The resorption of nutrients from senescent leaves allows plants to conserve and recycle nutrients. To explore the adaptation strategies of desert plants to nutrient-limited environments, we selected four typical desert plants (Populus euphratica Oliv., Tamarix ramosissima Ledeb., Glycyrrhiza inflata Batal., and Alhagi camelorum Fisch.) growing in the desert area of the northern margin of the Tarim Basin, China. The contents of nitrogen (N), phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), and Ferrum (Fe) in the leaves of these four typical desert plants and their resorption characteristics were analyzed. The relationship of nutrient resorption efficiency with leaf functional traits and soil physical-chemical properties in two different habitats (saline-alkali land and sandy land) was discussed. The results showed that the four plants resorbed most of the elements. Ca was enriched in the leaves of P. euphratica, G. inflate, and A. camelorum; Mg was enriched in the leaves of G. inflata; and Fe was enriched in the leaves of the four plants. The results of the redundancy analysis showed that leaf thickness, soil electrical conductivity, and soil P content were the major factors affecting the nutrient resorption efficiency of the four plants. Leaf thickness was negatively correlated with N resorption efficiency (NRE), P resorption efficiency, and Fe resorption efficiency; soil electrical conductivity was positively correlated with the resorption efficiency of most elements; and soil P content was negatively correlated with the resorption efficiency of most elements in the plant leaves. The results showed that soil physical-chemical properties and soil nutrient contents had an important impact on the nutrient resorption of plant leaves. The same species growing in different habitats also differed in their resorption of different elements. The soil environment of plants and the biological characteristics of plant leaves affected the resorption of nutrient elements in different plants. The purpose of this study is to provide small-scale data support for the protection of ecosystems in nutrient-deficient areas by studying leaf functional strategies and nutrient conservation mechanisms of several typical desert plants.