H24F-01
Fine Sediment Residency in Streambeds in Southeastern Australia.
A detailed understanding of channel forming and maintenance processes in streams requires some measurement and/or prediction of bed load transport and sediment mobility. Traditional field based measurements of such processes are often problematic due to the high discharge characteristics of upland streams. In part to compensate for such difficulties, empirical flow competence equations have also been developed to predict armour or bedform stabilising grain mobility. These equations have been applied to individual reaches to predict the entrainment of a threshold grain size and the vertical extent of flushing. In cobble- and boulder-bed channels the threshold grain size relates to the size of the bedform stabilising grains (eg. D84, D90). This then allows some prediction of when transport of the matrix material occurs. The application of Optically Stimulated Luminescence (OSL) dating is considered here as an alternative and innovative way to determine fine sediment residency times in stream beds. Age estimates derived from the technique are used to assist in calibrating sediment entrainment models to specific channel types and hydrological regimes. The results from a one-dimensional HEC-RAS model indicate that recurrence interval floods exceeding bankfull up to 13 years are competent to mobilise the maximum overlying surface grain sizes at the sites. OSL minimum age model results of well bleached quartz in the fine matrix particles are in general agreement with selected competence equation predictions. The apparent long (100-1400y) burial age of most of the mineral quartz suggests that competent flows are not able to flush all subsurface fine-bed material. Maximum bed load exchange (flushing) depth was limited to twice the depth of the overlying D90 grain size. Application of OSL in this study provides important insight into the nature of matrix material storage and flushing in mountain streams.
H24F-02
Vertical variation of vertical hydraulic conductivity in channel sediments
Vertical hydraulic conductivity (Kv) of streambed is of great importance in the analysis of stream-aquifer interactions. We used two methods to estimate the Kv in three rivers of Nebraska. The first method was in-situ permeameter test in river channels. We investigated streambed vertical hydraulic conductivity (Kv) in two depths, one incumbent to the other, immediately beneath the channel surface. Our results demonstrated that streambed Kv in the upper sediment layer was much higher than that in the sediment just beneath the upper layer at each test location. We speculate that hyporheic processes can result in a larger streambed Kv in the upper part of channel sediments. Specifically, water exchange through upwelling and downwelling zones can lead to bigger pore spaces and a more unconsolidated structure of sediments in the upper layer. The upward movement of gas produced by redox processes can loosen sediments and further enlarge pore spaces in the upper layers. Bio- activity in the upper part of the streambed can also expand pore space and thus increase hydraulic conductivity. In-situ permeameter tests penetrated the sediment to depth as much as 90 cm. We then used Geoprobe direct- push technique to generate electrical conductivity log and collect sequences of sediment cores from larger depths. Permeameter tests were conducted on these cores to determine Kv. Our results suggest that Kv values have a decrease tendency with the depth for sediments in these rivers.
H24F-03
Ground Water / Surface Water Exchange: Streambed Versus a Channel Bar
The streambed is important in controlling exchange of water, solutes, and heat between streams and ground water. Processes such as sedimentation, erosion, and fluctuations in diurnal temperatures can have significant effects on the streambed hydraulic conductivity, which in turn affects fluid velocities across the streambed. The objectives of this study are to quantify the difference in flux magnitude and direction within and around a channel bar. The focus of this presentation is to compare fluxes in channel bar sediments with fluxes in the streambed to determine the effect of the upper boundary conditions on sediment fluxes. A network of piezometers was installed on and around a channel bar located within the Truckee River, a dense 6th order river network, located primarily in northwest Nevada. Instruments used were temperature loggers, pressure transducers, and stage recorders. Several methods were simultaneously utilized to quantify water and heat fluxes and to interpret the hydrodynamic processes through the streambed sediments. Numerical simulations are being completed to quantify the spatial and temporal fluid flux and heat transport in relation to varied hydraulic parameters such as variable river stage, geometry, and hydraulic conductivity. In general, we have found that surface water exchange to the streambed occurs at the upstream portion of bed features and streambed discharge dominates at the downstream bed feature. This exchange is evidenced at the channel bar as well as localized riffles and point bars adjacent to the channel bar. We found that at least two separate hydraulic conditions are evident during our study. The range in water levels between the piezometers was altered from approximately 1.25 m to a minimum of 0.10 m and the mean potentiometric surface increased by 1 m. These variations are geomorphic responses due to a flood event, inundating the channel bar, and a channel restoration project both upstream and downstream of the study area. These alterations have caused a reversal in the vertical head gradient (VHG) in some locations by up to 0.14. There also appears to be a critical stage height that alters the flux direction or magnitude. In conclusion, stage boundary conditions coupled with streambed features significantly contribute to the exchange direction. Sedimentation and erosion from restoration activities and streambed evolution also impacted fluid flux patterns due to their impact on streambed surface hydraulic conductivity patterns.
H24F-04
Understanding Hyporheic Exchange and Groundwater Contribution in a Northeastern Stream Using Temperature as a Tracer
Understanding the role and extent of the hyporheic zone and groundwater/surface water interaction is becoming increasingly important in safeguarding surface water resources and stream ecosystem quality. Little research has been completed to date on characterizing the hyporheic zone in the Northeastern US. Research on a first order stream in coastal New Hampshire is being conducted to understand the potential physical extent of the hyporheic zone and to understand the flux between this zone and the surrounding landscape. Two primary characteristics, stream morphology and stream and subsurface temperature, are being utilized to understand these fluxes. Geophysical methods were used to characterize the physical extent of the hyporheic zone and potential areas of near surface flow and hill slope groundwater. Fiber Optic Distributed Temperature Sensing (FO-DTS) using a fiber optic cable is being used as a survey tool in combination with discrete temperature measurements using thermocouple and thermistor sensors, and traditional hydrologic evaluation, to provide a description of the hyporheic zone longitudinally and with depth along a 500-meter stream reach. Results of an intensive field campaign conducted in late summer/fall 2007 will be presented.
H24F-05
Spatial and Temporal Variability in Groundwater-Based Nitrogen Input to a Stream in an Agricultural Watershed in North Carolina
We used point measurements of streambed hydraulic head gradient (J), hydraulic conductivity (K), and nitrogen (N) concentrations to quantify groundwater-based N input to a large stream in an agricultural watershed. We focused on fluxes across the sediment-water interface by making all field measurements directly in the streambed (J and K apply to the top 36 cm, water samples for N measurement were drawn from a depth of 34 cm). Measurements were made at 38 points (Dec 2005) or 46 points (Feb, Apr, Jun, Aug, Oct, and Dec 2006) over a 263 m reach of West Bear Creek in the North Carolina Coastal Plain. Point values of groundwater flux (v) and total dissolved N flux (fTDN, mostly in the form of NO3-) were highly variable, with v ranging from - 0.80 to 3.1 m day-1 (mean v = 0.44 m day-1) and fTDN from -0.17 to 2.6 mol m-2 day-1 (mean fTDN = 0.18 mol m-2 day-1). On average, higher K and v, and lower groundwater concentration and fTDN, were measured in the center of the channel. Values of fTDN were about 10x larger than reported for uncontaminated forest streams and other agricultural streams. All point measurements showed groundwater seepage into the stream, except for the 2 or 3 points on the upstream side of a beaver dam that was present from Dec 2005-April 2006. Only 8 of 45 NO3- samples collected within 37 m downstream of the dam had NO3- concentrations above the detection level of 0.007 mM. This area of the streambed was likely a discharge zone for stream water that had infiltrated the streambed upstream of the dam, suggesting that the beaver dam may have contributed to NO3- loss. Analysis of dissolved gases (Ar, N2, CO2, CH4, O2) in streambed groundwater indicated significant lateral variability in redox potential and excess N2 (likely from denitrification), and suggests a difference in average groundwater recharge temperature between groundwater discharging into the left and right sides of the channel. Water and N fluxes through the ~1800 m2 streambed were spatial integrated from a 10-cm resolution grid interpolated from point measurements using the multiquadratic radial basis function method. Groundwater- based input of TDN to the reach (QTDN) was 186, 251, 248, 188, 368, 637, and 109 mol day-1 in Dec (2005), Feb, Apr, Jun, Aug, Oct, and Dec (2006) respectively, the majority of which was in the form of NO3-, with little contribution from dissolved organic N (25, 29, 5, 3, 55, 51, 13 mol day-1 in Dec 2005, Feb, Apr, Jun, Aug, Oct, and Dec 2006, respectively) and ammonium (<5 mol day-1). Changes in groundwater seepage explained much of the temporal variability in QTDN (r2=0.80) and QNO3-, the groundwater-based input of NO3- to the reach (r2=0.79). This relationship indicates strong control of these streambed N fluxes, at the scale of a large reach, by the groundwater flux through the streambed (even though some of the points of largest NO3- flux had among the lowest groundwater seepage rates but high NO3- concentrations). At the reach scale, groundwater seepage may be a reasonable predictor of QTDN and QNO3-. About 30% of the streambed accounted for 56% to 70% of QNO3- within the reach.
H24F-06
Artificial Streams, Distorted Processes: The Effect of Effluent on Stream-Aquifer Interactions
Treated wastewater, effluent, has been used as a remedy for overstressed aquifers and dry streambeds. Artificial recharge basins allow effluent to seep into the ground relieving stressed aquifers. These basins however, frequently become clogged due to physical obstacles resulting from chemical and biological reactions. Effluent can also be used to replace baseflow for dry streambeds. However, little is known about the effect of effluent on stream-aquifer interactions. The Upper Santa Cruz River, Arizona has effluent, from the Nogales International Waste Water Treatment Plant, as its dominant water input (excepting flood events). A series of monthly field campaigns were undertaken to understand the impact of effluent on the streambed at 16 different sites along a thirty kilometer stretch of the river. The San Pedro River, Arizona is a natural flow, perennial stream that was also tested monthly and used as a control reach. The field campaigns had two focuses: physical transformations in the streambed and water source identification using chemical composition. Physical transformations were measured through the use of 1) piezometers and seepage pans and 2) soil cores; both used to obtain saturated hydraulic conductivity values (KSAT) over time. Water sampling included: the monthly field campaigns, monsoon flows, and nearby wells. All samples were analyzed for cations, anions, and stable isotope ratios (δD and δ18O). Results indicate that the Santa Cruz River becomes disconnected from the aquifer adjacent to the point of effluent discharge. Furthermore, as the time between major flood events increases, the disconnection of the stream and aquifer extends further downstream. In total the results imply that water in the streambed is isolated from the groundwater, perhaps due to clogging and that barrier is removed after especially large flood flows (1000+ cfs).
H24F-07
Application of BERM - a bank erosion and retreat model
Bank Erosion and Retreat Model (BERM) distinguishes itself from other models in that the rate of basal erosion is a function of a gradient function of the longitudinal sediment transport rate and strength of the secondary flow rather than being proportional to the excessive near-bank velocity or shear stress. The rate of bank retreat is a resultant of basal erosion and bank collapse. Additionally, bank height, side slope, and thickness of each layer in the vertical structure of the bank are taken into consideration in the model. The feasibility of integrating BERM into a 2D hydrodynamic and sediment transport model has been verified by a few laboratorial cases and a meandering reach in the West Jordan River in Utah, USA. Therefore, BERM could be coupled with most existing hydrodynamic and sediment transport models and extend their ability to account for erodible banks.
H24F-08
Modeling groundwater-surface water interactions in an operational setting by linking object- oriented river basin management model (RiverWare) with 3-D finite-difference groundwater model (MODFLOW).
Accurate representation of groundwater-surface water interactions is critical to modeling low river flow periods in riparian environments in the semi-arid southwestern United States. As an example, over-appropriation of human water use in the Middle Rio Grande region adversely impacts the habitat of the endangered Rio Grande silvery minnow. Improved management practices during low flow conditions could prevent channel desiccation and habitat destruction. We present a modeling tool with significant potential for improved decision-making in stream reaches influenced by significant surface-groundwater interactions. While river basin management models typically represent operational complexities such as human elements of water demand and consumption with a high degree of sophistication, they often represent groundwater-surface water interactions semi-empirically or at coarse resolution. In contrast, distributed groundwater models, with an adequately fine grid represent groundwater-surface water interactions accurately, but seldom incorporate complex details of water rights and user demands. To best exploit the strengths of both classes of models, we have developed a link between the object-oriented river management software package RiverWare and the USGS groundwater modeling program MODFLOW. An interactive time stepping approach is used in the linked model. RiverWare and MODFLOW run in parallel exchanging data after each time-step. This linked framework incorporates several features critical to modeling groundwater-surface interactions in riparian zones, including riparian ET, localized variations in seepage rates and rule-based water allocations to users and/or environmental flows, and is expected to be an improved tool for modeling groundwater-surface water interaction in regions where groundwater storage repose to changing river conditions is rapid. The performance of the linked model is illustrated through applications on the Rio Grande in the vicinity of Albuquerque, New Mexico. Low river flow conditions are simulated in a high resolution linked model at several projected future scenarios of the region.