Water Quality Modeling

Objectives

The objective of this component of SRBEX is to construct a predictive model for the response of stream and river-water composition to forcings resulting from natural and anthropogenic causes. The Water Quality Model (WQM) derives its hydrological input from the Terrestrial Hydrology Model (THM). In addition, it uses a range of spatial data for needed information on bedrock lithology, temperature, and land cover distributions. The predictions it makes are important in their own right; they address the important question of how the quality of surface waters will respond to climate-induced changes in water supply and human-induced changes in land cover. They can also be used to test the hydrological model, providing tracers of various flow paths and contributing areas.

Results

Much of our effort has been focussed on elucidating the relationships between river composition and the controlling variables as a function of watershed size (Bluth and Kump, 1994; Richards and Kump, 1995). Our emphasis so far has been on the weathering-derived cations and bicarbonate. We are looking for scale-dependent and scale-independent characteristics which will allow us to predict changes in water quality at the various scales of the SRBEX experiments. We have compiled historical water-quality and discharge data from the USGS and Pennsylvania Dept. of Environmental Resources for 55 gage stations in the Pennsylvania portion of the SRB (Figure 23). These gage stations represent a characteristic range of watershed sizes, allowing us to assess whether one can build the water and dissolved load budgets at the scale of the full SRB from the subwatershed budgets. We have also produced GIS data-layers for bedrock geology (based on provisional data provided by the Pennsylvania Geological Survey), watershed divides, and land use/land cover. Figure 24 shows a summary of the scale, lithology, and land-use dependencies of magnesium flux. Note the following characteristics: Empirical relationships based on multiple regression analysis have been developed for use in the WQM. For example, the magnesium flux (FMg) was found to be well characterized by the following relationship:


where:

m = 3D fractional coverage by mining

c = 3D fractional coverage by carbonate lithologies

s = 3D fractional coverage by sandstone-shale lithologies

u = 3D fractional coverage by urban area


Figure 25 displays this relationship graphically for the two most important factors, m and c (which explain 70% of the variance). There is considerable discussion in the geochemical community about the temperature-dependence of chemical weathering rates in the field. More generally, we are interested in refining the WQM by including any temperature effects that may exist. Our approach has been both theoretical and empirical. We have developed a numerical model which treats the soil weathering environment as a plug-flow reactor. Temperature-related changes in pore-fluid chemistry are the result not only of the kinetics of reaction (which accelerate at higher temperatures according to the Arrhenius equation) but also of the change in pore-water residence time; as temperature increases the viscosity of water decreases, and thus so too does the residence time of water in the pore. We model the soil-water system as a network of pores of various diameters. As temperatures increase, not only is the residence time reduced, but increasingly smaller pores drain. Preliminary calculations suggest that the temperature dependence on weathering should be 15 kJ/mole less than what is observed in the laboratory (Richards et al., 1995).The empirical approach we have taken is to conduct a field experiment in the Mahantango subwatershed WE-38. Soil lysimeters emplaced along a hillslope and wells intercepting the groundwater system have been sampled over the course of several months to study how the yield of weathering-derived solutes changes in response to seasonal fluctuations in temperature, vegetation cover, rainfall, and storm frequency. The results of these experiments is currently being analyzed.

Future Plans

The focus of our research to date has been on the behavior of the weathering-derived solutes. The primary application of our results is in the areas of kinetics of surfacial processes and global geochemical cycles. We are now beginning the adaptation of the WQM necessary to add a predictive capability for the nutrient elements (nitrogen and phosphorus) and for those species associated with acid precipitation and runoff (e.g., sulfate). These solutes have analogs in the weathering-derived solutes, so we can use what we have learned to date to guide us in evaluating their behavior. Changes in the concentration of nutrients and pollutants are of broader, societal concern, and thus we feel it is appropriate to shift our emphasis in that direction in the future. There is a substantial amount of observational data relevant to these questions which we have begun to accumulate for the SRB. In the next few months we will be analyzing these data in terms of their relationship to the environmental factors described above.

Table of Contents Previous Section Next Section Download Report
Graphics and HTML text markup by Phil Kolb
Comments to: R. White, raw@essc.psu.edu