Predicting Future Urban Land Surface Temperature from Spatio-Temporal Land Use Land Cover Dynamics: A Case of Addis Ababa City and Surroundings, Ethiopia.
No Thumbnail Available
Date
2021-01
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
Although several studies were conducted on modeling future land use land cover (LULC) change in Ethiopia, predicting future urban land surface temperature (LST) from LULC dynamics is less common. The objective of the study was to predict the future LST of Addis Ababa city and its surrounding area from LULC dynamics. Four Landsat (TM, ETM+, and OLI-TIRS) images were used to extract and map LST and LULC distribution for the years 1991, 2002, 2011, and 2020. The performances of several land cover indices (LCI) to predict LST were assessed. CA-Markov chain analysis was applied to predict the LULC and LCI of 2050. The regression analysis was applied to project the future LST from LCI. The finding of the study indicated that the NDBI was the top selected for the prediction of the future LST with the highest correlation(R = 0.72). LST was predicted from NDBI with a mean absolute percentage error(MAPE) of 9.4%. The LULC changes between 1991 and 2020 showed that the built-up area was increased by 38000 ha whilst the water bodies, agriculture, and vegetation areas were decreased by 172 ha, 26905 ha, and 10920 ha, respectively from 1991 to 2020. Between 2020 and 2050, the coverage of the built-up area was predicted to increase by 26013 ha while the area coverage of water bodies, agricultural land, and vegetation area was predicted to decrease by 16 ha, 19133 ha, and 6864 ha, respectively. Between 1991 and 2020, the area coverage of the lowest LST range (10 –20 ⁰C) was decreased by 14728 ha while the area covered by the highest LST range(34 – 45 ⁰C) was increased by 16051 ha. The predicted LST result from NDBI showed that the lowest category was predicted to decrease by 544ha while the highest category was predicted to increase by 3353 ha from 2020 to 2050. And hence it was concluded that the rise of LST will continue in the future due to the LULC changes and global climate warming unless properly mitigated.
Description
Keywords
Addis Ababa City, Cellular Automata Markov, Land Cover Indices, Land Surface Temperature, Land Use Land Cover, Normalized Difference Built-up Index, Regression Analysis