Mapping the changes in land use/land cover and assessing its impact on drought characteristics in the Chi River Basin, Thailand | |
| Author | Nattanicha Meesaard |
| Call Number | AIT Thesis no.WM-20-07 |
| Subject(s) | Land use--Thailand--Maps Droughts--Thailand--Chi River Basin |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineeing in Water Engineering and Management |
| Publisher | Asian Institute of Technology |
| Abstract | This study aims to classify Landsat satellite images for land use/land cover (LULC) changes using supervised classification algorithms and to assess the impact of land use/land cover changes on drought events using the Standardized Precipitation Evapotranspiration Index (SPEI) in the Chi river basin. The land use/land cover was classified into four classes (urban, agriculture, forest, and water) in 2000, 2003, 2006, 2009, 2012, and 2014 from Landsat 5 TM and Landsat 7 ETM+ using Google Earth Engine (GEE). Moreover, the precipitation (P) and mean temperature (Tmean) data collected from ERA5 were used to calculate the SPEI using SPEI R package. The 6 and 12-month SPEI were used to analyze the drought events because this study tend to assess the long-term drought and land use changes in the study area. The deforestation and urbanization scenarios were considered to assess drought events. In addition, the correlation between monthly SPEI, Normalized difference vegetation index (NDVI), and Land surface temperature (LST) values were calculated. The results showed that the LULC changes classification with random forest method (RF) had a high accuracy than classification and regression tree (CART) method. Therefore, the RF method was used to classify the LULC changes. The agricultural areas were the most area in the Chi river basin and secondary was forestry areas. The agricultural areas were increasing whereas the forestry areas were decreasing in 2000-2009. The results also indicated that changes percentage between 2000 and 2014 concerned mainly water (-34.19%), agricultural areas (18.67%), forestry areas (-16.27%), and least changes urban (-2.18%). The overall accuracies were comparatively higher more than 93 percent. The spatial variations in drought properties results showed most of the consideration hotspots of deforestation and urbanization can have a high severity drought value. The highest severity drought was deforestation zone at 21.48-28.07 (based on SPEI-12 values). Moreover, the temporal variations in drought at consideration zone demonstrated the most extreme dry was - 2.2848 based on SPEI-6. However, most correlation coefficient (r) between monthly values of SPEI and NDVI was positive, and coefficient r between monthly values of SPEI and LST was negative. Furthermore, p-value result demonstrated that the relationship between three values at all consideration zone were significant (p<0.05) correlation. |
| Year | 2020 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Civil and Infrastucture Engineering (DCIE) |
| Academic Program/FoS | Water Engineering and Management (WM) |
| Chairperson(s) | Sundaram S. Mohana; |
| Examination Committee(s) | Shrestha, Sangam;Sarawut Ninsawat; |
| Scholarship Donor(s) | Ministry of Agriculture and Cooperatives (MOAC), Thailand;IT Fellowship; |
| Degree | Thesis (M. Eng.) - Asian Institute of Technology, 2020 |