Spatially explicit Waterbird Agent-based Model Program (SWAMP), Version 2.0

SWAMP provides a tool for wildlife management stakeholders to predict the effects of landscape changes on populations of foragers, specifically focused on waterfowl and other migratory birds. Specifically, we have developed SWAMP as a spatially-explicit agent based model to determine the carrying capacity and energy budgets of waterfowl foraging on moist-soil managed wetlands and flooded riceland during winter in the Central Valley (CV), with a goal to extend the application of SWAMP to other regions and species. SWAMP models the time-budget, foraging activities, and metabolic state of each bird individually throughout the season. While the rules governing patch selection and foraging behavior are user-defined (e.g. forage for 12 hours or until limited by gut capacity), variation in the habitat configuration and individual behavior (e.g. in patch selection and corresponding flight times) lead to interesting emergent dynamics. SWAMP also allows users to directly examine the energy intake and lipid reserve of each individual bird and compare that to the amount of energy it can attain at a given point in the season. The model output can predict the expected number of days a target population can be sustained depending on the amount and composition of available foraging habitats, whether flooded rice, managed moist-soil wetlands, or other landscape types. To meet the objective of creating an integrated decision support tool for this project, we needed to expand, refine and adapt the initial prototype version of SWAMP (Miller et al. 2014) to provide a realistic landscape context and to link southeast Oregon and northeastern California (SONEC) with the Central Valley. Adapting the original SWAMP model required a number of essential modifications and updates. These included: (a) updating the GIS landscapes and determining how habitat patches were identified and delineated; (b) creating a new water handling system to adapt to the very different pattern of water delivery, flooding and availability in the SONEC region (this new approach now makes SWAMP far more adaptable and flexible for other landscapes beyond the original CV model); and (c) revising patch selection rules from the original model to make to make these more realistic; (d) developing a new procedure to model migration.