利用月度肾综合征出血热发病人数和长时序月度NDVI值的相互关系,对肾综合征出血热的发病趋势及发病人数进行预测。研究区大杨树镇2001-2005年共有144例完整的HFRS病人资料,以及同期详细的宿主动物捕获数据。基于Landsat TM影像以及Google earth影像,大杨树地区土地利用分为4种类型,山地、林地、农田以及居民点。各类土地利用类型的NDVI数据由SPOT-4卫星的VGT-S10数据集(10d最大化合成的NDVI数据)提供。对HFRS病例与NDVI之间的关系进行图解分析、相关分析和回归分析。研究表明,NDVI的峰值多出现于8月,而HFRS发病人数的峰值多出现在11月。前朔3个月的农田NDVI值与HFRS病例数之间的相关系数为0.67(P值〈0.001)。农田NDVI峰值比HFRS病例的峰值提前了3个月。研究量化了NDVI与HFRS之间的关系,为HFRS早期预警系统的建立提供了依据。
In China, hemorrhagic fever with renal syndrome (HFRS) is still endemic in 31 of 32 provinces, autonomous regions and metropolis, and accounts for 90% of the HFRS cases reported globally. Some studies indicated that HFRS incidence was significantly different among different precipitation, temperature level, land-cover, soil types and normalized difference vegetation index (NDVI). The current study proposed to explore the temporal relationship between the inter-annual variation in the number of cases of human disease caused by hantavirus and time series of NDVI in order to predicate the trend and level of the case number of HFRS. Dayangshu district in Inner Mongolia Autonomous Region was selected as the study area. Based on Landsat TM image, the land-use of this district was divided into four categories: hilt, forest, farm-land and residence area. The relationship between the case number of HFRS and time series NDVI was explored by using graphic demonstration, correlation test and linear regression analyses. The study indicated that the highest correlation coefficient was 0.67 (P value〈0.001) between three months backward NDVI and the number of cases of HFRS in farmland. The result also indicated that the peak of cases happened 3 months later than the peak of monthly NDVI. This study quantified the temporal relationship between NDVI and the number of the cases of HFRS.