The spatial distribution of the Oriental migratory locust, Locusta migratoria manilensis (Meyen) egg pods was studied by integrating geostatistical analysis and GIS techniques. Data of egg pods were collected from both spatial scales over two years of extensive surveys: 450 m intervals for the whole study area, then 50 m grid for possible egg-laying areas which were located on the areas damaged by the locust plague or covered with sparse vegetation. Meanwhile, site ecological variables including vegetation coverage, and soil parameters (salinity, pH, water content at 5 cm depth, organic matter, etc.) were surveyed at these grid points. The results showed that three site variables including vegetation coverage, soil water content at 5 cm depth and soil salinity were significantly different between sites with and without egg pods, and the preferable ranges for L. m. maniliensis oviposition were 0-30%, 10.1％-20.0% and 0.09％-1.99% for these three variables, respectively. No eggs were laid when vegetation coverage exceeded 50%, soil water content at 5 cm depth was higher than 30%, or soil salinity was above 3%. Semivariograms indicated that the distribution of egg pods was in high heterogeneity. Spatial autocorrelation in egg pod distribution was at distances about 390 m. The spatial distribution of egg pods was patchy and aggregated in the two years and could be best described using spherical models. GIS risk assessment maps, derived by block kriging, displayed the probabilities of occurrence of the locust egg pods at an area-wide scale. The results may provide useful information on planning for sampling in the field, tracking embryo developments, monitoring targets and taking site-specific measures towards this locust.