Simulated water budget of a small forested watershed in the continental/maritime hydroclimatic region of the United States

Abstract

Annual streamflows have decreased across mountain watersheds in the Pacific Northwest of the United States over the last ~70 years; however, in some watersheds, observed annual flows have increased. Physically based models are useful tools to reveal the combined effects of climate and vegetation on long‐term water balances by explicitly simulating the internal watershed hydrological fluxes that affect discharge. We used the physically based Simultaneous Heat and Water (SHAW) model to simulate the inter‐annual hydrological dynamics of a 4 km2 watershed in northern Idaho. The model simulates seasonal and annual water balance components including evaporation, transpiration, storage changes, deep drainage, and trends in streamflow. Independent measurements were used to parameterize the model, including forest transpiration, stomatal feedback to vapour pressure, forest properties (height, leaf area index, and biomass), soil properties, soil moisture, snow depth, and snow water equivalent. No calibrations were applied to fit the simulated streamflow to observations. The model reasonably simulated the annual runoff variations during the evaluation period from water year 2004 to 2009, which verified the ability of SHAW to simulate the water budget in this small watershed. The simulations indicated that inter‐annual variations in streamflow were driven by variations in precipitation and soil water storage. One key parameterization issue was leaf area index, which strongly influenced interception across the catchment. This approach appears promising to help elucidate the mechanisms responsible for hydrological trends and variations resulting from climate and vegetation changes on small watersheds in the region. Copyright © 2015 John Wiley & Sons, Ltd.

What can we learn from multi-data calibration of a process-based ecohydrological model?

“…It is uncommon in catchment-scale calibration studies to use direct measurements of plot-scale tree transpiration (Du et al, 2014;Wei et al, 2016). This is not only because this type of measurements is rarely available, but also because most current hydrologic models cannot single out the transpiration fraction of evapotranspiration (M endez-Barroso et al, 2014;Paniconi and Putti, 2015).…”

The Added Value of Different Data Types for Calibrating and Testing a Hydrologic Model in a Small Catchment

“…Thyer et al (2004) and Kuras et al (2011) achieved a daily Nash Sutcliffe efficiency for runoff of approximately 0.90, and Thyer et al (2004) stated that runoff simulations were most sensitive to snowmelt characteristics as runoff was driven by spring snowmelt in their high elevated, forested study region. In another study, Wei et al (2016) used measurements of snow water equivalent, snow depth, transpiration, stomatal feedback to vapor pressure, soil and forest properties, and soil moisture to parameterize a physically based model without a flow routing module to simulate the water balance in a 4 km 2 catchment in the United States. Without using runoff data for model calibration, they could reproduce the annual and monthly variabilities of potential runoff (combined outputs of surface runoff and deep drainage) with a Nash Sutcliffe efficiency of 0.62 and 0.56, respectively.…”

Does fluctuation of meteorological conditions across years influence stand transpiration of Tectona grandis plantation?

“…More information about the water‐use of timber species is required to plan commercial plantations with more sustainable utilization of soil water (Aparecido, Miller, Cahill, & Moore, ; Murgue, Therond, & Leenhardt, ). Whole‐tree water use is correlated with tree size (diameter at 1.3‐m height; DBH), proportion of sapwood area, and leaf area index (LAI), which determine the total transpiration surface of a tree (Ma et al, ; Wei et al, ). Leaf phenology strongly affects transpiration through its effect on LAI (Dünisch & Morais, ; McJannet, Fitch, Disher, & Wallace, ).…”