Spatial heterogeneity is the basis of spatial pattern for a community, as well as the coexistence of a community. Many studies have uncovered the spatial heterogeneities of communities at relatively large scales, but it remains unclear at small scales, especially for artificial ecosystems. To determine the spatial heterogeneity dynamics of a community at a small scale, a study was conducted to investigate adult Coleoptera community in a Pinus koraiensis plantation at the Maoershan Ecosystem Research Station. The study site comprised a permanent plot was of 20 m × 20 m and was divided equally into 100 squares. Pitfall traps were used to collect the adult Coleoptera. In 2013 and 2014, 100 traps were set at the bottom-right corner of the 100 squares, with traps in 2013 containing 4% formalin solution, and traps in 2014 containing saturated salt solution. This study was conducted in June, July and September in 2013 and in August and October in 2014. Global Moran's I index was used to explain the spatial autocorrelation of the abundance of Coleoptera community and species. Spatial semi-variogram was selected to reveal the spatial heterogeneity of the abundance of Coleoptera community and species. A combination of a cross-variogram and a simple Mantel test was performed to identify the spatial relationships between different species. Ordinary kriging method was used to map the spatial heterogeneity of the abundances of Coleoptera community and species. In total, 2721 individuals belonging to 10 families and 42 species were collected. Community composition showed significant spatial and temporal variability at different times. In 2013, significant differences (P < 0.01) in abundances among June, July, and September were observed. And significant differences (p < 0.01) in abundances between August and October were detected in 2014. The results of Moran's I coefficient showed that abundances of adult Coleoptera communities, and a relatively small part of Coleoptera species in each season, had significantly spatial autocorrelations. The abundance of adult Coleoptera communities and some of the Coleoptera species in each season, showed spatial heterogeneous patterns, and those communities and species formed spatial aggregations at certain spatial scales. According to the values of (C₀/(C₀+C)), those spatial heterogeneities mainly were controlled by structured factors or combination of structured and random factors. The structure proportions of some species showed pure nugget effect resulting from sampling errors or spatial variability occurring within the minimum distance interval (2 m). For each community, the results of cross-variograme showed that the complex spatial relationships among different adult Coleoptera species existed at different spatio-temporal scales and were positive, negative or random spatial relationships. The values of (C₀/(C₀+C)) demonstrated that the spatial differentiations of adult Coleoptera species were mainly controlled by structured factors or random factors. However, significant spatial relationships between Coleoptera species were observed through the results of a simple Mantel test. The maps of ordinary kriging interpolation showed that the spatial patterns of communities with gradient distribution were obviously different in different seasons, and the strength of gradient distribution weakened from summer to autumn for each community. This study indicated that adult Coleoptera communities were obviously aggregated spatial distribution at a small scale (20 m). Spatial relationships between different species were complex, and the spatial relationships were seasonally variation. The results of this study will help us to understand the temporal and spatial variation mechanism of adult Coleoptera communities.