Predictive mapping for management and conservation of seagrass beds in North Carolina

1. This research extends techniques of predictive mapping from their application in terrestrial environments to marine landscapes by investigating the relationship between seagrass and hydrodynamics in Core Sound, North Carolina, USA. 2. An empirically derived logistic multiple regression model and a Boolean logic suitability model were used to produce several predictive map products, including: susceptibility of seagrasses to storms, probability of seagrass cover, and suitability of areas for restoration of seagrasses. A visual comparison between these maps and conventional seagrass polygon maps allows for a discussion of 'field' versus 'object' mapping, and the ramifications for management based on different cartographic techniques. 3. The predictive method used here showed that only a small portion (19%) of the seagrass bed in the study area would be expected to have a high probability of seagrass coverage. The majority of the seagrass habitat in the study area was predicted to have less than 50% probability of seagrass cover. In addition, 16% of the nearly 2000 ha of seagrass within the study area were predicted to be highly susceptible to acute storm events. Moreover, using a conservative set of site selection criteria, only 7% of the study area encompassed by seagrass habitat was predicted to have a high probability of successful restoration if injured. 4. This method provides for an inexpensive way to scale-up from high-resolution data to a coarser scale that is often required for conservation and management. Copyright (C) 2001 John Wiley Sons, Ltd.