Links
- Document: Report (html)
- Download citation as: RIS | Dublin Core
Abstract
Stochastic and structural uncertainties about forest dynamics present challenges in the management of ephemeral habitat conditions for endangered forest species. Maintaining critical foraging and breeding habitat for the endangered red-cockaded woodpecker (Picoides borealis) requires an uninterrupted supply of old-growth forest. We constructed and optimized a dynamic forest growth model for the Piedmont National Wildlife Refuge (Georgia, USA) with the objective of perpetuating a maximum stream of old-growth forest habitat. Our model accommodates stochastic disturbances and hardwood succession rates, and uncertainty about model structure. We produced a regeneration policy that was indexed by current forest state and by current weight of evidence among alternative model forms. We used adaptive stochastic dynamic programming, which anticipates that model probabilities, as well as forest states, may change through time, with consequent evolution of the optimal decision for any given forest state. In light of considerable uncertainty about forest dynamics, we analyzed a set of competing models incorporating extreme, but plausible, parameter values. Under any of these models, forest silviculture practices currently recommended for the creation of woodpecker habitat are suboptimal. We endorse fully adaptive approaches to the management of endangered species habitats in which predictive modeling, monitoring, and assessment are tightly linked.
| Publication type | Article |
|---|---|
| Publication Subtype | Journal Article |
| Title | Optimal regeneration planning for old-growth forest: addressing scientific uncertainty in endangered species recovery through adaptive management |
| Series title | Forest Science |
| Volume | 52 |
| Issue | 2 |
| Year Published | 2006 |
| Language | English |
| Contributing office(s) | Patuxent Wildlife Research Center |
| Description | 155-172 |
| Larger Work Type | Article |
| Larger Work Subtype | Journal Article |
| Larger Work Title | Forest Science |
| First page | 155 |
| Last page | 172 |
| Google Analytic Metrics | Metrics page |