Abstract: Identification of agricultural drought is a prerequisite for agricultural drought risk assessment and management.The key indicator for identifying agricultural drought is soil moisture content because the crop growth depends on moisture in soil in arid area.In this paper,a case study on maize drought in northwest Liaoning Province was carried out,and the whole growth period of maize was divided into six growing stages based on the information of FAO (Food and Agriculture Organization of the United Nations) and the observational data in northwest Liaoning Province during the growth period of maize for many years.The Cropland Soil Moisture Indexes (CSMIs) at different growing stages of maize were calculated using the MOD09A1 data.The correlation between CSMIs and the corresponding measured soil moisture content at 0-50 cm depth was analyzed.The results showed that all the absolute values of the correlation coefficients were higher than or equal to 0.5,and the confidence level reached 0.01.Taking the severe summer drought in northwest Liaoning Province in 2009 as an example,the CSMI indexes of maize in late-June and mid-August and at jointing and heading stages were calculated.The indexes at two growing stages were divided into four drought levels,i.e.the light drought,medium drought,heavy drought and severe drought by using the Natural Breakpoint Classification method.The drought level maps were drawn by using GIS (Geographical Information System) means.The results showed that the drought levels of maize at jointing and heading stages in northwest Liaoning Province in 2009 were mainly the light drought,medium drought and heavy drought.The area of medium drought was the largest compared with that at other two growing stages,and the area of heavy drought enlarged significantly at heading stage.Spatially,medium drought of maize occurred in Zhangwu County,Fuxin county,Kaiyuan City and north Shenyang,and heavy drought in Changtu County,Jianchang County,Beipiao City,Yi County and Linghai City,respectively.The distribution of maize drought levels in the most counties and cities in the study area was basically in line with the information of actual drought disaster situation in the same period in 2009.But there was a regional difference,for example,the drought in Chaoyang County and Jianping County might be caused by different regional climatic conditions and local disaster prevention and mitigation.The results of this study showed that the research method was feasible in identifying maize drought in northwest Liaoning Province,and could provide the technical support for the evaluation and management of agricultural drought disaster risk.

Key words: maize, grow stage, MODIS, CSMI, soil moisture content, drought identification, Liaoning Province