Geology topics

Thor E. Smith

Publications and source records attributed to Thor E. Smith.

8 recordsLinked to original sources

USGS2016

Groundwater level trends and drivers in two northern New England glacial aquifers

We evaluated long-term trends and predictors of groundwater levels by month from two well-studied northern New England forested headwater glacial aquifers: Sleepers River, Vermont, 44 wells, 1992-2013; and Hubbard Brook, New Hampshire, 15 wells, 1979-2004. Based on Kendall Tau tests with Sen slope determination, a surprising number of well-month combinations had negative trends (decreasing water levels) over the respective periods. Sleepers River had slightly more positive than negative trends overall, but among the significant trends ( p < 0.1), negative trends dominated 67 to 40. At Hubbard Brook, negative trends outnumbered positive trends by a nearly 2:1 margin and all seven of the significant trends were negative. The negative trends occurred despite generally increasing trends in monthly and annual precipitation. This counterintuitive pattern may be a result of increased precipitation intensity causing higher runoff at the expense of recharge, such that evapotranspiration demand draws down groundwater storage. We evaluated predictors of month-end water levels by multiple regression of 18 variables related to climate, streamflow, snowpack, and prior month water level. Monthly flow and prior month water level were the two strongest predictors for most months at both sites. The predictive power and ready availability of streamflow data can be exploited as a proxy to extend limited groundwater level records over longer time periods.

New Hampshire, Vermont↗

USGS2014

Simulated and observed 2010 floodwater elevations in the Pawcatuck and Wood Rivers, Rhode Island

Heavy, persistent rains from late February through March 2010 caused severe flooding that set, or nearly set, peaks of record for streamflows and water levels at many long-term U.S. Geological Survey streamgages in Rhode Island. In response to this flood, hydraulic models of Pawcatuck River (26.9 miles) and Wood River (11.6 miles) were updated from the most recent approved U.S. Department of Homeland Security-Federal Emergency Management Agency flood insurance study (FIS) to simulate water-surface elevations (WSEs) for specified flows and boundary conditions. The hydraulic models were updated to Hydrologic Engineering Center-River Analysis System (HEC-RAS) using steady-state simulations and incorporate new field-survey data at structures, high resolution land-surface elevation data, and updated flood flows from a related study. The models were used to simulate the 0.2-percent annual exceedance probability (AEP) flood, which is the AEP determined for the 2010 flood in the Pawcatuck and Wood Rivers. The simulated WSEs were compared to high-water mark (HWM) elevation data obtained in a related study following the March–April 2010 flood, which included 39 HWMs along the Pawcatuck River and 11 HWMs along the Wood River. The 2010 peak flow generally was larger than the 0.2-percent AEP flow, which, in part, resulted in the FIS and updated model WSEs to be lower than the 2010 HWMs. The 2010 HWMs for the Pawcatuck River averaged about 1.6 feet (ft) higher than the 0.2-percent AEP WSEs simulated in the updated model and 2.5 ft higher than the WSEs in the FIS. The 2010 HWMs for the Wood River averaged about 1.3 ft higher than the WSEs simulated in the updated model and 2.5 ft higher than the WSEs in the FIS. The improved agreement of the updated simulated water elevations to observed 2010 HWMs provides a measure of the hydraulic model performance, which indicates the updated models better represent flooding at other AEPs than the existing FIS models.

Rhode Island↗

USGS2012

Time series geophysical monitoring of permanganate injections and in situ chemical oxidation of PCE, OU1 area, Savage Superfund Site, Milford, NH, USA

In situ chemical oxidation (ISCO) treatment with sodium permanganate, an electrically conductive oxidant, provides a strong electrical signal for tracking of injectate transport using time series geophysical surveys including direct current (DC) resistivity and electromagnetic (EM) methods. Effective remediation is dependent upon placing the oxidant in close contact with the contaminated aquifer. Therefore, monitoring tools that provide enhanced tracking capability of the injectate offer considerable benefit to guide subsequent ISCO injections. Time-series geophysical surveys were performed at a superfund site in New Hampshire, USA over a one-year period to identify temporal changes in the bulk electrical conductivity of a tetrachloroethylene (PCE; also called tetrachloroethene) contaminated, glacially deposited aquifer due to the injection of sodium permanganate. The ISCO treatment involved a series of pulse injections of sodium permanganate from multiple injection wells within a contained area of the aquifer. After the initial injection, the permanganate was allowed to disperse under ambient groundwater velocities. Time series geophysical surveys identified the downward sinking and pooling of the sodium permanganate atop of the underlying till or bedrock surface caused by density-driven flow, and the limited horizontal spread of the sodium permanganate in the shallow parts of the aquifer during this injection period. When coupled with conventional monitoring, the surveys allowed for an assessment of ISCO treatment effectiveness in targeting the PCE plume and helped target areas for subsequent treatment.

New Hampshire↗

USGS2012

Century-scale perspective on water quality in selected river basins of the conterminous United States

Nutrient pollution in the form of excess nitrogen and phosphorus inputs is a well-known cause of water-quality degradation that has affected water bodies across the Nation throughout the 20th century. The recognition of excess nutrients as pollution developed later than the recognition of other water-quality problems, such as waterborne illness, industrial pollution, and organic wastes. Nevertheless, long-term analysis of nutrient pollution is fundamental to our understanding of the current magnitude of the problem, as well the origins and the effects. This report describes the century-scale changes in water quality across a range streams in order to place current water-quality concerns in historical context and presents this history on a national scale as well as for selected river reaches. The primary focus is on nutrient pollution, but the development and societal responses to other water-quality problems also are considered. Land use and agriculture in the selected river reaches also are analyzed to consider how these factors may relate to nutrient pollution. Finally, the availability of relevant nutrient and inorganic carbon data are presented for the selected river reaches. Sources of these data included Federal agencies, State-level reports, municipal public works facilities, public health surveys, and sanitary surveys. The availability of these data extends back more than a century for most of the selected river reaches and suggests that there is a tremendous opportunity to document the development of nutrient pollution in these river reaches.

Scientific Investigations Report↗

USGS2010

Effects of Highway Road Salting on the Water Quality of Selected Streams in Chittenden County, Vermont, November 2005-2007

A study of road-deicing chloride (Cl) concentrations and loads was conducted at three streams in Chittenden County, VT, from November 2005 to 2007. This study was done by the U.S. Geological Survey, in cooperation with the Vermont Agency of Transportation. The streams, Alder Brook, Allen Brook, and Mill Brook, were selected to represent different land uses in the upstream watershed, different road types and densities, and different geometric patterns of the roadway draining to the receiving stream to assess the relative contribution of and differences in state road-salt applications to stream Cl concentrations and loads. Water-quality samples were collected and specific conductance was measured continuously at paired stations upstream and downstream from State highways and related to Cl concentrations to assist in determining the effects of road-salting operations during winter maintenance on the levels of Cl in the streams. Mean concentrations of Cl ranged from 8.2 to 72 mg/L (milligrams per liter) in the water-quality samples collected at sampling stations upstream from State highway bridges and from 7.9 to 80 mg/L in those collected at sampling stations downstream of highway bridges. Mean Cl loads ranged from 1,100 to 4,090 lb/d (pounds per day) at upstream stations and from 1,110 to 4,200 lb/d at downstream stations. Estimated mean annual Cl loads ranged from 402,000 to 1,490,000 lb/yr (pounds per year) at upstream stations and from 405,000 to 1,530,000 lb/yr at downstream stations. Mean Cl concentrations in samples collected at the three paired stations were lowest at Mill Brook at VT 117 near Essex Junction, VT (7.9 mg/L) and highest at Allen Brook at VT 2A near Essex Junction, VT (80.7 mg/L). None of the monitored Cl concentrations in the water-quality samples collected at the three paired sampling stations exceeded either of the U.S. Environmental Protection Agency's (USEPA) recommended chronic and acute Cl toxicity criteria of 230 and 860 mg/L, respectively. A fourth stream site, a small tributary draining to Alder Brook between the upstream and downstream stations, was monitored from December 2006 to November 2007. This tributary collected runoff from a state highway and an interchange before flowing through a wetlands retention basin. The mean Cl concentration in water-quality samples collected at the tributary was 449 mg/L. The USEPA recommended chronic toxicity criterion of 230 mg/L was exceeded about 65 percent of the monitoring period. The USEPA recommended acute toxicity criterion of 860 mg/L was not exceeded. Estimated Cl loads below the State highway bridges exceeded loads above the bridges at all three paired stations during both years of the study. The differences in the annual loads between the upstream and downstream stations were 0.7, 3.0 and 14 percent at Mill, Allen, and Alder Brooks, respectively. Almost all of the difference (92 percent) at Alder Brook was due to the tributary. Cl applied by the State of Vermont for deicing purposes represented less than 20 percent of the annual estimated Cl load in all 3 streams below the state highways. The highest monthly Cl loads during the first year of the study were observed in January 2006 at all three stream stations because of an early snowmelt event. The highest monthly Cl loads during the second year of the study were observed in April 2007 at all three streams during spring snowmelt and were followed by decrease in Cl loading through the summer. Generally, the relation of Cl loads to runoff was similar at all three streams. In July and October 2007, loads increased slightly with an increase in runoff, indicating that Cl in the soils and groundwater may be contributing to the Cl levels during the summer and fall, well after the road-salting season. Cl loads in all three streams appear to be due primarily to sources in the watersheds upstream of the state highway bridge where road salt was applied and (or) Cl retained in soils and streambed

Scientific Investigations Report↗

USGS2006

Assessment of total nitrogen in the upper Connecticut River basin in New Hampshire, Vermont, and Massachusetts, December 2002–September 2005

A study of total nitrogen concentrations and loads was conducted from December 2002 to September 2005 at 13 river sites in the upper Connecticut River Basin. Ten sites were selected to represent contributions of nitrogen from forested, agricultural, and urban land. Three sites were distributed spatially on the main stem of the Connecticut River to assess the cumulative total nitrogen loads. To further improve the understanding of the sources and concentrations and loads of total nitrogen in the upper Connecticut River Basin, ambient surface water-quality sampling was supplemented with sampling of effluent from 19 municipal and paper mill wastewater-treatment facilities. Mean concentrations of total nitrogen ranged from 0.19 to 2.8 milligrams per liter (mg/L) at river sampling sites. Instantaneous mean loads of total nitrogen ranged from 162 to 58,300 pounds per day (lb/d). Estimated mean annual loads of total nitrogen ranged from 49,100 to 21.6 million pounds per year (lb/yr) with about 30 to 55 percent of the loads being transported during the spring. The estimated mean annual yields of total nitrogen ranged from 1,190 to 7,300 pounds per square mile per year (lb/mi 2 )/yr. Mean concentrations of total nitrogen ranged from 4.4 to 30 mg/L at wastewater-treatment sampling sites. Instantaneous mean loads of total nitrogen from municipal wastewater-treatment facilities ranged from 36 to 1,780 lb/d. Instantaneous mean loads of total nitrogen from paper mill wastewater-treatment facilities ranged from 96 to 160 lb/d. The median concentration of total nitrogen was 0.24 mg/L at forested sites, 0.48 mg/L at agricultural sites, 0.54 mg/L at urban sites, 0.48 mg/L at main-stem sites, and 14 mg/L at wastewater-treatment sites. Concentrations of total nitrogen at forested sites were significantly less than at all other site types (p<0.05). Concentrations of total nitrogen at agricultural, urban, and main stem sites were not significantly different among each other (p>0.05) but were significantly greater (p<0.05) than at forested sites and significantly less than concentrations at wastewater-treatment sites (p<0.05). Total nitrogen concentrations at wastewater-treatment sites were significantly different from all other site types (p<0.05). Annual yields of total nitrogen ranged from 732 to 1,920 (lb/mi 2 )/yr at forested sites; 1,550 to 2,980 (lb/mi 2 )/yr at agricultural sites; 1,280 to 1,860 (lb/mi 2 )/yr at urban sites that were not directly affected by wastewater effluent; 7,090 to 7,770 (lb/mi 2 )/yr at an urban site directly affected by wastewater effluent; and 1,300 to 2,390 (lb/mi 2 )/yr at main-stem sites. In this study, the mean annual load and yield of total nitrogen at the Connecticut River at Wells River, VT, was estimated at 4.47 million lb/yr and 1,690 (lb/mi 2 )/yr, respectively. The mean annual load and yield of total nitrogen at the Connecticut River at North Walpole, NH, was estimated at 9.60 million lb/yr and 1,750 (lb/mi 2 )/yr, respectively. The mean annual load and yield of total nitrogen leaving the upper Connecticut River Basin, as estimated at the Connecticut River at Thompsonville, CT, was 21.6 million lb/yr and 2,230 (lb/mi 2 )/yr, respectively.

Massachusetts, New Hampshire, Vermont↗

USGS2005

Effects of urbanization on stream quality at selected sites in the seacoast region in New Hampshire, 2001-03

A study of selected water-quality and macroinvertebrate community data was conducted at 10 stream sites in the Seacoast region of New Hampshire to determine if a relation is present between stream quality and the extent of urbanization in a watershed. Watersheds with similar characteristics, but varying in their degree of urban development, were studied. The percent of impervious surface, the percent of urban land use in a watershed, and the percent of urban land use in two types of stream buffers were compared and correlated with stream-quality variables. Specific conductance, turbidity, nitrite plus nitrate yields, and selected macroinvertebrate community data were significantly correlated with most measures of urbanization used in this study; however, concentrations and total phosphorus yields were not statistically correlated with most measures of urbanization in this study. The measures of urbanization that had the highest correlations with stream-quality variables were those measures that were associated with the percent of urban land in buffer zones near and upstream of a sampling site. A water-quality and habitat conditions score was negatively correlated with the percent of urban land in a 1-kilometer radial buffer of the sampling site (rho (r) = -0.86; p < 0.001), the percent of impervious surface (r = -0.70; p < 0.05), and the percent of urban land in the watershed (r = -0.67; p < 0.05). A biological condition score also was negatively correlated with the percent of urban land in a 1-kilometer radial buffer of the sampling site (r = -0.95; p < 0.0001), the percent of impervious surface (r = -0.75; p < 0.05), and the percent of urban land in the watershed (r = -0.79; p < 0.01). The percent of urban land in a 25-meter stream buffer along the stream corridor also had negative correlations with a water-quality and habitat conditions score (r = -0.80; p < 0.01) and a biological condition score (r = -0.86; p < 0.01). Mean Ephemeroptera, Plecoptera, and Trichoptera (EPT) taxa richness showed a response to urbanization in a watershed, indicating that EPT taxa richness may be an appropriate metric to evaluate the effects of urban land use on small streams in this region. Results from this study indicate that the percent of urban land use in buffer zones and the percent of impervious surface in a watershed can be used as indicators of stream quality.

Scientific Investigations Report↗

USGS2002

Travel times and dispersion of soluble dye in thirteen New Hampshire rivers

Dye was injected and traced in 13 New Hampshire rivers in 2000 to determine the velocity at which a soluble substance spilled into the river would travel to drinking-water supply intakes. Travel times were studied on the Ammonoosuc, Androscoggin, Connecticut, Contoocook, East Branch Pemigewasset, Exeter, Lamprey, Mascoma, Merrimack, Oyster, Piscassic, Salmon Falls, and Sugar Rivers. Dye was injected and sampled at low and mean flows, and the measured velocities extrapolated to provide an estimate of the velocity during a mean annual flood (mean of yearly peak flows for a specific river). Rivers were sampled downstream of the dye-injection sites to measure dye concentrations and the arrival and passage times of the dye cloud. This information was used to estimate a relation between river discharge and the expected dye concentrations and velocities for each river. Results of the dye tests were used to estimate a 6-hour travel distance at mean annual flood for each of the 13 rivers, and to create graphical relations that can be used to estimate the travel times and concentrations of solutes over a range of river discharges and distances. Observed transport velocities ranged from 0.03 to 2.4 feet per second, and the estimated 6-hour travel distances ranged from 5.8 to 43 miles. Rivers in upland areas had the highest velocities, whereas rivers in the low coastal region had the slowest velocities.

New Hampshire↗