Agricutural Drought Monitoring And Its Early Warning Based On Remotely Sensed Data NOAA/AVHRR

Drought is a phenomenon caused by the water shortage and the supply and demand unbalance.It is one of major environmental disasters in china and other countries in the world,which has caused the most severe loss of human and society,especially in significant loss of agriculture.Based on AnHui drought features,the drought of different climate zones in different seasons was studied:HuaiBai area in spring;HuaiBai area,YanHuai and JiangHuai area in summer;YanJiang and JiangNan area in fall;HuaiBai area in winter.This paper try to monitor and forecast drought in different climate zone and different seasons in AnHui province based on polar orbit meteorological satellite data and surface meteorological data,the result are listed as follows:1.The model between the Anomaly Vegetation Index(AVI),Vegetation Supply Water Index(VSWI),thermal inertia approach(ATI) and soil moisture(include 10cm and 20cm soil moisture) were established.The result shows that except fall-Yanjiang area,the model between AVI and 20cm soil moisture is significant at 0.05 level in other areas,but the model between AVI and 10 cm soil moisture is only significant at 0.05 level in summer-Huaibei,Yanhuai area and fall-Yanjiang,Jiangnan area;The model between VSWI and 20cm soil moisture is significant at 0.05 level or 0.01 level Only in April-Huaibei,August-Huaibei,Jianghuai,November-Yanjiang,Jiangnan areas;the model between VSWI and 10cm soil model is significant at 0.05 level only in August-Huaibei area;The model between ATI and 10cm soil model is significant at 0.05 level in Spring-Huaibei,Fall-Jiangnan,winter Huaibei areas,and the model between ATI and 20cm soil model is significant in Spring-Huaibei,Fall-Jiangnan areas2.In the case of spring and winter ATI approach has great effect,and the model between 10cm soil moisture and ATI value has better effect compared with 20cm.in the case of summer and fall,the model bettween AVI,VSWI and soil moisture is adaptable,and he model between 20cm soil moisture and ATI value has better effect compared with 10cm soil moisture,through testing and analysis,each monitoring model can be used in practice.3.Based on the observation data of soil humidity and meteorological data during 1981 to 2003,different seasons forecasting model was established by stepwise regression analysis in Huaibei area Anhui province.The results show that spring,summer,autumn,winter's average forecast precision is 88.92%,91.35%,91.96 %and 92.94%respectively.So we can draw a conclusion that the accumulating values of precipitation,sunlight in a period of ten days and the current soil humidity could be used preferably to estimate the soil humidity in ten days.