Abstract
This paper analyzes budget-constrained, nonpoint source (NPS) pollution control with costly information acquisition and learning. To overcome the inherent ill-posed statistical problem in NPS pollution data the sequential entropy filter, a cross entropy econometric approach, is applied to the sediment load management program for Redwood Creek, which flows through Redwood National Park in northwestern California. We simulate dynamic budget-constrained management with information acquisition and learning, and compare the results with those from the current policy. The analysis shows that when information acquisition increases overall abatement effectiveness the fiscally constrained manager can reallocate resources from abatement effort to information acquisition, resulting in lower sediment generation than would otherwise exist. In addition, with learning about pollution generation occurring over time the manager may switch from a high intensity of data collection to a lower intensity to further reduce sediment generation. Also, as sediment control proceeds at upstream sources, at some time in the future the marginal reduction in sediment for a given expenditure will equalize across the sources such that uniform abatement effort may occur across all sources.