RADIATIVE TRANSFER MODEL SIMULATIONS TO DETERMINE THE NIGHT TIME FOG DETECTION THRESHOLD
- Indian Institute of Remote Sensing, ISRO, Dehradun, India
Keywords: Fog detection, Radiative Transfer Model, SBDART, Indo-Gangetic plain, Brightness temperature difference threshold, Middle Infrared, Thermal Infrared
Abstract. Every winter the Indo-Gangetic plains (IGP) of northern India are severely impacted both socially and economically by fog. For night time fog detection, visible imagery cannot be used. Also, as emissions from ground and fog is almost similar in thermal infrared (TIR, 10.8 μm) channel, TIR channel cannot help in identifying fog. However, emission in middle infrared (MIR, 3.9 μm) channel is less than emission in TIR channel over foggy area. Therefore, brightness temperature difference (BTD) between TIR and MIR is positive during night time over fog area. This BTD technique cannot be directly used during day time as MIR channel is contaminated by solar radiations. In the present work, a spectral sensitivity analysis study has been done for these two spectral channels using radiative transfer model (RTM) simulations to determine a threshold BTD for night time fog detection. SBDART (Santa Barbara DISORT Radiative Transfer) model was used for this study to simulate brightness temperatures (BT). The RTM simulations of BT of the two spectral channels was carried out for different fog microphysical characteristics like fog optical depth (FOD) and fog droplet size (Re). The fog episode of January 2018 over IGP was studied by applying threshold BTD obtained from simulation results for INSAT-3D data. A threshold BTD value > 5 K detected night time fog over IGP with good accuracy. The threshold BTD obtained from satellite image is compared with different cases established from simulation result which gave idea about microphysical properties of fog over IGP during winter seasons.