Modis time series for irrigated-area mapping in hydrographic basins of the Brazilian Northeastern region

Abstract: The objective of this work was to evaluate the applicability of time series of the enhanced vegetation index (EVI), from the moderate resolution imaging spectroradiometer (Modis), in the mapping of irrigated areas in the Northeastern region of Brazil. Annual time series from 2006 to 2015 were classified with the iterative self-organizing data analysis technique (Isodata) algorithm, generating a binary map of irrigated and nonirrigated areas for each year. In the Sertão region, the classification showed an average kappa coefficient of 0.66, underestimating the irrigated area by 7.6%, compared with data of the 2006 agricultural census. In regions more humid than the Sertão, such as Agreste and Zona da Mata Nordestina, the methodology showed limitations to distinguish irrigated areas from natural vegetation, presenting average kappa coefficients of 0.26 and 0.00, respectively. The EVI time series from Modis are applicable for the mapping of irrigated areas in the Sertão of the Northeastern region of Brazil.

摘要：本研究旨在评估来自中分辨率成像光谱仪（MODIS）的增强型植被指数（EVI）时间序列在巴西东北部灌区制图中的适用性。研究采用迭代自组织数据分析算法（ISODATA）对2006至2015年的年度时间序列进行分类，逐年生成灌区与非灌区的二分类地图。相较于2006年农业普查数据，塞尔唐（Sertão）地区的分类结果平均科恩卡帕系数（Kappa coefficient）为0.66，灌区面积被低估7.6%。在比塞尔唐地区更为湿润的阿格雷斯特（Agreste）与东北部沿海林区（Zona da Mata Nordestina）等区域，该方法在区分灌区与自然植被方面存在局限，二者的平均科恩卡帕系数分别为0.26与0.00。MODIS的EVI时间序列适用于巴西东北部塞尔唐地区的灌区制图。