H11D-0321 0800h
Measurement Techniques for Determination of Groundwater Velocity in the German Scientific Borehole LUD-1
The scientific boreholes Lud 1/1A and additional observation wells in the surrounding are used to determine the groundwater velocity in the marsh area with three different methods.The boreholes are located in the marsch area south of the German town Cuxhaven 6 km far from the coast line. The here discussed scientific work is part of the Coastal Aquifer Test field (CAT-Field) activities. This Test field has the size of about 60 km to north-south and 30 km to east-west. A sandy hilly geest area with a high recharge is located in the middle of the CAT field. The fresh- saltwater transition zone is determined in a depth of 60m in the fully screeened boreholes LUD1/1A by logging (Panteleit et.al.,2003). Discussed are here the following investigations to determine the groundwater velocity. The push-pull-test during drilling the borehole Lud1a was a uranine mud tracer test. From this test first values for the dispersion coefficient and the groundwater velocity value were derived by Panteleit et. al. (2003). High resolution water level monitoring is done to determine the dynamics of the groundwater regime in the vicinity of the scientific borehole LUD1. They are performed in 12 observation wells in the vicinity of LUD1. The time dependent variations of the hydraulic gradients are derived from triangle interpretations. A density correction was necessary due to the influence of the deeper salt water body. The high resolution water level measurements reflect, the strong influence of the surface water flow in the drainage system. Colloid observations are carried out with the Phrealog technique in 10 depth horizons. The measurements show a very strong time dependent fluctuation in a time range of hours, but from long time averages useful data for the determination of ground water velocities are expected. With Multi electrode measurements on the borehole wall a tracer with a selected electric conductivity is injected near the electrodes and observed by geoelectric measurements. The application of the multi electrode technique is designed by borehole sleeves installed behind the casing (Kessels et. al., 2002) and the movable double wall packer system (Kessels, W.& Zoth, G., 1998). As the goundwaterflow is measured directly in the aquifer, and not in the borehole itself, the results of the double wall packer experiment are stable in direction and value. {References:} KESSELS, W., ZOTH, G.(1998): Doppelmantel - Packers mit geoelektrischer Meßtechnik zur Bestimmung der Abstandsgeschwindigkeit des Grundwassers, Patent Az:19855048.0,GGA-Hannover. PANTELEIT, B., KESSELS, W., BINOT, F.(2003): Mud Tracer Test while Drilling Soft Rock, submitted W.R.R., thesis, University Bremen KESSELS, W., RIFAI, H., THORENZ, C., ZOTH, G.(2002): Multi Electrode Geoelectric on the Borehole Wall -Determination of groundwater velocity and dispersion parameters-, AGU spring meeting, Washington
H11D-0322 0800h
Response of Water Levels in Devils Hole, Death Valley National Park, Nevada, to Atmospheric Loading, Earth Tides, and Earthquakes
Devils Hole, home to the endangered Devils Hole pupfish (Cyprinodon diabolis) in Death Valley National Park, Nevada, is one of about 30 springs and the largest collapse depression in the Ash Meadows area. The small pool leads to an extensive subterranean cavern within the regional Paleozoic carbonate-rock aquifer. Previous work has established that the pool level fluctuates in response to changes in barometric pressure, Earth tides and earthquakes. Analyses of these fluctuations indicate that the formation is a sensitive indicator of crustal strain, and provide important information regarding the material properties of the surrounding aquifer. Over ten years of hourly water-level measurements were analyzed for the effects of atmospheric loading and Earth tides. The short-term water-level fluctuations caused by these effects were found to be on the order of millimeters to centimeters, indicating relatively low matrix compressibility. Accordingly, the Devils Hole water-level record shows strong responses to the June 28, 1992 Landers/Little Skull Mountain earthquake sequence and to the October 16, 1999 Hector Mine earthquake. A dislocation model was used to calculate volumetric strain for each earthquake. The sensitivity of Devils Hole to strain induced by the solid Earth tide was used to constrain the modeling. Water-level decreases observed following the 1992 and 1999 earthquakes were found to be consistent with areas of crustal expansion predicted by the dislocation model. The magnitude of the water-level changes was also found to be proportional to the predicted coseismic volumetric strain. Post-seismic pore-pressure diffusion, governed by the hydraulic diffusivity of the aquifer, was simulated with a numerical model using the coseismic change in pore pressure as an initial condition. Results of the numerical model indicate that factors such as fault-plane geometry and aquifer heterogeneity may play an important role in controlling pore pressure diffusion in the Devils Hole area.
H11D-0323 INVITED 0800h
The Relationship Between Groundwater Monitoring Networks and the Chi-Chi Earthquake in Taiwan
A visco-elastic model is developed to simulate the ground water level changes in the Cho-Shui River alluvial fan in Taiwan after the Chi-Chi earthquake with magnitude 7.3 on the Richter scale. In addition, an analytical solution is derived with the assumption that no leakage occurred in confined aquifers during the co-seismic period. The solution is used to analyze the data collected from a high-density network of Groundwater monitoring wells in the Cho-Shui River alluvial fan. The simulated ground water level changes agree with the observations. The viscosity coefficient of the model was found to correlate with the hydraulic conductivity of the aquifer. The field observations and the simulations reveal the influence of geological structures and heterogeneity on the ground water changes and locations of sediment liquefactions in the alluvial fan during the Chi-Chi earthquake. Possible applications to imaging subsurface hydraulic heterogeneity are discussed using information about ground water level changes induced by earthquakes. Keywords-GChi-Chi earthquake, Groundwater level changes in wells, Visco-elastic model, Delineation, Sand/gravel formation.
H11D-0324 0800h
Application of Cross-Spectrum Analysis of the Barometric and Tidal Responses to Determinate the Hydrological Properties of Well-Aquifer system
In this paper we follows the Rojstaczer's (1988) method, apply the cross-spectrum method to estimate the transfer function between the water level and atmospheric pressure and water level and predicted earth tide. Because of our observation wells mostly located on alluvial layers, so we also curious about the differences of the characteristic responses with previous works. In the preliminary results of the cross-spectral estimation between water level and atmospheric pressure shows good distribution in frequency domain. With the good quality observation data the barometric response could easily estimate and apply for determinate the fluid flow properties. In the other way, the cross-spectral estimation between water level and predicted earth tide is more complicate and didn't show clear responses. The failure of extracted tidal response from water level fluctuation could be come from problem of predicted earth tide. Ocean load may act as main disturb force to theoretical tide locally. Except that, amplitude attenuation and phase shift due to poroelastic behavior of aquifer could be significant in alluvial deposits material. Future works will focus on overcome the above difficulties of extracted tidal response from observation data. Development of cross-spectral estimation techniques could reveal the characteristic responses of well-aquifer to different sources (seismic wave, vertical atmospheric loading, and volumetric strain). And determination of the hydrological properties of well-aquifer system will help us to clarify the mechanism of hydrological anomalies associated with the earthquake.
H11D-0325 INVITED 0800h
Mathematical and Computational Modeling of Multiphysics Couplings in the Geosciences
Multiphysics couplings can happen in different ways. One may have different physical processes (e.g. flow, transport, reactions) occurring within the same physical domain, or one may have different physical regimes (e.g., surface/subsurface environments, fluid/structure interactions) interacting through interfaces. We will discuss both of these types of multiphysics couplings during this presentation. Of particular interest will be the development of interpolation/projection algorithms for projecting physical quantities from one space/time grid to another, the investigation of mortar and mortar-free methods for coupling multiple physical domains, and the coupling of non-conforming and conforming finite element methods.
H11D-0326 0800h
Artificially Induced and Naturally Occurring Hydraulic Tomography for Characterizing Groundwater Basins
A hydraulic tomographic survey is an innovative approach to characterize the heterogeneity of hydraulic parameters in the subsurface. During hydraulic tomography, spatial and temporal pressure responses from a series of aquifer excitations at different locations and times are collected. Such a data set provides additional independent constraints and makes groundwater inverse problems better posed. Based on the sequential successive linear estimator approach (Yeh and Liu, 2000), we developed an iterative sequential successive linear estimator to analyze the data sets for estimating hydraulic conductivity and specific storage of three-dimensional groundwater basins. To account for the temporal correlation of transient heads, we treated head responses at different time steps in one pumping test as one dataset, whereas head responses from different pumping tests are incorporated into the inverse approach sequentially. After the sequential inclusion of the data sets from different excitations, we iterate the process to improve our estimates. Our study shows that the head is highly correlated with specific storage at early time during a pumping test and with hydraulic conductivity at late time. Consequently, to obtain good estimates of both hydraulic conductivity and specific storage, the head responses at both early and late time are needed. Our new inverse approach has been implemented on a parallel computing platform and applications of our new method for large-scale, naturally occurring hydraulic tomographic surveys are discussed using changes in external loadings on aquifers (such as river stages, barometric pressures, precipitations, surface water reservoirs, trains, etc.).
H11D-0327 INVITED 0800h
Using the response of streams to earthquakes to study hydrologic processes
The hydrologic response of streams to earthquakes provides a way of probing hydrologic processes and properties at spatial and temporal scales that might otherwise be inaccessible. Significant increases in streamflow are commonly observed following large earthquakes. Analysis of the data from a dense network of stream gauges after the 1999 (Mw = 7.5) Chi-Chi earthquake in central Taiwan suggests that the excess water may have originated from the nearby mountains (Wang et al., 2004). On the other hand, analysis of data for liquefaction and streamflow reveals a striking similarity between the distance-to-liquefaction sites (Rliq) versus M relation and the distance-to-streamflow increases (Rstr) versus M relation, suggesting a causal relation between liquefaction and streamflow increase (Manga et al., 2003). I estimate the volume of water released during the Chi-Chi earthquake from liquefaction and compare this with that released from the nearby mountains. The comparison shows that, while the volume of water released from the nearby mountains was sufficient to supply the extra water in the streamflow increase, that released from liquefaction may fall significantly short to supply the required extra water. The similarity between the Rliq ~ M and the Rstr ~ M relations may simply reflect that the basic mechanisms that control the earthquake-induced streamflow also control the earthquake-induced liquefaction.
H11D-0328 0800h
Adaptive Fusion of Stochastic Information for Imaging Fractured Vadose Zones
A stochastic information fusion methodology is developed to assimilate electrical resistivity tomography, high-frequency ground penetrating radar, mid-range-frequency radar, pneumatic/gas tracer tomography, and hydraulic/tracer tomography to image fractures, characterize hydrogeophysical properties, and monitor natural processes in the vadose zone. The information technology research will develop: 1) mechanisms and algorithms for fusion of large data volumes ; 2) parallel adaptive computational engines supporting parallel adaptive algorithms and multi-physics/multi-model computations; 3) adaptive runtime mechanisms for proactive and reactive runtime adaptation and optimization of geophysical and hydrological models of the subsurface; and 4) technologies and infrastructure for remote (pervasive) and collaborative access to computational capabilities for monitoring subsurface processes through interactive visualization tools. The combination of the stochastic fusion approach and information technology can lead to a new level of capability for both hydrologists and geophysicists enabling them to "see" into the earth at greater depths and resolutions than is possible today. Furthermore, the new computing strategies will make high resolution and large-scale hydrological and geophysical modeling feasible for the private sector, scientists, and engineers who are unable to access supercomputers, i.e., an effective paradigm for technology transfer.
H11D-0329 0800h
On the Possibility of Using River Stage Tomography to Characterize the Aquifer Properties of the Choshuishi Alluvial Fan, Taiwan
The fluctuations of groundwater level in a well can be caused by a flood propagating in nearby river due to the interaction of groundwater and surface water. Flood can serve as a natural stimulus without extra manmade excitation for a hydraulic test. Using the time series of groundwater level variation in a well with a known hydrograph of rive stage, the property of the subsurface is possible to be calculated by either deterministic or stochastic approach. The approach can be applied to wells in different locations and explore the spatial distribution of aquifer properties. In this study, the data of groundwater level for wells located from upstream to downstream of Choshu River were analyzed. The correlation between the groundwater fluctuation and river stage are investigated. We explore the possibility of using river stage tomograph as a tool in obtaining aquifer properties.
H11D-0330 0800h
Using Earthquake and Seasonal Variation of Precipitation to Characterize the Aquitard Properties of the Choshuishi Alluvial Fan, Taiwan
Characterizing the subsurface system is important to understand the processes of groundwater flow, solute transport or earthquake mechanism. The Choshuishi Alluvial Fan is one of the most important groundwater resources in Taiwan. The aquifers are mainly composed of gravel and coarse sand and are separated by aquitards of clay and silt. The existing monitoring system composing of 188 monitoring wells at 73 stations mainly provides hydrogeological information of aquifer for the purpose of water resources management. However, the properties of aquitards are required to construct a fully three dimensional hydrogeoloical model. Natural excitations such as earthquake, rainfall or atmospheric pressure change is useful to characterize the properties of aquitard. The post-seismic groundwater fluctuation of Chi-Chi earthquake with Mw 7.6 happened in 1999 is used to obtain the diffusivity of aquitard. Long-term record of groudwater level fluctuation is analyzed by harmonic analysis to obtain the vertical hydraulic conductivity. With the obtained aquitard properties, the hydrogeology of the Choshuishi Alluvial Fan becomes more lucid.
H11D-0331 0800h
Vertical Recharge of a Confined Aquifer in the Fish Farming Area
The rate of vertical groundwater recharge into a confined aquifer is often difficult to estimate due to poor information about confining beds and local pumping activities. Previous studies on a confined aquifer in the coastal plain of Taiwan suggested that its groundwater is primarily recharged laterally from the eastern highland. However, heavy rainfalls that cause all water wells to cease pumping and a widespread recovery of groundwater levels provide an unusual opportunity to re-examine the traditional concept of recharge process. Monitoring records during every heavy rainfall show that groundwater level rises in the fish farming area are far greater than those in the surrounding area. The abnormal rises imply the existence of a rapid cross-formational downward groundwater flow into the confined aquifer. We developed a 3-D finite-element groundwater flow model, based on core log, well log, and measured hydraulic parameters, to simulate the recovery process of groundwater levels during heavy rainfalls. Study results indicate that downward flow provides the major source of groundwater recharge in the fish farming area, while lateral flow provides the major recharge outside the area. Thousands of water wells developed in the fish farming area have been considered to be potential conduits for vertical recharge. The rapid vertical flow not only changes the conceptual model of regional groundwater flow system but poses a potential threat to the groundwater quality in the confined aquifer.