Reforestation is a primary end use for reconstructed soils following oil sands mining in northern Allberta, Canada. Limited soil water conditions in this climate will restrict plant growth. The objective of this study was to evaluate the effect of soil texture (gradation and layering) on plant available water and consequently on forest productivity for reclaimed coarse textured soils. A previously validated system dynamics (SD) model of soil waterr dynamics was coupled with ecophysiological and biogeochemical processes model, Biome-BGC-SD, to simulate fforest dynamics for different soil profiles. These profiles included contrasting 50 cm textural layers of finer sand and coarser sand in which the sand layers had either a well graded or uniform soil texture. These were compared to homog eneous profiles of the same sands. Two tree species of jack pine (Pinus banksiana Lamb) and trembling aspen (Popu lus tremuloides Michx.) were simulated using a 60-year climatic data base from northern Alberta. Available water h olding capacity (AWHC) was used to identify soil water regime, while leaf area index (LAI) and net primary product ion (NPP) were used as indices of forest productivity. Using the published and previously validated physiological parameters, the Biome-BGC-SD was used to study the responses of forest leaf area index and potential productivit y to AWHC on different soil profiles. Simulated results indicated that layering of uniform fine sand overlying co arse sand could significantly increase AWHC in the 1-m profile for coarse textured soils. This enhanced AWHC co uld result in an increase in forest LAI and NPP. The extent of the increase varied with coarse sand gradation and v egetative types The simulated results showed that the presence of 50 cm of uniform fine sand overlying 50 cm of gra ded coarse sand would provide an effective reclamation prescription to increase AWHC and forest productivity.