Spatial patterns of aboveground and belowground soil animal communities and their spatial correlations with environmental factors are important foundations for revealing spatial patterns and processes of aboveground and belowground ecosystems and mechanisms for maintaining biodiversity. However, the spatial relationships between soil animal communities and environmental factors in aboveground and belowground ecosystems are still not clear, especially in small scales. This study investigated the spatial distribution patterns of aboveground and belowground soil mite communities, for recognizing the spatial relationships between soil mite communities and environmental factors in a farmland planted with soybean in Sanjiang Plain, Northeast China. Aboveground and belowground soil mite communities were sampled at 100 uniformly distributed points in a 50 × 50 m² plot and environmental factor analyses performed at each point in August and October, 2011. Spatial distribution patterns and relationships were analyzed using geostatistical tools with semivariogram, crossvariogram and ordinary kriging. Simple Mantel test was selected to evaluate those spatial correlations between soil mite communities and environmental factors and to determine whether the spatial relationships were true or spurious. According to the results of semivariograms, aboveground and belowground soil mite communities in August, belowground soil mite community in October and most of the soil mite species from those communities showed clumped distributions from 5-40 m scales, revealing exponential, spherical or Gaussian models. Moreover, based on the analysis of ordinary kriging, obviously spatial structures were detected, and spatial heterogeneities were common characteristics of those communities and soil mite species from those communities above mentioned. The values of (C₀/C₀+C) demonstrated that the spatial differentiations of those soil mite communities and soil mite species above mentioned mainly controlled by structured factors or both structured and random factors. While for the aboveground soil mite community and most mite species in October, nugget effects were detected which might resulted from sampling errors or the spatial variability occurring within the minimum distance interval. In addition to the soil mite communities and soil mite species, environmental factors also showed spatial structured patterns, and those spatial differentiations were determined by structured factors. Nonetheless, the spatial distribution patterns of environmental factors were not consistent with those of the soil mite communities or soil mite species. And spatial distribution patterns of aboveground and belowground soil mite communities and environmental factors were dynamic from August to October. The results of crossvariograms showed spatial cross-correlations with different types (either negative or positive) between various soil mite communities and environmental factors, and between soil mite species and environmental factors in aboveground and belowground ecosystems, respectively. Simple Mantel test just confirmed a positively spatial relationship between Mesostigmata unidentified sp. 1 and average height of soybean from aboveground ecosystem in August. No true spatial relationships were identified between aboveground, or belowground soil mite communities, or other soil mite species and environmental factors. In total, the results of this study provided essential data for researching spatial patterns and processes in aboveground and belowground ecosystems and showed that belowground soil mite communities in summer and autumn and aboveground soil mite community in summer showed obviously spatial heterogeneity structures, while the spatial relationships between aboveground and belowground soil mite communities and most of the soil mite species from those communities and environmental factors were not significant. However, spatial analysis from this research only provide informations about spatial distribution patterns and spatial relationships at one point in two times and at a specific small scale (50 m × 50 m). Thus, further manipulative experiments are necessary to reveal the underlying processes for controlling the spatial patterns and relationships observed in the present research.