B43A-0520
Drought assessment for cropland of Central America using course-resolution remote sensing data

Thursday, 17 December 2015
Poster Hall (Moscone South)
Chi-Farn Chen1, Son Thanh Nguyen2, Cheng-Ru Chen3, Shou-hao Chiang2, Li-Yu Chang2 and Lau-Va Khin1, (1)NCU National Central University of Taiwan, Jhongli, Taiwan, (2)NCU National Central University of Taiwan, Center for Space and Remote Sensing Research, Jhongli, Taiwan, (3)Center for Space and Remote Sensing Research, Chung-Li, Taiwan
Abstract:
Drought is one of the most frequent and costliest natural disasters, which imposes enormous effects to human societies and ecosystems. Agricultural drought is referred to an interval of time, such as weeks or months, when the soil moisture supply of a region consistently falls below the appropriate moisture supply leading to negative impacts on agricultural production. Millions of households in Central America were dependent upon major food crops, including maize, beans, and sorghum, for their daily subsistence. In recent years, impacts of climate change through global warming in forms of higher temperature and widespread rainfall deficits have however triggered severe drought during the primera cropping season (April−August) in the study region, causing profound impacts on agriculture, crop production losses, increased market food prices, as well as food security issues. This study focuses on investigating agricultural droughts for cropland of Central America using the Moderate Resolution Imaging Spectroradiometer (MODIS) data. We processed the data for a normal year 2013 and an abnormal year 2014 using a simple vegetation health index (VHI) that is developed based on the temperature condition index (TCI) and vegetation condition index (VCI). The VHI results were validated using the Advanced Microwave Scanning Radiometer 2 (AMSR2) precipitation data and temperature vegetation dryness index (TVDI) that is developed based on the empirical analysis of TCI and VCI data. The correlation coefficients (r) obtained by comparisons between the VHI data and the AMSR2 precipitation and TVDI data were higher than 0.62 and –0.61, respectively. The severe drought was intensive during the dry season (January–April) and likely backed to normal conditions in May with the onset of rainy season. The larger area of serve drought was observed for the 2014 primera season, especially during April–July. When investigating the cultivated areas affected by severe drought in the primera season, the total cropland areas affected (by severe drought) observed for 2013 and 2014 were 2,463 km2 and 3,874 km2, respectively. This study demonstrates the applicability of MODIS data for agricultural drought monitoring in Central America.