In the College of Engineering, students and faculty are using data science to develop groundbreaking tools and tackle real-world challenges. Through the ongoing series, “Data for Good: Innovators in Action,” we highlight how these efforts are driving meaningful change across various societal issues. Join us in exploring these stories and discover how you can be a part of the movement to create a brighter, more equitable world.
Cal Poly will lead the creation of California’s new strategic plan for urban forestry, a major initiative in collaboration with Cal Fire to expand tree canopy and foster healthier, greener communities statewide.
The strategic plan, mandated by Assembly Bill 2251, will aim to increase California’s canopy cover by 10% by 2035. The effort will prioritize equitable access to urban forestry benefits, promote biodiversity, and address pressing challenges like climate resilience and air quality.
“Guiding this strategic plan gives Cal Poly students a direct path to using their data to solve real-world problems,” said biology Professor Jenn Yost, lead on the project. “We have the strongest datasets for tree mapping of anyone.”
With over 8 million data points powering its neural network, Cal Poly established itself as a central authority in urban tree information. Expanding the state’s tree canopy promises myriad benefits — from capturing carbon and reducing stormwater runoff to supporting wildlife, enhancing mental health and even lowering crime rates.
At its core, the interdisciplinary initiative reflects the groundbreaking power of data and collaboration. Biologists, computer scientists and social scientists are combining their expertise to craft policies and solutions that will drive significant, lasting change across California.
“We have computer science students learning botany and biologists learning to code,” said Jonathan Ventura, a computer science professor who helped map 40 million trees across California.
Under the leadership of Cal Poly’s Urban Forest Ecosystems Institute, this effort blends advanced technology with environmental stewardship to reimagine cityscapes and inspire a new generation of environmentally conscious engineers and scientists.
Before they could assess canopy cover, Cal Poly’s Urban Forest Ecosystems Institute (UFEI) team needed to answer a basic question: How many trees are there in California?
Mapping millions of trees was essential to UFEI’s mission to sustain and enhance green spaces. Initially a biology-focused effort, the institute soon brought in experts from computer science, social science, economics and geography — creating a broad approach to understanding urban forests.
Biology Professors Yost and Matt Ritter began by gathering data from major tree companies that had mapped municipal streets. Yet, with public trees accounting for only 20% of the urban forest, the team needed a fuller view.
“Why not just put AI magic to work?” suggested Ventura, who collaborates with Yost, Ritter and geography Professor Andrew Fricker — an expert in GIS and AI model refinement. “It seemed reasonable, but identifying individual trees from aerial images turned out to be harder than we expected.”
Cami Pawlak knows that all too well.
The biology master’s student who joined UFEI for her graduate work in urban forestry, coordinated what Ventura called his “undergrad army” — students tasked with manually identifying 100,000 trees using aerial imagery, a process that spanned two years.
“I spent my fair share of time putting points on trees,” said Pawlak, who hails from Playa Vista, California.
Among the team’s biggest challenges was deciding what counted as a tree. “What is a tree? What is a bush? Those were the kinds of debates we had,” Pawlak recalled, laughing at the unexpected conversations.
Ultimately, they marked tall, green trees until reaching 100,000 points, laying the groundwork for training their neural network. Running over two days, it mapped 40 million trees across urban areas statewide — a feat that would have taken lifetimes by hand.
With their mapping model established, the UFEI team next set out to understand the diversity within California’s urban forests.
Drawing from data collected by major tree companies, they compiled a statewide inventory of over 7.8 million trees, representing 552 species. This inventory offered an unprecedented view of California’s city forests, revealing species diversity by county and city and enabling exploration of foliage types, climate zones and water needs — all indicators of urban forest health.
Biodiversity, they knew, was crucial for resilience. “If you plant only one kind of tree and that species gets a disease, every one of those trees is at risk. More diverse urban forests can withstand threats better,” Yost explained.
Gathering all this information in one place was an undertaking that had never been attempted on this scale until UFEI took on the challenge.
Key to the effort was Daniel Gonzalez, a computer science graduate student from Lathrop, California, who played a pivotal role in building UFEI’s digital infrastructure. Gonzalez focused on improving SelecTree, an online tool that helps users choose trees based on various criteria. While functional, the tool often broke down, and Gonzalez was determined to improve its stability and reliability.
SelecTree already had a steady stream of users, from academics to nurseries to Cal Fire officials, but Gonzalez was stunned to discover 25,000 monthly visitors. Yost believes it’s the most-used tree website in the country.
As he enhanced SelecTree, Gonzalez mentored computer science undergraduates in cloud tools and code reviews, while also learning about California’s urban forests himself — including, for example, that the crape myrtle is the state’s most popular tree. “The magic is in the data,” he said, emphasizing the insights they had uncovered.
Today, SelecTree and all the UFEI tools are more reliable than ever, and Gonzalez, now a senior software engineer at Cisco, couldn’t be prouder.
The UFEI team’s mission expanded from mapping trees to measuring California’s canopy coverage — a crucial factor in guiding the state’s strategic plan.
“Understanding how much land is covered by trees is key to assessing the health and benefits of urban forests,” Ventura said.
In the summer of 2023, the team launched the canopy cover project with four students building the training data for their model. Throughout the year, the team honed the model, beginning with remote sensing data before refining it with hand-annotated images. By spring 2024, they had finalized the workflow, and by summer 2024, the model was ready for broader use.
“We’re close to running this across the entire state,” said Pawlak, who even picked up coding skills while working with Ventura and others. “Once fully operational, we’ll be able to track canopy changes year by year.”
Monitoring these changes will be invaluable for strategic planning, allowing the team to pinpoint areas where canopy cover is growing, stable or declining. These insights can shape policies to expand urban canopy, drawing lessons from communities with successful strategies.
Cities across California have clamored for the data, with Pacifica being the first to use UFEI’s canopy cover model to guide urban planning, showcasing the model’s tangible results.
For Yost, though, the true reward is watching students grow. “The coolest aspect is seeing computer science students start recognizing trees,” she said. “Sometimes I’ll get a text from one of them asking, ‘Isn’t this a certain type of tree?’ It’s such a great fusion of disciplines.”
Ventura echoed her sentiment: “It’s not just about finding the right algorithm; it’s what the data tells us and why it matters. This work connects students to real-world impact, far beyond abstract concepts.”
With a model primed for statewide application and cities eager for data, UFEI’s efforts are not only enhancing our understanding of California’s urban forests but also equipping emerging engineers and scientists to make a meaningful difference.
Housed within the College of Engineering, the Noyce School of Applied Computing brings together Electrical Engineering, Computer Science and Software Engineering, and Computer Engineering, with Statistics as an affiliate. The school provides a collaborative environment where students and faculty in fields like computer science apply data science to solve complex, real-world problems.
Call to Action: Plant the Future!
You can play a vital role in shaping California’s greener future. Use the SelecTree tool to choose and plant the right tree for your area, helping increase canopy cover and improve urban environments. By making informed choices, you’re not just planting a tree — you’re contributing to cleaner air, healthier communities and a more sustainable world.
Let’s grow a better future together — start by selecting your tree today!
By Emily Slater
