Abstract: Abstract: Insect infestation is one of the major biological disasters in crop production. To identify an insect-damaged area and to obtain its spatial distribution are important for agricultural disaster monitoring. These data are usually obtained through field investigation, collation, and summary. As an alternative, the remote sensing of insect infestation has advantages of large range, time savings, labor savings, and high speed. In summer of 2012, an outbreak of oriental armyworm (Mythimna seperata Walker) occurred in a vast area of northeast China. In order to examine the potential of remote sensing technique in monitoring such a migratory, fulminant, and devastating agricultural pest, several data processing and analysis procedures were carried out to assess the spatial distribution of oriental armyworm and its severity level, as follows. 1）Cornfield acreage was extracted in the study area using a decision tree classifier based on NDVI and single-band reflectance that was derived from multi-temporal HJ-1A/1B CCD images over the growing season of maize. 2) Based on field measurements, the pest severity level was associated with leaf biomass from several ground agronomic parameters; the aim was to find a certain remote sensing variable and its quantitative model with the ground agronomic parameter to monitor the oriental armyworm disaster severity level. 3) The relationship between vegetation indices that were derived from three temporal HJ-CCD satellite images on three different phases and agronomic parameters were established based on numerical analysis. 4) Using the relationship between agronomic parameters and oriental armyworm disaster severity level, it is possible to use remote sensing data to obtain the spatial distribution of oriental armyworm. The results showed that the leaf biomass was significantly correlated with oriental armyworm severity level (R2=0.905, n=51). Therefore, it is feasible to use leaf biomass as a surrogate of the hazard grade of oriental armyworm. The dynamical variation of the leaf biomass can be detected by the renormalized difference vegetation index (RDVI), which thus allows the remote sensing of this important agronomic parameter. A regression model was calibrated and validated against ground survey points. The determination coefficient (R2) of leaf biomass estimation and the root mean square error (RMSE) of the model achieved 0.7376 and 43.26 g·m-2, respectively. Based on this relationship, the oriental armyworm severity map was thus generated in the study area, which was in good agreement with our field observation. In conclusion, the present study illustrated the potential to use multi-temporal HJ-CCD images for monitoring maize oriental armyworm over vast area. Such a method may provide an opportunity to conduct yield loss assessments in a spatially continuous manner.