Volume XLII-4/W18
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-4/W18, 401–405, 2019
https://doi.org/10.5194/isprs-archives-XLII-4-W18-401-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-4/W18, 401–405, 2019
https://doi.org/10.5194/isprs-archives-XLII-4-W18-401-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

  18 Oct 2019

18 Oct 2019

SPATIOLTEMPORAL MODELING THE IMPACT OF SURFACE CHARACTERISTICS VARIATIONS ON LAND SURFACE TEMPERATURE VARIATIONS: A CASE STUDY OF SAMALGHAN VALLY

M. K. Firozjaei1, M. Makki2, J. Lentschke2, M. Kiavarz1, and S. K. Alavipanah1 M. K. Firozjaei et al.
  • 1Dept. of Remote Sensing and GIS, Geography Faculty, University of Tehran, Tehran, Iran
  • 2Department of Geography, Humboldt-Universität zu Berlin, Germany

Keywords: LST variation, surface parameters, PCA, PLSR, Samalghan Valley

Abstract. Spatiotemporal mapping and modeling of Land Surface Temperature (LST) variations and characterization of parameters affecting these variations are of great importance in various environmental studies. The aim of this study is a spatiotemporal modeling the impact of surface characteristics variations on LST variations for the studied area in Samalghan Valley. For this purpose, a set of satellite imagery and meteorological data measured at the synoptic station during 1988–2018, were used. First, single-channel algorithm, Tasseled Cap Transformation (TCT) and Biophysical Composition Index (BCI) were employed to estimate LST and surface biophysical parameters including brightness, greenness and wetness and BCI. Also, spatial modeling was used to modeling of terrain parameters including slope, aspect and local incident angle based on DEM. Finally, the principal component analysis (PCA) and the Partial Least Squares Regression (PLSR) were used to modeling and investigate the impact of surface characteristics variations on LST variations. The results indicated that surface characteristics vary significantly for case study in spatial and temporal dimensions. The correlation coefficient between the PC1 of LST and PC1s of brightness, greenness, wetness, BCI, DEM, and solar local incident angle were 0.65, −0.67, −0.56, 0.72, −0.43 and 0.53, respectively. Furthermore, the coefficient coefficient and RMSE between the observed LST variation and modelled LST variation based on PC1s of brightness, greenness, wetness, BCI, DEM, and local incident angle were 0.83 and 0.14, respectively. The results of study indicated the LST variation is a function of s terrain and surface biophysical parameters variations.