Mapping “Cold Dew Wind” damaged area of late paddy rice in Jiangxi Province using Moderate Resolution Imaging Spectroradiometer data

摘要

In south of China, double-season is popular pattern for paddy rice in order to utilizing the climate resources. But the late rice is damaged frequently by “Cold Dew Wind” (CDW) chilling damage, which is caused by the low temperature in late rice’s heading and flowering stage. At this time, once the temperature is below the low limits of the optimal temperature of paddy rice, it will cause empty shells and crusts, leading to yield reduction. This paper proposed a novel integrated framework to map the damage extent of CDW in Jiangxi Province, China to late rice from 2005 to 2018. Firstly, the daily mean air temperatures were estimated using the Moderate Resolution Imaging Spectroradiometer (MODIS) Land Surface Temperature (LST) products, the altitude information from Digital Elevation Model (DEM), MODIS Land Cover products and air temperature measurements at meteorological stations located in the Jiangxi Province from 2005 to 2018. Then all-weather and fully covered daily average temperature time series were constructed by employing multi-platform merging and temporal fusion. Meanwhile, late rice planting areas were identified based on the seasonal profiles of MODIS Enhanced Vegetation Index (EVI) and Land Surface Water Index (LSWI), and map errors from 2005 to 2016 at provincial scales were no more than 7.50%. Paddy rice growth characteristics were combined with EVI and LSWI to identify heading and flowering stages. Finally, the time series of daily mean air temperatures, paddy rice phenology information and planting area were used to estimate the rice crop area damaged by the CDW for late rice. The results show that from 2005 to 2018, the two years most affected by CDW were 2006 and 2011. The damaged areas in the study area were 89,691.99 hectares and 36,614.23 hectares, respectively, accounting for 6.38% and 2.34% of the total planted area of late rice. According to the duration of CDW, the disaster degree is divided into three levels: mild, moderate and severe. The range of disasters was the widest in 2006, but the area of mild disaster accounted for 93.12% of the total disaster area. The area affected in 2011 was second only to 2006, but the severely affected area accounted for 70.26% of the total affected area. These results indicate this paper realized the estimation of the late rice area affected by CDW, the assessment of the disaster level, and the spatial distribution map production of the damaged area and disaster level in the study area.