A Comparative Analysis of the Hydrological Performance of Reconstructed and Natural Watersheds.
The considered reconstructed watershed in this study (the flat top of the South Bison Hill, Fort McMurray, Alberta, which is about 6 years old) constitutes a thin layer of a peat-mineral mix (20 cm thick) overlying an 80 cm thick secondary (glacial till) layer on the shale formation, mimicking the natural soil horizons of undisturbed watersheds. As the reconstructed watershed is located in the boreal forest region, a mature boreal forest (Old Aspen site, about 88 years old) located in the Southern Study Area (SSA), BOREAS, Saskatchewan, Canada, is considered as a representative of natural watershed. The A-horizon with 25 cm of sandy loam texture, the B-horizon with 45 cm-thick sandy clay loam, and the C-horizon with 40 cm of a mixture of sandy clay loam and loam are considered in this study.
An existing System Dynamics Watershed (SDW) model (lumped and site-specific) is modified and adapted to model the hydrological processes of the reconstructed and natural watersheds, such as soil moisture, evapotranspiration, and runoff. The models are calibrated and validated on daily time scale using two years data (growing season) in each case. The hydrological processes are simulated reasonably well despite the high complexity involved in the processes of soil moisture dynamics and the evapotranspiration, for both study areas. Using the modified and calibrated models, long term simulations (48 years) are carried out on both the reconstructed and natural watersheds. Vegetation properties are switched between the reconstructed and natural watersheds and two scenarios are generated. Consequently, long term simulations are performed. With the help of a probabilistic approach, the daily soil moisture results are used to address the comparative soil moisture storage capability of the watersheds.
The results indicate that the selected reconstructed watershed is able to provide its designed store-and-release moisture of 160 mm (a requirement of the land capability classification for forest ecosystems in the oil sands) for the vegetation and meteorological moisture demands at a non-exceedance probability of 93%. The comparative study shows that the reconstructed watershed provides less moisture for evapotranspiration requirements than the natural watershed. The reconstructed watershed is able to provide less moisture than the natural watershed for both small and also mature vegetation scenarios. A possible reason for this may be that the reconstructed site is still in the process of restoration and that it may take a few more years to get closer to natural watersheds in terms of the hydrological sustainability. The study also demonstrates the utility of the system dynamics approach of modeling the case study under consideration. The future addition of a vegetation growth model to the hydrological model, and the development of a generic watershed modeling technique would be helpful in decision making and management practices of watershed reclamation.
Advisor:Elshorbagy, Amin; Sharma, Jitendra; Barbour, Lee; Guangul, Seifu; Shook, Kevin
School:University of Saskatchewan
School Location:Canada - Saskatchewan
Source Type:Master's Thesis
Keywords:statistical analysis boreal forests reconstructed watersheds natural hydrological modeling system dynamics approach
Date of Publication:09/05/2008