17:15 〜 19:15
[ACG41-P09] ASSESSMENT OF IRRIGATION AND COFFEE EXPANSION IMPACTS ON WATER RESOURCES IN THE SREPOK RIVER WATERSHED
キーワード：coffee area, machine learning, Srepok River Watershed, water supply
Coffee is one of the most important perennial crops in the Central Highlands of Vietnam, which is the second-largest coffee exporter globally. Water is considered a crucial factor in coffee production. Nevertheless, water resources have experienced numerous adverse impacts in recent years, such as climate change and anthropogenic activities, including land-use change and irrigation. Thus, the evaluation of water resources under the impacts of irrigation and coffee expansion faces numerous challenges owing to limitations in coffee distribution information. This study aimed to investigate the impact of coffee expansion leading irrigation requirements changing on water demand by mapping coffee land changes and simulating the water quantity in the Srepok River Watershed. The Random Forest algorithm was employed with optical and radar satellite images and ancillary data, such as elevation, transportation system, and slope, on the Google Earth Engine platform. The classified evaluation was subsequently conducted based on indices such as overall accuracy, user’s accuracy, producer accuracy, and kappa coefficient. Moreover, the hydrological components were simulated using the Soil and Water Assessment Tool with coffee irrigation information. The results indicate that the Random Forest algorithm is efficient for land use classification in coffee agroforestry areas, with an acceptance assessment. The coffee area in the land-use classification map was distributed in the middle of the basin, which was consistent with the observation data. Additionally, groundwater and surface water extraction increased under coffee area changes, potentially leading to water shortages for others, such as domestic and livestock usage.