Modeling seasonally frozen soils with two different datasets to identify the ability to reduce parameter uncertainty
(English)Manuscript (preprint) (Other academic)
Calibration and uncertainty analysis with selected parameters were conducted for two seasonally frozen soils in north China. A trade-off existed in good model performance indicators on water, heat or energy balance. Correlations between parameter values and model performances also showed a trade-off. Uncertainties in obtained parameter distributions were detected due to differences in calibration datasets, as well as complexity in frozen soils. Results showed that even with different datasets in calibration, most of the parameters calibrated showed common ranges. This indicated the availability in using common reasonable parameter sets for simulations in different frozen soils. Except for common ranges for most calibrated parameters, site-specified characteristics were also detected for each site, with totally different parameter ranges given two different datasets for calibration. Parameters related to soil hydraulic conductivity ( ) and surface aerodynamics ( ) were detected to be two site specific parameters in two sites, given datasets and calibration methods in this study. The uncertainty and sensitivity for site specific parameters should also be taken into consideration in choice of reasonable ranges for calibration. More detailed studies on site specific parameters would be of importance for better representations of water and energy balance in different seasonally frozen soils in cold regions.
Merged range; Seasonally frozen soils; Water-energy balance; Uncertainty; Correlation
Environmental Sciences related to Agriculture and Land-use
Research subject Land and Water Resources Engineering
IdentifiersURN: urn:nbn:se:kth:diva-184102OAI: oai:DiVA.org:kth-184102DiVA: diva2:914367
QC 201604052016-03-232016-03-232016-04-05Bibliographically approved