| Title |
Remote sensing approach to simulate the land use/land cover and seasonal land surface temperature change using machine learning algorithms in a fastest-growing megacity of Bangladesh |
| ID_Doc |
69080 |
| Authors |
Kafy, AA; Abdullah-Al-Faisal; Shuvo, RM; Naim, MNH; Sikdar, MS; Chowdhury, RR; Islam, MA; Sarker, MHS; Khan, MHH; Kona, MA |
| Title |
Remote sensing approach to simulate the land use/land cover and seasonal land surface temperature change using machine learning algorithms in a fastest-growing megacity of Bangladesh |
| Year |
2021 |
| Published |
|
| DOI |
10.1016/j.rsase.2020.100463 |
| Abstract |
Rapid urbanization across many regions in the world is altering the existing land use/land cover (LULC), which is significantly raising the land surface temperature (LST). The present study aims to estimate future LULC and seasonal (summer and winter) LST scenarios in one of the fastest-growing megacities and the business capital of Bangladesh, named as Chattogram. Support Vector Machine (SVM) algorithm and Landsat thermal bands were used to retrieve LULC and LST changing patterns for 1999, 2009 and 2019, respectively. The Cellular Automata (CA) and the Artificial Neural Network (ANN) machine learning algorithms were applied to simulate the LULC and seasonal LST scenarios for 2029 and 2039. The CA and ANN model were validated using simulated and estimated LULC and LST data for 2019. The CA model demonstrated an excellent accuracy with an overall kappa value of 0.82. Mean Square Error (MSE) and Correlation coefficient (R) values were used for ANN model validation, which also produced excellent results with 0.523 and 0.796 in summer and 0.6023 and 0.831 in winter season, respectively. Simulated LULC revealed a substantial increase in the urban built-up areas by 9.23% (2029) and 13.59% (2039), compared with the year 2019. Simulated summer LST demonstrates that 31.30% and 35.02% area will likely to face surface temperature more than 36 degrees C, followed by 1.28% and 29.53% area in winter season for year 2029 and 2039, respectively. The seasonal LST scenarios in different LULC demonstrated comparatively higher temperatures in built-up areas. Based on the results, a strong correlation was found between the changes in urban areas and raising LST by representing major encounters for environmental engineers and urban planners to mitigate the consequences of the surface urban heat island (SUHI) phenomenon. For ensuring sustainable urban growth and minimizing the SUHI effect in this fastest-growing megacity, future city master plan must focus on the importance of urban plantation and conservation of natural resources. |
| Author Keywords |
Urbanization; Land use/land cover change; Land surface temperature; Cellular automata; Artificial neural network |
| Index Keywords |
Index Keywords |
| Document Type |
Other |
| Open Access |
Open Access |
| Source |
Emerging Sources Citation Index (ESCI) |
| EID |
WOS:000654348000065 |
| WoS Category |
Environmental Sciences; Remote Sensing |
| Research Area |
Environmental Sciences & Ecology; Remote Sensing |
| PDF |
|