Panola Mountain Research Watershed Research Articles

A data-driven method for estimating the composition of end-members from stream water chemistry time series

Abstract. End-member mixing analysis (EMMA) is a method of interpreting stream water chemistry variations and is widely used for chemical hydrograph separation. It is based on the assumption that stream water is a conservative mixture of varying contributions from well-characterized source solutions (end-members). These end-members are typically identified by collecting samples of potential end-member source waters from within the watershed and comparing these to the observations. Here we introduce a complementary data-driven method (convex hull end-member mixing analysis – CHEMMA) to infer the end-member compositions and their associated uncertainties from the stream water observations alone. The method involves two steps. The first uses convex hull nonnegative matrix factorization (CH-NMF) to infer possible end-member compositions by searching for a simplex that optimally encloses the stream water observations. The second step uses constrained K-means clustering (COP-KMEANS) to classify the results from repeated applications of CH-NMF and analyzes the uncertainty associated with the algorithm. In an example application utilizing the 1986 to 1988 Panola Mountain Research Watershed dataset, CHEMMA is able to robustly reproduce the three field-measured end-members found in previous research using only the stream water chemical observations. CHEMMA also suggests that a fourth and a fifth end-member can be (less robustly) identified. We examine uncertainties in end-member identification arising from non-uniqueness, which is related to the data structure, of the CH-NMF solutions, and from the number of samples using both real and synthetic data. The results suggest that the mixing space can be identified robustly when the dataset includes samples that contain extremely small contributions of one end-member, i.e., samples containing extremely large contributions from one end-member are not necessary but do reduce uncertainty about the end-member composition.

Representation of Bi‐Directional Fluxes Between Groundwater and Surface Water in a Bucket‐Type Hydrological Model

AbstractIn most bucket‐type hydrological models, water can only flow from the groundwater to the stream and the flux is based on the groundwater storage. However, many catchments have losing stream sections, where streamflow recharges the groundwater. We developed a formulation to represent groundwater recharge by streamwater in a bucket‐type model and tested this formulation for the Panola Mountain Research Watershed to demonstrate its function and assess its performance. The upper reach of the Panola catchment is often dry and highly affected by flow from a bedrock outcrop; further downstream the stream is perennial. We simulated streamflow with the fully lumped version of a bucket‐type model and compared it to (a) a variant with sub‐catchments to more realistically represent the low storage and quick response from the bedrock outcrop and (b) a variant that also includes the bi‐directional exchange between the groundwater and the stream. For all three model variants, we compared simulated and observed streamflow and groundwater dynamics. Although the gain in overall model performance by including the bi‐directional exchange between the groundwater and the stream was small, the explicit representation of this exchange led to better streamflow simulations during drying‐down and wetting‐up periods. For Panola the fluxes along the stream appeared less important than subsurface drainage from the upper sub‐catchment to the downstream sub‐catchment. We recommend considering the bi‐directional fluxes between groundwater and the stream in bucket‐type hydrological models where these processes are important, and the focus of the simulations is on low flow conditions.

The evolving perceptual model of streamflow generation at the Panola Mountain Research Watershed

AbstractThe Panola Mountain Research Watershed (PMRW) is a 41‐hectare forested catchment within the Piedmont Province of the Southeastern United States. Observations, experimentation, and numerical modelling have been conducted at Panola over the past 35 years. But to date, these studies have not been fully incorporated into a more comprehensive synthesis. Here we describe the evolving perceptual understanding of streamflow generation mechanisms at the PMRW. We show how the long‐term study has enabled insights that were initially unforeseen but are also unachievable in short‐term studies. In particular, we discuss how the accumulation of field evidence, detailed site characterization, and modelling enabled a priori hypotheses to be formed, later rejected, and then further refined through repeated field campaigns. The extensive characterization of the soil and bedrock provided robust process insights not otherwise achievable from hydrometric measurements and numerical modelling alone. We focus on two major aspects of streamflow generation: the role of hillslopes (and their connection to the riparian zone) and the role of catchment storage in controlling fluxes and transit times of water in the catchment. Finally, we present location‐independent hypotheses based on our findings at PMRW and suggest ways to assess the representativeness of PMRW in the broader context of headwater watersheds.

Use of an Efficient Proxy Solution for the Hillslope‐Storage Boussinesq Problem in Upscaling of Subsurface Stormflow

AbstractThe hillslope‐storage Boussinesq (hsB) equation is a modification of the one‐dimensional Boussinesq representation of saturated subsurface flow that accommodates the impact of converging and diverging flows resulting from a hillslope with nonconstant width. Here, a surrogate model, the hsB Proxy, is developed which closely emulates the results of high‐quality numerical solutions of the hsB to reproduce the drainage response of a wide range of wedge‐shaped hillslopes in response to a recharge time series at a fraction of the computational cost of a numerical solution and with minimal loss of fidelity. The efficacy of the Proxy in emulating the hsB response of a single hillslope and the aggregate response of a network of hillslopes is demonstrated. The Proxy applies to wedge‐shaped hillslopes with homogenous hydraulic conductivity under uniform recharge, with the expectation that these hillslopes may inform future investigations into the upscaling behavior that results in the known power‐law relationship between subsurface storage and outflow in a basin. The utility of the hsB as a tool for evaluating upscaling problems is demonstrated in two applications. First, the Proxy response is compared to the known basin‐scale recession behavior in the Panola Mountain Research Watershed, under both transient and steady‐state conditions. Second, the Proxy is applied to derive the single best effective hillslope in a basin. The Proxy is expected to be a useful tool for rapid simulation of subsurface flow from hillslopes for use in hydrological models and land surface schemes.

Effects of climate‐related variability in storage on streamwater solute concentrations and fluxes in a small forested watershed in the Southeastern United States

AbstractStreamwater quality can be affected by climate‐related variability in hydrologic state, which controls flow paths and affects biogeochemical processes. Thirty‐one years of input/output solute fluxes at Panola Mountain Research Watershed, a small, forested, seasonally water‐limited watershed near Atlanta, Georgia, were used to quantify the effects of climatic‐related variability in storage on streamwater solute concentrations and fluxes. Streamwater fluxes were estimated for ten solutes from weekly and event sample concentrations using regression‐based methods. The most pertinent storage attribute (current or antecedent watershed, shallow, and deep storage) for each solute was determined by fitting separate concentration relationships. The concentration‐discharge relationships varied more for reactive solutes such as potassium, sulfate, and DOC and less for weathering products (base cations and dissolved silica) and conservative chloride. Many solutes exhibited higher concentrations when storage levels were lower or wetting up, which was likely the result of the concentrating effects of evapotranspiration and/or the buildup and flushing of weathering products associated with longer residence times. The impacts of storage modeling on annual fluxes varied by solute, ranging from about 5% (magnesium) to 52% (nitrate) as relative standard deviations, and sufficiently removed climate‐related patterns observed in streamwater concentrations. Sulfate was particularly mobilized following growing season droughts but only if deep storage was sufficiently recharged, possibly indicating that sulfides in the deep storage pool were oxidized to sulfate during droughts and mobilized when re‐wetted. The lack of streamwater sulfate response to 61% declines in atmospheric deposition indicates the importance of watershed biogeochemical processes on controls of streamwater export of sulfate. The approach of explicitly incorporating storage in the streamwater concentration modeling elucidated the effects of climate on streamwater water‐quality and may provide insight into the effects of climatic change on future fluxes.

Quantifying Climate‐Related Interactions in Shallow and Deep Storage and Evapotranspiration in a Forested, Seasonally Water‐Limited Watershed in the Southeastern United States

AbstractThe Southeastern United States experiences recurring hydrological droughts, which can reduce water availability and result in water deficits. Long‐term monitoring data from Panola Mountain Research Watershed, a small, forested, seasonally water‐limited watershed near Atlanta, GA, was used to quantify interactions of climatic variability with shallow and deep storage and evapotranspiration. Watershed storage (WS) and actual evapotranspiration (AET) were estimated monthly from 1985 through 2015 using a relationship with baseflow and a water‐budget approach, respectively. A watershed shallow storage (SS) water budget, based on soil moisture profile data, indicated a transition from recharge to surplus as dormant season SS increased to field capacity, and a transition from utilization of stored water to water deficits as growing season SS declined toward its wilting point. Deeper storage appeared unavailable to AET during dry conditions. The majority of deeper storage recharge occurred during the dormant season and required SS to be near field capacity. WS was an effective drought indicator. Growing season droughts typically occurred when WS was below normal at the end of the dormant season and growing season precipitation (P) was below or near normal. A hydrologic persistence analysis showed that monthly month‐standardized WS was significantly correlated (p‐value < 0.05) with past monthly month‐standardized WS for the previous 19 months and with past monthly P for the previous 11 months, indicating the importance of past hydrologic conditions on WS. These findings allowed assessment of expected climatic changes on the water cycle, which included effects on future recharge, water deficits, and drought occurrence.

Approaches to stream solute load estimation for solutes with varying dynamics from five diverse small watersheds

AbstractEstimating streamwater solute loads is a central objective of many water‐quality monitoring and research studies, as loads are used to compare with atmospheric inputs, to infer biogeochemical processes, and to assess whether water quality is improving or degrading. In this study, we evaluate loads and associated errors to determine the best load estimation technique among three methods (a period‐weighted approach, the regression‐model method, and the composite method) based on a solute's concentration dynamics and sampling frequency. We evaluated a broad range of varying concentration dynamics with stream flow and season using four dissolved solutes (sulfate, silica, nitrate, and dissolved organic carbon) at five diverse small watersheds (Sleepers River Research Watershed, VT; Hubbard Brook Experimental Forest, NH; Biscuit Brook Watershed, NY; Panola Mountain Research Watershed, GA; and Río Mameyes Watershed, PR) with fairly high‐frequency sampling during a 10‐ to 11‐yr period. Data sets with three different sampling frequencies were derived from the full data set at each site (weekly plus storm/snowmelt events, weekly, and monthly) and errors in loads were assessed for the study period, annually, and monthly. For solutes that had a moderate to strong concentration–discharge relation, the composite method performed best, unless the autocorrelation of the model residuals was <0.2, in which case the regression‐model method was most appropriate. For solutes that had a nonexistent or weak concentration–discharge relation (model R2 < about 0.3), the period‐weighted approach was most appropriate. The lowest errors in loads were achieved for solutes with the strongest concentration–discharge relations. Sample and regression model diagnostics could be used to approximate overall accuracies and annual precisions. For the period‐weighed approach, errors were lower when the variance in concentrations was lower, the degree of autocorrelation in the concentrations was higher, and sampling frequency was higher. The period‐weighted approach was most sensitive to sampling frequency. For the regression‐model and composite methods, errors were lower when the variance in model residuals was lower. For the composite method, errors were lower when the autocorrelation in the residuals was higher. Guidelines to determine the best load estimation method based on solute concentration–discharge dynamics and diagnostics are presented, and should be applicable to other studies.

On the transition of base flow recession from early stage to late stage

This paper is focused on the transition of base flow recession from early stage to late stage. The volume flow rate that takes place when such a transition occurs is identified for each of the twenty-three recession events observed at the Panola Mountain Research Watershed (PMRW) in Georgia, USA, using a newly developed cumulative regression analysis method. Meanwhile, the flow at the watershed outlet, which was recorded when the discharge at the perennial stream head diminishes to zero, is identified for each recession event. As evidenced by a correlation coefficient of 0.90, these two characteristic flows are found to be highly correlated, suggesting a fundamental linkage between the transition of base flow recession from early stage to late stage and the drying up of ephemeral streams. During the early stage, the contraction of ephemeral streams largely controls the recession behavior, whereas in the late stage when perennial streams dominate the flowing streams, the contraction of flowing streams is minimal and groundwater hydraulics governs the recession behavior.

Hillslope–riparian‐stream connectivity and flow directions at the Panola Mountain Research Watershed

AbstractTo obtain a better understanding of water's journey from the hillslope to the stream, water level time series from 26 groundwater wells in a hillslope–riparian study area in the Panola Mountain Research Watershed, Georgia, USA, were examined with respect to the occurrence of perched groundwater and groundwater flow directions. A connected groundwater table was taken as an indication of a hydrologic connection between the hillslope and the stream. Flow directions were determined based on observed water levels at different combinations of three neighbouring wells. The analyses indicated that the hillslope was disconnected from the stream most of the time but that almost the entire hillslope became connected to the stream during large rainfall events. This coincided with more sustained streamflow during these events. There were clear differences in the timing of the groundwater response between the upper hillslope, lower hillslope and riparian zone, with water levels near the stream and on the upper hillslope rising earlier than on the midslope. The flow directions were generally downslope and were most variable around a bedrock depression on the midslope. On the hillslope, the direction of the water table followed the local bedrock topography when the water level was low at the start and end of an event and the overall surface topography when water levels were high. In the riparian zone, flow directions either did not change significantly during events or turned from a more downstream (i.e. downvalley) direction to a more downslope (i.e. towards the stream) direction. These results highlight the competing influence of surface and bedrock topography on hillslope flow directions and improve our understanding of hillslope–riparian‐stream connectivity. Copyright © 2015 John Wiley & Sons, Ltd.

Evaluation of High-Frequency Mean Streamwater Transit-Time Estimates Using Groundwater Age and Dissolved Silica Concentrations in a Small Forested Watershed

Many previous investigations of mean streamwater transit times (MTT) have been limited by an inability to quantify the MTT dynamics. Here, we draw on (1) a linear relation (r2 = 0.97) between groundwater 3H/3He ages and dissolved silica (Si) concentrations, combined with (2) predicted streamwater Si concentrations from a multiple-regression relation (R2 = 0.87) to estimate MTT at 5-min intervals for a 23-year time series of streamflow [water year (WY) 1986 through 2008] at the Panola Mountain Research Watershed, Georgia. The time-based average MTT derived from the 5-min data was ~8.4 ± 2.9 years and the volume-weighted (VW) MTT was ~4.7 years for the study period, reflecting the importance of younger runoff water during high flow. The 5-min MTTs are normally distributed and ranged from 0 to 15 years. Monthly VW MTTs averaged 7.0 ± 3.3 years and ranged from 4 to 6 years during winter and 8–10 years during summer. The annual VW MTTs averaged 5.6 ± 2.0 years and ranged from ~5 years during wet years (2003 and 2005) to >10 years during dry years (2002 and 2008). Stormflows are composed of much younger water than baseflows, and although stormflow only occurs ~17 % of the time, this runoff fraction contributed 39 % of the runoff during the 23-year study period. Combining the 23-year VW MTT (including stormflow) with the annual average baseflow for the period (~212 mm) indicates that active groundwater storage is ~1,000 mm. However, the groundwater storage ranged from 1,040 to 1,950 mm using WY baseflow and WY VW MTT. The approach described herein may be applicable to other watersheds underlain by granitoid bedrock, where weathering is the dominant control on Si concentrations in soils, groundwater, and streamwater.

The effect of spatial throughfall patterns on soil moisture patterns at the hillslope scale

Abstract. Improving the understanding of the controls on subsurface stormflow generation has been the goal of numerous experimental and modeling studies. However, the effect of the spatial variability of throughfall on soil moisture patterns and subsurface stormflow (SSF) generation has not yet been studied in detail. The objectives of this study are three-fold: (1) to investigate the influence of a spatially variable throughfall pattern on soil moisture; (2) to investigate if soil moisture patterns reflect a balance between a throughfall and bedrock topography patterns; and (3) to investigate how this balance changes when soil depth, storm size and slope angle are varied. Virtual experiments are used to address these questions. A virtual experiment is a numerical experiment driven by collective field intelligence. It provides a learning tool to investigate the effect of individual processes in a complex system. In our virtual experiment we combined spatial throughfall data from the Huewelerbach catchment in Luxembourg with the topography of a well-studied hillslope within the Panola Mountain Research Watershed, Georgia, USA. We used HYDRUS-3D as a modeling platform. The virtual experiment shows that throughfall patterns influence soil moisture patterns, but only during and shortly after a storm. With a semi-variogram analysis we showed how the effective range of the soil moisture pattern (i.e., the main descriptor of a spatial pattern in case of a small nugget to sill ratio), is similar to the effective range of the throughfall pattern during the storm and gradually returns to the effective range of the bedrock topography after throughfall has ceased. The same analysis was carried out to investigate how this balance changes due to changes in storm size, soil depth, and slope. The analysis showed that the throughfall pattern is more important during large storms on gentle slopes. For steeper slopes the bedrock topography becomes more important.

Water storage at the Panola Mountain Research Watershed, Georgia, USA

AbstractStorage is a major component of a catchment water balance particularly when the water balance components are evaluated on short time scales, that is, less than annual. We propose a method of determining the storage–discharge relation using an exponential function and daily precipitation, potential evapotranspiration (PET) and baseflow during the dormant season when evapotranspiration (ET) is low. The method was applied to the 22‐year data series of the 0.41‐ha forested Panola Mountain Research Watershed, Georgia. The relation of cumulative daily precipitation minus daily runoff and PET versus baseflow was highly significant (r2 = 0.92, p < 0.0001), but the initial storage for each year varied markedly. For the 22‐year study period, annual precipitation and runoff averaged 1240 and 380 mm, respectively, whereas the absolute catchment storage range was ~400 mm, averaging 219 mm annually, which is attributed to contributions of soil water and groundwater. The soil moisture of a catchment average 1‐m soil depth was evaluated and suggests that there was an active (changes in soil storage during stormflow) and passive (a longer‐term seasonal cycle) soil water storage with ranges of 40–70 and 100–120 mm, respectively. The active soil water storage was short term on the order of days during and immediately after rainstorms, and the passive or seasonal soil storage was highest during winter when ET was lowest and lowest during summer when ET was highest. An estimate of ET from daily changes in soil moisture (ETSM) during recessions was comparable with PET during the dormant season (1.5 mm day−1) but was much lower during the growing season (June through August); monthly average SMET and PET ranged from 2.8 to 4.0 mm day−1 and from 4.5 to 5.5 mm day−1, respectively. The growing season difference is attributed to the overestimation of PET. ETSM estimates were comparable with those derived from hillslope water balances during sprinkling experiments. Master recession curves derived from the storage–discharge relation adjusted seasonally for ET (1.5 and 4.0 mm day−1 during the dormant and growing seasons, respectively) fit actual recessions extremely well. Copyright © 2011 John Wiley & Sons, Ltd.

On the base flow recession at the Panola Mountain Research Watershed, Georgia, United States

There are two types of nonlinearity for recession slope curves: (1) the data cloud of observed ln(–dQ/dt) versus ln(Q) along a straight line where the slope is not equal to 1, and (2) the data cloud of observed ln(–dQ/dt) versus ln(Q) is concave or convex. From the 0.1 ha hillslope, the 10 ha headwater subwatershed, and the 41 ha watershed at the Panola Mountain Research Watershed (PMRW), the observed ln(–dQ/dt) versus ln(Q) changes from convex to concave. This has been explained by the spatial heterogeneity of recession properties and nonlinearity of hillslope hydraulics. Hillslope bedrock leakage and return flow to the perennial riparian zone aquifer are important factors affecting the ln(–dQ/dt) versus ln(Q) relations, and an assessment of this mechanism for explaining the observed base flow recessions at the PMRW is the sole focus of this research. As a result, the lower envelope and the upper envelope should be used to estimate the recession parameters for the 0.1 ha hillslope and the 41 ha watershed, respectively, and the recession at the 10 ha watershed is a mixture of them. Based on the recession analysis, the leakage to the bedrock on the hillslope and return flow to the perennial stream riparian groundwater can be estimated, and the estimation can be verified by other studies.

Gypsies in the palace: experimentalist's view on the use of 3‐D physics‐based simulation of hillslope hydrological response

AbstractAs a fundamental unit of the landscape, hillslopes are studied for their retention and release of water and nutrients across a wide range of ecosystems. The understanding of these near‐surface processes is relevant to issues of runoff generation, groundwater–surface water interactions, catchment export of nutrients, dissolved organic carbon, contaminants (e.g. mercury) and ultimately surface water health. We develop a 3‐D physics‐based representation of the Panola Mountain Research Watershed experimental hillslope using the TOUGH2 sub‐surface flow and transport simulator. A recent investigation of sub‐surface flow within this experimental hillslope has generated important knowledge of threshold rainfall‐runoff response and its relation to patterns of transient water table development. This work has identified components of the 3‐D sub‐surface, such as bedrock topography, that contribute to changing connectivity in saturated zones and the generation of sub‐surface stormflow. Here, we test the ability of a 3‐D hillslope model (both calibrated and uncalibrated) to simulate forested hillslope rainfall‐runoff response and internal transient sub‐surface stormflow dynamics. We also provide a transparent illustration of physics‐based model development, issues of parameterization, examples of model rejection and usefulness of data types (e.g. runoff, mean soil moisture and transient water table depth) to the model enterprise. Our simulations show the inability of an uncalibrated model based on laboratory and field characterization of soil properties and topography to successfully simulate the integrated hydrological response or the distributed water table within the soil profile. Although not an uncommon result, the failure of the field‐based characterized model to represent system behaviour is an important challenge that continues to vex scientists at many scales. We focus our attention particularly on examining the influence of bedrock permeability, soil anisotropy and drainable porosity on the development of patterns of transient groundwater and sub‐surface flow. Internal dynamics of transient water table development prove to be essential in determining appropriate model parameterization. Copyright © 2010 John Wiley & Sons, Ltd.

Power law catchment‐scale recessions arising from heterogeneous linear small‐scale dynamics

Power law recession behavior has typically been attributed to the hydraulics of subsurface flows in the landscape (e.g., the Boussinesq equation). Many catchments exhibit enormous heterogeneity in their characteristics, such as the permeability of soils, hillslope length and gradient, and soil mantle thickness. In the paper we demonstrate using a simple linear flow model that the between‐hillslope heterogeneity alone can give rise to observed apparently nonlinear recession slope curves. The degree of nonlinearity (represented by the exponent fitted to the recession slope curve) is shown to increase with the degree of heterogeneity of catchment hydraulic properties. In sufficiently heterogeneous catchments, the recession is also sensitive to the history of recharge events, tending to increase when the past events have built up a storage of water in the ‘slow’ parts of the catchment. The models based on the catchment heterogeneity were fitted to data from the Panola Mountain Research watershed, yielding a good fit, and allowing us to infer how the heterogeneity in catchment properties varies with scale. The range of exponents explained by this theory overlaps with the range explained by the hydraulics of subsurface flow, which raises doubts about our ability to extract catchment hydraulic parameters on the basis of recession curves so long as the heterogeneity is not accounted for. In general the results presented here have enabled us to show how the heterogeneity of landscape characteristics is upscaled and manifested in the apparent nonlinear relationship at the catchment scale.

Consistency between hydrological models and field observations: linking processes at the hillslope scale to hydrological responses at the watershed scale

AbstractThe 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, with a view to building a parsimonious model of catchment processes. The focus is 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 conceptual models of streamflow recession are defined at both the hillslope scale and the watershed scale, and an assessment made as to whether models at the hillslope scale can be aggregated to be consistent with models at the watershed scale.Results from this study show that a model with parallel linear reservoirs provides the most plausible explanation (of those tested) for both the linear hillslope response to rainfall and non‐linear recession behaviour observed at the watershed outlet. In this model each linear reservoir is associated with a landscape type. The parallel reservoir model is consistent with both geochemical analyses of hydrological flow paths and water balance estimates of bedrock recharge. Overall, this study demonstrates that standard approaches of using recession analysis to identify the functional form of storage–discharge relationships identify model structures that are inconsistent with field evidence, and that recession analysis at multiple spatial scales can provide useful insights into catchment behaviour. Copyright © 2008 John Wiley & Sons, Ltd.

A reference data set of hillslope rainfall‐runoff response, Panola Mountain Research Watershed, United States

Although many hillslope hydrologic investigations have been conducted in different climate, topographic, and geologic settings, subsurface stormflow remains a poorly characterized runoff process. Few, if any, of the existing data sets from these hillslope investigations are available for use by the scientific community for model development and validation or conceptualization of subsurface stormflow. We present a high‐resolution spatial and temporal rainfall‐runoff data set generated from the Panola Mountain Research Watershed trenched experimental hillslope. The data set includes surface and subsurface (bedrock surface) topographic information and time series of lateral subsurface flow at the trench, rainfall, and subsurface moisture content (distributed soil moisture content and groundwater levels) from January to June 2002.

A Seismic Inversion Experiment to Identify Scatterers in a Granite Outcrop

A scattering inversion experiment was conducted on exposed Panola granite located southeast of Atlanta, Georgia, in the Panola Mountain Research Watershed. The objective was to test a near-surface seismic technique for locating subsurface discontinuities, such as fractures, that can be strong scatterers of seismic waves. Shallow water-productive fractures in the crystalline and metamorphic rocks underlying expanding suburban and urban areas, like those in central Georgia, are important because they have the potential of significantly augmenting surface freshwater supplies. Sixteen vertical [Formula: see text] geophones were placed in an array with aperture of [Formula: see text]. A weight-drop source was deployed around the array at distances of [Formula: see text]. We successfully recorded seismic waves in the frequency range of [Formula: see text] traveling through the unweathered Panola granite. The lack of overburden suppressed the interference from dispersed surface waves and enhanced coupling of the high-frequency energy source to the granite. We calculated semblance coefficients for waves propagating across the array as a function of time, back-azimuth and apparent velocity and used these to locate the source of the scattered waves within a [Formula: see text] area extending [Formula: see text] below the surface. Zones that were strong scatterers of surface waves correlated with topographic irregularities on the rock outcrop and the location of a small creek. Scattered [Formula: see text]- and [Formula: see text]-waves identified a three-dimensional distribution of scatterers which outlines the downward extension of fractures below the stream and the contact of the granite with amphibolite biotite gneiss.

Rainfall threshold for hillslope outflow: an emergent property of flow pathway connectivity

Abstract. Nonlinear relations between rain input and hillslope outflow are common observations in hillslope hydrology field studies. In this paper we use percolation theory to model the threshold relationship between rainfall amount and outflow and show that this nonlinear relationship may arise from simple linear processes at the smaller scale. When the rainfall amount exceeds a threshold value, the underlying elements become connected and water flows out of the base of the hillslope. The percolation approach shows how random variations in storage capacity and connectivity at the small spatial scale cause a threshold relationship between rainstorm amount and hillslope outflow. As a test case, we applied percolation theory to the well characterized experimental hillslope at the Panola Mountain Research Watershed. Analysing the measured rainstorm events and the subsurface stormflow with percolation theory, we could determine the effect of bedrock permeability, spatial distribution of soil properties and initial water content within the hillslope. The measured variation in the relationship between rainstorm amount and subsurface flow could be reproduced by modelling the initial moisture deficit, the loss of free water to the bedrock, the limited size of the system and the connectivity that is a function of bedrock topography and existence of macropores. The values of the model parameters were in agreement with measured values of soil depth distribution and water saturation.

Effect of bedrock permeability on subsurface stormflow and the water balance of a trenched hillslope at the Panola Mountain Research Watershed, Georgia, USA

AbstractThe effect of bedrock permeability on subsurface stormflow initiation and the hillslope water balance is poorly understood. Previous hillslope hydrological studies at the Panola Mountain Research Watershed (PMRW), Georgia, USA, have assumed that the bedrock underlying the trenched hillslope is effectively impermeable. This paper presents a series of sprinkling experiments where we test the bedrock impermeability hypothesis at the PMRW. Specifically, we quantify the bedrock permeability effects on hillslope subsurface stormflow generation and the hillslope water balance at the PMRW. Five sprinkling experiments were performed by applying 882–1676 mm of rainfall over a ∼5·5 m × 12 m area on the lower hillslope during ∼8 days. In addition to water input and output captured at the trench, we measured transpiration in 14 trees on the slope to close the water balance. Of the 193 mm day−1 applied during the later part of the sprinkling experiments when soil moisture changes were small, <14 mm day−1 was collected at the trench and <4 mm day−1 was transpired by the trees, with residual bedrock leakage of >175 mm day−1 (91%). Bedrock moisture was measured at three locations downslope of the water collection system in the trench. Bedrock moisture responded quickly to precipitation in early spring. Peak tracer breakthrough in response to natural precipitation in the bedrock downslope from the trench was delayed only 2 days relative to peak tracer arrival in subsurface stormflow at the trench. Leakage to bedrock influences subsurface stormflow at the storm time‐scale and also the water balance of the hillslope. This has important implications for the age and geochemistry of the water and thus how one models this hillslope and watershed. Copyright © 2006 John Wiley & Sons, Ltd.