Adaptation of a process based model to capture wetland dynamics in the prairie pothole region
The importance of wetlands is well documented for many ecosystem services, including floodwater storage, reduction of sedimentation and nutrient loading, and wildlife and plant diversity. Development of a general framework to model wetlands and their benefits is difficult due to the complex nature of wetland dynamics. The goal of this study is to enhance the functionality of and systematically parameterize the general agroecosystem model, Agricultural Policy/Environmental eXtender (APEX), to simulate wetland hydrology and associated environmental impacts in the prairie pothole region. APEX requires detailed information about landscape position, hydrologic function, soil properties, climate regime, and management, factors that play a large role in natural, restored, and agricultural wetland landscapes. We utilized the extensive body of literature on the hydrology of the Cottonwood Lake area (CWLA) in Stutsman County, ND, along with long-term data collected on climate and water depth from 1980 to 2012. Functionality of the APEX model was enhanced to include snowmelt run-off on frozen soils without water infiltration and evapotranspiration as a function of wetland geometry (perimeter and surface area) and volume. Modeled water volume dynamics were compared to measured water volume dynamics for two wetlands in CWLA, P6 and T9, throughout the 32-year study period.
The APEX model adequately captures the fluctuations in wetland pond depth. The correlation coefficient between measured and simulated values over 32 years is 0.78 for wetland P6 and 0.83 for wetland T9. The APEX model is capable of capturing the general trends in water fluctuations during dry years (less than 400 mm of precipitation per year) and wet years (greater than 550 mm of precipitation per year), however the model currently more accurately captures wet years (RMSE = 0.13) than dry years (RMSE = 0.31). Next, the calibrated APEX model will be tested on two additional highly monitored wetlands, St. Denis National Wildlife Area in Saskatchewan and Orchid Meadows in South Dakota. These sites represent different soil types and precipitation regimes and will be used to test the transferability of the APEX model to other wetland locations within the prairie pothole region. Developing the tools to simulate wetland hydrology is necessary to understand how current conservation practices are impacting wetland function and to interpret how changes in management and climate conditions may impact wetland function and associated ecosystem services.