High-resolution thermal retrieval with OTES from HyTES data: validation in NETSENSE-2023 and urban application over Mesa, Colorado

Extended abstract / summary of an ongoing research article prepared for submission to a peer-reviewed scientific journal. This article presents the high-resolution thermal retrieval methodology (OTES – Optimised Temperature and Emissivity Separation) applied to hyperspectral HyTES (Hyperspectral Thermal Emission Spectrometer) data, demonstrating its capability for characterising urban thermal heterogeneity at metric scale. The methodology is rigorously validated using the NETSENSE-2023 field campaign at Grosseto (Italy), where in situ radiometric measurements over water, bare soil, mixed vegetation and periurban pixels constrained uncertainty propagation models for emissivity and temperature. The OTES workflow integrates ERA5 atmospheric profiles processed through MODTRAN radiative transfer simulation, pixel-level spectral transmissivity filtering, and per-pixel retrieval of land surface temperature (LST) and land surface emissivity (LSE). Validation against NETSENSE-2023 in situ data shows mean absolute differences of approximately 1.5 K for LST and 0.025 for LSE, with differential performance across surface types. The validated method is applied to a nocturnal HyTES overpass over Mesa, Colorado (USA) in 2022, generating Level 2 LST/LSE products and Level 3 urban indicators (SUHI, UTFVI, DI and hotspot density) at 2–3 m spatial resolution. Results reveal detailed thermal structure of the urban fabric, with distinct micro-hotspots linked to material properties and urban geometry that remain unresolvable at kilometre-scale. The OTES methodology, complementary to the moderate-resolution UVEG-TES branch, establishes a coherent multiscale framework for urban thermal analysis and is positioned as a candidate operational approach for next-generation high-resolution thermal missions such as LSTM and TRISHNA.