Ecohydrologic modeling in a boreal deciduous forest: model evaluation for application in non-stationary climates
Abstract
In upland boreal regions, deciduous forest types are projected to become increasingly prevalent in future climates with more frequent fire, but soil moisture may limit their growth. To quantify energy and water balance and soil moisture dynamics in a deciduous boreal forest, we applied the physically-based, one-dimensional ecohydrological Simultaneous Heat and Water (SHAW) model, loosely coupled with the Geophysical Institute Permafrost Laboratory (GIPL) permafrost model, to a post-fire deciduous forest stand in interior Alaska over a 13-year period. Using a Monte Carlo model parameterisation and global uncertainty analysis, we found that SHAW-GIPL reproduced soil moisture dynamics well across multiple years and depths. We identified many parameter sets that performed comparably during a four-year calibration period, suggesting the presence of equifinality. Model performance declined in a nine-year validation period, indicating some overfitting and demonstrating the need to account for interannual variability in model evaluation. A parameter set chosen solely to maximize performance during the calibration period showed greater error in very wet/warm years, implying increased uncertainty under future climate scenarios. In contrast, a second parameter set was selected to reduce the decrease in model performance from the calibration to validation period. This set achieved comparable performance to the initial parameter set during the calibration period and significantly reduced the sensitivity of model performance to interannual climate variability. These findings illustrate (1) that the SHAW model, coupled with GIPL, can adequately simulate soil moisture dynamics in this boreal deciduous region, (2) the importance of interannual variability in model parameterisation, and (3) a simple method for selecting model parameters to improve applicability in non-stationary climates. Building on this work, SHAW and GIPL will be coupled with the LANDIS-II forest simulation model to improve our understanding of how vegetation, soil moisture, and permafrost dynamics may be altered under climate change in interior Alaska.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2020
- Bibcode:
- 2020AGUFMB022.0002M
- Keywords:
-
- 0414 Biogeochemical cycles;
- processes;
- and modeling;
- BIOGEOSCIENCES;
- 0428 Carbon cycling;
- BIOGEOSCIENCES;
- 0702 Permafrost;
- CRYOSPHERE