The United States Geological Survey and the University of Washington collaborated on a series of initial experiments on the Lewis, Toutle, and Cowlitz Rivers during September 2000 and a detailed experiment on the Cowlitz River during May 2001 to determine the feasibility of using helicopter‐mounted radar to measure river discharge. Surface velocities were measured using a pulsed Doppler radar, and river depth was measured using ground‐penetrating radar. Surface velocities were converted to mean velocities, and horizontal registration of both velocity and depth measurements enabled the calculation of river discharge. The magnitude of the uncertainty in velocity and depth indicate that the method error is in the range of 5 percent. The results of this experiment indicate that helicopter‐mounted radar can make the rapid, accurate discharge measurements that are needed in remote locations and during regional floods.

No abstract

The suitability of common‐offset ground‐penetrating radar (GPR) to detect free‐phase hydrocarbons in bedrock fractures was evaluated using numerical modeling and physical experiments. The results of one‐ and two‐dimensional numerical modeling at 100 megahertz indicate that GPR reflection amplitudes are relatively insensitive to fracture apertures ranging from 1 to 4 mm. The numerical modeling and physical experiments indicate that differences in the fluids that fill fractures significantly affect the amplitude and the polarity of electromagnetic waves reflected by subhorizontal fractures. Air‐filled and hydrocarbon‐filled fractures generate low‐amplitude reflections that are in‐phase with the transmitted pulse. Water‐filled fractures create reflections with greater amplitude and opposite polarity than those reflections created by air‐filled or hydrocarbon‐filled fractures. The results from the numerical modeling and physical experiments demonstrate it is possible to distinguish water‐filled fracture reflections from air‐ or hydrocarbon‐filled fracture reflections, nevertheless subsurface heterogeneity, antenna coupling changes, and other sources of noise will likely make it difficult to observe these changes in GPR field data. This indicates that the routine application of common‐offset GPR reflection methods for detection of hydrocarbon‐filled fractures will be problematic. Ideal cases will require appropriately processed, high‐quality GPR data, ground‐truth information, and detailed knowledge of subsurface physical properties. Conversely, the sensitivity of GPR methods to changes in subsurface physical properties as demonstrated by the numerical and experimental results suggests the potential of using GPR methods as a monitoring tool. GPR methods may be suited for monitoring pumping and tracer tests, changes in site hydrologic conditions, and remediation activities.

S U M M A R YCrosshole-radar velocity tomography is increasingly being used to characterize the electrical and hydrologic properties of the Earth's near-surface. Because radar methods are sensitive to the water content of geologic materials, velocity tomography is a good proxy for imaging soil water retention in the vadose zone and porosity in the saturated zone. In many near-surface environments, radar velocity varies over a few orders of magnitude. Common velocity tomography applies ray theory that assumes infinite frequency propagation. The ray approximation may induce velocity modelling artefacts and loss of localization. We propose an alternative method for computing velocity tomogram sensitivities using Fresnel volumes based on first-order scattering. The Fresnel volume sensitivities account for the finite-frequency of the crosshole radar signal and model the physics of radar propagation more accurately than the ray theory approximation.We demonstrate that applying finite-frequency Fresnel volume sensitivities provides improved radar velocity tomograms in low contrast environments. Analysis of the singular value decomposition of the sensitivity matrix demonstrates how the finite-frequency inversion recovers and localizes velocity heterogeneities better than ray theory. The singular value spectrum obtained from the full waveform sensitivities matches well with the Fresnel volume results. Furthermore, these basis functions are smooth and localized because the kernels capture the first order wave propagation effect compared to ray based sensitivity, which is a high frequency approximation. Through forward modelling experiments, we validate the finite-frequency sensitivity for crosshole radar velocity. In the Fresnel volume approach, the traveltime picking is more efficient because the datum is the peak of the first pulse rather than the first arrival, and therefore, data pre-processing is simpler and may be easily automated. The synthetic Fresnel volume inversion results show improvements in the final model and the data fits are better when compared to the ray theoretical inversions.

Vertical-radar profiles (VRPs) and neutron porosity logs were acquired at two sites in New England -Haddam Meadows State Park in Connecticut and Massachusetts Military Reservation on Cape Cod. Both sites include boreholes drilled to depths from 30 to 50 meters into unconsolidated fluvial or glacial sediments. The VRP data are inverted using Tikhanov regularization to obtain interval radar propagation velocities. Of the two sites, the radar velocities at Haddam Meadows State Park show more variability with depth because this site consists of poorly sorted fluvial sediments, whereas the radar velocities at Cape Cod show much less variability because this site consists of well-sorted glacial sediments.The interval radar propagation velocities from the VRPs are converted to estimates of saturated sediment porosity using the Topp and time-propagation petrophysical models. VRPderived porosities are compared to neutron log-derived porosities and yielded a correlation between values derived from the two methods. Lack of correlation between the VRP-derived porosities and the neutron log-derived porosities at some depths may be explained by discrepancy in the sample volume of each method, by problems in the petrophysical models, or by differences in borehole construction methods used at each site. Overall correlation between the VRP-derived porosities and the neutron log-derived porosities supports the advantage of deriving porosities from VRP data due to the decreased cost and ease of data acquisition, and simple processing and inversion routines.