Capturing Small-Scale Surface Temperature Variation across Diverse Urban Land Uses with a Small Unmanned Aerial Vehicle

Abstract

Urbanization increases the urban land surface temperature (LST), challenging society and the environment. This study measured the LST of diverse land uses (LU) in Dallas–Fort Worth (DFW) using a high-resolution (8 cm) thermal infrared sensor onboard a small, unmanned aerial vehicle (UAV). LUs included park (PA), industrial (IA), residential low-cost (RLC), and residential high-cost (RHC) areas. LST was collected by the UAV at different times on eight nonconsecutive days. UAV-collected LST was compared with that from Landsat 8-9 and in situ measurements. RHC reported the highest mean LST, and PA showed the lowest mean LST. Dark-colored asphalt shingle roofs in RHC had the highest mean LST range at 35.67 °C. Lower LST was measured in shaded areas and under thick green cover, whereas areas with thin green cover occasionally reported higher LST than pavements. The micro-urban heat island (MUHI) was calculated between LUs and within land cover types (roof, pavement, green, and water). The MUHI varied from 4.83 °C to 15.85 °C between LUs and 0.2 °C to 23.5 °C within LUs for the less than 1 km2 study area. While the UAV thermal sensor and Landsat demonstrated a similar trend of LST variation, the UAV sensor reported more intense MUHI. An average percent bias (PBIAS) of 5.1% was calculated between the UAV sensor and in situ measurements. This study helps inform the urban design process by demonstrating how land use decisions impact LST locally and provides valuable insight for studies concerned with fine-scale urban LST variability.