Estimates of nonpoint source pollution to watersheds in Belize using the AGNPS model.

摘要

There are serious pressures on developing countries to convert forest resources to agriculture that could have significant impacts on watershed hydrology. Simulations of rainstorm events using AGNPS (Agricultural Nonpoint Source) for the Double Run subwatershed and the North Stann Creek watershed show significant risk of high nitrogen (N) and sediment loads to Belize's marine environment. Excessive N and sediment loads may adversely affect marine ecosystems and eventually imperil the largest barrier reef in the western hemisphere. Field sampling show annual N loads from the Double Run subwatershed of approximately 4.96 × 107 kg. Empirical relationship between [N] in the river and depth of rainstorm events had an R squared of 0.94. This correlation provides a reliable, cost-effective estimate of [N] in the river based on easily available precipitation data.; AGNPS predictions of [N] in runoff for specific rainstorms are much lower than experimentally determined values. Reasons for this are: AGNPS apportion nitrogen both in the sediment phase and in runoff, assumptions made for calculating [N] in runoff based on [N] in the river, and the large cell size used for the simulation. However, AGNPS provides reliable predictions for storms of depths between 10 mm and 51 mm (R2 is 0.67).; AGNPS users must consider the following limitations of the model; large data requirements, nutrient concentration is apportioned in runoff and adsorbed to sediment, and subsurface flow is not used in stormflow generation. However, AGNPS can be a very effective watershed management tool if used along with extensive field work.