Spatial patterns of Liriomyza trifolii (Burgess) and Trialeurodes vaporariorum (Westwood) populations were analyzed in order to explore their spatial population dynamics. Geostatistical analysis was used to map the two populations in commercial tomato greenhouses. Semivariograms were calculated to model the change in spatial correlation with increasing distance between samples. For L. trifolii data sets, none of the theoretical variogram models used in this study was properly fit for the adult semivariogram data, but the spherical and Gaussian models were properly fit for the immature semivariogram data (based on cross-validation). However, the range (maximum distance of spatial dependence) for the leafmine (i.e. immature infestations) differed between two models; ≈ 33m based on spherical model, and ≈ 21m based on Gaussian model. For T. vaporariorum data sets, the spherical model provided the best fit for the adult semivariogram data, and the exponential model was the best fit for the immature semivariogram data (based on cross-validation). The range of the adult was ≈ 21m based on spherical model, and the immatures were also spatially correlated at distances up to ≈ 22m apart based on exponential model. The degree of spatial association between the distributions of L. trifolii and T. vaporariorum was determined using Spatial Analysis by Distance IndicEs (SADIE). Adult L. trifolii showed overall spatial dissociation (negative relationship) with adult T. vaporariorum for all sampling periods. However, association analysis found a spatial association (positive relationship) between the distributions of the leafmine by immature L. trifolii and immature T. vaporariorum for all sampling periods.