[Objective] The knowledge about the spatial distribution characteristics of salinization is the basis for amelioration of saline-alkali soils. Compared with research based on a single scale, nested multi-scale research can be more effectively applied to analyzing structural characteristics of the spatial variation of soil salinity with higher accuracy in describing scale-dependent self-correlated variation of salinization. [Method] In the Anjihai Irrigation Zone of Xinjiang, soil samples were collected by soil layer in the film-mulched cotton fields for analysis of rules and variabilities of the spatial distribution of salt content in the root zone of the drip-irrigated cotton field under film mulch with the method combining conventional statistics and geo-statistics, as affected by nested scales of 4 km, 500 m and 100 m. [Result] Results show that, for any scale, soil salinity in the root-zone of the cotton fields in the zone is generally low, with the mean being in the range of 1.52-1.87 g·kg^-1, moderate in variability, and characterized by continuous variation and accumulation at the bottom of the root zone. Non- and mildly-salinized soils dominate the irrigation zone and are mainly located in the south-east part of the zone, where the surface water is low in salinity and the ground is quite high with unobstructed drainage, while relatively seriously salinized soils are mainly distributed in areas where the ground water is shallow and overflows, and in areas that are affected by the leakage from reservoirs in the plains and trunk canals. As influenced by some structural factors (e.g. terrain landform) and stochastic factors (e.g. artificial activities), soil salt content in this area shows a significant spatial self-correlation in distribution. With increasing scale, nugget-to-sill ratio of the geo-statistical analysis model decreases, and related distance, self-correlation and impact of structural factors increase, but the influence of random factors weakens. On the contrary with declining research scale, spatial distribution becomes more explicit, and the impact of random factors intensifies. [Conclusion] For those relatively seriously salinized areas, it is advisable to employ the nested multi-scale root-zone sampling strategy so as to get a full knowledge about characteristics of the spatial distribution and variability of soil salinity in the drip-irrigated cotton fields under film mulch.