Application of a GIS-based distributed hydrology model for prediction of forest harvest effects on peak stream flow in the Pacific Northwest

Abstract

Abstract:Spatially distributed rainfall±runo models, made feasible by the widespread availability of land surface characteristics data (especially digital topography), and the evolution of high power desktop workstations, are particularly useful for assessment of the hydrological eects of land surface change. Three examples are provided of the use of the Distributed Hydrology-Soil±Vegetation Model (DHSVM) to assess the hydrological eects of logging in the Paci®c Northwest. DHSVM provides a dynamic representation of the spatial distribution of soil moisture, snow cover, evapotranspiration and runo production, at the scale of digital topographic data (typically 30±100 m). Among the hydrological concerns that have been raised related to forest harvest in the Paci®c Northwest are increases in¯ood peaks owing to enhanced rain-on-snow and spring radiation melt response, and the eects of forest roads. The ®rst example is for two rain-on-snow¯oods in the North Fork Snoqualmie River during November 1990 and December 1989. Predicted maximum vegetation sensitivities (the dierence between predicted peaks for all mature vegetation compared with all clear-cut) showed a 31% increase in the peak runo for the 1989 event and a 10% increase for the larger 1990 event. The main reason for the dierence in response can be traced to less antecedent low elevation snow during the 1990 event. The second example is spring snowmelt runo for the Little Naches River, Washington, which drains the east slopes of the Washington Cascades. Analysis of spring snowmelt peak runo during May 1993 and April 1994 showed that, for current vegetation relative to all mature vegetation, increases in peak spring stream¯ow of only about 3% should have occurred over the entire basin. However, much larger increases (up to 30%) would occur for a maximum possible harvest scenario, and in a small headwaters catchment, whose higher elevation leads to greater snow coverage (and, hence, sensitivity to vegetation change) during the period of maximum runo. The third example, Hard and Ware Creeks, Washington, illustrates the eects of forest roads in two heavily logged small catchments on the western slopes of the Cascades. Use of DHSVM's road runo algorithm shows increases in peak runo for the ®ve largest events in 1992 (average observed stream¯ow of 2 . 1 m 3 s 71 ) averaging 17 . 4% for Hard Creek and 16 . 2% for Ware Creek, with a maximum percentage increase (for the largest event, in Hard Creek) of 27%. #

Modeling the effect of glacier recession on streamflow response using a coupled glacio-hydrological model

“…The main objective of the model is to simulate the spatial distribution of soil moisture, snow cover, evapotranspiration and runoff production over a range of spatial scales, at hourly to daily time intervals. DHSVM uses a two-layer canopy representation for evapotranspiration (overstory and understory), a two-layer energy balance model for snow accumulation and melt, a multilayer unsaturated soil model, and a saturated subsurface flow model (Storck et al, 1998). The two layer energy and mass balance approach in simulating snow accumulation and melt is similar to that described by Anderson (1968), and is described in detail by Andreadis et al (2009).…”

Description and Evaluation of a Hydrometeorological Forecast System for Mountainous Watersheds

“…The version used in this study incorporates ex- plicit channel routing, allowing for the forecasting of streamflow at any point within the channel network. DHSVM includes an energy balance snowpack model that explicitly represents the effect of a forest canopy on snow accumulation and snowmelt (Storck and Lettenmaier 1999). The hydrologic model was configured at 150-m horizontal resolution and all hydrologic forecasts used a 1-h time step.…”

Grays River Watershed and Biological Assessment Final Report 2006.

“…The model has been used also to study the interactions between climate and hydrology (Wigmosta et al 1995;Arola and Lettenmaier 1996;Nijssen et al 1997) and the potential impacts of climate change on water resources (Leung et al 1996;Leung and Wigmosta 1999;Wigmosta and Leung 2001). Furthermore, the model has proven to be an important tool Storck et al 1998Bowling Storck 2000;Bowling and Lettenmaier 2001;Wigmosta and Perkins 2001;Vanshaar and Lettenmaier 2001;Vanshaar et al 2002;Waichler et al 2005). Waichler et al (2005) conducted a detailed test of the DHSVM in small watersheds in the H. J. Andrews Experimental Forest, Oregon, examining treatment effects on streamflow.…”