Abstract

This report by the U.S. Geological Survey, in cooperation with the San Antonio Water System, describes the results of a statistical analysis of major ion and trace element geochemistry of water at seven wells transecting the freshwater/saline-water interface of the Edwards aquifer in San Antonio, Texas, either over time or in response to variations in hydrologic conditions. The data used in this report were collected during 1986–2006. The seven monitoring wells are screened at different depths in the aquifer at three sites that form a generally north-to-south transect. The three wells of the southern site and the deeper of the two middle-site wells are open to the freshwater/saline-water transition zone, which contains saline water. The shallower well of the middle site and the two wells of the northern site are open to the freshwater zone.

Mean specific conductance (SC) values were greater at transition-zone wells than at freshwater-zone wells, but SC did not vary systematically with depth. Concentrations of all major ions except bicarbonate were greater at transition-zone wells than at freshwater-zone wells, but concentrations tended to be more variable at freshwater-zone wells. Mean molar ratios of magnesium:calcium, sulfate:chloride, and sodium:chloride were similar at transition-zone wells and freshwater-zone wells. Concentrations of trace elements for many water samples at the seven transect wells were below the laboratory analytical reporting level. Detections of trace elements were more frequent at transition-zone wells, and mean concentrations of cadmium, chromium, copper, lead, and silver were elevated at transition-zone wells relative to freshwater-zone wells.

All strong correlations between SC and major ions were positive, and in general there were more and stronger correlations between SC and major ions in the water from the freshwater-zone wells than from the transition-zone wells. Except for the shallowest transition-zone well, the transition-zone wells had relatively few strong correlations overall. The lack of a strong correlation indicates that much of the variability in the major ion concentrations at these wells might be a result of analytical variability caused by the multiple laboratory analytical methods used. In most cases, strong correlations between concentrations of trace elements were positive, and transition-zone wells and freshwater-zone wells had water with a similar number of significant correlations.

Principal components analysis indicates dilution of ground water by low-ionic-strength meteoric water at the three freshwater-zone wells and at the shallowest transition-zone well. At the two deeper transition-zone wells at the southern site, principal components analysis indicates that there is no systematic variation in major ion concentrations. At three transition-zone wells, there was a general trend toward less salinity over the 21-year period of sampling. Trends in SC at the freshwater-zone wells were less consistent. There is no systematic change in the direction of trend in SC by water type (saline or fresh), between sites, or with depth. In general, trends in major ion concentrations corresponded to those in SC. For each trace element over the 21-year sampling period, there was either no trend or a downward trend.

Relations between SC, major ions, and major ion molar ratios and hydrologic indicators (concurrent or prior time-averaged measures of water level and effective rainfall) were investigated. Correlations between geochemical variables and measures of water level in the freshwater-zone wells were much more frequent than correlations between geochemical variables and measures of water level in the transition-zone wells. There were correlations between SC and all measures of water level at the two freshwater-zone wells at the northern site, but there were no correlations between SC and any measures of water level at any transition-zone wells. SC was correlated with effective rainfall at all freshwater-zone wells and at one transition-zone well.

The statistical analyses taken together indicate that the geochemistry at the freshwater-zone wells is more variable than that at the transition-zone wells. The geochemical variability at the freshwater-zone wells might result from dilution of ground water by meteoric water. This is indicated by relatively constant major ion molar ratios; a preponderance of positive correlations between SC, major ions, and trace elements; and a principal components analysis in which the major ions are strongly loaded on the first principal component. Much of the variability at three of the four transition-zone wells might result from the use of different laboratory analytical methods or reporting procedures during the period of sampling. This is reflected by a lack of correlation between SC and major ion concentrations at the transition-zone wells and by a principal components analysis in which the variability is fairly evenly distributed across several principal components. The statistical analyses further indicate that, although the transition-zone wells are less well connected to surficial hydrologic conditions than the freshwater-zone wells, there is some connection but the response time is longer.

First posted February 10, 2009