H11G-01 INVITED
Physics-Based Continuous Simulation of Long-Term Near-Surface Hydrologic Response for the Coos Bay Experimental Catchment
The study reported here employed the physics-based InHM to simulate continuous hydrologic response from 1990 through 1996 for the Coos Bay (CB1) experimental catchment. InHM dynamically simulates 3D variably- saturated subsurface flow using Richards equation and 2D surface and open channel flow using the diffusion- wave approximation to the depth-integrated shallow-water equations. The uniqueness of the boundary-value problem (BVP) used in a previous study to successfully simulate three sprinkling experiments was assessed, via model performance evaluation against piezometric and discharge data, for 33 events extracted from the seven- year continuous record. The simulations conducted in this effort suggest the potential for interaction between the deeper water table and near-surface hydrologic response, which is in agreement with the detailed field observations made during the CB1 sprinkling experiments. The InHM simulations could not adequately reproduce the observed pore-water pressures, suggesting that detailed characterization of the locations and connectivities of bedrock fractures would be necessary to simulate distributed hydrologic response at locations where bedrock fracture flow is important. The results from this study suggest that uniqueness is a problem for physics-based models when employing a BVP used successfully for smaller magnitude storms to simulate larger storms. The long-term simulations conducted here, combined with previous event-based hydrologic- response simulations and field-based observations, highlight the challenges in characterizing / simulating fractured bedrock flow at small catchments like CB1.
H11G-02
Characterizing Long-Term Near-Surface Soil Moisture Variability in Natural Environments using Electrical Resistivity Imaging and Hydrological Modeling
The sensitivity of electrical conductivity to changes in moisture content of subsurface materials makes time-lapse electrical resistivity imaging (ERI) an ideal method to monitor transient moisture conditions in natural settings. Recent work has highlighted the usefulness of the ERI method for monitoring infiltration and solute transport processes in the vadose zone, yet most of these studies have been conducted in controlled settings. As a result, only limited knowledge exist on the applicability of ERI for monitoring long-term characteristics of near-surface soil moisture under natural conditions. Here we present results from a year-long application of ERI to monitor and quantify the effects of seasonal climate variability and vegetation dynamics on soil moisture beneath different vegetation types at a natural field site in Mid-Michigan. The site has been equipped with a permanent array of 84 surface resistivity electrodes spanning a 124 meter transect, 4 sets of 14 borehole electrodes, and multiple soil moisture and temperature sensors. Subsurface changes in resistivity were interpreted using standard 1D inversions and 2D difference inversions and have been corrected for temperature fluctuations. Laboratory measurements of the conductivity- soil water content relationships were used in conjunction with differential resistivity inversions to estimate changes in soil moisture distribution. Our combined vadose zone soil water infiltration modeling results and ERI interpretations show that ERI can successfully monitor the influence of seasonal vegetation changes and climatic variability on soil moisture and groundwater recharge. We will discuss challenges and uncertainties associated with the use of ERI method for environmental monitoring and hydrogeological applications.
H11G-03
Characterization of Watershed Model Behavior and Sensitivity-Guided Parameter Estimation Across a Hydroclimatic Gradient
A fundamental tradeoff exists in watershed modeling between a model's flexibility for representing a range of watersheds with varying characteristics versus its potential for over-parameterization. This study uses global sensitivity analysis and multiobjective optimization to evaluate how a medium-complexity model, the Sacramento Soil Moisture Accounting Model (SAC-SMA), represents a wide range of watersheds with varying physical and hydroclimatic characteristics. Initially, Sobol's sensitivity analysis has been used to evaluate SAC-SMA's behavior in twelve Model Parameter Estimation Experiment (MOPEX) watersheds in the US. The watersheds span a wide hydroclimatic gradient from arid to humid systems. Four evaluation metrics reflecting base flows, mid-range flows, peak flows, and long-term water balance were used to comprehensively characterize trends in sensitivity and model behavior. Results show significant variation in parameter sensitivities that are strongly correlated to the hydroclimatic characteristics of the MOPEX watersheds. The sensitivity patterns are consistent with the expected dominant processes in each case and demonstrate the need for moderate model complexity to adequately represent different hydroclimatic regimes. As a final step, the parameter sensitivities have been used to guide an adaptive multiobjective calibration framework and assess the impacts of reducing parameterization requirements for each of the MOPEX watersheds.
H11G-04
Consistency between hydrological models and field observations: Linking processes at the hillslope scale to hydrological responses at the watershed scale.
The purpose of this paper is to identify simple connections between observations of hydrological processes at the hillslope scale and observations of the response of watersheds following rainfall. We focus on the well- studied Panola Mountain Research Watershed (PMRW), Georgia, USA. Recession analysis of discharge Q shows that while the relationship between dQ/dt and Q is approximately consistent with a linear reservoir for the hillslope, there is a deviation from linearity that becomes progressively larger with increasing spatial scale. To account for these scale differences we define conceptual models of streamflow recession at both the hillslope scale and the watershed scale, and assess how models at the hillslope scale can be aggregated to be consistent with models at the watershed scale. Results show that it is possible to reproduce the non-linear behaviour of the watershed following rainfall by adding flow from a small set of linear reservoirs. Indeed, the non-linear recession behavior at the 41-ha Panola watershed can be reproduced by combining linear time constants from the 0.1-ha hillslope and the 10-ha upper catchment with a linear time constant that reflects a slowly draining aquifer. These simple links between linear processes at the hillslope scale and non-linear watershed scale should be considered carefully when calibrating distributed hydrological models, especially in cases when the transmissibility varies considerably between different hydrological response units.
H11G-05
Use of Residence Time and Hydrograph Component Information to Test Effective Parameters in a Physically Based Rainfall-Runoff Model
Many rainfall-runoff models exist that can reproduce the behavior of storm hydrographs. As these models become the basis for landuse and climate change prediction, it becomes essential that they work for the right process reasons. Much ambiguity still exists in how model parameters relate to field measurements and how model structure capture dominant runoff processes. Recent work has demonstrated that stream water mean residence time and storm hydrograph components provide measures of watershed behavior that are orthogonal and non- redundant to the stream hydrograph. However, to date, only conceptual models have been tested in this way. This paper reports on an analysis of a distributed physically based model to represent watershed behavior. We ask the questions: Can effective parameters calibrated from large river basins be used to reproduce small scale watershed behavior in terms of reproducing stream flow hydrograph, residence time and source components? And what level of detail is warranted in the model to represent these components? We used a distributed rainfall-runoff model applied to a small catchment (Maimai catchment, New Zealand, 3.7 ha) and field observed data including stream flow, isotopic / geochemical tracers. We first investigated whether field measured soil properties can be used directly as a model parameter to reproduce reasonable hydrographs at the outlet. We then applied effective parameter sets calibrated from a large river basin to see how the parameter sets could be used at the smaller scale. Finally, we examined the ability to reproduce mean residence time and new/old water component with parameters obtained from soil property and calibrated from the large river basin to assess the model can represent the watershed behavior.
H11G-06
HMF-Geophysics - An Update
There is growing recognition of the challenges we face, in many parts of the world, in finding and maintaining clean sources of water for human consumption and agricultural use, while balancing the needs of the natural world. Advancements in hydrologic sciences are needed in order to develop an improved understanding of the controls on the quantity, movement, and quality of water, thus enhancing our ability to better protect and manage our water resources. Geophysical methods can play a central role in these investigations. CUAHSI (Consortium of Universities for the Advancement of Hydrologic Sciences) is developing, with the support of the National Science Foundation, a Hydrologic Measurement Facility (HMF), which contains a Geophysics module, referred to as HMF-Geophysics. The Geophysics module will support and advance the use of geophysics for hydrologic applications. Currently in second year of a 3 year pilot study, the main aim of HMF-Geophysics is to develop the infrastructure necessary to provide geophysical techniques and the expertise to apply them correctly for the hydrological community. The current working model consists of a central HMF-Geophysics facility and a number of volunteer nodes. The latter consists of individuals at universities who have volunteered to be part of HMF-Geophysics by using their equipment, and/or software, and expertise, in research partnerships with hydrologists. In response to an inquiry the central facility takes on the evaluation of the potential of geophysics to the area of research/watershed. The central facility can then undertake a feasibility study to determine how/if geophysical methods could be of use, and to evaluate the "value-added" by geophysics to the science. Once it is clear that the geophysics can contribute in a significant way to addressing the science questions the central facility works with the hydrologist to set up the next step. Our assumption is that at this point, the hydrologist (perhaps with a geophysics partner) will apply to NSF, or elsewhere, for funding. With the feasibility study complete, they will be in a position to show that geophysics can provide valuable information at their site. This is a very important step, and a key contribution that can be made by HMF-Geophysics. We present here some initial findings from 4 such feasibility studies conducted over summer 2007 in collaboration with a number of the WATERS test-bed sites. These were conducted in a range of environments, from urban to pristine headwater watersheds, with a range of differing hydrogeological aims.
H11G-07
Modeling Hydrologic Response to Land Cover Change in the Inland Pacific Northwest
Although physically based hydrologic models have been applied to understand the mechanisms by which land use change affects watershed hydrology, these models are not always directly transferable from region to region. This is partly because many different mechanisms may be responsible for producing runoff alterations. Perfect fitting of the hydrograph does not necessarily mean that all the internal hydrologic mechanisms have been accurately simulated. A detailed study has been designed to validate internal watershed mechanisms simulated by the Distributed Hydrology Soil Vegetation Model (DHSVM), to assess the hydrologic effects of land use change an interior Pacific Northwest experimental watershed. Hydrological measurements in the experimental area include streamflow, snowpack properties, canopy throughfall, soil moisture, and sap flow to assess the simulated hydrologic components, and hence the model's ability of predict the effects of land cover change. Model simulations span a 5-year pre-treatment, 4-year post-road construction without harvesting, and 5-year post-treatment period to ensure that the model parameterizations accurately quantify the effects of land cover alteration. The validated model was used to make a retrospective simulation of when the entire watershed was clear-cut to predict historical flow regimes. The historical fully clear-cut scenario was then used to provide a baseline to compare to contemporary harvest patterns characterized by sequential canopy removal and regrowth over smaller spatial units. Preliminary modeling results will be presented to illustrate the capability of hydrology model in predicting and forecasting hydrological responses to a range of contemporary forest practices.