Abstract:

 Land Surface Temperature (LST) is an important variable for understanding land–atmosphere interactions and ecosystem processes. This study uses Landsat 8 and 9 observations to examine spatial and temporal variability in LST across Surface Energy Budget Network (SEBN) sites in 2024: the Audubon Research Ranch semiarid grassland in Arizona (Ameriflux ID: US-Aud), the Chestnut Ridge deciduous forest in eastern Tennessee (US-ChR), and the Bondville cropland in Illinois (US-Bo1). The Land Surface Temperature, Normalized Difference Vegetation Index (NDVI), and surface albedo were analyzed across spatial scales ranging from 100 m to 5 km from the tower sites. Results showed that LST was generally spatially homogeneous near tower footprints. During the growing season, LST showed stronger relationships with NDVI, particularly at Chestnut Ridge and Bondville, while surface characteristics became more evident during fall and winter. Comparisons between satellite and tower observations showed overall good agreement. This work supports NOAA ARL’s Surface Energy Budget Network (SEBN), which provides long-term observations to better understand energy, water, and carbon exchanges across different ecosystems. Through this internship, the student deepened their understanding of NOAA mission areas and strengthened their skills in remote sensing, GIS, and data analysis, including gaining experience with RStudio.