Detecting heat-inducing urban built-up surface material with multi remote sensing datasets using reflectance and emission spectroscopy

Abstract

This study aims to evaluate the spatial and spectral characteristics of dominant heat-inducing urban built-up surface materials (UBSM) in Kalaburagi, Karnataka State (India) using multi remote sensing datasets and field/lab instruments. A total of 16 UBSM were collected at different locations of the study area and spectra were generated using ASD Field spec 3 Spectro radiometer operated at VNIR and SWIR regions (0.35 μm to 2.5 μm). The collected spectra were then resampled with the spectral range of 0.3–2.5 μm and used for classifying dimensionally reduced AVIRIS-NG (Airborne Visible and Infrared Imaging Spectrometer – Next Generation) dataset through Spectral Angle Mapper (SAM) classifier for understanding spatial and spectral characteristics of heat inducing UBSM in the study area. The classification shows that concrete, basalt stones and asbestos are the dominant UBSM in the study area and its spectral reflectance are comparatively low in the VNIR and SWIR regions. The land surface temperature (LST) of the study area was retrieved from the Landsat-8 TIRS (Thermal Infrared Sensor), and compared with the ECOSTRESS (ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station) LST data product and Fluke Infrared thermometer measurements operated at LWIR (8–14 μm). All these LST estimation and comparison processes were synchronized at the most possible manner (±7 days). Additionally, emissivity of collected UBSM were determined using FTIR (Fourier Transform Infrared Radiometer) emissivity in the laboratory. The amalgamation of results shows that barren land (mostly basaltic), basalt stones and concrete surfaces cover dominantly the high LST region of the study area. The FTIR measurements reveal that, among 16 dominant UBSM, basalt stones had recorded high spectral radiance (0.0309 Wm−1μm−1sr−1) and emissivity (0.98) followed by concrete (emissivity = 0.94). Whereas marble had the lowest spectral radiance (0.0246 Wm−1μm−1sr−1) and emissivity (0.49). The results agree with the SWIR reflection spectra (where basalt stones, concrete and barren rocks shows lower reflectance) and principle of Wien's displacement law (where radiance spectra display series of peaks that corresponding to the temperature measurements). Besides proposing a newer methodology for understanding spatial and spectral distribution of dominant heat inducing urban built-up surface materials, the results offer baseline data for studying microclimatic variations in the study area.