In desert ecosystems, precipitation shifts dramatically influence biological soil crusts (BSCs) and the phosphorus dynamics within the topsoil layer. However, the impact of variations in winter snowfall, an essential desert water source, on BSCs and phosphorus composition in the soil below remains to be fully understood. Based on this, our research focuses on the crusts formed by Syntrichia caninervis, a species indicative of mature BSCs development. We implemented three distinct treatments: snow removal (-S), natural snowfall (CK), and enhanced snowfall (+S). Through assessing the physicochemical characteristics of the soil within the S.caninervis crust and the underlying 0-5 cm soil layer including measures of nitrate nitrogen, ammonium nitrogen, total nitrogen, organic carbon and total carbon alongside the assessment of phosphorus availability (Resin-P, NaHCO₃-Pi, and NaHCO₃-Po), moderate availability phosphorus (NaOH-Pi, NaOH-Po) and phosphorus with non-availability phosphorus (HCl-P, Residual-P), as well as the activity of enzymes involved in phosphorus cycling (alkaline phosphatase, phosphodiesterase and acid phosphatase), our study seeks to elucidate the effects of snowfall variation on these essential ecological parameters. Our findings reveal that: (1) in comparison to the CK group, the -S and +S treatments significantly impact the phosphorus availability and its utilization difficulty within the S.caninervis crust layer, yet they did not significantly affect the phosphorus makeup in the 0-5 cm sub-crust soil layer. Notably, the -S and +S interventions decreased Resin-P and NaHCO₃-Pi levels while elevating NaHCO₃-Po concentrations within the crust. Furthermore, the -S treatment notably reduced the levels of phosphorus forms that were challenging to utilize (HCl-P and Residual-P), highlighting that snow alterations predominantly affected phosphorus components at the crust layer of S.caninervis. (2) Across the -S, CK, and +S snow gradient, we observed a rising trend in the total inorganic phosphorus content within the crust layer, whereas organic phosphorus content initially increased before diminishing. The 0-5 cm sub-crust soil layer displayed similar patterns for both inorganic and organic phosphorus content, characterized by an initial rise followed by a decline. (3) The predictive modeling via random forest analysis identified soil moisture content, alkaline phosphatase activity, and soil total nitrogen content as the primary determinants influencing phosphorus availability, moderate availability phosphorus and non-availability phosphorus under the studied snow treatments. The results underscore the pivotal role of winter snowfall variations in shaping BSCs and phosphorus dynamics within the surface soil of water constrained desert ecosystems.