In this talk, we study how varying environmental conditions influence the evolution of seed banks in plants. Our model is a modification of the Wright–Fisher model with finite-age seed bank, introduced by Kaj, Krone and Lascoux. We distinguish between wild type individuals, producing only nondormant seeds, and mutants, producing seeds with finite dormancy. To understand how environments shape the establishment of seed banks, we analyse the process under diffusive scaling. The results support the biological insight that seed banks are favoured under adverse and fluctuating environments. Mathematically, our analysis reduces to a stochastic dynamical system forced onto a manifold by a large drift, which converges under scaling to a diffusion on the manifold. By projecting the system onto its linear counterpart, we derive an explicit formula for the limiting diffusion coefficients. This provides a general framework for deriving diffusion approximations in models with strong drift and nonlinear constraints. This is a joint work with Alison Etheridge.