NS34A-01 INVITED
Long Term Resistivity Monitoring: Imaging Natural Infiltration in the Vadose and Transition Zones of a Shallow Aquifer
A long-term resistivity monitoring system was designed and installed in the summer of 2005 in Van Buren County, Michigan. The field site is located near a small, man-made pond. This location was selected to facilitate observation of the characteristics of natural infiltration with increasing distance from a surface water body. Four vertical resistivity probes of varying lengths, between 0.5 and 1.5 meters, are positioned in a line perpendicular to the pond. Electrodes spaced at five centimeters are used to measure the subsurface resistivity. Normal data collection occurs every four hours and consists of 92 pole-pole resistivity measurements from electrode pairs along the lengths of the probes. Hourly and daily precipitation totals are taken from an on-site rain gauge. Data collection occurs at a rate of one cycle (92 resistivity measurements) every three minutes during and through four hours following rainfall events, thus providing high-resolution infiltration data. The subsurface at this site is heterogeneous making conversions from resistivity to absolute water content difficult. However, the characteristics of infiltration are easily imaged with this system at the four probe locations by converting the data to percent change of resistivity. There has been no significant disruption of the collection cycles since December 2005, providing over two years of continuous subsurface resistivity data from this site. The long-term nature of this experiment and the high resolution of subsurface resistivity measurements, with data points every 5 centimeters along the length of each probe, make this experiment somewhat unique. The most recent analysis of the data shows the variation of the infiltration with season and saturation state over a period of two years. Seasonal characteristics of infiltration have remained similar over the two-year period of data collection, providing a measure of reliability for the data.
NS34A-02
Measuring Unsaturated Soil Hydraulic Properties Rapidly by the Moment Method
Measurement of unsaturated soil hydraulic properties is difficult, time consuming, and expensive. We are conducting research on a new platform of tests that rely on the moment created by changing moisture content profiles within a soil core. Utilizing inexpensive off-the-shelf sensors and controls, the approach can simultaneously measure the water retention curve and the unsaturated hydraulic conductivity curve for either wetting or drying scans. The tests require only 10's of minutes to hours to perform and are completely autonomous. Validation of results is being performed by gamma-ray attenuation.
NS34A-03 INVITED
Characterizing Tidally Influenced Submarine Ground-Water Discharge in an Estuary Using Fiber-Optic Distributed Temperature Sensing and Marine Electrical Resistivity
Submarine ground-water discharge (SGD) is an important source of nutrients and pollutants to ecologically sensitive estuaries. We observe SGD and the movement of freshwater/saltwater interfaces by measuring two physical properties that differ between the coastal aquifer and estuarine waters—temperature and electrical resistivity. SGD from the fresh coastal aquifer is colder in the summer than estuarine waters, and warmer in the winter. Additionally, fresh ground water is more resistive than high salinity estuarine water. We evaluate the utility of two geophysical methods, fiber-optic distributed temperature sensing (FO-DTS) and marine electrical resistivity (MER) to observe SGD and the freshwater/saltwater interface in submarine sediments. Recent advances in FO-DTS technology allow for measurement of temperature with spatial resolution of 1 m, temporal resolution on the order of 1 min, and temperature precision of 0.01 °C, although tradeoffs exist between temporal resolution and precision. The FO-DTS instrument used in this study is based on analysis of Raman scatter. Raman scattering occurs as light interacts with the fiber materials, producing backscatter energy that is temperature-dependent (anti-Stokes scattering) and temperature-independent (Stokes scattering). The ratio of the magnitudes of the anti-Stokes to Stokes scattering depends exponentially on the fiber temperature. MER is a well documented approach to observing subsurface electrical resistivity contrasts associated with porewater salinity, porosity, or lithology. A 48-electrode cable with 1-m spacing was used; full surveys including reciprocal data were conducted in about 50 min. Our study includes two field deployments at Waquoit Bay National Estuarine Research Reserve (WBNERR), East Falmouth, Massachusetts. The first deployment was a reconnaissance effort in May-June 2006 to observe the areal distribution of SGD. This deployment used a 1.3-km fiber-optic cable deployed in a 80 by 60-m grid The cable was weighted and submerged into the bay sediment to a depth of several centimeters. From the FO-DTS data, we inferred a zone of tidally-driven, near-shore, fresh SGD with minimal shore-parallel variation at Waquoit Bay. We returned to the site for a second deployment in June-July 2007, which used FO-DTS in tandem with MER. We instrumented a single transect extending offshore about 50 m. The cables were weighted and permanently installed at a depth of approximately 0.5 m to ensure replication of data collection geometries on return visits to understand seasonal variability. The FO-DTS temperature time-series data at near-shore locations are dominated by a semi-diurnal (tidal) signal, whereas the temperature at off-shore locations are dominated by a diurnal signal (day/night heating and cooling). MER surveys produced high-resolution time-lapse tomograms, providing insight into the temporal variations of the subsurface freshwater/saltwater interface. The tomograms show a high-resistivity zone near the shore at low tide, indicative of fresh SGD, and consistent with the FO-DTS results.
NS34A-04
Electrical Resistivity Imaging of the Architecture of Sub-stream Sediments
The modeling of fluvial systems, including biogeochemical fluxes and exchange, is constrained by a lack of spatial information about the continuity and structure of stream bed sediments. There are few methods for noninvasive characterization of stream beds. Invasive methods using piezometers and cores fail to provide detailed spatial information on the prevailing architecture and its continuity. Geophysical techniques play a pivotal role in providing spatial information on subsurface properties and processes across many other environments, and we have extended the use of one of those techniques to stream beds. This new application has been realized in part by the recent advances in inversion algorithms. For our modeling in the examples here we use a triangular finite element-based forward solution coupled with an ‘Occams' style inversion. The model allows specification of electrode locations at any node within the finite element mesh. Regions within the inverse model may be ‘disconnected' to permit enhancement of sharp boundaries, such boundaries exist in the present studies, e.g. at the known stream - stream bed interface. We demonstrate, through two examples, how electrical resistivity imaging can be utilized for characterization of sub-channel architecture. In the first example, electrodes installed in riparian boreholes and on the stream bed are used for imaging under the river bed the thickness and continuity of a highly permeable alluvial gravel layer overlying chalk. In the second example, electrical resistivity images, obtained using electrodes installed on the stream bed, provide a constrained estimate of the sediment volume behind a log jam, which had eluded measurement using conventional methods due to the boulder content of the stream. The two examples show that noninvasive electrical resistivity imaging is possible in complex stream environments and provide valuable information about the subsurface architecture beneath the stream channels.
NS34A-05 [WITHDRAWN]
Ultra-Wideband Vivaldi Antenna Array for High Resolution Subsurface Imaging
Use of ultra-wideband electromagnetic waves to image the subsurface yields enhanced resolution, provided sources, antennas and recording equipment can be developed and calibrated over the complete bandwidth of interest. We present a demonstration of the latest microwave transmission and recording technology to obtain high-resolution images. Our transmitter and receiver electronics are embodied in the vector network analyzer (PNA series) from Agilent, an eight-port vector network analyzer that records amplitude and phase. The analyzer is connected through a microwave multiplexer and microwave switch to an 8 element, balanced, antipodal Vivaldi antenna array, which can transmit or receive data over a bandwidth from 1.3 to 20 GHz. The bandwidth of the integrated system is determined by the bandwidth of the microwave switch, from DC to 18GHz, which interfaces the multiplexer to the PNA. The capabilities of a microwave multiplexer are employed to collect multi-channel data, by using one channel for transmission and reception on all 8 channels. The demonstration of this integrated system will be focussed on scattering from a single conducting cylinder as well as two cylinders, a dielectric and conducting cylinder, spaced at different intervals. The increased bandwidth, over that obtained in conventional GPR recording will result in pulses that have little ringing, allowing the detection of deeper reflections and eliminating any post-processing distortions that arise from deconvolving the traditional oscillatory waveform. Although the demonstration will be presented in the GHz bandwidth, suitable for imaging over a length scale to 1m, this integrated system will scale to lower bandwidths and can operate from 100 MHz to 3 GHz, with a resultant penetration depth of 10 to 20 m depending on the subsurface properties. Given the electronic constraints of the switch and the PNA, this scaling is simply achieved by enlarging the Vivaldi antenna dimensions. The advantage of using the PNA is the complete programmability, with a typical average noise level of -147 dBc and a dynamic range of 128 dB at the test-set ports. Scattering data obtained in this configuration will be imaged using time reversal techniques and an animation of a data set obtained using a previous generation of vector network analyzers will be presented. A minimum entropy stopping condition will be employed.
NS34A-06
3D and 4D GPR for Stratigraphic and Hydrologic Characterization of Field Sites
In a time of almost unlimited mobility, information, and connectivity it is surprising how our knowledge of natural systems becomes fragmented as soon as we enter the ground. Excavation, drilling, and 2D geophysics are unable to capture the spatio-temporal variability inside soil and rock volumes at the 1-10m scale. The problem is the lack of efficient and high-resolution imaging for the near surface domain. We have developed a high- resolution 3D Ground Penetrating Radar (GPR) system suitable for data acquisition at field sites. To achieve sharp and repeatable subsurface imaging we have integrated GPR with a rotary laser/IR strobe system. With 40 xyz coordinate updates per second, continuously moving GPR antennae can be tracked centimeter precise. A real-time LED guidance system shows the GPR antenna operator how to follow pre-computed survey tracks. Without having to stake out hundreds of survey tracks anymore one person now can scan an area of up to 600m2 per hour with a dual GPR antenna at 1m/s with 0.1m line spacing. The coordinate and GPR data are fused in real-time providing a first look of the subsurface in horizontal map view for quality control and in-field site assessment during data acquisition. The precision of the laser positioning system enables centimeter accurate repeat surveys to image and quantify water content changes in the vadose zone. To verify quantitative results of such 4D GPR we performed a controlled pond infiltration injecting 3200L of water from a 4x4m temporary pond with a thin soil layer and 5m of unsaturated porous limestone below. A total of sixteen repeated 3D GPR surveys were acquired just before the infiltration and in the following 2 weeks. All data were recorded with 250MHz antennae on a 5x10cm grid covering an area of 18x20m. Data processing included 3D migration and extraction of time shifts between pairs of time- lapse 3D GPR surveys. From the time shifts water content changes were computed using the Topp equation. The results show how zones experiencing the largest water content changes follow select stratigraphic boundaries. Many of these intervals correspond to thin shell hash layers sampled in core holes. After 42hrs the horizontal diameter of the wetting bulb was 7m. Background soil evaporation determined from GPR data recorded outside the infiltration area was 25mm of water column. The evaporation corrected GPR massbalance for the ponding experiment yields 3160L. This indicates that the 4D GPR estimates of water content changes have better than 5% accuracy.
NS34A-07
HMF-Geophysics: A Model for Collaborative Research in Hydrogeophysics
CUAHSI is developing, with the support of the NSF, a Hydrologic Measurement Facility (HMF). HMF-Geophysics is the near-surface geophysics module of HMF. Over the three years of the NSF grant (2005-08) we will determine, through broad community consultation, how best to utilize geophysical instrumentation and engage geophysical expertise in addressing key challenges in the hydrologic sciences. Our goal is to put in place the infrastructure needed to develop and maintain partnerships between the hydrologic and geophysical communities so that geophysical methods are used in a way that represents the state-of-the-science. Our current model consists of a central “node” that conducts feasibility studies to determine how/if geophysical methods could be of use in a hydrologic research project. In addition to the central node we have developed a system of affiliated nodes, individuals at 14 institutions who have committed to support HMF-Geophysics activities by offering equipment, software, and expertise. Once a feasibility study has shown the value of geophysics at a particular site, we match the hydrology PI with one of the nodes to develop the full-scale research project. We have conducted feasibility studies at 6 sites: Reynolds Creek Watershed, the H.J. Andrews Experimental Forest, and four WATERS test- beds, the latter are described below. The objective of the Baltimore test-bed is to quantify the urban water cycle, with an emphasis on groundwater, using the Gwynns Falls watershed as a pilot study area. Electrical resistivity imaging, ground penetrating radar, and seismic refraction were assessed as a means of determining depth to bedrock or to the water table within the riparian zone of urbanized streams. A regional time-lapse microgravity survey was conducted at the 200 sq- km watershed scale to infer the storage change in the underlying aquifers. Research in the Crown of the Continent test-bed in Montana is focused on understanding the interactions between the river channel and the subsurface water of its associated bars and floodplain. We used a combination of surface and borehole electrical resistivity tomography to image the geometry and internal structure of the sediment packages in these floodplain deposits. By transforming the geophysical parameters, using high- resolution aquifer testing conducted in the boreholes, our plan is to identify preferential flowpaths known to exist in the heterogeneous sediments. One of the interests of the Sierra Nevada test-bed is the surface–groundwater interaction that occurs in the alpine meadows. The purpose of the geophysics was to assist in identifying the subsurface flowpaths. We provided subsurface characterization using electrical resistivity and ground penetrating radar and conducted a series of time-lapse electromagnetic induction surveys during the snow melt season. The Clear Creek test-bed in Iowa is an intensively managed watershed comprising wetlands, agricultural and urbanized catchments. We used electromagnetic induction mapping to obtain high-resolution ground conductivity maps. These will be calibrated against infiltration experiments run at numerous locations to produce a spatial map of hydraulic properties of the soil. This will assist in the study of the link between land use and high soil erosion rates in the watershed.