Volume XLII-4/W18
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-4/W18, 1035–1040, 2019
https://doi.org/10.5194/isprs-archives-XLII-4-W18-1035-2019
© 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, 1035–1040, 2019
https://doi.org/10.5194/isprs-archives-XLII-4-W18-1035-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

  19 Oct 2019

19 Oct 2019

MODELLING URBAN HEAT ISLAND USING REMOTE SENSING AND CITY MORPHOLOGICAL PARAMETERS

M. P. Taheri Otaghsara and H. Arefi M. P. Taheri Otaghsara and H. Arefi
  • School of Surveying and Geospatial Engineering, College of Engineering, University of Tehran, Tehran, Iran

Keywords: Remote sensing, Urban heat island, LIDAR, Partial least square, Landsat, Sky view factor, Land surface temperature

Abstract. The aim of this study was to model the surface urban heat island (SUHI) based on remote sensing data, urban morphology parameters and partial least square (PLS) regression using Santa Rosa, California, USA as a case study. Night-time Land surface temperature (LST) was estimated for all the available Landsat 8 night-time data from august to November for the year 2013. Urban morphology parameters such as Building Volume (BV) and Sky View Factor (SVF) were calculated using available LIDAR data of the study area and Normalized Difference Vegetation (NDVI), Index based Built-up Index (IBI) were calculated using Landsat 8 data sets and were used to identify the impact of urban surface characteristics on land surface temperature. Partial least square (PLS) regression analysis was used to assess the correlation and statistically significance of the variables on LST and model the night time LST of the study area. The results of the analysis showed that the LST has a strong positive relationship with IBI and BV and negative relationship with SVF and NDVI and also at night-time, results showed that SVF has a stronger impact on LST differences than NDVI where areas with high density trees had higher temperature than obstacle free vegetated areas. The result of night time LST modelling of the study area was R-square with value of 0.81 between estimated and predicted LST and RMSE with value of 1.02 °C.