abundance percentages of prevalent street tree species and genera were found to be greatest for small cities and declined from small to midsize cities and from midsize to large cities (Table 4). Additionally, diversity statistics for all street tree species and genera, except for the evenness of street tree species, indicated that street tree diversity was lowest for small cities and increased from small cities to midsize cities and from midsize cities to large cities (Table 5). A diversity t-test confirmed that the differences in street tree diversity for city size were statistically significant (Table 6). Comparable findings for street tree diversity on the basis of city size cannot be made from the 1978 dataset because data were not collected for midsize and large cities. Wray and Jungst (1981) stated that a study similar to the one conducted for smaller cities in 1978 would be conducted for larger cities, but to the best of our knowledge this study for larger cities was not conducted.

This paper was also able to assess street tree age based on trunk diameter in the 2008 to 2024 datasets and make comparisons for city size. DBH distribution profiles for all street trees by city size showed distribution peaks for small and midsize cities in the 30.48 cm to 45.72 cm (12 in to 18 in) DBH size class, while larger cites showed a consistent descending trend line from small to large DBH size classes (Figure 3). These profiles suggest that small and midsize cities contained a greater percentage of older street trees relative to younger street trees than larger cities. Conversely, for the 40 small cities comprising the 1978 dataset, the DBH distribution profile for all street trees showed a descending trend line from the 0 cm to 7.62 cm (0 in to 3 in) DBH size class to the 15.24 cm to 30.48 cm (6 in to 12 in) DBH size class, followed by a peak in the 30.48 cm to 50.80 cm (12 in to 20 in) size class before descending again in the > 50.80 cm (> 20 in) size class (Figure 2). As stated above, Wray and Jungst (1981) believed that the large number of trees contained in the smallest DBH classes reflected a surge in the planting of trees following the removal of elms killed by DED. Davis (1993), in a longitudinal study of 22 communities in Kansas which had been inventoried 2 or 3 times over 20 years, likewise identified a surge in tree planting in the wake of DED as communities responded to dramatic street tree population losses that occurred in a relatively short period of time. However, Davis (1993) also stated that the surge in tree planting had not been sufficient to keep pace with tree mortality in many communities, which then resulted in aging street tree populations and greater maintenance costs. Wade and Kielbaso (2014) made a similar finding for 6 Midwestern United States cities where the average trunk diameter of publicly planted trees increased between 1980 and 2005, which they attributed to the cities not planting enough new trees. This paper identified for the 2008 to 2024 datasets a descending DBH profile for large Iowa cities that suggests a large number of new trees have been planted relative to the rest of the street tree population (Figure 3). It is beyond the scope of this paper to judge whether the number of new trees will be sufficient to keep pace with tree mortality or, if not, if aging street tree populations and increased maintenance costs will occur. However, peaks in the 30.48 cm to 45.72 cm (12 in to 18 in) DBH size class for small and midsize cities suggest that aging street tree populations will likely result over time and create greater maintenance costs for many of the inventoried cities.

This paper also assessed street tree condition in the 2008 to 2024 datasets and made comparisons by city size. Findings made for street tree condition were mixed. The percentage of trees in all cities receiving good condition wood ratings steadily declined as trunk diameter increased, consistent with the comment made by Wray and Mize (1985) that overall tree condition tended to deteriorate as trees aged and trunk diameter increased (Figure 4). However, the percentage of trees receiving good condition leaf ratings showed little if any decline as trunk diameter increased (Figure 5). Results for tree condition by city size were similarly mixed. Small cities were found to have the lowest percentage of trees receiving good condition wood ratings, and a modest increase was found for the percentage of trees receiving good condition wood ratings as city size increased. A similar relationship was not found for condition leaves and city size (Table 7).

The mixed findings for tree condition and city size are not surprising and may be due to dataset limitations. Data for tree diversity and trunk size are fairly straightforward to collect. For example, accurate trunk size data can be obtained using a measuring tape, although there is some inconsistency in how multi-stemmed trees are measured (Magarik et al. 2020). However, tree condition ratings involve percentage estimates which can involve more subjectivity than identifying tree species and determining trunk diameter (Gartner et al. 2002; Ma et al. 2021). This subjectivity can be further impacted by inter-operator variability if, as was the case for the 2008 to 2024 datasets, multiple individuals are involved in data collection over many years. Given potential issues associated with collecting tree condition data, perhaps more weight should be given to tree diversity and age when making findings for datasets comprised of multiple street tree inventories conducted at different periods of time, or perhaps more weight should be given to wood condition than to leaf condition. However, there is also the possibility that the inconsistencies in tree condition ratings are informative and suggest nuances in the data that require further study.

If results for tree condition are mixed, results for tree diversity and age appear more instructive and suggest work needs to be done to both increase tree diversity and plant new trees, especially in small cities. These objectives can be closely related. In the aftermath of DED when many trees needed to be planted to replace elms that had died, Wray and Mize (1985) were concerned about tree species selection and which trees were best suited as street trees in Iowa. High percentages of less frequently planted tree species, such as Q. macrocarpa (bur oak), Platanus occidentalis (American sycamore), and T. cordata (littleleaf linden) were found in the 1978 inventories to be in better condition relative to other tree species as trunk diameter increased. Wray and Mize (1985) therefore reasoned that these species were good candidates for street tree plantings that would not only diversify street tree populations but would also have greater survivability and therefore require less maintenance than other tree species, including many of the species most prevalent in the 1978 dataset. Selecting tree species for street tree planting with greater survivability and needing less maintenance was important for small cities because, as Wray and Jungst (1981) had found, small cities had limited financial resources and a lack of trained personnel dedicated to street tree management; tree maintenance was a low priority for city officials and typically occurred only in cases of severe need or when everything else needing to be done had been done.

When Wray and Mize (1985) evaluated tree species for street tree plantings, the EAB and bur oak blight (Tubakia spp.) had not yet become problems in Iowa, nor had the Asian longhorned beetle (ALB) (Anoplophora glabripennis) started threatening trees elsewhere in the United States. The ALB is especially worrisome because it is a polyphagous pest that has a preference for the tree species and genera it attacks but will attack multiple species and genera; therefore, the strategy of minimizing tree losses to an invasive pest by selectively planting some tree species and avoiding others may be ineffective (Laćan and McBride 2008). The ALB has yet to be found in Iowa (Iowa Department of Agriculture & Land Stewardship 2025), but it is a potentially serious threat to the state’s street tree population because it prefers trees in the Acer genus, and there is an overabundance in Iowa of Acer spp. street trees, particularly in the state’s small cities.

The IDNR similarly believes that there are too many Acer spp. street trees in Iowa and that the state’s street trees and trees growing on private property should be diversified. It has therefore created a guide for planting trees besides those in the Acer genus titled “Rethinking Maples: A Case for Cultivating Tomorrow’s Canopy” (Iowa DNR 2023). After stating that “maples make up more than one third of all tree in Iowa communities” and that “a diverse mix of trees is necessary for maintaining a healthy and resilient community forest,” the guide recommends 71 species besides Acer spp. for planting, both native and non-native to Iowa, including, as suggested by Wray and Mize (1985), Q. macrocarpa and P. occidentalis, which are native, and T. cordata, which is not. Other species contained in the guide include Q. ellipsoidalis (Northern pin oak), Q. muehlenbergii (chinkapin oak), and Q. shumardii (Shumard oak), as well as Aesculus glabra (Ohio buckeye), Liquidambar styraciflua (sweetgum), and Ostrya virginiana (hophornbeam). These species have been successfully planted as street trees in Iowa but not in large numbers. “Right tree, right place” guidance is also given to “homeowners and city staff” regarding species requirements and preferences for soil type, spacing, and sun versus shade exposure. This guidance is essential, not only because a tree species well-suited to site conditions is more likely to grow well and survive, but because this information may not be known by municipal officials responsible for planting decisions, especially in small and midsize cities. Only 11 Iowa cities employ a municipal forester and none with a population less than 23,982 (Hannigan 2025, personal communication; Iowa State Data Center 2025). Guidance is also given to select tree species resistant to storm damage. This topic is relevant in Iowa due to derechos, rapidly moving thunderstorms with damaging straight-line wind gusts at least 58 mph (93.3 kph) occurring over an area at least 250 mi (402.3 km) long that strike the state periodically (National Weather Service 2025). A derecho in August 2020 was particularly severe. In Cedar Rapids, Iowa’s second largest city, sustained winds of 70 mph (112.7 kph) and wind gusts up to 140 mph (225.3 kph) destroyed 18 percent of the city’s municipally managed trees (Jordan 2021; Confluence Inc. 2022).

Recommending both native and non-native tree species can also be important to increasing tree diversity. Because consumer preferences for commonly planted tree species are persistent (Simons and Hauer 2014) and nurseries tend to stock tree species which sell well and not less popular ones, underutilized tree species can be difficult to source (Sydnor et al. 2010; Conway and Vecht 2015). In fact, Iles and Vold (2003) found that Iowa nurseries overproduced, and Iowa landscape professionals specified, a disproportionately small number of tree species and cultivars. Therefore, although native tree species may be better adapted for local climates and offer better wildlife habitat (Conway et al. 2019), the choice of available tree species may be further reduced and diversifying tree populations made more difficult if native species are prioritized to the exclusion of non-native species (Sjöman et al. 2016).

Finally, while this paper’s findings for tree diversity and the overabundance of Acer spp. agree with the IDNR and its general recommendation to plant trees other than those belonging to the Acer genus, there are limitations with the datasets analyzed in this paper that should be acknowledged since they could potentially impact its findings. Of particular significance, and primarily in the inventories conducted by IDNR, is the large number of trees identified at only the genus level or not assigned to a species or genus. Many trees, following i-Tree protocols, were grouped as Broadleaf Deciduous Other, Coniferous Evergreen Other, etc., because their species were not contained in i-Tree’s Midwest tree list when the IDNR inventories were conducted. Examples include A. × freemanii (Freeman maple), which was inventoried as Acer spp.; Populus alba (white poplar) and P. grandidentata (big-tooth aspen), which were inventoried as Populus spp.; and Abies concolor (white fir) and Picea glauca (white spruce), which were inventoried as CEL OTHER (Conifer Evergreen Large) (Hannigan 2025, personal communication). Therefore, the diversity of species and genera contained in the IDNR dataset for small and midsize cities is almost certainly greater than can be reported based on the inventory data. In addition, given that some tree species such as A. × freemanii, P. alba, and P. grandidentata were grouped at the genus level rather than being accounted for at the species level, genus diversity is most likely a more accurate indicator of diversity than species diversity for small and midsize cities. However, even if the relative abundance of all Acer spp. in small cities is more accurately 40% than 42%, or in midsize cities is more accurately 38% than 40%, Acer spp. remain significantly over-represented in those street tree populations, and greater tree diversity is still warranted. The situation with Fraxinus spp. is more nuanced, since EAB has been present in Iowa since at least 2010 and has been confirmed in all of the state’s 99 counties (Iowa Department of Agriculture & Land Stewardship 2024). It is reasonable to assume that, given that large numbers of Fraxinus spp. have already been killed by the EAB and more will be killed in the future and that DBH distributions for all city sizes indicate few Fraxinus spp. are being planted, the over-representation of F. pennsylvanica in Iowa street tree populations, relatively minor compared to South Dakota, USA, will decline and at some point no longer be true.

Additional limitations in the data include the following. Cities inventoried in 1978 were randomly selected, but the cities inventoried between 2008 and 2024 were not; therefore, findings based on the 2008 to 2024 datasets could be vulnerable to selection bias and might not comprise a representative sample. However, the 2008 to 2024 datasets are comprised of a much larger number of cities and trees than the 1978 dataset, and a larger sample size is more likely to contain less variability and thereby lead to more accurate tree population estimates. The definitions of small cities based on population size also differ between the 1978 and 2008 to 2024 datasets, which creates some doubt about construct validity and the accuracy of comparisons made between the datasets for tree diversity, age, and condition in small cities. The 1978 dataset, unlike the 2008 to 2024 datasets, contains summary data only; this greatly limits the ability to rework the 1978 dataset and thereby enable, for example, comparisons to be made for tree diversity, age, and condition for small cities of equivalent population size. Lastly, the 1978 dataset only contains data for cities with populations of 10,000 or less; analyses comparable to those made for these cities cannot be made for cities with larger populations. By contrast, the 2008 to 2024 datasets contain individual data records for trees including data for midsize and large cities which facilitates comparisons to be made for variables of interest on the basis of city size.

Conclusion

More than 50 years have passed since Wray and Prestemon (1983) wrote that information about street tree management in Iowa’s small cities was negligible. The street tree inventories conducted in 1978 began providing that information, and the inventories conducted by IDNR advanced that process significantly. It needs to advance still further. This paper was able to make findings from the 2008 to 2024 datasets for street tree diversity, age, and condition on the basis of city size and to make limited comparisons with the 1978 dataset for small cities. It was unable to make statistically significant longitudinal findings for these variables in small cities to determine trends associated with them. The data required to do so, ideally from a randomly selected set of small cities geographically dispersed to account for variations in temperature and precipitation that are re-inventoried on a regular periodic basis, are not available. IDNR has started to re-inventory cities that had been inventoried previously, but it is unclear if data are being collected in a manner that will facilitate longitudinal analyses or if longitudinal analyses are a priority. The importance of making such findings has increased with climate change. Street trees are more vulnerable to increases in temperature, drought, and precipitation intensity due to already stressful streetscape conditions, and adaptability to climate change has become an important factor in assessing the health and survivability of street tree populations (Lohr et al. 2016) and the continued provision of the many benefits that street trees provide (Morgenroth et al. 2016). Brandt et al. (2021) have projected that increases in the number of hot days and in precipitation in the Upper Midwest due to climate change will likely modify the suitability of some tree species currently planted along streets and favor those tree species that are more heat and drought tolerant. Unfortunately, the need to re-inventory street trees to facilitate longitudinal analyses and more fully understand trends in street tree structure and composition with the impact of climate change corresponded in Iowa with a reduction in state government funding for natural resource management (Iowa DNR 2020). The ability of IDNR’s Urban and Community Forestry program to work with cities in the state to improve the sustainability of their urban tree resources has therefore depended increasingly on federal grants and partnerships with nonprofit organizations. However, recent events indicate that many federal grants supporting street tree management are in jeopardy of being reduced or eliminated. For example, a $630,000 USD grant awarded to Decorah, Iowa, in 2023 to fund tree trimming and pruning, new tree plantings, and a street tree inventory was cancelled in 2025 due to a nationwide federal funding freeze (Decorah News 2025).

When Wray and Jungst (1981) interviewed officials from the cities inventoried in 1978 and asked what type of assistance would encourage more street tree planting and maintenance, the primary response was that more financial assistance was needed. Similarly, in 2010, nearly 70% of respondents to a statewide survey of Iowa communities commissioned by Trees Forever, an Iowa nonprofit and volunteer-based tree planting organization, identified lack of funds as the most significant challenge in carrying out tree planting and tree care projects (Iowa State University 2010). Funding for street tree management continues to be an issue for Iowa’s small cities. Although IDNR provided a detailed 6 year tree management plan tailored to the budget resources of each inventoried community, finding funds to remove all Fraxinus spp. and hazardous trees and also maintain and prune existing trees has been difficult for many cities (Murrow 2025, personal communication). Since 67.7% of Iowa cities with a population of 5,000 or less in the 2020 US Census experienced a decline in population from the 2010 US Census (Iowa State Data Center 2025), it appears unlikely that the financial resources available to Iowa’s small cities for street tree management will increase substantially anytime soon. Despite this concern, the prognosis for street tree management in Iowa’s small cities seems more positive than in the early 1980s. Whereas Wray and Jungst (1981) found that street tree management plans were nonexistent in Iowa’s small cities, in 2024 IDNR reported that 352 of Iowa’s small cities had management plans (Iowa DNR 2024b). Wray and Jungst (1981) also found discrepancies between real tree maintenance needs and maintenance needs as perceived by city officials. For small cities where IDNR conducted inventories and developed management plans, city officials now have an accurate understanding of real tree maintenance needs. Even if funds are not available to act fully upon those plans, accurately understanding maintenance needs facilitates more efficient allocation of resources and increases the prospects for improvement in street tree health and condition. Wray and Prestemon (1983) additionally found that funding issues reduced tree planting. Beginning in 1989, the Iowa legislature required utilities doing business in the state to promote energy efficiency; subsidizing a tree planting program was one way a utility could meet this mandate, and many did and continue to do so (Vitosh and Thompson 2000). Iowa’s small cities have participated in these programs and benefitted from them, although the extent to which these programs have positively impacted their street tree populations is unclear.

Therefore, while this paper found in following up on the 3 papers written in the 1980s that Iowa’s small cities may still have some issues in managing their street trees as reflected in the findings made for tree diversity, age, and condition, it appears that progress is being made in addressing them, although the extent and pace of that progress cannot be fully ascertained due to a lack of data. Moving forward, IDNR’s Urban and Community Forestry program will hopefully have sufficient resources to continue assisting Iowa’s small cities in managing their street trees. The assistance to small cities seems especially warranted in a state characterized and in many ways defined by them so that these cities will be able to better manage their street trees and maintain if not increase the benefits these trees provide.

Conflicts of Interest

The author reported no conflicts of interest.

Acknowledgements

The author thanks E. Hannigan and C. Morrow, Iowa Department of Natural Resources, for graciously making available the IDNR street tree inventory data without which the paper could not have been written.

Does Excess Mulch Depth Lead to Poor Tree Growth and Condition, Root Girdling, and Decay? A Systematic Literature Review

Thu, 01 Jan 2026 11:00:00 GMT

Abstract

Mulch is placed around the base of trees to improve soil conditions, water conservation, and tree growth while decreasing weed competition, mower damage, and soil compaction. Current industry best practices and trade magazine articles recommend a mulch depth of 5 to 10 cm (2 to 4 inches) and caution against exceeding this depth, warning of issues affecting stem tissue like stem girdling roots and pathogens. To examine scientific support for this threshold, we conducted a systematic review of the peer-reviewed literature on excess mulch depth. We identified 11 studies that examined the effects of increasing depths of mulch on tree and soil physiology. All but two studies tested mulch depths exceeding the 5- to 10-cm (2- to 4-inch) range. The impact of deep mulch is unclear; methodological differences, including mulch type and examined variables, limit comparisons between studies. It is possible that fine mulch with low porosity results in deleterious effects similar to planting trees too deeply, explaining observations by practitioners. While further research should determine the effects of mulch depth beyond 10 cm (4 inches) on tree physiology, there are often negative side effects reported for exceeding 10 cm (4 inches) but few negative effects reported for mulch depths within 5 to 10 cm (2 to 4 inches).

Introduction

Applying a layer of mulch around the base of a tree prevents or reduces many of the common abiotic disturbances that impact trees in the urban environment. Mulch can improve soil moisture, reduce soil erosion and compaction, prevent soil temperature fluctuations, increase soil nutrition, decrease weed competition, and decrease the impact of salts and environmental contaminants (Chalker-Scott 2007; Marble et al. 2015; Saha et al. 2020). A ring of mulch around the base of a tree also helps to protect it from mechanical damage caused by mowers and grass or string trimmers (Morgenroth et al. 2015; Blair et al. 2019; Lilly et al. 2022). When managing trees through construction, proactively applying a layer of mulch can reduce soil compaction caused by the movement of heavy equipment (Lichter and Lindsey 1994; Fite et al. 2011). All these benefits of mulch contribute to improved tree establishment and growth (Chalker-Scott 2007; Maggard et al. 2012).

Despite these many benefits, excessive mulching may cause problems. Within the arboriculture industry, there has been significant emphasis in trade magazines and extension publications on the negative impacts of excessive mulch depth, including stem girdling roots, early tree mortality, excessive temperatures, and reduced gas exchange (Billeaud and Zajicek 1989b; Rakow 1992; Ball 1999; Herms et al. 2001; Carlson 2002; Jackson 2018; Smith 2020; Boggs 2024). These problems were noted in unsubstantiated articles in trade magazines as early as the 1980s when Gouin (1983, 1986) wrote that over-mulching—both excess mulch depth and mulch against the trunk—allegedly caused the suffocation of shallow roots, the decomposition of roots, the production of roots from the trunk, the formation of trunk cankers, and created environments conducive to pathogens like Phytophthora cinnamomi (cinnamon fungus) and Botryosphaeria ribis. These cumulative impacts were said to produce symptoms like foliar chlorosis, small leaves, poor growth, and dieback (Gouin 1986). Gouin (1986) wrote that “often the problems are irreversible, and the plants are doomed.” These sentiments have been echoed in trade articles (e.g., Ball 1999; Carlson 2002), extension fact sheets (e.g., Boggs 2020; Appleton 2021; Rainey et al. 2022; Boggs 2024), and online resources (e.g., Johnson et al. 2008) ever since.

Because trade magazines and extension publications serve as important methods of knowledge transfer for arborists (Martin et al. 2024), the sentiment that too much mulch is deleterious has been widely accepted, despite commonly lacking reference to formal peer-reviewed and replicable research (Giblin 2013). Presently, industry best management practices (BMPs) suggest a layer of 5 to 10 cm (2 to 4 inches) of mulch applied around the tree but kept away from the trunk (Herms et al. 2001; Watson 2014a; Costello et al. 2017). After settling, the mulch should be 5 to 7.5 cm (2 to 3 inches) deep (Watson and Himelick 2013; Watson 2014a).

The recommendation for a shallow depth of mulch must be balanced against the needed depth to achieve desired outcomes. Some of the many benefits of mulch, including weed control and construction damage mitigation, are dependent upon a mulch depth that exceeds the BMP’s 5 to 10 cm (2 to 4 inch) depth recommendation. For example, Greenly and Rakow (1995) found that increased mulch depth, which they tested up to 25.5 cm (10 inches), reduces weed density and diversity. In another example, Lichter and Lindsey (1994) recommended a mulch layer of 15 cm (6 inches) or greater to reduce soil compaction during construction. Presently, it is unclear what scientific evidence supports the standard 5 to 10 cm (2 to 4 inch) mulch depth recommendation nor what should be considered an excessive mulch depth for routine applications.

To better understand the negative impacts on tree physiology associated with mulching in excess of the 5 to 10 cm (2 to 4 in) depth currently recommended in industry BMPs and textbooks, we conducted a systematic review of the peer-reviewed literature. These types of “mini reviews” that focus on specific research questions are increasingly common, summarizing nascent literature on commonly held practices that may lack scholarly support and providing straight-forward summaries for practitioners (Elfar 2014). In environmental sciences, mini reviews have examined limited literature on emergent topics like biochar in agriculture (Wang et al. 2022), green nanotechnology (Salem 2023), wastewater bioremediation (Nyika and Dinka 2022), plastic marine debris (Wayman and Niemann 2021), social sciences in dry forest restoration (Powers 2022), and the impact of nitrogen emissions on environmental and human health (de Vries 2021). This format is important in arboriculture where many practices and commonly held perceptions lack scientific review (Gillman 2008; Chalker-Scott and Downer 2020).

In this systematic mini review of mulch depth, we examined the methods, tested depths, overall findings, and limitations of studies on the impact of mulch depth on tree physiology. This review benefits arborists, horticulturists, and landscapers by providing a summary of evidence supporting—or contrasting—their understanding of deep mulch applications. This review is also beneficial for professional associations like the International Society of Arboriculture (ISA), which produce BMPs and standards that reflect both industry consensus and the known science. For researchers interested in tree physiology and interactions with physical, chemical, and biological processes of soils, this review highlights research gaps and the opportunity for future research.

Materials and Methods

Search Strategy and Screening

This review was conducted according to the procedures for systematic reviews outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)(Page et al. 2021b). First, a search was conducted in 5 common databases for keywords in study titles, abstracts, and keywords (Figure 1). The search terms and strings with Boolean operators were: “mulch” AND (“depth” OR “volcano” OR “girdling” OR “rot” OR “decay” OR “insect” OR “pathogen” OR “pest” OR “rodent”) AND “tree.” Akin to Martin and Conway (2025), we also searched the 10 arboriculture and urban forestry journals that are most often referenced by arborists and urban foresters (Martin et al. 2024) as several are not indexed in the 5 databases (Hauer and Koeser 2018). In addition to the 5 databases and 10 arboriculture and urban forestry journals, we also searched the Journal of Environmental Horticulture. Any researcher funded through the Horticulture Research Institute Grant is required to be published in the Journal of Environmental Horticulture (Horticultural Research Institute [date unknown]), so it was likely that mulch depth studies have been published in this venue. The search of the databases and first 10 journals was conducted on 2024 December 12. We updated the review to include the Journal of Environmental Horticulture on 2025 March 19.

Figure 1.

PRISMA diagram of the systematic literature review for studies on the effect of mulch depth on tree physiology (Page et al. 2021a). Journals marked with an asterisk (*) did not offer .ris export options for search results; therefore, study records were manually compiled into a .ris file and uploaded to Covidence.

A total of 1,200 studies were identified from the databases and journals (Figure 1). All studies identified during the search were imported into Covidence, a systematic review platform, that was used for study screening. After the removal of duplicates, 1,055 studies were screened based on titles and abstracts. Following screening, 89 studies were read in full to assess their eligibility for inclusion.

Only English-language studies were included in this review. To be included, studies must have examined trees or woody shrubs and considered mulch depth as a factor. At least one type of mulch used must have been a woody mulch type. Mulch depth must have been varied in the study design. Therefore, studies that compared no mulch (control) to one depth of mulch (e.g., 10 cm) were excluded. If studies had multiple experiments, the study was included if at least one experiment met the inclusion criteria, in which case the relevant experiment is examined in the Results. To ensure that studies of high-quality were included, articles published in non-peer-reviewed journals (e.g., Arborist News) were excluded. No automated filters were used to exclude studies based on these criteria.

After 10 studies were determined to be eligible for inclusion, the included studies were read to identify any additional studies cited in the included study that meet the inclusion criteria, a literature review process called backward citation searching (Jalali and Wohlin 2012). Through backward citation searching, we identified one study that met the inclusion criteria and was therefore included as the 11th study in our review (Pellett and Heleba 1995)

Analysis of Included Articles

For each study, we identified bibliographic details (publication year and journal), study methods (mulch type and depth and tree species tested), and the variables examined. Because some species can be classified as trees or woody shrubs, the term “tree” is used throughout to avoid differentiation between woody shrubs and trees. Several studies included multiple experiments. Only the experiments that met the inclusion criteria were examined. We present an overview of the findings of the 11 studies and examine their limitations.

Results

A total of 11 studies were identified that examined the impacts of mulch depth on tree physiology. The 11 studies are identified in the Appendix. The publication dates ranged from 1989 to 2023, although 9 studies were published before 2010. Only two journals had published more than one study. In the Journal of Environmental Horticulture, 5 studies were published, and 3 studies were published in the Journal of Arboriculture—now publishing under the name Arboriculture & Urban Forestry.

Mulch Depths and Types

All but two studies—Bartley et al. (2016) and Richardson et al. (2008)—tested woody mulch depths that exceed the 5 to 10 cm (2 to 4 inches) recommendation for normal mulch depths by at least 5 cm (2 inches) (Figure 2). All studies included a control depth—that is, an area of no mulch (0-cm [0-inch] mulch depth). Excluding the no-mulch control depth, 4 studies tested 2 mulch depths while 7 studies tested 3 mulch depths. The most common interval between tested depths was 5 cm (2 inches)(n = 8), followed by 10 cm (4 inches)(n = 5), and 7.5 cm (3 inches)(n = 4). Only two studies, Greenly and Rakow (1995) and Sun et al. (2023), tested mulch depths with inconsistent intervals.

Figure 2.

Mulch depths tested in the 11 reviewed studies.

Studies varied in how many mulch types were examined. There were 5 studies that only tested 1 mulch type and 3 tested 2 mulch types. Beyond 2 mulch types, Billeaud and Zajicek (1989a) tested 4 mulch types while Foshee et al. (1996) and Bartley et al. (2016) tested 5 mulch types. This totals to 25 mulch tests across 11 studies and 17 unique mulch types. Of the 17 unique mulch types, only 3 were used in more than one study: pine (Pinus spp.) bark nuggets (n = 5), pine (Pinus spp.) wood and bark mulch (n = 4), and hardwood mulch (n = 2).

Tested mulches were made from 12 different source materials, including 7 tree genera and species. Across the 25 mulch tests, pine (Pinus spp.) was the most commonly used mulch material, used in 10 studies. This includes the pine species loblolly pine (Pinus taeda). The other tree genera and species used as mulch material—baldcypress (Taxodium distichum), cypress (Cupressus spp.), eastern redcedar (Juniperus virginiana), pecan (Carya illinoinensis), Chinese privet (Ligustrum sinense), and sweetgum (Liquidambar styraciflua)—were each only used once. There were 3 studies that used undefined wood mulch, classifying it as hardwood mulch (n = 2), softwood mulch (n = 1), chipped limbs (n = 1), and shredded mulch (n = 1). Grass clippings, garden waste, hardwood leaves, and newspaper were each used once.

Tree Species

After updating botanical names to reflect current nomenclature per Royal Botanic Gardens‚ Kew (2024), a total of 22 species were used across the 11 studies. In examining the impact of mulch depth, 6 studies tested only 1 tree species and 3 studies tested 2 tree species. Of the 3 studies that tested more than 2 tree species, Pellett and Heleba (1995) tested 3 species, Richardson et al. (2008) tested 4 tree species, and Hild and Morgan (1993) tested 5 tree species. Wax-leaf/Japanese privet (Ligustrum japonicum) was the only species to appear in more than one study.

Replicants and Plot Designs

The total number of trees planted in any study varied between 49 and 400 trees (Table 1). Six studies tested multiple tree species and reported the number of trees. Greenly and Rakow (1995), Hild and Morgan (1993), Richardson et al. (2008), and Bartley et al. (2016) used the same number of tree species per treatment whereas Arnold et al. (2005) and Pellett and Heleba (1995) used different numbers of tree species per treatment.

Table 1.

Number of trees, treatments, and replicants and block design used in the 11 reviewed studies on mulch depth.

Plot designs were most often randomized complete block design where each tree was an individual unit planted into the block (n = 5)(Table 1). In two studies, a randomized complete design was used with trees in individual nursery containers. Split-split plots, split plots, random plots within a forest, and randomized complete block design with 7-m2 plots were all used in one study each.

In 2 cases, the total number of trees did not match the number of replicants and blocks. In Greenly and Rakow (1995), a total of 60 trees, divided between 30 Quercus palustris and 30 Pinus strobus, were planted in 4 blocks with an equal number of Q. palustris and P. strobus per 3 mulch depths and two mulch types plus one unmulched control per block. With 7 treatments per species (6 mulch treatments plus 1 control) replicated across 4 blocks, that would total 28 trees per species; whether the 2 remaining trees per species were kept in reserve was not mentioned. Arnold et al. (2005) described having 8 blocks of Koelreuteria bipinnata with “single plant replication of each treatment combination” and 12 treatments (4 mulch depths, 3 planting depths). This would total 96 K. bipinnata, 12 greater than the 84 trees mentioned earlier in their study.

Variables Examined

The 11 studies explored variables grouped as tree physiology, soils, and pests (Table 2). Within these 3 broad groupings, variables were classified into 9 tree physiology classes, 7 soil classes, and 2 pest classes. Tree physiology classes were the most common, of which trunk diameter and area was most commonly measured (n = 6).

Table 2.

Classification of variables measured by the 11 reviewed studies. Open circles indicate variables included in authors’ own indices. Totals for each class include only the filled black circles.

Methods for measuring variables within these classes varied. For example, trunk diameter was measured at just above the root flare by Greenly and Rakow (1995), at 1.3 cm (0.5 inches) above ground level by Hild and Morgan (1993), and at 15 cm (6 inches) above the soil surface by Arnold et al. (2005). In another example, tree condition was measured using a scale of 0 (very poor quality) to 5 (“high quality container plant with excellent color and aesthetic appearance”) by Billeaud and Zajicek (1989a) and a 0 (no observed injury) to 10 (dead plant) phytotoxicity rating by Bartley et al. (2016).

Three studies also included a composite index. In all cases, this index included crown (canopy) width and tree height. Richardson et al. (2008) and Bartley et al. (2016) used the same index but formulated differently. The index can be given as

where h is tree height and w1 and w2 are the perpendicular crown width measurements. Greenly and Rakow (1995) used a different composite index, the formula for which is given as

where h is tree height, w is a singular crown width measurement, and c is the tree caliper.

Summary of Findings

Results across the studies varied; study findings were conflicted on whether increased mulch depth is detrimental to tree physiology. In addition, many of the variables in Table 2 were only examined by one study or, if examined by multiple studies, were examined under different biophysical conditions or using different methods. There are also data handing issues (e.g., data lost for phosphorous [Stafne et al. 2009] and reporting limitations or gaps that preclude further analysis). For example, Billeaud and Zajicek (1989a) examined 4 different mulch types but examined the effect of mulch depth on new growth without a mulch type factor. With these limitations in mind, we summarize the 11 studies’ findings in the following subsections.

Impact of Mulch on Tree Physiology

Tree Mortality: Arnold et al. (2005) found that mulch depth had a significant effect on tree mortality, increasing with increasing depth. Mulch depths as little as 7.6 cm (3 inches) led to increased tree mortality, and mortality rates were more profound for Koelreuteria bipinnata. Stafne et al. (2009) found that tree mortality increased with mulch depth, increasing from less than 5% mortality in 5 cm (2 inches) or less of mulch to 15% in 10 cm (4 inches) of mulch and 35% in 15 cm (6 inches) of mulch. They attributed this to atypically high precipitation rates with greater soil moisture under the 10- and 15-cm (4- and 6-inch) mulch depths. In contrast, Hild and Morgan (1993) found that tree mortality did not differ with mulch depth after 18 months.

Tree and Foliage Condition: Arnold et al. (2005) found that mulch thickness had a significant effect on the percentage of the canopy with symptoms of foliar stress (including “chlorosis, marginal necrosis, and/or premature leaf senescence”), generally increasing with mulch depth. Billeaud and Zajicek (1989a) found that differences in visual appearance were not statistically significant between mulch and non-mulch treatments, except for pine mulch, which had a significantly lower rating than the four other mulch types and the control. The effect of mulch depth was not reported. Bartley et al. (2016) measured phytotoxicity ratings, finding no phytotoxicity, although the statistical results are not reported for mulch depth. Hild and Morgan (1993) found that canopy fill did not significantly differ between mulch depths.

Trunk Diameter: No study reported an effect of mulch depth on changes in trunk diameter. Arnold et al. (2005) and Hild and Morgan (1993) found that mulch thickness had no significant effect on trunk diameter. Gilman and Grabosky (2004) found that there was no difference in caliper growth between 0-cm (0-inch), 7.5-cm (3-inch), and 15-cm (6-inch) mulch depths when mulch was applied over the root ball. Greenly and Rakow (1995) found that caliper was greater for mulched versus unmulched trees, but the effect of mulch depth on caliper was not reported.

Trunk Area: Foshee et al. (1996) and Stafne et al. (2009) both examined trunk cross-sectional area (TCSA). Foshee et al. (1996) found that TCSA increased with increasing mulch depth, although there was little difference between 20 and 30 cm (8 and 12 inches) of mulch depth. Stafne et al. (2009) found that TCSA was higher with mulch than without, but TCSA between mulch depths did not significantly differ.

Shoot Growth: Greenly and Rakow (1995) found that shoot growth was significantly greater at 7.5-cm mulch depth than, in order of greatest to least shoot growth, 15, 0, and 25.5 cm (6, 0, and 10 inches) of mulch. The differences for oak shoots were more profound than pine shoots, and oak shoots had notably lower growth than 0 cm (0 inches) of mulch depth. Pellett and Heleba (1995) found that 10-cm (4-inch) mulch depth had reduced growth rate versus 5-cm (2-inch) mulch depth. Billeaud and Zajicek (1989a) found that shoot dry weights did not differ between mulch types and bare soils except for pine mulch, which had the lowest shoot dry weights, but did not report the effect of mulch depth. However, Billeaud and Zajicek (1989a) found that increasing mulch depth from 0 to 15 cm (0 to 6 inches) resulted in decreasing new growth.

Crown Width: Hild and Morgan (1993) measured crown width as the average of the greatest horizontal crown width and its perpendicular crown width. They found no significant difference in crown width across 0, 7.5, and 15 cm (0, 3, and 6 inches) mulch depths.

Tree Height: Arnold et al. (2005) found that mulch thickness had a significant effect on the height of Fraxinus pennsylvanica but did not have a significant effect on the height of Koelreuteria bipinnata. The study did not include a post-hoc test for pair-wise comparisons, but the authors reported that mulching had a negative effect on F. pennsylvanica height at mulch depths greater than 7.6 cm (3 inches) when trees were planted at or above grade. The lack of height reduction for below-grade planting under 7.6 cm (3 inches) of mulch depth was attributed to the low survival rate of the trees planted below grade (Arnold et al. 2005). Hild and Morgan (1993) found no significant difference in tree height across 0, 7.5, and 15 cm (0, 3, and 6 inches) of mulch depths. Stafne et al. (2009) found that tree heights were greater for mulched than bare soil but found no significant difference in tree height across 5, 10, and 15 cm (2, 4, and 6 inches) of mulch depth.

Roots: Sun et al. (2023) found that 5 cm (2 inches) of mulch produced the most positive effects on fine root morphology at 6 months of mulching but was surpassed by 10 and 20 cm (4 and 8 inches) of mulch in the 9th and 12th months of mulching. Sun et al. (2023) also found that specific root length, specific surface area, and fine root biomass were lower in 5 cm of mulch while root tissue density was higher in 5 cm (2 inches) of mulch than the 10 and 20 cm depths (4 and 8 inches). Billeaud and Zajicek (1989a) found that root dry weights did not differ between mulch types and bare soils but did not report the effect of mulch depth.

Composite Indices: Bartley et al. (2016) and Richardson et al. (2008) found no impact of mulch depth on tree growth measured via their shared composite index. Greenly and Rakow (1995) found that mulch depth had a significant influence on growth, measured via their index (width × height × caliper), but the directionality of the association was not reported.

Impact of Mulch on Soil Chemistry and Temperature

Soil Moisture and Water Content: Broadly, soil moisture and water content were found to be higher under a layer of mulch, although the optimal layer of mulch was unclear. No mulch leads to water loss from evaporation, but deep mulch depths may inhibit irrigation or precipitation (Arnold et al. 2005) or retain too much water (Stafne et al. 2009). Arnold et al. (2005) found that soil water potential was least negative under 7.6 cm (3 inches) of mulch compared to bare soil or mulch depths up to 22.9 cm (9 inches). Similarly, Sun et al. (2023) found that mean soil water content at 0 to 20 cm and 20 to 40 cm was generally higher under 10 cm (4 inches) of mulch than 0, 5, and 20 cm (0, 2, and 8 inches). In contrast, Greenly and Rakow (1995) found that soil moisture increased with increasing mulch depth from 9.8% mean soil moisture with no mulch to 17.4% mean soil moisture with 25 cm (10 inches) of mulch, although differences between mulch depths were not significant. These findings may be due to the testing approach: Greenly and Rakow (1995) tested soil moisture one week after precipitation whereas Arnold et al. (2005) monitored soil water potential throughout the first growing season and Sun et al. (2023) measured at 6, 9, and 12 months after mulch application.

Soil pH: The impact of mulch depth on soil pH was equally unclear. Sun et al. (2023) found that soil pH generally increased with increasing mulch depth while Billeaud and Zajicek (1989a) found that soil pH was decreased with increasing mulch depths. Stafne et al. (2009) found that soil pH decreased with mulch than without, although there were no consistent trends with increasing mulch depth. Greenly and Rakow (1995) found pH to be unaffected by mulch depth. Similarly, Hild and Morgan (1993) found that soil pH was typically unaffected by mulch depth (n = 15), although in the few cases where it did (n = 3), soil pH decreased with increasing mulch depth.

Soil Salinity: Two studies tested soil salinity. Hild and Morgan (1993) found that soil electrical conductivity was significantly lower under mulch versus no mulch, but there were seldom differences between the 2 mulch depths of 7.5 and 15 cm (3 and 6 inches). Greenly and Rakow (1995) tested soluble salt concentration but did not examine it in the context of mulch depth.

Soil Nutrient Levels: Four studies examined soil nutrient levels, producing conflicting results. Greenly and Rakow (1995) found no relationship between mulch depth and nitrates. Billeaud and Zajicek (1989a) found that soil nitrogen was significantly lower under mulch than bare soils but found no significant differences between mulch depths of 5, 10, and 15 cm (2, 4, and 3 inches). Sun et al. (2023) did not report all soil nutrients analyzed, but 10 cm (4 inches) of mulch generally had higher soil organic carbon, ammonium, nitrate, microbial biomass carbon, and microbial biomass nitrogen while 20 cm (8 inches) of mulch generally had higher total phosphorus. Soil tests by Pellett and Heleba (1995) “showed little difference” for 6 nutrients and “small differences” for 7 heavy metals under mulch versus bare soil, but the effect of mulch depth was not reported.

Soil Oxygen: Greenly and Rakow (1995) were the only study to examine percent oxygen for mulch depths, finding a non-significant decline in percent oxygen with increasing mulch depths. The difference between 0 cm (0 inches) of mulch and 25.5 cm (10 inches) of mulch was only 0.9% soil oxygen.

Soil Temperature: The effect of mulch depth on soil temperature was examined in 3 studies. Greenly and Rakow (1995) found that the mean soil temperature in summer decreased with increasing mulch depth, although the difference became less pronounced into late September. There was often no significant pair-wise difference between 0 and 7.5 cm (0 and 3 inches) of mulch nor 15 and 25 cm (6 and 10 inches) of mulch. Pellett and Heleba (1995) found that soil temperature decreased with increasing mulch depth during afternoon temperatures but had little difference around sunrise. The difference between 5 cm (2 in) and 10 cm (4 inches) of mulch was less than the difference between 0 cm (0 inches) and 5 cm (2 inches). Sun et al. (2023) recorded no consistent trend in soil temperature under various mulch depths.

Impact of Mulch on Pathogens and Weed Control

Pathogens: After piling up mulch around the trunk, Greenly and Rakow (1995) found no visible discoloration or colonization by pathogens or cankers. Greenly and Rakow (1995) also cut wounds into the trees prior to planting, which had grown normal preventative calluses.

Weed Control: There were 5 studies that found that deep mulch applications provided the greatest weed control (Billeaud and Zajicek 1989a; Greenly and Rakow 1995; Pellett and Heleba 1995; Richardson et al. 2008; Stafne et al. 2009) with maximum mulch depths ranging from 7.5 cm (3.0 inches)(Richardson et al. 2008) to 25 cm (10 inches)(Greenly and Rakow 1995). However, at least some weed control benefits were provided with as little as 5 cm (2 inches) and 7.5 cm (3 inches) of mulch (Billeaud and Zajicek 1989a; Greenly and Rakow 1995).

Influence of Environmental Factors on Results

Several studies noted the potential impact of adverse weather. Arnold et al. (2005) noted that dieback on Koelreuteria bipinnata may have resulted from tip dieback in a late winter freeze. Gilman and Grabosky (2004) noted in their methods that a period of very hot and dry weather required supplemental irrigation. Some trees dropped a portion of their foliage, one lost all its leaves, and one died. Greenly and Rakow (1995) had the opposite issue: an above average year of precipitation during their no-irrigation study year likely reduced the effect of the mulch layer on water conservation. Similarly, Stafne et al. (2009) reported increasing tree mortality with increasing mulch depth but attribute the primary reasons to record rainfalls in the second study year. These influences reflect the general struggles of arboricultural and botanical research outside a controlled greenhouse experiment. This is reflected by Hild and Morgan (1993) who warn that results would differ under moisture limited environments, faster mulch decomposition, and in the absence of weed control.

Discussion

Popular ideas without scientific observation can become commonly held perceptions, a pervasive occurrence in the landscaping industry (Gillman 2008; Chalker-Scott and Downer 2020). These perceptions occasionally become so well-regarded that they appear in extension publications that are used by practitioners for educational purposes, despite lacking the support of scientific observations (Chalker-Scott and Downer 2018). Our mini review of the peer-reviewed literature on the impacts of deep mulch on tree physiology finds that while data show mortality and some growth parameters can be negatively impacted by excessive mulch depth, there is inconclusive evidence to establish a definitive threshold beyond which tree health is compromised. Whether deep mulching impacts trees, and the depth at which impacts occur, is likely dependent on soil and mulch type, soil water quantity, and species niches, including tolerance to water logging and organic matter deposition. Study designs, statistical approaches, and extraneous and nuisance variables preclude comparisons between studies. However, while studies differed in whether trees grew better with mulch depths beyond 10 cm (4 inches), no study found that tree survival was negatively impacted by planting within 5 to 10 cm (2 to 4 inches) of mulch.

A notable difference between studies is the type of mulches used. Not all mulches perform the same, and there are great differences in the effects of mulch type on plant growth and soil characteristics (Litzow and Pellett 1983). Not included in our review is a study by Watson and Kupkowski (1991) that compared 45 cm (18 inches) of coarse mulch to areas without mulch. While this study was not included in the review because it did not compare multiple mulch depths, Watson and Kupkowski (1991) found that the 45-cm-deep (18-inch-deep) mulch did not impact trees. Based on their findings, it is possible that the negative impacts of deep mulching are a consequence of finer mulch that becomes akin to soil. This is supported by Billeaud and Zajicek (1989a) who found that pine bark mulch, which was the smallest, densest mulch examined, resulted in the poorest visual plant rating, growth, and shoot dry weight. This aligns with a tree planting book by Watson and Himelick (2013), who wrote that fine textured mulch could reduce soil oxygen under wet conditions, even when applied with as little as 5 cm (2 inches).

Fine textured mulch would essentially raise the soil profile, causing the stem encircling or girdling roots noted in studies where trees were planted too deeply (Wells et al. 2006; Day and Harris 2008; Harris et al. 2016; Hauer and Johnson 2021). These girdling roots negatively affect trunk taper (Day and Harris 2008). Wells et al. (2006) also found reduced survival rates in deeply planted trees. This would explain the observations by practitioners and extension specialists who may be inferring that girdling occurs as a consequence of the soil profile being raised.

Additionally, the application of mulch increases soil moisture in both the mulch and underlying soil when compared to unmulched, grass areas (Watson 1988). This can improve survivability under drought conditions, but can cause mortality under extended flood conditions, as was observed by Stafne et al. (2009). As a result, studies on mulch depth undertaken during drought conditions will likely reach different conclusions than studies on mulch depth undertaken during flood conditions. Similarly, Drilias et al. (1982) proposed that deep planting (root collar 15 to 25 cm [6 to 10 inches] below grade) may lead to increased canker occurrence. If the decay of fine mulch leads to the more soil-like properties, this may explain the correlation between excess mulch depth and an increased incidence of pathogens reported by practitioners and extension specialists. Further study is needed on pathogen prevalence in both deeply planted trees and deeply mulched trees.

Additional Literature

Outside of the peer-reviewed literature, other formal studies have been conducted but remain unpublished. In 2010, the Tree Research & Education Endowment (TREE) Fund (Naperville, IL, USA) funded a research project titled “Evaluating damage resulting from volcano mulching”. Based on the project’s examination of 84 trees in Chicago, what appeared to be mulch volcanoes were actually thin mulch layers covering soil mounds, resulting from improper planting or maintenance practices (Watson 2014b). Both the soil mounds and mulch were well-aerated, and 80% of surveyed trees showed no disease symptoms. However, one-third of the trees had problematic encircling or stem girdling roots. Additional research on container-grown trees revealed that trees with volcano mulch exhibited more wound-related discoloration, suggesting this mulching practice may increase vulnerability to damage. However, the study found that while disease-causing fungi were isolated from volcano mulched trees, there was no indication that the fungi were aggressive pathogens, regardless of volcano mulching. Watson (2014b) concludes that “while volcano mulching is not considered best practice, when data from this study is fully analyzed, it is not expected to support the assertion that volcano mulching results in widespread and health-threatening disease development in landscape trees.”

A similar study is described in the MS thesis of Giblin (2013), which investigated the impact of wounding and mulch depth on the growth and internal discoloration of 48 Northwoods red maple (Acer rubrum ‘Northwoods’). Shredded wood mulch was applied with and without contact with the stem through 3 variations: 15 cm (6 inches) of mulch with no contact to the stem, 15 cm (6 inches) of mulch with contact to the stem, and 30 cm (12 inches) of mulch with contact to the stem (Giblin 2013). After 30 months, destructive harvesting found that the trees with 30 cm (12 inches) of mulch against the stem had significantly greater number of twig nodes and greater final stem area than the two 15-cm (6-inch) mulch treatments. Across the mulch treatments, there was no significant difference in wound occlusion nor stem discoloration cross-sectionally or vertically, in contrast to the findings reported by Watson (2014b).

Limitations of Our Review

Because we excluded non-English studies, this review may exclude additional studies on the impacts of deep mulch applications on trees. However, because the included studies are conflicting, we are accurately presenting the existing dichotomy. One or two additional non-English studies would further emphasize this existing dichotomy, rather than providing support for a clear maximum mulch depth. Similarly, we conducted our search in 5 databases and 11 journals, following the search protocols of Martin and Conway (2025), which includes searching the commonly read journals of arborists and urban foresters (Martin et al. 2024). This approach therefore identifies relevant studies in non-indexed journals like Arboriculture & Urban Forestry (Hauer and Koeser 2018). This can introduce bias into the results as it may exclude studies in non-indexed journals outside of the 11 journals that we searched. However, backward citation searching helped address this bias by identifying literature cited by studies that we found through our search.

While our review examined peer-reviewed literature, we have identified two unpublished studies in the Discussion (Giblin 2013; Watson 2014b). We have also excluded articles published in informally reviewed trade magazines like Arborist News. Many such articles, including those in the Introduction, attempt to synthesize practitioner observations or scientific literature (although often without citations). These articles are not review papers nor research studies themselves. While both unpublished or informally reviewed studies can contribute to the development of best practices, they have not undergone the rigour of the peer review process and are often remiss of the required methodological elements of a publishable scientific study (Gastel and Day 2016).

Future Research and Recommended Study Design

While studies have identified the minimum mulching depth required to improve soil health, plant growth, and weed control (Chalker-Scott 2007), there is no clear maximum depth from peer-reviewed literature. Further research should seek to identify maxima based on mulch type. We recommend that future studies consider 5 items in their research design and reporting.

First, researchers must report the type of mulch used (e.g., pine bark mulch), the time that it has been dried (e.g., 3 months of drying before application), and its physical and chemical properties when applied and at the time of excavation. Sun et al. (2023) provide a good table of the properties of their mulch.

Second, researchers should consider the biophysical factors beyond the tree and mulch. The soil properties, weather and climate data, and use of supplemental irrigation should be included in the study. Physical and chemical properties of the soil influence tree health, root defects, and moisture retention. The study area’s climatic norms and weather data for the duration of the study are also important for interpreting the results for other climatic zones.

Third, researchers must include both a control group (no mulch) and several depths of mulch (e.g., 10, 20, and 30 cm [4, 8 and 12 inches] of mulch). The depths should include depths both within and above the recommended mulch depths from best management practices and standards, often 5 to 10 cm (2 to 4 inches).

Fourth, researchers should consider the width and slope of the mulch ring and its proximity to the trunk. These are additional factors that may be cofounding variables. A deep, widely mulched ring may cause more girdling. Mulch that is piled with a high slope may cause rainwater to run away from the trunk instead of infiltrate through the mulch pile, although there are no research studies that introduce these concepts. The proximity of mulch to the trunk may impact occlusion and growth (Giblin 2013).

Fifth, researchers should note the planting depth of the trees. The previous findings cited in our Discussion that highlight the impact of planting depth (e.g., Wells et al. 2006; Day and Harris 2008; Hauer and Johnson 2021) indicate that this could be a confounding factor in root encircling or constricting the trunk.

In addition to these 5 research design and reporting recommendations, it is also important that researchers published non-significant results. The finding that there is no difference between conventional and deep mulching is, in and of itself, a substantial result and one that will be beneficial for practitioners.

Conclusion

While there are many benefits to applying mulch around the base of urban trees, practitioners and extension specialists often warn about the deleterious impacts of excessive mulch depth. It is commonly believed that mulch exceeding 10 cm (4 inches) can lead to girdling or encircling roots and poor tree growth. Our literature review identified 11 studies that tested the impact of mulch depth on tree physiology, as well as soil physiology, water conservation, pathogens, and weeds. Due to differences in study methods and tested variables, the impact of deep mulch remains inconclusive though there is evidence that mulch depth over 10 cm (4 inches) can have negative impacts on growth in some instances. Differences in biophysical and climatic factors, tree species, and mulch type and porosity explain many of the observed differences in outcomes, resulting in some researchers finding greater mulch depth beneficial while others report negative impacts. At present, there is no clear maximum mulch depth that leads to declining tree condition. We emphasize the need for future research that focuses on mulch depth and tests physiological variables of trees. Future research should follow the 5 research and design recommendations for ensuring quality studies as presented in this review. By expanding the literature on mulch depth, extension papers, educational materials, and industry BMPs and standards can better reflect scientific findings.

Conflicts of Interest

The authors reported no conflicts of interest.

Acknowledgements

Thank you to Anna Heckman for her help in searching for archival resources. Thank you to Lindsey Mitchell of the International Society of Arboriculture (ISA) for providing scanned copies of Arborist News from the ISA archives.

Urban Tree Diversity: Key Lessons from the UTD5 Conference

Thu, 01 Jan 2026 11:00:00 GMT

Urban forests and trees constitute a critical component of sustainable urban ecosystems, offering ecological, social, and climatic benefits that are increasingly indispensable in the context of global environmental change. As cities confront the compounded challenges of climate change, extreme weather events, population growth, and urban densification, the strategic enhancement of tree diversity emerges as a fundamental approach to strengthening resilience and ensuring long-term urban livability.

The 5th International Urban Tree Diversity Conference (UTD5), convened in Madrid, Spain, in October 2024, served as a pivotal forum for advancing this agenda. Building upon the legacy of previous meetings in Sweden, Australia, Canada, and the United States, UTD5 assembled 150 participants from 26 countries, encompassing researchers, practitioners, and policymakers. The scientific program comprised 8 keynote lectures, 44 oral presentations, a round table, 16 posters, and 3 technical visits, fostering interdisciplinary dialogue and knowledge exchange across multiple domains of urban forestry and arboriculture.

The thematic structure of UTD5 reflected the multifaceted nature of urban tree diversity and its role in addressing contemporary urban challenges. Contributions were organized around four principal topics:

Using Tree Diversity to Mitigate Risks from a Changing Climate

Urban forests represent a nature-based solution for climate adaptation and risk mitigation. Presentations examined strategies such as species diversification and integrated green infrastructure to enhance resilience against rising temperatures and extreme weather events, emphasizing the ecological and functional benefits of diverse urban tree populations. The papers included in this section provide empirical evidence and conceptual frameworks for implementing tree diversity as a core strategy for climate resilience in urban environments.

Novel Approaches to Urban Tree Selection

This theme explored innovative methodologies for selecting species suited to complex urban environments. Case studies illustrated the application of advanced data analytics, predictive modeling, and biodiversity principles to optimize tree selection processes, thereby ensuring ecological stability and aesthetic quality under future climatic scenarios. Selected contributions demonstrated how these approaches can be operationalized to guide species choice and planting design in diverse urban contexts.

Emerging Technologies and Their Role in Urban Forest Management

Technological innovation is reshaping urban forestry practices. Contributions highlighted the integration of remote sensing, drone-based monitoring, and data-driven decision-support systems to improve diagnostic accuracy, facilitate early detection of pests and diseases, and optimize resource allocation for sustainable management. The papers featured here showcase practical applications of these technologies and evaluate their potential to transform urban forest governance and maintenance.

Harnessing Open Data and Citizen Engagement to Support Urban Forestry

Effective governance of urban forests increasingly relies on participatory approaches and transparent information systems. Presentations addressed the potential of open-data platforms and citizen-science initiatives to strengthen collaborative management, enhance public awareness, and foster shared responsibility for urban tree diversity. Research included in this section illustrates how participatory frameworks and digital tools can bridge gaps between scientific knowledge and community action, reinforcing inclusive urban forestry strategies.

The three papers selected for this special section, all from North America, exemplify the breadth and depth of research presented at UTD5. Collectively, they advance theoretical frameworks and practical applications that underscore the importance of interdisciplinary collaboration in shaping resilient, biodiverse urban landscapes. By integrating ecological science, technological innovation, and participatory governance, these contributions offer critical insights for the future of urban forestry and arboriculture. They all offer some specific recommendations as well for enhancing urban forest and urban tree diversity.

Sara Barron and colleagues investigate, with the assistance of student researchers, the role of university courtyard areas as diverse and inclusive areas for promoting social connection and wellbeing. The contribution by Andrew Koeser et al. explores ways of enhancing urban tree diversity, quality, and abundance in the Chesapeake Bay Watershed of the United States. The paper by Philip John Potyondy, finally, discusses street tree planting guidelines in the city of Minneapolis from a tree diversity enhancement perspective.

Conflicts of Interest

The authors reported no conflicts of interest.

Minneapolis Street Tree Planting Guidelines to Increase Diversity

Thu, 01 Jan 2026 11:00:00 GMT

Abstract

The threat and eventual loss of ash trees (Fraxinus) to emerald ash borer (Agrilus planipennis) in Minneapolis, MN, USA, was a major opportunity to establish a more diverse and resilient public street tree population. This generational opportunity was embraced. With the goal of increasing urban forest resiliency against future pests and conditions, Minneapolis developed and applied multiscaled tree selection guidelines to systematically select and plant a diverse mix of trees. Within a relatively short amount of urban forest time, the diversity of the Minneapolis public street tree population greatly increased. Within the past two decades, the number of genera that make up 1% or more of the public street tree population has nearly doubled. Before the guidelines, maple (Acer) comprised 30% of the public street tree population. Currently, there are no genera that comprise 20% or more of the public street tree population in Minneapolis. As a result of the guidelines, there is more diversity across the whole city, within neighborhoods, and along individual street block segments. The benefits of this diversification will hopefully lessen the exposure to and impact of future urban forest pests and other challenging conditions.

The myriads of benefits from urban trees have been well documented ( Tyrväinen et al. 2005 ; Turner-Skoff and Cavender 2019 ). Cost benefit analysis ( Song et al. 2018 ) and the need for targeted site design have also been shared ( Pataki et al. 2021 ). Concerns have been raised about the potential insect and disease impact from exotic pests to the Minneapolis urban forest ( Nowak et al. 2006 ). The potential costs and budgetary impacts of emerald ash borer (Agrilus planipennis) have been estimated ( Kovacs et al. 2010 ; Hauer and Peterson 2017 ). Management challenges, tactics, and options to address emerald ash borer infestation have also been investigated ( McKenney and Pedlar 2012 ; Flower et al. 2015 ; McCullough 2020 ). The often referred to ‘rule of thumb’ urban forest diversity guidance credited to Frank S. Santamour, Jr. recommends having no more than 10% of a single species, 20% of a genus, and 30% of a family ( Santamour 1990 ). Implementation of this 10/20/30 rule has been credited with increasing diversity for the sake of resilience across urban forests and has also been a topic of discussion and analysis ( Kendal et al. 2014 ). A stringent 5% limit of a genus has more recently been proposed to reduce the impact of urban forest threats ( Ball 2015 ). In the face of the impending impact of emerald ash borer to the urban forest, Minneapolis developed and implemented public street tree diversity guidelines to increase resilience to future threats at the city, neighborhood, and street block segment scales.

Similar to many cities across the Eastern United States prior to 1970, Minneapolis’ street tree population was primarily comprised of elm (Ulmus). Following significant public street tree losses resulting from Dutch elm disease (caused by Ophiostoma novoulmi), Minneapolis engaged community groups across the city and developed the “Minneapolis Neighborhood Boulevard Reforestation Plan” in 1978 ( Sand et al. 1978 ). The plan assigned a tree species to every street segment across Minneapolis on a block-by-block basis. The plan was organized by neighborhood. There are currently 87 neighborhoods across Minneapolis. This plan aimed to make a tenfold increase in the diversity of the Minneapolis public street tree population ( Figure 1 ).

Figure 1.

Chart of Minneapolis public street tree percent by genus over multiple years. Pre-1970 was reported in the “Minneapolis Neighborhood Boulevard Reforestation Plan” (Sand et al. 1978). The 1978 data was the intended population from the “Minneapolis Neighborhood Boulevard Reforestation Plan.” The 2004 data is from an i-Tree analysis. The data from 2010 to 2024 is from the Minneapolis Park & Recreation Board public tree inventory system.

Data was collected in 2004 for a sampling study of the Minneapolis public street tree population using protocols for STRATUM (Street Tree Resource Analysis Tool for Urban Forest Managers), which eventually became i-Tree Streets (McPherson et al. 2005). Results from this USDA Forest Service analysis confirm that the reforestation plan from 1978 had been successfully realized by 2004 (Figure 1). In terms of susceptibility of the Minneapolis public street tree population to a single urban forest pest, the 2004 forest was in much better shape than the forest pre-1970. When Dutch elm disease arrived in Minneapolis in the 1960s and 1970s, over 90% of the public street tree population was susceptible to the devastating disease. By 2004, only 17% of the public street tree population was susceptible to the soon to arrive emerald ash borer. Although a potential 17% city scale loss was a much more operationally achievable outlook, a full street block of all susceptible trees was still daunting and alarming from the perspective of an individual property. Emerald ash borer was confirmed in Minneapolis in 2010. By this time, Minneapolis had also completed a tree inventory census of public trees and started using an online tree inventory system. Tree inventory census data from 2010 was very similar to the sampled results from the 2004 analysis, with ash comprising 18% of the public street tree population (Figure 1).

Many cities in North America have been impacted by emerald ash borer with each city striving to provide the best preparation and care for the forests in communities they serve. There were cities that were unfortunately taken by surprise close to “ground zero.” Many communities formed plans to treat a subset of their public ash trees to “buy time” while gradually replacing nontreated ash trees. Other cities launched plans to treat all their public ash trees through routine maintenance investments to perpetuate tree benefits as long as possible. At the same time, cities like Minneapolis enacted plans to gradually and systematically replace public ash trees while allowing individuals, block clubs, and neighborhood organizations to treat ash trees in alignment with their local goals and budgets. The Ash Canopy Replacement Plan in Minneapolis was formed following many considerations and stakeholder engagement, including: assessment of the level of infestation; the perceived rate of spread; long-term maintenance costs; scenario analysis; community pesticide concerns; neighborhood engagement; elected official input; advisory commission involvement; research partnership; and agency consultation.

In 2014, the Minneapolis Park & Recreation Board launched a plan to systematically replace 40,000 public ash trees (30,000 street trees and 10,000 park trees) over an 8-year period. Emerald ash borer infested trees were removed and replaced as the highest priority. Proactively nonsymptomatic public ash trees were also gradually replaced. This was facilitated by selecting a small number of ash trees from each block and park across the city to be replaced over all 8 years of the plan. Said another way, full street block segments of nonsymptomatic ash trees were not removed at the same time.

Public trees in Minneapolis are selected and planted by Minneapolis Park & Recreation Board staff arborists and foresters. From 2004 to 2013, Minneapolis planted an average of about 4,500 public trees annually. Trees were purchased with general operating funds from property taxes. From 2014 to 2022, Minneapolis planted an average of about 9,000 public trees annually. The annual cost to purchase trees was about $1 million. Additional funds, beyond general operating funds, were secured via a Tree Preservation and Reforestation Levy which enabled the Minneapolis Park & Recreation Board to facilitate its Ash Canopy Replacement Plan. Besides removing ash trees, the plan doubled Minneapolis’s planting capacity. Since 2022, Minneapolis has been planting more than 9,000 trees per year; however, the Tree Preservation and Reforestation Levy has not continued as a funding source. Finding the balance of funds to maintain this level of public tree planting has become challenging and less certain. In 2023 and 2024, instead of coming from the Tree Preservation and Reforestation Levy, half of Minneapolis’s tree purchase funds came from the American Rescue Plan Act, a federal economic stimulus bill passed in 2021 to aid in the recovery from the COVID-19 pandemic. In 2025, with still only half of the needed funds coming from general operating funds, the other half of the tree purchase budget is coming from a combination of two grants, a philanthropic gift, and revenue from the sale of carbon credits. Since 2021, Minneapolis has annually registered tree planting carbon offset projects with the national nonprofit carbon registry and certification organization, City Forest Credits, via a public-private partnership with the project operator, Green Cities Accord, to help fund future tree planting and maintenance.

In Minneapolis, public tree planting stock is sourced through an annual bid process where each variation in cultivar, root type, and size is individually bid. For the past decade, there have been approximately 200 individually bid line items on the annual bid document. Minneapolis purchases trees from about a dozen different nursey vendors to achieve the desired quantity in each category. Minneapolis has not utilized contract growing as a means to acquire a desired level of diversification. Thus far, Minneapolis has been utilizing the annual bid listing to describe and signal to growers what the public tree planting needs are in Minneapolis. In addition, the Minneapolis Park & Recreation Board has a strong long-term research and outreach partnership with the University of Minnesota to trial cultivars, stock types, planting methods, and climatic shifts before transitioning into full production.

In the lead up to the ash tree replacement plan, public street tree selection guidelines were developed to increase diversity within a street block and at the neighborhood scale. These guidelines are applied by referencing tree inventory data at multiple scales (city, neighborhood, and street block segment). Species selection is essentially limited based on the existing genera level diversity that exists at each scale.

Starting at the neighborhood scale, the guideline restricts the planting of any genus that comprises 10% or more of the neighborhood public street tree population. As an example, if there is already 12% linden (Tilia) within a neighborhood, then linden is restricted from being planted in that neighborhood. Dynamic reports are generated within Minneapolis’s tree inventory system to guide arborists and foresters on which genera are overrepresented.

There are two guidelines that further restrict tree selection at the street block scale. Tree selections are restricted along a street block segment from any genus that comprises 5 or more trees from a given genus. As an example, if there are already 2 bicolor oaks (Quercus bicolor) and 3 northern pin oaks (Quercus ellipsoidalis) on a block, all oaks (Quercus) are restricted from being planted along that street block segment.

The other street block scale guideline is aimed at achieving resiliency to a potential future infestation from Asian long-horned beetle (Anoplophora glabripennis). Public street tree selections of Asian long-horned beetle preferred host genera include: maple (Acer), buckeye (Aesculus), birch (Betula) and elm (Ulmus). These are restricted if there are already a combined 5 trees from any of those preferred host genera along a block segment. As an example, if there are already 4 maples and 1 elm along a street block segment, then maple and elm along with birch and buckeye are all restricted from being planted on that street block segment.

Guideline compliance and quality control has been annually assessed to ensure compliance and to determine ongoing arborist and forester training needs and improvements. The guidelines have also been incorporated into Minneapolis standard specification documents to ensure contractors, consultants, and City Divisions comply with the guidelines when planning and building public infrastructure projects within the urban forest.

The Minneapolis public street tree selection guidelines applied at the neighborhood and block scales have increased diversity at the city scale (Figure 1). In 2010, 11 genera made up 1% or more of the public street tree population with 4 genera each comprising more than 10% of the population (maple 28%, ash 18%, linden 16%, elm 12%), combined representing 74% of the population. In 2024, 21 genera made up 1% or more of the public street tree population with only 2 genera each comprising more than 10% of the population (maple 19%, linden 12%), combined representing only 31% of the population. In 2024, no genera made up over 20% of the public street tree population in Minneapolis.

The Minneapolis urban forest is more diverse and more resilient to future pests and conditions as a result of utilizing neighborhood and street block scale tree selection guidelines.

Cities interested in increasing public urban forest street tree diversity could apply the same or similar neighborhood and block scale guidelines to achieve similar increases in diversity. The guidelines could be used as is or further honed to meet a given community’s needs. One way to improve or further customize the guidelines would be to choose an amount more specific than limiting to 10% of a genus by neighborhood. Said another way, the neighborhood guideline or limit could be set individually by genus. For example, instead of limiting all genera to 10% at the neighborhood scale, a community may choose to limit Catalpa to 8% and Malus to 5%, or to limit Quercus to 11%. A community could also choose to set some species or species group limits instead of stopping at genus. As an example, a community could limit red oak group Quercus with a different limit than white oak group Quercus from a need or concern related to the rate of spread of oak wilt (Bretziella fagacearum). Further customizing the block scale guidelines could also be considered to best fit another city.

Neighborhood and street block segment have been meaningful and intuitive scales for applying the guidelines in Minneapolis, however other scales and geographies could also be considered. One problem with a political boundary such as neighborhood is that it could still allow for overpopulation within a vicinity where neighborhood boundaries meet. For instance, Tilia may be overrepresented in an adjacent neighborhood and heavily populated on multiple streets within close proximity to a given planting location and would therefore not be an ideal selection, yet it might still be an allowable selection within the guideline based on where the neighborhood boundaries happen to align. Utilizing simple GIS functions within a tree inventory application, a neighborhood rule could be customized to each planting site based on a radius measurement. For example, instead of using political neighborhood boundaries to designate what genera are overrepresented in an area, a proximity with a set distance of a half mile radius could be used to dynamically determine what genera are already overrepresented within a “neighborhood” proximity to each planting location.

Similarly, street block segment length might not be an ideal scale, considering blocks are not uniform across most cities. The same type of spatial GIS function as described above could be utilized to determine a street segment length as a radius. For example, instead of using a street block segment from one intersection to another, a proximity with a set radius distance of 500 ft (152.4 m) could be dynamically used to determine the best near scale “block” diverse selection at each planting location.

Conversationally, concerns are occasionally raised about block scale diversity reducing a community’s desire for symmetry. Minneapolis has not experienced resistance from community members about lack of symmetry and has very rarely fielded dissent from managers of small highly designed project areas. Minneapolis has also not experienced resident concerns about planting a higher level of diversity. From time-to-time, additional resources, conversations, and education have been found to be helpful to familiarize residents with a tree type that might be new to them. The Minneapolis Tree Advisory Commission has been especially supportive of the Minneapolis Park & Recreation Board’s public street tree diversity accomplishments.

Conflicts of Interest

The author reported no conflicts of interest.

Acknowledgements

This paper is based on a presentation given at the 5th International Conference on Urban Tree Diversity (UTD5), held in Madrid, Spain, 24–25 October 2024. The conference was organized by Arbocity, the Forestry Engineering School from the Technical University of Madrid (UPM), and the Nature Based Solutions Institute (NBSI). The author would like to acknowledge Craig Pinkalla for significant contributions to the development of the Minneapolis street tree planting guidelines. The author would also like to acknowledge Ralph Sievert and the rest of the Minneapolis Park & Recreation Board Forestry Department for embracing and successfully implementing the guidelines.

Increasing Urban Tree Diversity, Quality, and Abundance in the Chesapeake Bay Watershed: Challenges and Opportunities

Thu, 01 Jan 2026 11:00:00 GMT

Background

While many tree professionals recognize the importance of planting quality stock and a diversity of species to enhance longevity and increase urban forest resiliency, the availability of such stock is often limited.

Methods

To address this disconnect, we conducted 3 focus groups with growers, designers, urban foresters, and other technical experts from the Chesapeake Bay watershed region (USA) to identify challenges and opportunities for growing greater numbers of high-quality, underused species.

Results

Contract growing was seen as a key opportunity for increasing quality and diversity. Additionally, increased communication between growers and tree purchasers, as well as potential partnerships with nonprofit or state nurseries, were identified as potential solutions where the marketability of underused species was limited. There were differences among participants regarding their preferences for native species, non-native species, cultivars, and non-cultivars.

Introduction

Tree species diversity is a crucial factor in enhancing urban forest resilience and ecosystem stability, especially in the face of changing climates and the emergence of pests and pathogens ( Morgenroth et al. 2016 ; Huff et al. 2020 ). The overreliance on a single species or a few species in managed urban forests increases the risk of widespread mortality during outbreaks if the dominant species are vulnerable to emerging biotic or abiotic threats ( Laćan and McBride 2008 ). Numerous guidelines for increasing species diversity at the species, genus, and family taxonomic levels have been proposed and promoted among urban forest managers ( Santamour 1990 ; Raupp et al. 2006 ), and history has seen many cases of widespread die-offs of overly abundant urban tree species ( Jernelöv 2017 ; Klooster et al. 2018 ; Bahder et al. 2019 ).

There are approximately 73,300 tree species worldwide, with over 1,800 species found in North America ( Cazzolla Gatti et al. 2022 ). Despite this abundance of diversity, many urban areas rely heavily on a handful of “go-to” tree species. In their survey of urban forest managers across the United States, Ma et al. (2020) found that just 6 species account for the majority (62%) of street and park trees in any given city. Similar results were observed by Cowett and Bassuk (2017) , who compiled tree inventories in the Northeastern United States and found that one species, Acer platanoides, represented over 16% of street trees recorded. Moreover, the Acer genus accounted for nearly 39% of the total population compiled during their search.

Outside the United States, Lohr et al. (2016) observed that a single tree species comprised 20% of the tree population in an average city. In Helsinki, Finland, and Bangkok, Thailand, this figure rose to 40%, with a single species dominating the urban forest inventory. Similarly, Galle et al. (2021) found that the Ulmus genus represented nearly half of the trees in the city center of Amsterdam, Netherlands. In a similar study, Sjöman et al. (2012) reported that the Tilia genus accounted for 13.3% to 46.3% of urban trees across 10 Nordic cities.

Planting a diverse array of tree species is a valuable strategy to mitigate losses from future, unknown threats. Additionally, selecting high-quality nursery stock can enhance the potential longevity of urban trees, regardless of what the future may bring ( Allen et al. 2017 ). Less commonly produced species may not be available in the size or quality desired by tree purchasers, prompting them to choose from more abundant (and potentially overused) alternatives ( Burcham and Lyons 2013 ). In their study of tree planting decisions in Toronto, Canada, Conway and Vander Vecht (2015) observed that species selection was ultimately influenced by the available supply. Quantity often dictated what was planted in park settings, while quality constrained options for residential planting purchases.

These constraints on urban tree selection have been documented and analyzed to improve diversity, quality, and general availability ( Sydnor et al. 2010 ; Burcham and Lyons 2013 ). Hilbert et al. (2023b) outlined the numerous layers of decision-making during production, design, procurement, and installation that filter out potential planting options and collectively reinforce the status quo, where relatively few species dominate planted urban landscapes. Some of these decisions fall outside the typical purview of green industry professionals, such as urban design and planning practices that limit site conditions ( Nitoslawski et al. 2016 ). That said, conflicting goals within the green industry itself often unintentionally restrict tree selection options ( Hilbert et al. 2023b ). For example, urban forest policies that mandate the use of large-caliper stock for mitigation or planting projects generally limit the species that can be procured.

As part of the Chesapeake Bay Watershed Agreement, a strategic goal was set to expand urban tree canopy by 2,400 acres between 2014 and 2025. Partners in this regional agreement have additional goals at local levels, and have experienced challenges procuring trees of the species, size, quality, and/or quantity needed for some projects. In Maryland, which accounts for roughly one-fifth of the land area within the Chesapeake Bay watershed, 5 million trees are set to be planted by 2031, with 500,000 of those being planted in underserved urban areas ( Gilbert 2023 ). Building on past research that has characterized regional tree supply chains and identified key constraints and opportunities, and acknowledging that tree production varies widely across the United States and beyond ( Hilbert et al. 2023b ), this study aimed to explore the challenges and possibilities of establishing a sustainable supply of high-quality, climate-resilient tree species suited for urban areas within the Chesapeake Bay watershed.

Motivated by a growing need for diverse, high quality nursery stock to meet urban forestry planting goals across the Chesapeake Bay watershed region, the research focused on two main questions: first, “What barriers impact the quantity and quality of trees commercially grown in this region?” and second, “What opportunities could enable growers to produce a sufficient number of climate-ready, lesser-used tree species to meet consumer needs?”

While the findings from this study are inherently tied to the region of inquiry, tree production systems worldwide face similar challenges ( Burcham and Lyons 2013 ; Conway and Vander Vecht 2015 ; Avolio et al. 2018 ). As such, the challenges and opportunities identified in this study extend beyond the Chesapeake Bay watershed, contributing to the global discourse on the complex issue of low urban tree diversity.

Methods

Study Area

The Chesapeake Bay watershed is a large and ecologically significant drainage basin spanning approximately 16.6 million hectares (64,000 square miles) across parts of 6 states—Delaware, Maryland, New York, Pennsylvania, Virginia, and West Virginia—as well as the District of Columbia. Land uses vary from rural farms and forests to major East Coast cities including Washington, DC; Richmond, Virginia; Annapolis, and Baltimore, Maryland; and Harrisburg, Pennsylvania. There are 207 jurisdictions (mostly counties) in the watershed, and from 2013 to 2018 only 26 (12.5%) reported net increases in tree canopy cover ( Chesapeake Tree Canopy Network 2025 ). It is the largest estuary in the United States ( NOAA 2021 ) and encompasses both humid subtropical and humid continental climate zones ( Kottek et al. 2006 ). More than 18 million people reside in the region ( Chesapeake Bay Foundation 2025 ).

Participant Selection

We conducted 3 focus group sessions in Spring 2024 with professionals involved in urban tree production, specification, and procurement across the Chesapeake Bay watershed. To ensure a balanced, cross-disciplinary dialogue, we purposively sampled 5 stakeholder categories: wholesale nursery growers, landscape architects, municipal foresters and arborists, representatives from environmental nonprofits and government agencies, and private-sector practitioners (e.g., landscape contractors)( Morgan 1997 ; Femdal and Solbjør 2018 ).

In total, 52 individuals were invited (15 Nursery, 12 Landscape Design, 11 Municipal Arborist/Urban Forester, 10 Nonprofit/Government, 4 Private Industry). Invitations were drawn from prior projects, regional NGOs, and referrals from the Chesapeake Bay Trust and USDA Forest Service Urban & Community Forestry program. We supplemented these lists via snowball sampling whenever a category’s scheduled slots remained unfilled.

Of those invited, 20 (38.5 %) agreed to participate and were scheduled into 1 of 3 sessions:

Focus Group 1 : Scheduled n = 6 (2 Nursery; 0 Landscape Design; 1 Municipal Arborist/Urban Forester; 3 Nonprofit/Government; 0 Private Industry)
Focus Group 2: Scheduled n = 7 (1 Nursery; 0 Landscape Design; 2 Arborist/Urban Forester; 3 Nonprofit/Government; 1 Private Industry)
Focus Group 3: Scheduled n = 7 (3 Nursery; 1 Landscape Design; 1 Municipal Arborist/Urban Forester; 1 Nonprofit/Government; 1 Private Industry)

In total, 17 participants ultimately attended (85% of those scheduled; 32.7% of invitees), with per-category recruitment rates ranging from 8.3 % (Landscape Design) to 60 % (Nonprofit/Government) ( Table 1 ). Final group sizes were as follows: FG1 n = 4, FG2 n = 7, FG3 n = 6. This stratified, multi-stakeholder composition allowed us to explore decision-making around tree production, design, and maintenance from complementary professional perspectives.

Focus Group Logistics

A total of 3 focus groups were held. Each meeting lasted approximately 2 hours, with subsequent meetings spaced 1 to 2 weeks apart to allow time for preliminary analysis and any necessary improvements (clarifying questions, adjusting delivery timing, etc.). All meetings were conducted virtually using an online meeting service (Zoom), which facilitated recording and transcription. A shared virtual whiteboard (Google Jamboard) was used to facilitate discussion by recording and organizing responses to questions in real-time ( Appendix ). Zoom’s chat feature also provided participants with the opportunity to write complex questions or responses.

Focus group discussions began by asking tree growers to identify factors influencing their production decisions and rank these by importance, followed by soliciting reactions from purchasers and planters ( Appendix ). We then reversed this process, asking purchasers and planters about their selection criteria and priorities, with growers providing feedback. Participants were asked to identify preferred tree species for urban areas and explain their choices based on important attributes. We introduced climate change projections for the region (4.5 to 10 °F warming with increased extreme weather events by the 2080s) and asked how these might alter their previous selections. After a short break, the discussion shifted to barriers in the urban tree supply chain from both the purchaser/planter and grower perspectives. The session concluded by exploring potential solutions to these challenges, including industry changes, cross-sector collaboration, and examples of successful tree procurement arrangements. At the end of each focus group, the research team presented a slide summarizing the key discussion points. Participants were asked to review this bulleted list of key challenges and opportunities, making modifications, additions, or deletions as they saw fit.

Data Analysis

The transcripts were automatically downloaded from Zoom and cleaned through a multi-step process. Cleaning the transcripts was an iterative review process, including a full listen-through with pausing and rewinding to ensure accuracy. Consecutive responses from the same speaker were consolidated into a single block of text with the aid of AI (ChatGPT 4) then manually reviewed for errors. During the first review, subtle verbal cues (e.g., sarcasm, emphasis) were annotated by italicizing these words to provide contextual understanding and were revisited during subsequent analysis. Tags were inserted to indicate significant nonverbal actions, such as gestures or laughter, using square brackets. Verbal tics such as “um,” “you know,” and repeated words were removed without altering the original language or grammar. Fragmented or incomplete phrases that did not contribute meaning were also eliminated. Run-on sentences were broken up, and punctuation was added to improve readability. Instances of unintelligible speech were marked with a timestamp for reference.

The finalized transcripts were uploaded into Quirkos (Quirkos Software, Edinburgh, United Kingdom) for qualitative thematic analysis by a primary researcher. A combination of inductive and deductive coding approaches was used. Deductive coding focused on responses that directly addressed the research questions, for example grouping responses into themes such as “factors influencing production.” Inductive coding captured new themes that emerged during the discussions, which were organized as sub-themes (e.g., “customer demand”) within the larger thematic groups (e.g., “factors influencing production”). The analysis was an iterative process, involving multiple rounds of reviewing and refining thematic groupings as patterns emerged.

A second researcher reviewed the transcripts within a word processing software and added comments to code for themes and to add annotations. This was compared with the first round of coding completed by the primary researcher to ensure accuracy and consistency. Additionally, coded themes were compared to the key summary points drafted by the researchers and participants at the end of each focus group. This comparison helped contrast the initial takeaways with insights gained over time and through a more in-depth analysis of the conversations.

Throughout the analysis, a collaborative approach was maintained to ensure consistency in the coding process. Regular discussions between the primary researcher and the secondary researcher helped refine the analysis and thematic organization, validating emerging themes against the raw data. Despite these safeguards, all researchers are most familiar with the tree supply chain from a purchaser’s perspective, which may have influenced our interpretation of certain aspects of the conversations.

Results

Challenges to Enhancing Urban Tree Diversity, Quality, and Availability

Market Demand

Focus group participants indicated that market demand was the primary driver for nursery growers, with inventory shaped by consumer trends in design and home gardening. While growers may have preferred species to sell, customer preferences dominate. As one grower explained, “We’re obviously in the business for profit, so we are trying to predict what the market is gonna demand. We’re at the mercy of the landscape architects, and the contractors and our customers.” Participants explained that to minimize financial risks, growers often focus on species and size classes with consistent sales. Although municipalities sometimes have specific requirements, these clients make up a smaller share of the market, which is largely influenced by commercial developers prioritizing price and availability over species diversity and quality. This leaves little incentive for growers to meet stricter specifications or contracts.

The pressure to produce low-cost trees is intensified by external market forces. Growers face competition from large-scale nurseries outside the Chesapeake Bay region, while closures and liquidation sales add a supply of cheap stock priced unsustainably low. These closures often result from financial struggles or the retirement of nursery owners, with no succession plans in place. This can limit the local availability of high-quality trees.

Seed and Liner Availability

Growers in our focus group highlighted challenges in sourcing seedlings and liners (i.e., small plants grown from a rooted cutting, seedling, plug, or tissue culture plantlet that is grown to a marketable size) for new species. Many nurseries rely on purchasing liners from a small group of large-scale producers in Tennessee and the Western USA, which can lead to supply bottlenecks. One grower mentioned struggling for 5 years to purchase 400 liners of a preferred species, only receiving a tenth of his request annually. Additionally, the centralization of liner production makes it harder to obtain native species with local provenance, which are often preferred or required by municipal ordinances. As one grower noted, “It is because of the liners…we’re having major issues in the pipeline with trying to get trees.”

Funding vs. Production Cycles

Our focus groups highlighted the mismatch between the multi-year growth cycle of trees and the shorterterm nature of most urban forestry funding sources, which presents a significant challenge in shaping the market through tree purchases. Growing trees from seedlings to a size suitable for planting, especially larger caliper stock for urban areas, takes several years. However, funding cycles driven by local government budgets and planting grants are often annual or short-term, with even long-term grants typically lasting only 3 years, which is not enough time to impact nursery planting decisions. As one nonprofit expert noted, “To get what we need, the species we need, and the quality we need, we have to start 5 years in advance.”

Participants further explained that the uncertainty of future funding from grants and budgets makes it difficult to commit to long-term tree production contracts or large-scale planting projects. Organizations often face pressure to meet short deadlines for grants, which may not align with optimal planting seasons or tree availability. Bureaucratic delays in fund distribution further complicate meeting seasonal planting windows. Additionally, the inconsistency of grant funding limits the ability to influence tree nursery production, as there’s no guarantee of future resources to support changes in tree availability or species diversity, creating a cycle that hinders long-term market impact.

Technical Staff

Participants expressed a challenge not previously highlighted in earlier focus groups (Koeser et al. 2022; Hilbert et al. 2023a), which is the shortage of qualified technical staff who understand tree quality and species diversity. Municipalities and nonprofits often struggle to develop and maintain the capacity for effective tree management, with non-competitive pay scales exacerbating the issue by hindering the recruitment and retention of skilled professionals. Participants stressed the importance of these technical roles in ensuring the survival of newly planted trees, with one respondent from the nonprofit sector emphasizing, “[Regarding our earlier discussion of technical expertise] I really wanted to hit home that those positions need to be highly valued.”

Opportunities for Enhancing Urban Tree Diversity, Quality, and Availability

Partnerships

Partnerships with growers were identified as a key strategy for improving tree availability and quality in urban forestry. Participants stressed the value of closer relationships between end-users and nurseries to better align production with demand. These collaborations range from informal arrangements, where foresters provide advance notice of needs, to structured contract growing agreements. Some organizations have successfully partnered with nurseries by placing advance orders or using brokers to tag trees meeting quality standards. As one arborist participant noted, “Relationships help you smooth all these little bumps out.” Collaborations with industry, NGOs, and government entities were also seen as critical. Partnerships with universities, botanical gardens, zoos, and local agencies expand resources and expertise. One respondent shared efforts to work with universities to propagate specific species, while another described providing temporary storage for local groups to hold stock before planting.

Contract Growing

Contract growing emerged as a key solution for addressing tree availability challenges in urban forestry. It offers advantages like better alignment between grower production and end-user needs, securing specific species and qualities, and potential cost savings. It also minimizes risks for growers producing new or specialized trees—a point highlighted by nursery operators. One grower explained:

“What would be most helpful for growers is more of a partnership with end users to find out what’s in the pipeline, what they’re going to need. Ideally, even some sort of contract grow situation where sizes, specs, and varieties are listed. Otherwise, as growers, we’re taking the risk of trying to grow something we think will sell.”

Some organizations have successfully used short-term contracts for smaller materials, particularly for large projects, pre-funded initiatives, and hard-to-find or native species.

However, longer-term contracts for larger trees pose challenges. Participants noted the 5- to 7-year commitment required for urban trees brings risks, including business uncertainties and natural factors like weather or pests. As one grower shared, “From a grower’s perspective, it scares me a little, committing to grow these trees, and then something happens—weather, nature, [pests], or whatever.” Additional obstacles to contracts include short grant windows, funding uncertainties, and the lengthy growth cycle of trees. Despite these concerns, contract growing was recognized for its potential to improve tree availability and quality while reducing growers’ financial risks. Participants indicated that stable, long-term funding and careful planning are critical for expanding these arrangements, especially for larger urban trees.

State Nurseries

As mentioned in the challenges section, many nurseries noted they outsource the propagation of seeds and cuttings to companies specializing in seedling or liner production. Respondents highlighted the underfunding of some state nurseries in recent years, with one arborist noting, “[the] Virginia State Forest Service, its nurseries have been up and down”; and a nonprofit expert sharing, “We go with [the] State of Maryland State Nursery because we got rid of our Delaware nursery years ago.”

Respondents saw an opportunity to reinvest in these facilities. Such reinvestment might enable state nurseries to produce seedlings and liners for desirable tree species that are currently considered unprofitable by commercial nurseries. Additionally, these nurseries may focus on cultivating local provenances of broadly distributed species and offer more detailed information about seed sources than what is typically available from commercial stock.

Nonprofit Nurseries

Several participants described their experiences or interest in creating small-scale, in-house nurseries to grow hard-to-find tree species. These efforts often focused on producing smaller-sized stock, such as in propagation gardens or nurseries yielding a few thousand trees annually. Goals included cultivating rare species, reforesting former agricultural lands, and advancing local conservation initiatives. Approaches ranged from starting plants from locally collected seeds to purchasing and raising liner stock. This strategy enables organizations to combine self-grown stock with purchases from established nurseries, offering greater control over species selection and local adaptation.

Enhancing Communication Among Industry Associations

Improved communication between tree producers and buyers was seen by participants as being vital for addressing nursery tree availability. Trade shows, particularly the Mid-Atlantic Nursery Trade Show (MANTS), were highlighted as key networking opportunities, with one participant describing MANTS as “heaven on earth” for industry connections. Another grower echoed this in saying, “The state nursery organizations are probably the best place to go.” These events provide a platform for face-to-face meetings, relationship-building, and staying informed about market demands. State-level organizations, such as the Pennsylvania Landscape Nurserymen’s Association and the Maryland Nursery and Greenhouse Association, were also noted as critical resources for tackling industry challenges and fostering local connections.

While designers, arborists, and urban foresters benefit from networking and education opportunities within their own organizations, participants expressed a need for greater collaboration across green industry sectors. Expanding communication beyond these groups can address shared challenges and ensure a more coordinated approach to improving tree availability and quality.

Creation of a Centralized Plant Finder Database

Participants highlighted the need for nursery stock lists or some sort of centralized database to address challenges in finding up-to-date information on nursery tree availability. They envisioned a system where users could search for trees within a specific radius, with details on sizes and quantities, streamlining the buying process. As one nonprofit purchaser noted, “...it’d be great to have some central database where you can see…what is available from the nurseries… Who has what? Of what size? It would make buying trees a lot easier.” However, concerns were raised about maintaining accurate, real-time updates, especially for smaller nurseries that update inventories infrequently. Growers indicated that nurseries vary widely in their use of technology, with some adopting online ordering and commercial plant-finder apps, while others rely on traditional methods. This inconsistency in inventory management may pose significant obstacles to creating a unified database.

Other Insights—The Use of Natives and Cultivars

Many respondents expressed a preference for native tree species, with some exclusively purchasing or using them. One participant said that within their area, “all the jurisdictions are trying to get 90%, if not 100%, native trees.” One participating grower noted that their retail garden business is receiving increased demand from customers. However, the preference for natives was not universal. Some respondents continue to use non-native species, recognizing their value when they are non-invasive and well-suited to urban environments. Relying solely on native species presents challenges, including limited variety and difficulty sourcing certain trees. As one municipal arborist explained, “We have about a list of 12 trees that we plant, [laughs] and it seems to be because we’re limiting ourselves to purely native trees.” This approach reflects the absence of many native species in nursery production, particularly in larger sizes, rather than a lack of local tree diversity in the Chesapeake Bay area.

Focus group participants explained that cultivars can offer benefits such as improved drought tolerance, winter hardiness, consistency in production, and disease resistance, like with American elm (Ulmus americana) cultivars. However, concerns about limited genetic diversity arise due to cloning and the overuse of similar genetic material, as pointed out by some participants. The market for cultivars is largely demand-driven, with commercial services often prioritizing them over straight species. As one grower mentioned, “It’s demand driven, and maybe it’ll change, but right now what’s crossing my desk [are requests for] mostly ‘Red Sunset’ maples and ‘October Glory’ maples.” This preference presents a challenge for nurseries balancing ecological considerations, as straight species are favored for restoration projects, and genetic diversity is crucial for ecosystem health. Additionally, producing cultivars in urban environments typically requires nurseries to balance the availability of disease-resistant varieties with site constraints.

Discussion

As observed in past focus groups conducted in Florida ( Hilbert et al. 2023a ) and across the United States ( Koeser et al. 2022 ), economic factors drive growers’ decisions about which species to produce. Growers are primarily concerned with producing trees they have confidence they can sell. Moreover, Hilbert et al. (2023a) highlighted that the time required for a tree to reach a marketable size significantly impacts production costs. Slower-growing trees, although often ideal for compact urban planting sites, take longer to reach the larger sizes commonly specified in commercial and municipal contracts. This creates market pressure favoring the production of faster-growing, weedier tree species. Market pressure also comes in the form of supply chain disruptions and limitations ( GoMaterials 2021 ; Koeser et al. 2022 ). Growers in this study shared the challenge of relying on large-scale liner producers that drive the tree palette for nurseries further down the supply chain. They also pointed out the challenge of being out-competed by low prices from nursery liquidation sales, something not found in similar studies of urban tree supply and likely exacerbated by major market disruptors like COVID-19 and extreme weather events ( Sallin 2021 ).

Another challenge brought up in our groups that has not been discussed in similar urban tree supply chain studies is the lack of technical staff. This staffing gap can impact all areas of urban tree projects, from planning to long-term care ( Roman et al. 2013 ; Hargrave et al. 2023 ). Having staff who understand nursery tree quality standards and procurement processes, and are able to build long-term relationships with growers is essential to getting desired trees in the ground.

Contract growing has been discussed in multiple focus groups involving both growers and urban tree purchasers ( Koeser et al. 2022 ; Hilbert et al. 2023a ). Longer-term contracts provide additional stability, which is valued by growers producing long-term crops while navigating year-to-year fluctuations in the housing market or shifts in design trends. They also provide tree purchasers a communication pathway for specifying the species, sizes, and quality they desire ( Stephens 2010 ). The mismatch between the multi-year growth cycle of trees and the shorter-term nature of most urban forestry funding sources can create a challenge to contract growing, something also discussed in similar focus group work ( Koeser et al. 2022 ; Hilbert et al. 2023a ). Urban forestry programs have successfully engaged in contact growing in the USA through implementing it as part of large-scale planting projects, such as with New York City Parks ( Stephens 2010 ); through regional initiatives coordinated by nonprofits like Chicago Region Trees Initiative (T. Brannen, personal communication) or through intergovernmental cooperative purchasing agreements as with the Suburban Tree Consortium ( West Central Municipal Conference 2020 ). While contracts were frequently mentioned in our discussions, participants noted that not every agreement needed to be formal. Establishing long-term business relationships and fostering a shared history were often sufficient for many growers to adjust their production to include preferred species for repeat clients.

Increased communication has also been a common theme in past focus group research ( Koeser et al. 2022 ; Hilbert et al. 2023a ). This includes informal exchanges between growers and purchasers that go beyond what is immediately necessary for one-time transactions. Focus group participants often express appreciation for their involvement in the research process, as it provides an opportunity to sit down and engage with other green industry professionals outside their immediate field. Attending and presenting at conferences beyond those directly relevant to one’s work can also be an effective way to engage new audiences and networks across sectors.

A solution discussed by participants that has not come up in similar research is the role of state nurseries. There are 38 state, territorial and tribal nurseries in the USA that primarily focus on growing native, locally adapted species for reforestation efforts ( USDA Forest Service 2025 ). State-operated nurseries could reinvest resources to grow seedlings and liners of urban-suitable species that private growers currently deem unprofitable. They could also prioritize local provenances of widely planted trees and provide far more detailed provenance information than is usually offered by commercial suppliers.

Although it may seem that the issue of low urban tree diversity is well understood and addressed by green industry professionals, actual diversity observed in urban landscapes tells a different story ( Ma et al. 2020 ). Past research underscores the need for continued efforts to educate and engage those involved in growing, specifying, planting, and caring for urban trees ( Polakowski et al. 2011 ; Lohr 2013 ). This includes fields often omitted from discussions, such as civil engineering, which significantly contribute to urban reforestation but are less likely to specify diverse planting palettes ( Thompson et al. 2021 ). While urban tree supply chains are complex and slow to change, research projects like the one we present here provide a range of actionable opportunities that can be championed at the local level.

Future research may examine the barriers and solutions for local governments seeking to implement grower agreements. Several cities have successfully navigated budget and purchasing office processes to establish long-term agreements with local nurseries. However, many others perceive this as an unattainable ideal. Interviews with successful cases could highlight key strategies or indicators that a community may be receptive to such arrangements.

Similarly, future research may explore how non-profit and state nurseries can address gaps in existing supply chains. This is a potentially sensitive issue as there may be tension between commercial nurseries and state-run operations in the marketplace. That said, certain species and nursery products (e.g., liners and finished trees) currently lack a viable market. State and nonprofit nurseries could play a crucial role in filling these gaps, helping to establish demand and allowing for commercial operations to eventually scale up production.

Conclusion

The urban tree nursery industry faces challenges that limit tree diversity and quality, largely driven by market demand and external price competition. The limited availability of seedlings and liners, coupled with the mismatch between long-term tree growth cycles and short-term funding, further complicates efforts to meet the needs of urban forestry. Additionally, the shortage of qualified technical staff in municipalities and nonprofits exacerbates these issues. However, there are opportunities to address these challenges. Contract growing offers a promising solution to align nursery production with the specific needs of end-users, reducing risks associated with producing new or uncommon species. Collaborative partnerships between growers, municipalities, nonprofits, and research institutions could enhance the availability and quality of trees, particularly those that are underrepresented in commercial production but are essential for urban environments. Furthermore, the development of a centralized plant finder database and improved communication through industry events could streamline the tree-buying process, although implementing these solutions presents its own set of challenges. Overall, a combination of strategic partnerships, improved communication, and innovative growing models can help overcome current barriers and ensure the sustainable growth of urban forests.

Conflicts of Interest

The authors reported no conflicts of interest.

Acknowledgements

This paper is based on a presentation given at the 5th International Conference on Urban Tree Diversity (UTD5), held in Madrid, Spain, 24–25 October 2024. The conference was organized by Arbocity, the Forestry Engineering School from the Technical University of Madrid (UPM), and the Nature Based Solutions Institute (NBSI). This work was funded as the project “Scope of Work 4: Addressing Regional Tree Supply Challenges and Opportunities” by the Chesapeake Bay Trust using funding from the United States Environmental Protection Agency. The findings and conclusions in this publication are those of the authors and should not be construed to represent any official USDA or US Government determination or policy. Additional insights can be found in an expanded report titled, “Addressing Regional Tree Supply Challenges and Opportunities: A Rapid Assessment” (Koeser et al. 2024).