Integrated wetland management for waterfowl and shorebirds at Mattamuskeet National Wildlife Refuge, North Carolina

Open-File Report 2017-1052
By: , and 

Links

Abstract

Mattamuskeet National Wildlife Refuge (MNWR) offers a mix of open water, marsh, forest, and cropland habitats on 20,307 hectares in coastal North Carolina. In 1934, Federal legislation (Executive Order 6924) established MNWR to benefit wintering waterfowl and other migratory bird species. On an annual basis, the refuge staff decide how to manage 14 impoundments to benefit not only waterfowl during the nonbreeding season, but also shorebirds during fall and spring migration. In making these decisions, the challenge is to select a portfolio, or collection, of management actions for the impoundments that optimizes use by the three groups of birds while respecting budget constraints. In this study, a decision support tool was developed for these annual management decisions.

Within the decision framework, there are three different management objectives: shorebird-use days during fall and spring migrations, and waterfowl-use days during the nonbreeding season. Sixteen potential management actions were identified for impoundments; each action represents a combination of hydroperiod and vegetation manipulation. Example hydroperiods include semi-permanent and seasonal drawdowns, and vegetation manipulations include mechanical-chemical treatment, burning, disking, and no action. Expert elicitation was used to build a Bayesian Belief Network (BBN) model that predicts shorebird- and waterfowl-use days for each potential management action. The BBN was parameterized for a representative impoundment, MI-9, and predictions were re-scaled for this impoundment to predict outcomes at other impoundments on the basis of size. Parameter estimates in the BBN model can be updated using observations from ongoing monitoring that is part of the Integrated Waterbird Management and Monitoring (IWMM) program.

The optimal portfolio of management actions depends on the importance, that is, weights, assigned to the three objectives, as well as the budget. Five scenarios with a variety of objective weights and budgets were developed. Given the large number of possible portfolios (1614), a heuristic genetic algorithm was used to identify a management action portfolio that maximized use-day objectives while respecting budget constraints. The genetic algorithm identified a portfolio of management actions for each of the five scenarios, enabling refuge staff to explore the sensitivity of their management decisions to objective weights and budget constraints.

The decision framework developed here provides a transparent, defensible, and testable foundation for decision making at MNWR. The BBN model explicitly structures and parameterizes a mental model previously used by an expert to assign management actions to the impoundments. With ongoing IWMM monitoring, predictions from the model can be tested, and model parameters updated, to reflect empirical observations. This framework is intended to be a living document that can be updated to reflect changes in the decision context (for example, new objectives or constraints, or new models to compete with the current BBN model). Rather than a mandate to refuge staff, this framework is intended to be a decision support tool; tool outputs can become part of the deliberations of refuge staff when making difficult management decisions for multiple objectives.

Suggested Citation

Tavernia, B.G., Stanton, J.D., and Lyons, J.E., 2017, Integrated wetland management for waterfowl and shorebirds at Mattamuskeet National Wildlife Refuge, North Carolina: U.S. Geological Survey Open-File Report 2017–1052, 43 p., https://doi.org/10.3133/ofr20171052.

ISSN: 2331-1258 (online)

Study Area

Table of Contents

  • Abstract 
  • Introduction
  • Purpose and Scope
  • Objectives
  • Alternatives
  • Predictive Models
  • Tradeoffs Using Portfolio Analysis
  • Future Changes to the Decision Framework
  • References Cited
  • Appendix 1. Glossary of Hydroperiod Terms
  • Appendix 2. Waterfowl Habitat Modeling
  • Appendix 3. Building Predictive Models with Expert Judgment 
  • Appendix 4. Expert Elicitation of Conditional Probability Tables
  • Appendix 5. Bird-Use Day Estimates
  • Appendix 6. Genetic Algorithm Approach to Portfolio Analysis 
  • Appendix 7. Management Action Costs
Publication type Report
Publication Subtype USGS Numbered Series
Title Integrated wetland management for waterfowl and shorebirds at Mattamuskeet National Wildlife Refuge, North Carolina
Series title Open-File Report
Series number 2017-1052
DOI 10.3133/ofr20171052
Year Published 2017
Language English
Publisher U.S. Geological Survey
Publisher location Reston, VA
Contributing office(s) Patuxent Wildlife Research Center, Eastern Ecological Science Center
Description vii, 43 p.
Country United States
State North Carolina
Other Geospatial Mattamuskeet National Wildlife Refuge
Online Only (Y/N) Y
Additional Online Files (Y/N) N
Google Analytic Metrics Metrics page
Additional publication details