基于 FY3B／MERSI水稻长势监测及其与 AQUA／MODIS数据对比分析

摘要

In order to extract the area information and monitor the growth of rice using FY-3B/MERSI da-ta,the difference in single channel reflectance between FY-3B/MERSI and AQUA/MODIS was analyzed.Sec-ondly,the spatial distribution of the double-cropping early rice in Poyang Lake area was extracted by means of visual interpretation and remote sensing supervised classification method.Finally,the growth of double-crop-ping early rice was monitored using NDVI ( Normalized Difference Vegetation Index) .The results showed that, (1) the red band reflectance and near-infrared band reflectance had highly significant correlations between the data of FY3B/MERSI and AQUA/MODIS in the rice-growing area.Among them,the red band reflectance and near-infrared band reflectance of MERSI were even smaller than MODIS.(2) The rice extraction accuracy overall reached 92%,which can satisfy the needs of growth remote sensing monitoring.(3) There were signifi-cant positive correlation between FY3B/MERSI-NDVI and AQUA/MODIS-NDVI,and the monifored growth was consistent with the actual data of the seedlings,so the result of the classification rice seedlings based on FY3B/MERSI can meet the needs of rice growth RS operational monitoring.%为利用中分辨率光谱成像仪FY3B/MERSI数据进行水稻种植面积信息提取及长势监测，首先对FY3B/MERSI与AQUA/MODIS单通道反射率进行对比分析；其次，通过目视解译和遥感监督分类方法提取了环鄱阳湖区双季早稻的空间分布；最后，基于面积识别结果，利用NDVI对环鄱阳湖区双季早稻长势进行监测。结果表明：（1） FY3B/MERSI与AQUA/MODIS水稻种植区红光和近红外通道反射率具有极显著相关性，且MERSI红光波段及近红外波段反射率都略小于对应MODIS反射率；（2）水稻的提取精度总体达到了92％，可以满足其生育期长势遥感监测的需求；（3） FY3B/MERSI和AQUA/MODIS计算的水稻区NDVI存在显著正相关，且二者测得的水稻长势均与实际苗情数据基本吻合，因此，由FY3B/MERSI遥感影像直接得到的水稻苗情分类基本可以满足业务化需要。