<p>In the context of climate change adaptation, sustainable urban development, and environmental justice, local civil services must meet a range of dynamic demands. To do so, municipalities require both qualitative and quantitative knowledge about the current state and the development of urban structures such as impervious surfaces and vegetated areas within their boundaries. However, obtaining this data through surveys is costly and time-consuming, and the frequency of these surveys is too low to capture changes consistently. As a result, data availability to support urban planning strategies often depends on financial priorities and is not consistently available to all local authorities in Germany. Remote sensing data, which has extensive spatial coverage and is regularly available, offers a more viable option for effective monitoring of urban structures. However, local authorities often lack knowledge about the benefits and limitations of using such data. The UrbanGreenEye project aims to bridge this gap by developing urban climate indicators based on Earth Observation data that meet the needs of local authorities. Drawing on the experience of nine partner municipalities, the project will demonstrate the use and implementation of these indicators in planning processes and strategies. It will also help create digital twins for urban planning applications and provide a free, regularly updated indicator-geodata foundation for Germany to support decision-making, particularly for climate change adaptation. The indicators will help identify locations experiencing high thermal and hydrological stress and quantify the relief provided by vegetated and pervious areas. Land surface temperature (LST) derived from satellite data from the US Landsat program will be used to monitor thermal stress, while the urban green volume and vegetation vitality indicators, derived from EU Copernicus Sentinel-2 satellite data, will contribute to thermal stress relief. The imperviousness indicator will also be derived from Sentinel-2 data using spectral models. Artificial intelligence algorithms, including Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) and attention-based transformer models, will be used to extract complex information from the urban surfaces, which require large amounts of reference data to capture necessary details. This reference data will be generated from high-resolution aerial images using CNN, supported by local ground truth data from municipal authorities and citizen science projects. It will then be upscaled and used as a reference for satellite-level models to provide nationwide consistent products. The satellite-based indicators will be validated for error ranges and at different spatial scales using the micro-scale climate model PALM-4U. Eventually, the indicators will be used to create a model for urban green volume deficiency to identify hot spots for adaptation measures and support planning strategies.</p>

Urban Green Infrastructure (UGI) provides ecosystem services such as cooling of temperatures and is majorly important for climate change adaptation. Green Volume (GV) describes the 3-D space occupied by vegetation and is highly useful for the assessment of UGI. This research uses Sentinel-2 (S-2) optical data; vegetation indices (VIs); Sentinel-1 (S-1) and PALSAR-2 (P-2) radar data to build machine learning models for yearly GV estimation on large scales. Our study compares random and stratified sampling of reference data, assesses the performance of different machine learning algorithms and tests model transferability by independent validation. The results indicate that stratified sampling of training data leads to improved accuracies when compared to random sampling. While the Gradient Tree Boost (GTB) and Random Forest (RF) algorithms show generally similar performance, Support Vector Machine (SVM) exhibits considerably greater model error. The results suggest RF to be the most robust classifier overall, achieving highest accuracies for independent and inter-annual validation. Furthermore, modelling GV based on S-2 features considerably outperforms using only S-1 or P-2 based features. Moreover, the study finds that underestimation of large GV magnitudes in urban forests constitutes the biggest source of model error. Overall, modelled GV explains around 79% of the variability in reference GV at 10m resolution and over 90% when aggregated to 100m resolution. The research shows that accurately modelling GV is possible using openly available satellite data. Resulting GV predictions can be useful for environmental management by providing valuable information for climate change adaptation, environmental monitoring and change detection.

Urban Green Infrastructure (UGI) provides ecosystem services such as cooling of temperatures and is majorly important for climate change adaptation. Green Volume (GV) describes the 3-D space occupied by vegetation and is highly useful for the assessment of UGI. This research uses Sentinel-2 (S-2) optical data; vegetation indices (VIs); Sentinel-1 (S-1) and PALSAR-2 (P-2) radar data to build machine learning models for yearly GV estimation on large scales. Our study compares random and strati ed sampling of reference data, assesses the performance of different machine learning algorithms and tests model transferability by independent validation. The results indicate that strati ed sampling of training data leads to improved accuracies when compared to random sampling. While the Gradient Tree Boost (GTB) and Random Forest (RF) algorithms show generally similar performance, Support Vector Machine (SVM) exhibits considerably greater model error. The results suggest RF to be the most robust classi er overall, achieving highest accuracies for independent and inter-annual validation. Furthermore, modelling GV based on S-2 features considerably outperforms using only S-1 or P-2 based features. Moreover, the study nds that underestimation of large GV magnitudes in urban forests constitutes the biggest source of model error. Overall, modelled GV explains around 79% of the variability in reference GV at 10m resolution and over 90% when aggregated to 100m resolution. The research shows that accurately modelling GV is possible using openly available satellite data. Resulting GV predictions can be useful for environmental management by providing valuable information for climate change adaptation, environmental monitoring and change detection. 1 Introduction Urbanization and climate change are considered global megatrends that will continue to affect life on this planet (Retief et al. 2016). The United Nations suggest that already today, 55% of the world's population live in urban areas and that this number is estimated to rise to 68% by 2050 (United Nations, Department of Economic and Social Affairs, Population Division 2018). Human induced climate change leads to continuously rising average temperatures and poses risks through increased climate andweather extremes including oods, heatwaves and droughts (IPCC 2021). Both phenomena further increase pressures on the natural environment, including biodiversity and ecosystem resilience. Thus, in urban and environmental planning, these megatrends and their interconnected effects need to be considered (Retief et al. 2016;Gill et al. 2007;Mathey et al. 2011).Looking at climate change adaptation in urban contexts, green spaces function as urban green infrastructure (UGI) that provide a variety of ecosystem services (Gill et al. 2007;Mathey et al. 2011; Frick et al. 2020;Palliwoda et al. 2020;Matzarakis 2001). Studies show that greater abundance of UGI, including increased green volume and number of green roofs, has strong positive effects on reducing peak summer temperatures in citi...