Parts of the Raritan River basin in central New Jersey have undergone increasing development over the last several decades. The increasing population relies on the region's ground water and surface water sources for its residential, commercial, and industrial water supply. Urbanization, regionalized wastewatertreatment facilities, stream channel alterations, and interbasin transfers of water can all affect water availability. This pilot study was conducted to determine whether significant trends exist in the base-flow and overland-runoff characteristics of streams in two subbasins with different percentages of urban/built-up land (Anderson et at., 1976). Changes in flow characteristics that could indicate future reductions in safe water yield of the Raritan River basin were examined. Flow and flow variability of the steams draining these two subbasins have increased over time. Many of the flow measures studied experienced pronounced trend shifts about 1960. The cause of these changes cannot be readily determined from the data, nor is it clear whether the increased flow variability lies outside the natural range of flow variability of the streams draining the subbasins. (KEY TERMS: land use; trend analysis; time series; LOWESS smoothing.)

Efforts to relate shallow ground‐water quality to the land use near a well lead to several statistical difficulties. These include potential uncertainty in land‐use categorical data due to misclassification, data closure, distributional skewing, and spatial autocorrelation. Methods of addressing these problems are, respectively, the establishment of limits on minimum buffer radius, the estimation of contrasts, rank‐based tests of association, and sub‐sampling to prevent buffer overlap. Relations between the presence of purgeable organic compounds in ground water and land use are used to illustrate these problems and methods.

No abstract

Electromagnetic (EM) logging provides an efficient method for high‐resolution, vertical delineation of electrically conductive contamination in glacial sand‐and‐gravel aquifers. LM. gamma, and lithologic logs and specific conductance data from sand‐and‐gravel aquifers at five sites in the northeastern United States were analyzed to define the relation of KM conductivity to aquifer lithology and water quality. Municipal waste disposal, septic waste discharge, or highway deicing salt application at these sites has caused contaminant plumes in which the dissolved solids concentration and specific conductance of ground water exceed background levels by as much as 10 to 20 limes. The major hydrogeologic factors that affected KM log response at the five sites were the dissolved solids concentration of the ground water and the silt and clay content in the aquifer. KM conductivity of sand and gravel with uncontaminated water ranged from less than 5 to about 10 millisiemens per meter (mS/m); that of silt and clay zones ranged from about 15 to 45 mS/m: and that of the more highly contaminated zones in sand and gravel ranged from about 10 to more than 80 mS/m. Specific conductance of water samples from screened intervals in sand and gravel at selected monitoring well installations was significantly correlated with KM conductivity. CM logging can be used in glacial sand‐and‐gravel aquifer investigations to (1) determine optimum depths for the placement of monitoring well screens: (2) provide a nearly continuous vertical profile of specific conductance to complement depth‐specific water quality samples; and (3) identify temporal changes in water quality through sequential logging. Detailed lithologic or gamma logs, preferably both, need to be collected along with the F.M logs to define zones in which elevated EM conductivity is caused by the presence of sill and clay beds rather than contamination.

No abstract