[1] Sea level rise during the 21st century will have a wide range of effects on coastal environments, human development, and infrastructure in coastal areas. The broad range of complex factors influencing coastal systems contributes to large uncertainties in predicting long-term sea level rise impacts. Here we explore and demonstrate the capabilities of a Bayesian network (BN) to predict long-term shoreline change associated with sea level rise and make quantitative assessments of prediction uncertainty. A BN is used to define relationships between driving forces, geologic constraints, and coastal response for the U.S. Atlantic coast that include observations of local rates of relative sea level rise, wave height, tide range, geomorphic classification, coastal slope, and shoreline change rate. The BN is used to make probabilistic predictions of shoreline retreat in response to different future sea level rise rates. Results demonstrate that the probability of shoreline retreat increases with higher rates of sea level rise. Where more specific information is included, the probability of shoreline change increases in a number of cases, indicating more confident predictions. A hindcast evaluation of the BN indicates that the network correctly predicts 71% of the cases. Evaluation of the results using Brier skill and log likelihood ratio scores indicates that the network provides shoreline change predictions that are better than the prior probability. Shoreline change outcomes indicating stability (−1 < rate < 1 m/yr) or erosion (rate < −1 m/yr) tend to occur for two sets of input scenarios. Stable shoreline change rates occur mainly for low rates of relative sea level rise and occur in low-vulnerability geomorphic settings. Rates indicating erosion result for cases where the rate of relative sea level rise is high and moderate-to-high vulnerability geomorphic settings occur. In contrast, accretion (rate > 1 m/yr) was not well predicted. We find that BNs can assimilate important factors contributing to coastal change in response to sea level rise and can make quantitative, probabilistic predictions that can be applied to coastal management decisions.Citation: Gutierrez, B. T., N. G. Plant, and E. R. Thieler (2011), A Bayesian network to predict coastal vulnerability to sea level rise,

[1] Resource managers need to make decisions to plan for future environmental conditions, particularly sea level rise, in the face of substantial uncertainty. Many interacting processes factor in to the decisions they face. Advances in process models and the quantification of uncertainty have made models a valuable tool for this purpose. Long-simulation runtimes and, often, numerical instability make linking process models impractical in many cases. A method for emulating the important connections between model input and forecasts, while propagating uncertainty, has the potential to provide a bridge between complicated numerical process models and the efficiency and stability needed for decision making. We explore this using a Bayesian network (BN) to emulate a groundwater flow model. We expand on previous approaches to validating a BN by calculating forecasting skill using cross validation of a groundwater model of Assateague Island in Virginia and Maryland, USA. This BN emulation was shown to capture the important groundwater-flow characteristics and uncertainty of the groundwater system because of its connection to island morphology and sea level. Forecast power metrics associated with the validation of multiple alternative BN designs guided the selection of an optimal level of BN complexity. Assateague island is an ideal test case for exploring a forecasting tool based on current conditions because the unique hydrogeomorphological variability of the island includes a range of settings indicative of past, current, and future conditions. The resulting BN is a valuable tool for exploring the response of groundwater conditions to sea level rise in decision support.

We used a numerical model to investigate how a barrier island groundwater system responds to increases of up to 60 cm in sea level. We found that a sea-level rise of 20 cm leads to substantial changes in the depth of the water table and the extent and depth of saltwater intrusion, which are key determinants in the establishment, distribution and succession of vegetation assemblages and habitat suitability in barrier islands ecosystems. In our simulations, increases in water-table height in areas with a shallow depth to water (or thin vadose zone) resulted in extensive groundwater inundation of land surface and a thinning of the underlying freshwater lens. We demonstrated the interdependence of the groundwater response to island morphology by evaluating changes at three sites. This interdependence can have a profound effect on ecosystem composition in these fragile coastal landscapes under long-term changing climatic conditions. Published 2013. This article is a U.S. Government work and is in the public domain in the USA.

Habitat studies that encompass a large portion of a species' geographic distribution can explain characteristics that are either consistent or variable, further informing inference from more localized studies and improving management successes throughout the range. We identified landscape characteristics at Piping Plover nests at 21 sites distributed from Massachusetts to North Carolina and compared habitat selection patterns among the three designated U.S. recovery units (New England, New York-New Jersey, and Southern). Geomorphic setting, substrate type, and vegetation type and density were determined in situ at 928 Piping Plover nests (hereafter, used resource units) and 641 random points (available resource units). Elevation, beach width, Euclidean distance to ocean shoreline, and least-cost path distance to low-energy shorelines with moist substrates (commonly used as foraging habitat) were associated with used and available resource units using remotely sensed spatial data. We evaluated multivariate differences in habitat selection patterns by comparing recovery unit-specific Bayesian networks. We then further explored individual variables that drove disparities among Bayesian networks using resource selection ratios for categorical variables and Welch's unequal variances t-tests for continuous variables. We found that relationships among variables and their connections to habitat selection were similar among recovery units, as seen in commonalities in Bayesian network structures. Furthermore, nesting Piping Plovers consistently selected mixed sand and shell, gravel, or cobble substrates as well as areas with sparse or no vegetation, irrespective of recovery unit. However, we observed significant differences among recovery units in the elevations, distances to ocean, and distances to low-energy shorelines of used resource units. Birds also exhibited increased selectivity for overwash habitats and for areas with access to low-energy shorelines along a latitudinal gradient from north to south. These results have important implications for conservation and management, including assessment of shoreline stabilization and habitat restoration planning as well as forecasting effects of climate change.