Proceedings of the 2007 Georgia Basin Puget Sound Research Conference
Session 3C: Modeling and Decision Support Tools II
Chair: Carol Maloy
Development of a High Resolution 3-D Unstructured Grid Circulation Model of the Whidbey Basin in Puget Sound
The Whidbey Basin is Puget Sound’s largest supplier of freshwater and sediments and has been identified as a key region of ecological importance, especially for salmon. It consists of three major estuaries (Skagit River, Stillaguamish River and Snohomish River) that are characterized by large intertidal zones and multiple tidal channels which provide important habitat for the production of Chinook populations within the basin.
The tidal mixing, inundation, and tidal circulation processes in each estuary, and the interaction among these three estuaries play important role on nearshore habitat restoration and fish migratory pathways. This paper describes the development of a three-dimensional (3-D) hydrodynamic model for the Whidbey Basin. A companion paper titled Application of a 3-D Hydrodynamic Model of Whidbey Basin to Assess Cumulative Effects of Restoration Projects in the Snohomish River Estuary provides an example of the utility of the tool developed.
The Whidbey Basin hydrodynamic model was developed using the unstructured, finite volume, 3D coastal ocean model (FVCOM). A fine grid resolution, as small as 10 m, was specified in the in the nearshore areas. The model was driven by tides, salinity and temperature along the open boundaries, and wind and solar heating at the surface. The model was successfully calibrated against observed data for a neap-spring tidal cycle collected during fall of 2006. Model results demonstrated that the Whidbey Basin model has the capability of simulating the physical processes at the required resolution and accuracy to help assess the feasibility of nearshore restoration projects and provide useful information for marine fishery habitat research.
Application of a 3-D Hydrodynamic Model of Whidbey Basin to Assess Cumulative Effects of Restoration Projects in the Snohomish River Estuary
The Whidbey Basin is Puget Sound’s largest supplier of freshwater and sediments and has been identified as a key region of ecological importance, especially for salmon. With a high percentage of shallow, low-gradient, shorelines which are important for juvenile salmonid rearing, Whidbey Basin provides critical habitats to several Chinook populations within this area and supports populations from other parts of Puget Sound. However, over the last century, considerable near-shore tidal marshland habitat within the Basin has been lost due to dike construction.
Efforts are underway to restore the degraded estuary habitat and re-establish migratory pathways for salmon. To help assessing the feasibility of improving these estuarine functions through proposed restoration actions, a hydrodynamic model, specifically suited for near-shore restoration design was developed for the Whidbey Basin. Details of this model are described in a companion paper titled Development of a High Resolution 3-D Unstructured Grid Circulation Model of the Whidbey Basin.
In this paper we describe the application of this model to the Snohomish River estuary. The fresh water plume interaction within the river delta region, inundation of the marsh-lands and the salinity variability in the estuary were simulated for the existing conditions. The model was then applied for the Qwuloolt Marsh, Smith Island, Union Slough, and Biringer Farm restoration projects. Located relatively close to one another near the mouth of the estuary, these projects form one of the largest restoration efforts in Washington. The results provide valuable information regarding potential affects on morphology, and marine habitat conditions under the influence of restoration projects.
Landscape Metrics for Monitoring Landscape Change in Western Washington
This study identifies and applies a set of landscape metrics for monitoring landscape change in Western Washington over the period of 1991 to 2001. The objective was to select metrics that may be used as indicators to quantify and monitor landscape patterns trends at the basin and sub-basin scale. The study is based on a multi-year NOAA Coastal Change Analysis Program (C-CAP) Land Cover, a land cover classification of Landsat Thematic Mapper 5 imagery developed for the National Land Cover Database (NLCD). The analysis was performed at three time-steps (1991, 1996, and 2001) for Western Washington. A multi-scale analysis was conducted using two watershed basin scales using USGS hydrologic units (HUCs), 4 and 6, to evaluate landscape trends at two different scales. We quantified the sensitivity of the selected metrics to changes in land cover and population density that occurred in Western Washington to assess their usefulness as indicators for monitoring regional landscape change.
This work was conducted with support from the U.S. Environment Protection Agency, Region 10, Regional Geographic Initiative.
Sensitivity Analysis of Spatial Population Viability Models for Critical Habitat Identification
Metapopulation dynamics are strongly influenced by the size, quality, and spatial arrangement of populations, the dispersal rates and correlations among populations and the structure of habitat surrounding populations. Varying such spatial parameters in sensitivity analyses of population viability analyses (PVAs) helps evaluate the influence of parameter uncertainty on model predictions, but few studies conduct such analyses, perhaps because no computationally-efficient tools exist to facilitate comprehensive sensitivity analyses for spatial PVAs. This paper presents GRIP 1.0 (Generation of Random Input Parameters), a program that varies both spatial and non-spatial input parameters for PVAs created in RAMAS Metapop 4.0. GRIP is used here to explore the consequences of spatial variation on the persistence of 40 threatened species and rank the influence of spatial and non-spatial parameters on extinction probability using standardized regression coefficients. Our results underscore the importance of including spatial parameters in sensitivity analyses; the most influential parameters were either spatial in nature (e.g., distance among and number of populations) or linked to the number and configuration of populations (e.g., correlation of vital rates among populations). GRIP supports computationally-efficient sensitivity analyses for spatial PVAs for prioritizing research and identifying critical habitat for species that occupy habitat patches of varying size, quality and distribution.
A Simple Population Forecast Model for Purple Martins in British Columbia
Western Purple Martins (Progne subis arboricola Behle 1968) have been the subject of an extensive and highly successful mainly volunteer-based nest box recovery program in the Georgia Depression Ecoprovince of southwest British Columbia since 1985, following major population declines after the mid-1900s, with over 600 pairs nesting in BC in 2006. Since 1997 a scientific monitoring program was implemented involving documented annual nest box inspections at almost all nesting colonies within the province to monitor and track progress of the recovery and obtain reliable data on abundance, productivity, nesting success and fledgling production for this geographically confined population. Analysis of the resulting data accumulated over a decade has allowed development of a simple population forecast model based on nesting pair abundance, fledgling production and annual survival that fairly accurately (within 5-10% so far) predicts return of breeding birds the following year. Input criteria (number of nesting pairs and number of nestlings fledged) are easily obtained from nest box inspections prior to fledging and average overall annual survival (all age classes combined) can be calculated from previous years’ data. These data can be used with average fledging success to interactively predict returns in subsequent years, with accuracy subject to primarily weather-related fluctuations in nesting and fledging success, as well as variation in annual survival.