This dataset contains georeferenced multi‑source data including field sampling measurements, satellite‑derived vegetation indices, and atmospheric dust deposition simulations collected in the Hotan Oasis, Xinjiang, China. Field sampling was carried out from 9 to 15 August 2025, covering maize and cotton fields. A total of 77 individual plants were sampled, including 57 maize plants (42 for model construction and 15 for validation) and 15 cotton plants for model development. Measured parameters include foliar dust retention per unit leaf area expressed in grams per square meter (g/m²), determined using standardized field sampling and laboratory weighing procedures. Complementary satellite data consist of the Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) derived from Landsat‑9 (acquired 11 August 2025) and Sentinel‑2 (acquired 21 August 2025). Atmospheric dust deposition flux at 1 km spatial resolution was simulated using the WRF‑Chem model. All data were processed following standardized protocols to ensure consistency and quality. The dataset was produced to examine the spatial heterogeneity of vegetation dust retention and its environmental drivers, and supports further studies in agricultural monitoring, ecological protection, and dust deposition impacts on vegetation in arid desert‑oasis ecosystems.