|Document Type:||Journal Article|
|Title:||Making the best use of modeled data: multiple approaches to sensitivity analysis of a fish-habitat model|
|Author:||E. Ashley Steel, Paul McElhany, N. J. Yoder, M. D. Purser, K. Malone, B. E. Thompsen, K. A. Avery, D. Jensen, G. Blair, C. Busack, M. D. Bowen, J. Hubble, T. Kantz|
Fisheries management has become increasingly dependent on large and complex models; however, tools and technologies for model evaluation have lagged behind model development and application. Sensitivity analyses can test the degree to which particular model inputs or internal parameters affect model outputs and are recommended in model construction, calibration, and assessment. We describe three parallel sensitivity analyses of the Ecosystem Diagnosis and Treatment (EDT) model and draw combined conclusions. The details of how each agency conducted and utilized sensitivity analyses are outlined and the trade-offs between simpler and more intensive sensitivity analyses are described. Combined insights on the EDT model include identification of input parameters to which the model is surprisingly insensitive and quantification of prediction intervals. We conclude that known uncertainties in input data and internal parameters lead to large prediction intervals around estimates of population abundance and productivity but that the identification of high priority reaches for restoration and protection is relatively robust to these uncertainties. We recommend that sensitivity analyses are applied to all models used in fisheries management.