Urban forests play a vital role in making cities more resilient to climate change while improving the quality of life for urban residents. From cooling overheated neighborhoods to capturing carbon and supporting biodiversity, urban trees deliver a wide range of ecosystem services. However, understanding the distribution, composition, and functioning of these urban forests remains challenging due to the complexity and heterogeneity of urban landscapes. This topic is addressed by our external PhD student Carolin Rempfer at EORC, she is working at DLR within the Department of Geo-Risks and Civil Security.
Carolin’s research focuses on harnessing remote sensing techniques to better identify, analyze, and characterize urban forests. By working with very high-resolution aerial and satellite imagery, she develops innovative approaches that allow researchers and urban planners to monitor trees at detailed spatial scales and derive meaningful insights for sustainable city planning.
Detecting Urban Forests from High-Resolution Imagery
A central aspect of Carolin’s work is the identification of urban forests using traditional image processing techniques applied to very high-resolution aerial imagery. These data sources provide the level of detail required to detect individual tree crowns, delineate canopy structures, and differentiate vegetation from surrounding urban surfaces.
Using advanced segmentation and filtering methods, individual trees can be detected and mapped across complex urban environments. Such approaches enable the efficient creation of spatial inventories of urban trees—an essential prerequisite for urban forest management and climate adaptation strategies. Remote sensing approaches are particularly valuable because traditional field inventories are often costly, time-consuming, and difficult to implement at city scale.
Tree Species Classification with Machine Learning
Beyond detecting trees, Carolin investigates how machine learning algorithms can be used to classify tree species using very high-resolution aerial and satellite data. Combining spectral, structural, and textural information from remote sensing imagery allows the identification of different species or species groups.
Machine learning methods can capture subtle differences in canopy characteristics, making it possible to distinguish species composition within urban forests. These data provide critical insights into the diversity and health of urban green infrastructure and help support more informed urban ecosystem management.
Characterizing Urban Forest Diversity
Another focus of Carolin’s research is the characterization of urban forests based on multiple ecological dimensions. Urban forests are not uniform—rather, they consist of diverse patches with varying species composition, structural complexity, and spatial distribution patterns.
By integrating remote sensing with ecological analysis, her work aims to better understand these patterns across cities. Mapping the structural and compositional diversity of urban forests can reveal how green spaces are distributed, identify areas lacking vegetation, and highlight locations where biodiversity could be enhanced. Remote sensing technologies increasingly allow researchers to quantify structural attributes such as canopy height, density, and spatial configuration at high spatial resolution.
Urban Forests and Climate Adaptation
Urban forests are a key component of climate adaptation strategies in cities. Trees mitigate urban heat, improve air quality, store carbon, and enhance biodiversity. Carolin’s research explores how remote sensing data can help assess the current and future contributions of urban forests to climate resilience.
By analyzing spatial patterns of trees and their ecological characteristics, it becomes possible to model how urban forests contribute to temperature regulation, carbon sequestration, and environmental quality. These analyses support evidence-based decision-making for urban planning and climate adaptation policies.
Quantifying Ecosystem Services
A final component of Carolin’s work focuses on evaluating the ecosystem services provided by urban trees. These include carbon sequestration, air pollution reduction, temperature regulation, stormwater management, and support for urban biodiversity.
High-resolution remote sensing enables the assessment of these ecosystem services at detailed spatial scales by providing accurate information on tree location, canopy size, and species composition. Such spatially explicit data help quantify how urban forests contribute to healthier and more sustainable cities.
Towards Data-Driven Urban Green Infrastructure
Carolin Rempfer’s research highlights the potential of combining remote sensing, image analysis, and machine learning to improve our understanding of urban forests. By generating detailed, spatially explicit information on urban tree populations, her work contributes to more effective urban planning, better management of green infrastructure, and stronger climate resilience in cities.
As remote sensing technologies continue to advance, they will play an increasingly important role in supporting sustainable urban development and unlocking the full potential of urban forests.
