Present and Future Soil Moisture Variations at an Arctic Wetland: Implications for Water Vapor and Carbon Fluxes
Anna Liljedahl1, Larry D. Hinzman2, Craig E. Tweedie3, Donatella Zona4, Walter C. Oechel5
1International Arctic Research Center, University of Alaska Fairbanks, PO Box 753851, Fairbanks, AK, 99775, USA, Phone 907-474-1951, akliljedahl [at] alaska [dot] edu
2International Arctic Research Center, University of Alaska Fairbanks, Fairbanks, AK, USA, lhinzman [at] iarc [dot] uaf [dot] edu
3Biological Sciences, University of Texas El Paso, El Paso, TX, USA, ctweedie [at] utep [dot] edu
4Center for Spatial Technologies and Remote Sensing, University of California Davis, Davis, CA, USA, dzona [at] ucdavis [dot] edu
5Global Change Research Group, San Diego State University, San Diego, CA, USA, oechel [at] sunstroke [dot] sdsu [dot] edu
Soil moisture exerts a strong control on the amounts and types of greenhouse gases released to the atmosphere. Here we aim to reduce the uncertainty of soil water projections at fine scales in order to refine our understanding of the future fate of arctic wetlands as well as the strength of positive feedbacks to the Earth system. We applied the Water Balance Simulation Model ETH (WaSiM-ETH), a deterministic spatially distributed hydrological model, to an intensively studied wetland in northern Alaska. End-of-21st century hydrological projections were forced with the ECHAM5 720 ppm stabilization experiment. The water balance of these wetlands is highly dependent upon small scale topographical features. A large portion of the study watershed is represented by a drained thaw lake basin that exhibits low hydraulic gradients with the main topographical variations represented by high and low centered polygons. The low-centered polygons complicate hydrological modeling efforts by favoring ponding and reducing runoff. Evapotranspiration currently represents the major pathway of water loss from these wetlands, although the rates are somewhat suppressed by the presence of maritime air masses and in some years also by near-surface soil moisture. We applied and validated WaSiM-ETH on measured runoff, evapotranspiration and spatially distributed water table observations. Important processes represented in WaSiM-ETH includes a) a simple empirical formula representing the seasonal freezing and thawing of the active layer that is essential to include in order to successfully simulate the hydrologic regime in permafrost regions, b) a dynamic linkage between soil moisture and evapotranspiration and, c) a surface routing module allowing a dynamic generation of ponds. We show that WaSiM-ETH allows for a realistic representation of the water balance in these wetlands and is a powerful tool to project future hydrological stores and fluxes.