Seasonal evaluation of downscaled land surface temperature: A case study in a humid tropical city

Abstract

The present study evaluates the seasonal variation of estimated error in downscaled land surface temperatures (LST) over a heterogeneous urban land. Thermal sharpening (TsHARP) downscaling algorithm has been used with a separate combination of four selected remote sensing indices. This study assesses the capability of TsHARP technique over mixed land use/land covers (LULC) by analyzing the correlation between LST and remote sensing indices, namely, normalized difference built-up index (NDBI), normalized difference vegetation index (NDVI), normalized difference water index (NDWI), and normalized multi-band drought index (NMDI) and by determining the root mean square error (RMSE) and mean error (ME) produced by downscaled LST. Landsat 8 OLI (Operational Land Imager) and TIRS (Thermal Infrared Sensor) images have been used for pre-monsoon, monsoon, post-monsoon, and winter seasons in 2014 covering the whole Raipur City, India. The RMSE of the downscaled LST decreases from 120 to 480 m spatial resolution in all the four seasons. It is concluded that NDBI is the most effective LULC index having the least error produced in TsHARP downscaling technique, irrespective of any season. Post-monsoon season reflects the most successful result followed by monsoon season. Even in the monsoon season of high vegetation coverage, NDBI presents a lower range of downscaled error compared to NDVI. This indicates better performance of NDBI in detecting the spatial and temporal distribution of mixed urban land.

Relationship between land surface temperature and normalized difference water index on various land surfaces: A seasonal analysis

“…The area is under a tropical wet and dry climate with four typical seasons (pre-monsoon, monsoon, post-monsoon, and winter). Hot and dry premonsoon season extends from March to May (Govil et al 2019). June to September (rainy months) is significantly considered under the monsoon season.…”

“…Various land surface biophysical parameters were applied to specify different types of land surface features (Govil et al 2019(Govil et al , 2020. In this study, special emphasis was given on NDWI (McFeeters 1996(McFeeters , 2013 for determining the relationship with LST.…”

An investigation on seasonal variability between LST and NDWI in an urban environment using Landsat satellite data

“…Satellite image helps us in monitoring these rapid urban changes by visible and near-infrared (VNIR), shortwave infrared (SWIR) and thermal infrared (TIR) bands (Chen et al 2006;Mondal et al 2011;Das et al 2013;Ghobadi et al 2014;Guha 2016;Alexander 2020). The normalized difference remote sensing indices are invariably used in detecting several environmental changes in urban landscape (Chen et al 2006;Guha, Govil, Dey et al 2018, 2020Govil et al 2019;Guha andGovil 2019, 2020;He et al 2019). Apart from the LST estimation, TIR region of electromagnetic spectrum is also very useful in identifying land surface related changes in any region and is significantly applied in almost every sector of earth science domain (Wen et al 2017;Ferelli et al 2018;Guha et al 2018aGuha et al , 2018bAlexander 2020;Govil et al 2020aGovil et al , 2020bNimish et al 2020;Sultan and Satyanarayana 2020).…”

“…act as positive catalysts for rising up the LST. Recently, some important research works on LST were conducted in most of the Indian sub-continent countries like India, Pakistan and Sri Lanka (Senanayake et al 2013;Guha et al 2017Ranagalage et al 2017;Arshad et al 2019;Ranagalage, Estoque, Handayani, et al 2018;Dissanayake et al 2019c;Govil et al 2019;Mehmood and Butt 2019;Ranagalage et al 2019;Imran and Mehmood 2020). Normalized difference water index (NDWI) is the most popular index for the extraction of water body that is invariably applied in land use and LST related studies (McFeeters 1996(McFeeters , 2013Chen et al 2006;Weng and Quattrochi 2006;Essa et al 2012;Chen et al 2013;Yuan et al 2017).…”

“…In many current research articles, the relationship between LST and NDWI have been constructed using TIR remote sensing (Chen et al 2006;Ezzine et al 2014;Debnath et al 2018;Zhang et al 2017;Choudhury et al 2019;Feng et al 2018;German et al 2018;Govil et al 2019;Wu et al 2019). But, the seasonal analysis of the LST-NDWI relationship in tropical India is rare.…”