Our urban-ecology research asks how the functional and structural diversity of urban greenspace shapes ecosystem services and human well-being. This matters because cities concentrate people, infrastructure, and heat exposure, yet urban vegetation is highly heterogeneous at fine spatial scales. Most widely used greenspace products are too coarse to resolve the patchwork of different function types of greenspace across neighborhoods, and they rarely capture the structural attributes or functional traits that determine how greenspace actually works. As a result, key outcomes—how greenspace cools the urban heat island, supports birds and pollinators, and influences pollen loads and allergy seasons—remain difficult to quantify mechanistically and to compare across cities. We aim to fill this gap by integrating advanced remote-sensing products with complementary ground and community data streams, including citizen science observations, targeted field measurements, and ecological monitoring networks. By combining these datasets in a common modeling framework, we can map fine-scale variation in urban vegetation, link it to measurable ecosystem services, and identify where and for whom urban greenspace delivers the greatest benefits. Currently, there are a couple of works in progress and one work in review.

• Functional traits of urban street trees and their cooling efficiency across US neighborhoods (Su et al., in review)
• Deep learning framework to generate sub-meter mapping of greenspace in cities across the U.S. (Li et al., in preparation)

We have generated building function types.

We have also generated submeter greenspace products.