H33B-01
The Impact of Land use on the Quantity and Quality of Groundwater Recharge into the Coastal Plain Aquifer, Israel
A study designed to examine the impact of urbanization on the quantity and quality of groundwater recharge in the Coastal Plain Aquifer, Israel, is being carried out in the city of Ashdod, where urbanization in the last 40 years has been one of the most rapid in Israel. Infiltration processes under different urban land setups are being compared to those beneath nearby cultivated areas and the undeveloped sand dunes. Temporal variations in water content at various depths along the entire vadose zone were measured for two consecutive years at five sites that represent typical land uses. Water-content measurements were conducted using a monitoring system that includes flexible time-domain reflectometry (FTDR) sensors installed in slanted boreholes throughout the entire vadose zone. The variation in water content enables direct tracking of the wetting- front propagation velocities, variations in vadose-zone water storage and calculation of the infiltration fluxes. In addition to the FTDR system, a new monitoring system that allows continuous measurements of the water potential and frequent sampling of the sediment pore water was installed at a sand dune site. This new system is comprised of vadose-zone sampling ports (VSPs) and FTDR sensors installed in slanted boreholes. Preliminary results show that in addition to the expected changes in pore-water pressure due to changes in water content caused by the infiltration process, increases in pore-water pressure were detected simultaneously throughout the entire vadose zone after each of the major rain storms. This pressure wave was detected each time fast and significant propagation of the wetting front was observed. Vadose-zone pore water was frequently sampled by a second VSP system and analyzed for chemical and isotopic compositions presenting changes in the vadose zone chemical and isotopic profiles throughout the year.
H33B-02
Dynamics of Floodwater Infiltration and Groundwater Recharge Under Ephemeral Channels in Arid Regions
Shallow alluvial aquifers underneath ephemeral streams are often the only reliable source of water that can sustain human habitation in arid environments (e.g. Arava Valley, Israel; Rio Andarax, Spain; Kuiseb River, Namibia). The main source of replenishment of these alluvial aquifers is recharge from floodwater infiltration. Accordingly, effective management of surface water and groundwater in arid regions requires a better understanding of the processes controlling floodwater infiltration and recharge of alluvial aquifers. This study focuses on understanding the dynamic process of floodwater infiltration from ephemeral channels while implementing innovative methods specifically designed to quantify the recharge fluxes. The monitoring system provides real-time continuous measurements of the hydraulic conditions in all three domains involved in the recharge process: (a) the flood, (b) water-content variations along the unsaturated profile, (c) the groundwater response to the recharge event. Water-content variations along the unsaturated profile were monitored using flexible TDR (FTDR) probes installed along slanted boreholes underneath the stream channel. Water levels and salinity of both the flood and the groundwater were measured simultaneously. Two study sites were selected for this work: the Buffels River, South Africa and the Kuiseb River, Namibia. The monitoring stations installed at those sites recorded several flood events during 2005/2006. Data collected during this period revealed the dynamic process in which floodwater percolates through the vadose zone and recharges the groundwater. Each flood initiated an infiltration event expressed by wetting of the vadose zone and a rise in the water table. The sequential wetting of the vadose zone allowed direct calculations of the wetting-front propagation velocities and percolation fluxes from land surface down to the groundwater. With the arrival of the wetting front to the water table, groundwater began to rise, indicating an increase in groundwater storage in response to the recharge event. Water fluxes were calculated using several independent methods: (a) combining the calculated wetting-front propagation velocity with the change in moisture profile, (b) the rate at which the water table rises as an indication of the percolation rate, and (c) the final increase in groundwater storage through the measured change in groundwater levels. Interestingly, the calculations performed for all of the floods yielded corresponding flux values of approximately 1 cm/h. Aquifer dimensions, as well as total recharge estimations, were also derived from the data. Salt-transport dynamics at each site and the positive influence of the flood events on groundwater quality were revealed from the EC measurements.
H33B-03
Real-time Soil Characterization with a Stochastic Data Assimilation Approach
Data assimilation (DA) is an estimation framework to characterize the states of a system by merging the information in measurements and physical models. The purpose of this study is to explore the feasibility of DA methods in maximizing the information content of sensor networks deployed in a complex environmental system, a test bed for wastewater re-use in Palmdale, CA. However, to protect the groundwater from pollution, a real-time monitoring system, consisting of a sensor network and coupled flow and transport models, especially for nitrate concentration, is implemented to ultimately provide feedback for irrigation operation. Synthetic experiments, which also consider sensor measurement errors, with the Ensemble Kalman Filter (EnKF) are performed to estimate the space-time evolution of the soil states under scenarios of successively more input uncertainty. The results from the EnKF are compared with the performance of an open-loop forward modeling simulation, which is not updated with sensor measurements, and show significant improvement in the estimate of soil moisture profiles under all uncertainty scenarios. Based on the improved estimation of water flow in vadose zone, improved estimates of nitrate concentrations are expected as well. Work to assimilate nitrate and soil temperature measurements and estimate sensor error characteristics online are ongoing.
H33B-04
Multimodel Prediction of Water Flow in a Field Soil Using Pedotransfer Functions
Combining predictions using various independent models, often called multimodel prediction, has become a very popular technique in climate prediction and is now increasingly being used also in subsurface hydrology. The objectives of this work were (a) to compare different methods of multimodel prediction of the field soil water regime using pedotransfer functions, and (b) to see whether the calibration of a flow model with field data can be replaced by multimodel predictions. The multimodel prediction in this work consisted of running the Richards model with outputs of individual PTFs and then combining the obtained outputs into a single prediction. We compared weighing predictions from individual models by (1) using only the best model, (2) assigning equal weights, (3) using the unconstrained superensemble (i. e. regressing measured values to outputs of individual models), (4) using singular value decomposition in the regression, (5) using Bayesian model averaging, and (6) applying weights derived from the Kullback-Leibler information for each model. We evaluated the weighing methods in terms of their accuracy (i. e. errors in reproducing the training, or hindcast, datasets), and reliability (i.e., errors in reproducing the test datasets). The two best weighing methods (Bayesian model averaging and regression with singular value decomposition) had average accuracy and reliability RMSE values of about 0.01 cm3cm-3 at 35 cm depth, and of about 0.005 cm3cm-3 at larger depths for one month monitoring and 13 months of testing. Calibrating the Richards model resulted in RMSE values of 0.009 cm3cm-3 at 35 cm depth and from 0.004 to 0.006 cm3cm-3 at larger depths. This indicates that monitoring of the soil water regime in combination with multimodel prediction instead of calibrating the flow model can be a viable approach to simulating field water flow in the vadose zone.
H33B-05
Diffusion of Nutrients in an Isolated Wetland Resulting From Shallow Pore Water Gradients Affected by Antecedent Soil Conditions.
Historically sequestered nutrients in wetland soils may be gradually released to the water column through the process commonly referred to as internal loading. The watershed for Lake Okeechobee, FL (USA) is heavily agricultural and excess nutrients in this area are drained to the Lake by ditches and canals. Formerly isolated, wetlands in this area have also been extensively ditched and drained. In this study, diffusive fluxes of nutrients were calculated using Fick's First Law from shallow pore water gradients, and later compared to fluxes measured by an incubated laboratory experiment on 10-cm intact soil cores from the same sites. Three intact soil cores from a wetland located on an operational beef farm were used to measure total phosphorus (TP), along with soil properties such as porosity, bulk density, and pH. Simultaneously, pore water concentrations of total organic carbon (TOC), total Kjeldahl nitrogen (TKN), and soluble reactive phosphorus (SRP) were also measured at the same three sites for a period of twelve months, and compared to surface water concentrations during flooded periods. A strong correlation between concentration gradients in pore water SRP and those observed in soil TP, suggests that shallow pore water concentrations reflect antecedent soil conditions. If this is true, then fluxes associated with diffusion and advection could greatly affect the total ground water fluxes across the soil-water interface. Fickian diffusive fluxes, estimated six times over a twelve month sampling period, were found to vary between 7-38 mg.m-2.d-1 for TOC, 1-18 mg.m-2.d-1 for TKN, and 0.04-0.86 mg.m-2.d-1 for SRP. While factors such as wetland stage and hydroperiod may have affected the fluxes, it is ultimately the concentration gradients across the soil-water interface that drives diffusive fluxes.
H33B-06
Reactive Transport Modeling of Vadose Zone Contamination: Feedback between Reactions and Gas-Phase Transport
The unsaturated zone acts as a buffer zone for contaminants on their way to the water table but can also attenuate the emission of contaminants leaving the subsurface environment through the gas phase. A reactive transport model that includes multicomponent gas transport has been developed to investigate the processes that contribute to the generation and attenuation of contaminants in the unsaturated zone. In particular, the model is suitable to study the feedback processes between advective-diffusive gas transport and geochemical reactions. The model is also able to estimate diffusive and advective contributions to gas transport in multicomponent systems. Two model applications are presented that investigate gas transport and reactions in mine tailings and at a site with organic contamination. In mine tailings, atmospheric oxygen transported into the sediment column is consumed in the oxidation of sulfide minerals. Gas volume loss caused by the consumption of atmospheric oxygen drives advective fluxes. In the absence of carbonate minerals, the advective component accounts for 16 % of the net oxygen flux into the column, while, in a carbonate-rich system, advection accounts for 10 % of the net oxygen flux. Dissolution of carbonate minerals has a moderating effect on advective gas transport since carbon dioxide can partially compensate for the depletion of oxygen. At an oil spill site, volatilization and degradation of organic contaminants cause advective and diffusive fluxes of organic vapors away from the source zone. At early stages, volatilization dominates and oxidation of these organic vapors attenuates the emission of contaminants to the atmosphere. The contribution of advection to organic vapor fluxes is significant initially but decreases with time. At later stages, the oil source becomes depleted of its most volatile fraction, and anaerobic degradation of aromatic compounds and heavier n-alkanes results in the production of methane. Up to 15 % of methane produced is transported away from the source zone by advection. Subsequently, methane is oxidized in the presence of atmospheric oxygen. A reduction of total gas pressure during oxidation results in the development of advective gas fluxes toward the oxidation zone. As a consequence, the distribution of non-reactive gases such as argon or nitrogen in the unsaturated zone can be used as indicators of these processes.
H33B-07
Transport of TNT and DNT in Soil Under Infiltration and Evaporation Events
The effects of water infiltration and evaporation on the fate and transport of explosive-related chemicals (ERCs) in soils were studied. The experiments were conducted in a 100 cm uniform cylindrical sand column packed with homogeneous sand, and instrumented with air and water pressure sensors and sampling ports to monitor hydraulic conditions and ERCs concentration profiles in soil. TNT and DNT crystals were placed in a porous membrane and buried as a point source near the surface of the soil. Infiltration of water containing a conservative tracer was followed by evaporation periods. Spatial and temporal concentration distributions of conservative solutes were used to determine physical transport parameters of the soil under different water contents and flow conditions. Results indicate that physical transport is spatially variable and that TNT and DNT transport is rate- limited. Evaporation events result in accumulation of salts near the surface of the soil, thus affecting the fate and transport of ERCs.
H33B-08
Using Soil Moisture as a Guide in Controlling the Amount of Irrigated Water on Grass Lawns
Soil moisture content is one of the factors that controls the infiltration capacity of soils. Precipitation and irrigation increase soil moisture which in turn reduces infiltration capacity. This results in increased runoff during subsequent storm events. Increased stormwater runoff may cause adverse environmental problems such as increased soil erosion, increased bed and suspended loads in streams, and increased non-point source pollution. Monitoring soil moisture on irrigated plots can be used as a guide for efficient use of irrigated water. Thus, irrigation systems only will be turned on when soil moisture falls below a threshold value for the respective soil type. However, landscapers at Northern Kentucky University (NKU) schedule the irrigation of grass lawns without taking into consideration the level of soil moisture. This has resulted in incidences of irrigation of the lawns during or immediately after a heavy storm event. Effective monitoring of the soil moisture of irrigated fields has been shown to help in controlling cost of irrigation and conserving valuable resources. This can be achieved by using instruments such as tensiometers and neutron probes to monitor soil moisture (Manning, 1992). On an irrigated field such as a grass lawn, the ideal condition will be to maintain soil moisture between field capacity and wilting point. The objective of this study is to investigate the effect of soil texture and slope on the amount of irrigated water used on selected grass lawns on NKU campus at Highland Heights, Kentucky. The grass lawns were selected based on low slope (0 to 10¬0), medium slope (10¬0 to 15¬0), and high slope (more than 15¬0 ). Two plots were selected for each slope category. The soil texture of each grass lawn was determined by performing standard particle size distribution analysis of samples taken during the installation of the tensiometers. A survey instrument and a GIS software were used to analyze the slopes. The tensiometers were monitored daily and NKU Grounds Department was advised to irrigate those plots only when the soil moisture fell below a specified threshold level. The threshold value was between 70 and 80 centibars for the range of soil textures at the site. Temperature and precipitation data were gathered from NKU's Department of Physics and Geology weather center and the Northern Kentucky Airport Weather Station. The second low slope plot (LSII) has the highest percent sand of 40% whilst the second medium slope (MSII) has the lowest sand of 30%. Although the textures of the low slope plots are significantly different, there was not much difference between the moisture readings. However, a slight difference in the texture of the high slope plots tends to affect water infiltration and moisture retention capacities. It takes longer for water to infiltrate the finer grained, high slope plot but it retains the moisture longer once it is saturated. Air temperatures of 850 F and above was the controlling factor as all plots dried faster even after irrigation or precipitation. Overall, the soil moisture monitoring resulted in less amount of irrigated water use; less than half the normal amount.
H33B-09
Horizontal Flow and Transport in the Vadose Zone
Flow and transport in the vadose zone is sensitive to sediment heterogeneities. Similar to the previous work on capillary barriers that exploited this sensitivity, the present study focuses on taking advantage of substantial differences in unsaturated hydraulic conductivity in coarse versus fine sediments under unsaturated conditions. In the present work we demonstrate that horizontal flow can be induced and controlled in the vadose zone, in regions where fine sediments exist above coarse sediments. Laboratory studies have been conducted under a number of media geometries to demonstrate the potential for inducing flow through use of coupled (injection/withdrawal) suction-cup lysimeters. Numerical studies were run to compliment and extend the ongoing lab work. HYDRUS_2D, a finite element model, was utilized to simulate both water flow and chemical transport within a fine-sediment layer embedded in a coarse-sediment matrix. Published parameter values for the coarse and fine sands used in the laboratory experiments, when applied in HYDRUS_2D, allowed successful reproduction of the laboratory results. Using the sediment-type parameter sets available in HYDRUS_2D, the model has allowed identification of a range of sediment combinations that show promise for inducing and controlling horizontal flow and chemical transport within the vadose zone.
H33B-10
Effects of Field Data Collection Frequency on Soil Water Modeling and Parameter Estimation
Inverse parameter estimation is becoming an increasingly popular tool for calibrating and evaluating hydrologic models. In this study, PEST (Paramater ESTimation) software was used in conjunction with the one-dimensional soil transport model SHAW (Simultaneous Heat and Water Transport) to evaluate the impacts of winter climate variations on spring snowmelt and infiltration. Lysimeter, soil moisture, and frost data were collected during winter months on an agricultural field in south-central Wisconsin. The data was used to create a one- dimensional soil column model in SHAW. PEST was used to optimize the hydraulic conductivity and air entry potential parameters in the SHAW model based on winter weather patterns. The results show that the frequency of field data and observations affect the ability of PEST to accurately reproduce a set of known parameters in the SHAW model.
H33B-11
Soil Moisture Memory and Predictability of Seasonal Streamflow in Sri Lanka
Global water and energy budget studies as well as long term soil moisture observation records have shown that the time scales of soil moisture anomaly dissipation are about 2-3 months. This implies that the baseflow component of streamflow should, in principle, be partly predictable at seasonal time scales. Here we investigate the hypothesis of seasonal streamflow predictability for Sri Lanka. Gridded 0.25 degree monthly precipitation observations for Sri Lanka were merged with 0.5 degree 6-hourly bias-corrected global reanalysis data for the period 1979-1993 and used to force the NASA Catchment Land Surface Model (CLSM) over Sri Lanka. CLSM simulations of monthly minimum and maximum surface temperatures and streamflow compared well with station observations. One-month lagged autocorrelation of simulated soil moisture was high (spatial average was about 0.75), with anomaly dissipation time scales of about 3-5 months. The lagged cross-correlation between simulated total runoff efficiency and soil moisture showed predictability of streamflow up to 3 months in advance in the drier seasons (April to June and July to September).
H33B-12
Evapotranspiration Parameterizations at a Grass Site in Florida, USA
In spite of the fact that grasslands account for about 40% of the ice-free global terrestrial land cover, their contribution to the surface exchanges of energy and water in local and regional scale is so far uncertain. In this study, the sensitivity of evapotranspiration (ET) and other energy fluxes to wetness variables, namely the volumetric Soil Water Content (SWC) and Antecedent Precipitation Index (API), over a non-irrigated grass site in Central Florida, USA (28.049 N, 81.400 W) were investigated. Eddy correlation and soil water content measurements were taken by USGS (U.S. Geological Survey) at the grass study site, within 100 m of a SFWMD (South Florida Water Management District) weather station. The soil is composed of fine sands and it is mainly covered by Paspalum notatum (bahia grass). Variable soil wetness conditions with API bounds of about 2 to 160 mm and water table levels of 0.03 to 1.22 m below ground surface, respectively, were observed throughout the year 2004. The Bowen ratio exhibited an average of 1 and values larger than 2 during few dry days. The daytime average ET was classified into two stages, first stage (energy-limited) and second stage (water- limited) based on the water availability. The critical values of API and SWC were found to be about 56 mm and 0.17 respectively, with the second one being approximately 33% of the SWC at saturation. The ET values estimated by the simple Priestley-Taylor (PT) method were compared to the actual values. The PT coefficient varied from a low bound of approximately 0.4 to a peak of 1.21. Simple relationships for the PT empirical factor were employed in terms of SWC and API to improve the accuracy of the second stage observations. The results of the ET parameterizations closely match eddy-covariance flux values on daily and longer time steps.
H33B-13
Storage and Hysteresis in the Soil Plant Atmosphere Continuum
We explore potential models of the effects of storage and hysteresis in plant tissues on transpiration and water stress. We extend the traditional circuit analogy for the soil-plant-atmosphere continuum to include the coupling of the plant with a stochastic soil moisture potential representing rainfall input. Analytic solutions for the soil- moisture dry-down coupled with the transpiration pathway and diurnal stomatal control are presented along with approximate solutions of the full stochastic problem. We also examine numerically the effect of introducing nonlinearities into the simple plant model (e.g., variable capacitance as a method of simulating xylem cavitation) and the potential for resulting hysteretic behavior in the transpiration stream.
H33B-14
Land use as a Parameter of Distributed Hydrological Modeling at the CATIE Farm, Turrialba, Costa Rica.
As distributed hydrological models continue to develop, their amount of spatial detail requires the evaluation of a larger set of variables. Deforestation is often cited as a principal cause of changing hydrological regimes in the tropics. However, many studies debate the exact mechanism of change. Also, much of the tropics have been permanently deforested for agricultural expansion. Therefore, in this study we instrumented fields (1-6 ha) of four common land uses (forest, coffee agroforestry, sugar cane, and pasture) with meteorological stations, soil moisture probes, and H-flumes. Additional field measurements have shown differences in bulk density, saturated hydraulic conductivity, and soil moisture dynamics between land uses. Hydrograph analysis suggests that the pasture site responds differently to rainfall than the other land uses. Runoff from the pasture site results in higher intensity, greater volume, and shorter duration runoff events than the other land uses. However, the other land uses respond more frequently with lower maximum event intensities, lower volumes and longer durations. In the forest and coffee sites, soil moisture dynamics suggest the importance of lateral preferential flow paths due to root influenced soil structure for runoff response. Therefore, while vertical Ksat values may be greater at all sites than most rainfall intensities, lateral Ksat values may differ significantly between sites. Field measurements and the distributed physically based Soil Moisture Routing (SMR) model are being used test hypotheses, and direct further field research. These results will prove important to understand hydrological connectivity in fragmented landscapes, and the potential recovery of hydrological services within a typical humid tropical environment.