SEARCH · Search USGS

Results for “Input”

Search indexed USGS publications on groundwater, aquifers, geologic maps, mineral resources and earthquakes. Explore source records by subject and place.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 253 records · Page 14Linked to original sources

USGS2001

Effect of canopy removal on snowpack quantity and quality, fraser experimental forest, Colorado

Snowpack peak water equivalent (PWE), ion concentration, content, and spatial distribution of ion load data from spring 1987-1996 in a 1 ha clearcut and adjacent forested plots vegetated by mature Picea engelmannii and Abies lasiocarpa in the Fraser experimental forest (FEF), Colorado are presented. Our objectives were: (1) to see if a forest opening might redistribute snowfall, snowpack moisture, and snowpack chemical content, and (2) to examine the importance of canopy interception on snowpack quantity and chemistry. On an average, the canopy intercepted 36% of snowfall. Interception was correlated with snowfall amount, snowpack PWE beneath the canopy, and air temperature. Canopy removal increased snowpack PWE to >90% cumulative snowfall inputs. Snowpack K-, H-, and NH4+ concentrations on the clearcut were lower and NO3- higher than in the snowpack beneath the forested plots. Cu mulative snowfall K+ input was less than in the clearcut snowpack; H+ inputs were greater in snowfall than in the snowpack of any plot; and inorganic N (NO3- and NH4+) inputs from snowfall to the clearcut were greater than to the forested plots. Processes accounting for the differences between snowfall inputs and snowpack ion content were leaching of organic debris in the snowpack, differential elution of the snowpack, and canopy retention. There were significant trends by year in snowpack ion content at PWE without similar trends in snowfall inputs. This finding coupled with snowpack ion elution bring into question the use of snowpack chemistry as an indicator of winter atmospheric inputs in short-term studies. ?? 2001 Elsevier Science B.V.

Journal of Hydrology↗

USGS2009

Quantifying the spatial details of carbon sequestration potential and performance

Upscaling the spatial and temporal changes of carbon stocks and fluxes from sites to regions is challenging owing to the spatial and temporal variances and covariance of driving variables and the uncertainties in both the model and the input data. Although various modeling approaches have been developed to facilitate the upscaling process, few deal with error transfer from model input to output, and error propagation in time and space. The author has developed the General Ensemble Biogeochemical Modelling System (GEMS) for upscaling carbon stocks and fluxes from sites to regions with measures of uncertainty. This chapter describes the GEMS model, its application to regional- and larger-scale areas, and the new results that demonstrate the challenges of upscaling. GEMS relies on site-scale biogeochemical models to simulate carbon dynamics at the site scale. The spatial deployment of the site-scale model in GEMS is based on the spatial and temporal joint frequency distribution of major driving variables (e.g., land cover and land use change, climate, soils, disturbances, and management). At the site scale, GEMS uses stochastic ensemble simulations to incorporate input uncertainty to quantify uncertainty transfer from input to output, and to identify trends in both input data and simulation results. It permits one to simulate the range of possible permutations of input values and identify the trends and variance in both the input data and results. Using data assimilation techniques, GEMS simulations can be constrained by field and satellite observations, including estimates of net primary production (NPP) from the Moderate Resolution Imaging Spectroradiometer (MODIS), grain yield and cropping practices, and forest inventories. The modeling philosophy embedded in GEMS makes it ideal for assimilating information with various uncertainties to support estimating the spatial details of carbon sequestration potential as well as dynamic monitoring of the performance of carbon sequestration activities over large areas. As a case study, GEMS is applied to simulate the spatial and temporal details of carbon sources, sinks, and uncertainty in the Ridge and Valley ecoregion in the eastern United States

Book chapter↗

USGS1993

Analysis of summer phosphorus fluxes within the pelagic zone of Eau Galle Reservoir, Wisconsin

Major phosphorus (P) fluxes to and from the pelagic zone (i.e., open water region including epilimnion, metalimnion, and hypolimnion) were estimated from data collected over a 6 year period during the summer in Eau Galle Reservoir, Wisconsin. P inputs to the pelagic zone included profundal sediments, the watershed, groundwater, and transport of P from the littoral zone. P outputs from the pelagic zone included discharge from the reservoir, deposition, and transport of P to the littoral zone. Nighttime convective circulation was assumed to be the dominant mechanism of P exchange between the littoral and pelagic zones. Littoral P inputs, often neglected from budgetary analyses, accounted for 15% of the total measured P input and 25% of the internal P input to the pelagic zone. External P inputs were greatest, accounting for 42% of the total measured P input to the pelagic zone. These results emphasize the need for control of various sources of P inputs in the development of lake and reservoir management strategies.

Lake and Reservoir Management↗

USGS1999

Long-term consequences of disturbance on nitrogen dynamics in an arid ecosystem

Anthropogenic activity is causing dramatic changes in the nitrogen (N) cycle in many ecosystems. Most research has focused on the increase in N input caused by atmospheric deposition and invasion of N-fixing species, and on their effects on resource availability and species composition. However, in contrast to many ecosystems experiencing large increases in N input, many arid ecosystems are experiencing loss of nutrients due to land-use change. An important component of many arid ecosystems on a worldwide basis is the microbiotic crust, a biological soil crust composed of lichens, cyanobacteria, mosses, and algae. Nitrogen fixation by lichens and cyanobacteria comprising the crust is the primary source of N input in many of these ecosystems. We quantified the long-term consequences of surface disturbance in an arid ecosystem on the Colorado Plateau by comparing pristine sites with those of known disturbance history. Disturbance caused an increase in the abundance of cyanobacteria and a decrease in lichens within the microbiotic crust. Carbon isotope composition (δ 13 C) of the crust reflects this shift in species composition; values for disturbed sites were 4.5‰ higher than undisturbed sites. Nitrogen isotope composition (δ 15 N) of the microbiotic crust was 1.5–2.2‰ higher for disturbed sites, probably resulting from relatively greater gaseous N loss from the crust. Historic disturbance has caused a long-term decrease in rates of N fixation by the microbiotic crust; nitrogenase activity in pristine sites was 250% greater than sites intermittently disturbed 30 yr ago. The decrease in N input from fixation and continued gaseous N loss has caused a 25–75% decrease in soil N content. Altering relative rates of N input and loss, coupled with input of N from microbiotic crusts with relatively higher δ 15 N, has caused an increase in soil and plant δ 15 N at disturbed sites. This decrease in soil N caused by disturbance will likely cause changes in species composition similar to those observed in ecosystems that have been disrupted by excess N input from atmospheric deposition.

Ecology↗

USGS1999

Seasonal relationships between precipitation, forest floor, and streamwater nitrogen, Isle Royale, Michigan

The Upper Great Lakes receive large amounts of precipitation-NH + 4 and moderate NO - 3 inputs. Increased atmospheric inorganic N input has led to concern about ecosystem capacity to utilize excess N. This paper summarizes a 5-yr study of seasonal N content and flux in precipitation, snowpack, forest floor, and streamwater in order to assess the source of inorganic N outputs in streamflow from a small boreal watershed. Average precipitation N input was 3 kg ha -1 yr -1 . The peak snowpack N content averaged 0.55 kg ha -1 . The forest floor inorganic N pool was ≈ 2 kg ha -1 , eight times larger than monthly precipitation N input. The inorganic N pool size peaked in spring and early summer. Ninety percent of the forest floor inorganic N pool was made up of NH + 4 -N. Forest floor inorganic N pools generally increased with temperature. Net N mineralization was 15 kg ha -1 yr -1 , and monthly rates peaked in early summer. During winter, the mean monthly net N mineralization rate was twice the peak snowpack N content. Streamwater NO - 3 concentration peaked in winter, and inorganic N output peaked in late fall. Beneath the dominant boreal forest species, net N mineralization rates were positively correlated ( P < 0.05) with streamwater NO - 3 concentrations. Forest floor NO - 3 pools beneath alder [ Alnus rugosa (Du Roi) Spreng] were positively correlated ( P < 0.01) to streamwater NO - 3 output. At the watershed mouth, streamwater NO - 3 concentrations were positively correlated ( P < 0.05) with precipitation NO - 3 input and precipitation amount. The relatively small snowpack N content and seasonal precipitation N input compared to forest floor inorganic N pools and net N mineralization rates, the strong ecosystem retention of precipitation N inputs, and the seasonal streamwater NO - 3 concentration and output pattern all indicated that little streamwater NO - 3 came directly from precipitation or snowmelt.

Michigan↗

USGS2013

Hydrology and water quality of Shell Lake, Washburn County, Wisconsin, with special emphasis on the effects of diversion and changes in water level on the water quality of a shallow terminal lake

Shell Lake is a relatively shallow terminal lake (tributaries but no outlets) in northwestern Wisconsin that has experienced approximately 10 feet (ft) of water-level fluctuation over more than 70 years of record and extensive flooding of nearshore areas starting in the early 2000s. The City of Shell Lake (City) received a permit from the Wisconsin Department of Natural Resources in 2002 to divert water from the lake to a nearby river in order to lower water levels and reduce flooding. Previous studies suggested that water-level fluctuations were driven by long-term cycles in precipitation, evaporation, and runoff, although questions about the lake’s connection with the groundwater system remained. The permit required that the City evaluate assumptions about lake/groundwater interactions made in previous studies and evaluate the effects of the water diversion on water levels in Shell Lake and other nearby lakes. Therefore, a cooperative study between the City and U.S. Geological Survey (USGS) was initiated to improve the understanding of the hydrogeology of the area and evaluate potential effects of the diversion on water levels in Shell Lake, the surrounding groundwater system, and nearby lakes. Concerns over deteriorating water quality in the lake, possibly associated with changes in water level, prompted an additional cooperative project between the City and the USGS to evaluate efeffects of changes in nutrient loading associated with changes in water levels on the water quality of Shell Lake. Numerical models were used to evaluate how the hydrology and water quality responded to diversion of water from the lake and historical changes in the watershed. The groundwater-flow model MODFLOW was used to simulate groundwater movement in the area around Shell Lake, including groundwater/surface-water interactions. Simulated results from the MODFLOW model indicate that groundwater flows generally northward in the area around Shell Lake, with flow locally converging toward the lake. Total groundwater inflow to Shell Lake is small (approximately 5 percent of the water budget) compared with water entering the lake from precipitation (83 percent) and surface-water runoff (13 percent). The MODFLOW model also was used to simulate average annual hydrologic conditions from 1949 to 2009, including effects of the removal of 3 billion gallons of water during 2003–5. The maximum decline in simulated average annual water levels for Shell Lake due to the diversion alone was 3.3 ft at the end of the diversion process in 2005. Model simulations also indicate that although water level continued to decline through 2009 in response to local weather patterns (local drought), the effects of the diversion decreased after the diversion ceased; that is, after 4 years of recovery (2006–9), drawdown attributable to the diversion alone decreased by about 0.6 ft because of increased groundwater inflow and decreased lake-water outflow to groundwater caused by the artificially lower lake level. A delayed response in drawdown of less than 0.5 ft was transmitted through the groundwater-flow system to upgradient lakes. This relatively small effect on upgradient lakes is attributed in part to extensive layers of shallow clay that limit lake/groundwater interaction in the area. Data collected in the lake indicated that Shell Lake is polymictic (characterized by frequent deep mixing) and that its productivity is limited by the amount of phosphorus in the lake. The lake was typically classified as oligotrophic-mesotrophic in June, mesotrophic in July, and mesotrophic-eutrophic in August. In polymictic lakes like Shell Lake, phosphorus released from the sediments is not trapped near the bottom of the lake but is intermittently released to the shallow water, resulting in deteriorating water quality as summer progresses. Because the productivity of Shell Lake is limited by phosphorus, the sources of phosphorus to the lake were quantified, and the response in water quality to changes in phosphorus inputs were evaluated by means of eutrophication models. During 2009, the total input of phosphorus to Shell Lake was 1,730 pounds (lb), of which 1,320 lb came from external sources (76 percent) and 414 lb came from internal loading from sediments in the lake (24 percent). The largest external source was from surface-water runoff, which delivered about 52 percent of the total phosphorus load compared with about 13 percent of the water input. The second largest source was from precipitation (wetfall and dryfall), which delivered 19 percent of the load compared to about 83 percent of the water input. Contributions from septic systems and groundwater accounted for about 3 and 2 percent, respectively. Increased runoff raises water levels in the lake but does not necessarily increase phosphorus loading because phosphorus concentrations in the tributaries decline during increased flow, possibly because of shorter retention times in upstream wetlands. Phosphorus loading to the lake in 2009 represented what occurred after a series of dry years; therefore, this information was combined with data from 2011, a wet year, to estimate phosphorus loading during a range of hydrologic conditions by estimating loading from each component of the phosphorus budget for each year from 1949 to 2011. Comparisons of historical water-quality records with historical water levels and applications of a hydrodynamic model (Dynamic Lake Model, DLM) and empirical eutrophication models were used to understand how changes in water level and the coinciding changes in phosphorus loading affect the water quality of Shell Lake. DLM simulations indicate that large changes in water level (approximately 10 ft) affect the persistence of stratification in the lake. During periods with low water levels, the lake is a well-mixed, polymictic system, with water quality degrading slightly as summer progresses. During periods with high water levels, the lake is more stratified, and phosphorus from internal loading is trapped in the hypolimnion and released later in summer, which results in more extreme seasonality in water quality and better clarity in early summer. Results of eutrophication model simulations using a range in external phosphorus inputs illustrate how water quality in Shell Lake (phosphorus and chlorophyll a concentrations and Secchi depths) responds to changes in external phosphorus loading. Results indicate that a 50-percent reduction in external loading from that measured in 2009 would be required to change phosphorus concentrations from 0.018 milligram per liter (mg/L) (measured in 2009) to 0.012 mg/L (estimated for the mid-1800s from analysis of diatoms in sediment cores). Such reductions in phosphorus loading cannot be accomplished by targeting septic systems or internal loading alone because septic systems contribute only about 3 percent of the phosphorus input to the lake, and internal loading from the sediments of Shell Lake contributes only about 25 percent of phosphorus input. Complete elimination of phosphorus from septic systems and internal loading would decrease the phosphorus concentrations in the lake by 0.003–0.004 mg/L. Therefore, reducing phosphorus concentration in the lake more than by 0.004 mg/L requires decreasing phosphorus loading from surface-water contributions, primarily runoff to the lake. Reconstructed changes in water quality from 1860 to 2010, based on changes in the diatom communities archived in the sediments and eutrophication model simulations, suggest that anthropogenic changes in the watershed (sawmill construction in 1881; the establishment of the village of Shell Lake; and land-use changes in the 1920s, including increased agriculture) had a much larger effect on water quality than the natural changes associated with fluctuations in water level. Although the effects of natural changes in water level on water quality appear to be small, changes in water level do have a modest effect on water quality, primarily manifested as small improvements during higher water levels. Fluctuations in water level, however, have a larger effect on the seasonality of water-quality patterns, with better water quality, especially increased Secchi depths, in early summer during years with high water levels.

Wisconsin↗

USGS2016

Water quality and hydrology of Silver Lake, Oceana County, Michigan, with emphasis on lake response to nutrient loading

Executive Summary Silver Lake is a 672-acre inland lake located in Oceana County, Michigan, and is a major tourist destination due to its proximity to Lake Michigan and the surrounding outdoor recreational opportunities. In recent years, Silver Lake exhibited patterns of high phosphorus concentrations, elevated chlorophyll a concentrations, and nuisance algal blooms. The U.S. Geological Survey (USGS), in cooperation with the Silver Lake Improvement Board and in collaboration with the Annis Water Resources Institute (AWRI) of Grand Valley State University, designed a study to assess the hydrologic and nutrient inputs to Silver Lake in order to identify the events and conditions that affect the nutrient chemistry and production of algal blooms in the lake. This information can inform water-resource managers in developing various management strategies to prevent or reduce the occurrence of future algal blooms. USGS and AWRI scientists collected data from November 2012 to December 2014 to provide information for future management decisions for Silver Lake. Silver Lake can be classified as a polymictic lake and has a residence time of approximately 223 days. Based on the mean lake Secchi depth, total phosphorus, and total nitrogen concentrations, Silver Lake is classified as a eutrophic lake. In-situ bioassay results indicate that algal growth in Silver Lake is colimited by both nitrogen and phosphorus. The nutrient budget for Silver Lake was calculated using the BATHTUB model based on 2 years of water-quality data collection. The BATHTUB model, developed by the U.S. Army Corps of Engineers, treats the lake as a well-mixed system with multiple inputs and outlets for both water and dissolved constituents, such as nutrients. Based on results of the BATHTUB model, which were conditioned on observed concentrations and flows, the mean annual input of phosphorus to Silver Lake was approximately 1,342 pounds (lb); the mean annual input of nitrogen to Silver Lake was approximately 51,998 lb. The major measured sources of phosphorus loading to Silver Lake were groundwater and Hunter Creek, whereas the major measured sources of nitrogen to Silver Lake were Hunter Creek, groundwater, and atmospheric deposition. The largest loading of phosphorus and nitrogen to Silver Lake occurred during the spring. Minimal phosphorus deposition (if any) occurred in the lakebed sediment; however, of the nitrogen that entered Silver Lake, approximately 42.2 percent was deposited in the lakebed sediment as simulated by the BATHTUB model. In addition to measured sources, a septic load model was used to estimate the likely range of septic contribution to groundwater and adjacent surface waters. The likely septic loading scenario estimates that septic systems contribute 47.8 percent of the phosphorus to groundwater and 22.3 percent of phosphorus to Hunter Creek. These results indicate that septic systems are a major source of phosphorus loading to Silver Lake. The likely septic loading scenario indicated that septic systems account for 0.95 percent of the nitrogen load to Hunter Creek and 1.1 percent of the contribution of nitrogen to groundwater. The BATHTUB model was used to estimate future nutrient loading and eutrophication scenarios based on water-quality data collected from Silver Lake, groundwater, major tributaries, and atmospheric deposition. A separate septic load model was used to estimate the septic contribution to groundwater or directly to surface water, and the nutrient load estimates were modeled using the BATHTUB model to determine subsequent water-quality changes to Silver Lake. BATHTUB model scenarios based on measured data: The first BATHTUB scenario evaluated the condition of Silver Lake and the change to lake water quality (trophic status) as a result of changes in nutrient loading from different sources. Based on BATHTUB model simulations, if groundwater loading of phosphorus and nitrogen only were decreased by 75 percent, and all of the other nutrient inputs stayed the same, the future condition of Silver Lake would most likely remain highly mesotrophic to eutrophic (the current [2014] condition of Silver Lake). If nutrient loading continued to increase in groundwater, the lake would continue to remain eutrophic with more frequent algal blooms. If nutrient loading from Hunter Creek only decreased by 50–75 percent, and all of the other nutrient inputs stayed the same as the baseline dataset, Silver Lake would remain eutrophic to highly mesotrophic. By reducing the input of manageable nutrient sources (Hunter Creek, groundwater, and lawn runoff) by 75 percent, the BATHTUB model simulation indicates that Silver Lake would be classified as mesotrophic, which is indicative of improved water quality, water clarity, and reduced algal bloom frequency. Simulations also were run using the BATHTUB model to evaluate the number of days Silver Lake could experience algal blooms (algal blooms are defined as modeled chlorophyll a in excess of 10 micrograms per liter [µg/L]) as a result of an increase/decrease in phosphorus and nitrogen loading from groundwater, Hunter Creek, and (or) a combination of sources. If the phosphorus and nitrogen loading from Hunter Creek is decreased (and all other sources are not altered), Silver Lake will continue to experience algal blooms, but less frequently than what is currently experienced. The same scenario holds true if the nutrient loading from groundwater is decreased. Another scenario was simulated using a combination of sources, which includes increases and decreases in phosphorus and nitrogen loading from sources that are the most likely to be managed, and includes groundwater (as a result of conversion of household septic to sewers), Hunter Creek (conversion of household septic to sewers), and lawn runoff. Results of the BATHTUB model indicated that a 50-percent reduction of phosphorus and nitrogen from these sources would result in a considerable decrease in algal bloom frequency (from 231 to 132 days) and severity, and a 75-percent reduction would greatly reduce algal bloom occurrence on Silver Lake (from 231 to 57 days). BATHTUB model scenarios based on septic load model: A scenario also was conducted using the BATHTUB model to simulate the conversion of septic to sewer and included a low, high, and medium (likely) scenario of nutrient loading to Silver Lake. Simulations of the BATHTUB model indicated that, under the likely scenario, the conversion of all onsite septic treatment to sewers would result in an overall change in lake trophic status from eutrophic to mesotrophic, thereby reducing the frequency of algal blooms and algal bloom intensity on Silver Lake (chlorophyll a >10 µg/L, from 231 to 184 days per year, or chlorophyll a >20 µg/L, from 80 to 49 days per year).

Michigan↗

USGS2018

Water-quality response to changes in phosphorus loading of the Winnebago Pool Lakes, Wisconsin, with special emphasis on the effects of internal loading in a chain of shallow lakes

The Winnebago Pool is a chain of four shallow lakes (Lake Poygan, Lake Winneconne, Lake Butte des Morts, and Lake Winnebago) that are fed primarily by the Fox and Wolf Rivers, two large agriculturally dominated rivers in Wisconsin, United States. Because the lakes have received extensive phosphorus inputs from their watershed, they have become highly eutrophic with much phosphorus in the water column as well as trapped in their sediments. Each of the four Winnebago Pool lakes has been included on the Wisconsin Department of Natural Resources impaired waters list because of their high total phosphorus concentrations, water-quality use restrictions, and excess algal growth. The study described in this report is part of a Total Maximum Daily Load investigation to determine what actions are needed to improve the water quality (trophic status) of these lakes and thus be able to be removed from the impaired waters list and restore their designated uses. As part of this study, data were collected to describe the existing water quality of the lakes, detailed phosphorus budgets were developed for each of the lakes to describe the different sources of the phosphorus, and two eutrophication models (BATHTUB and Jensen models) were used to determine how much of the phosphorus being input to the lakes needs to be reduced for the lakes to be removed from the impaired waters list and restore their designated uses. In-lake water-quality data indicated that each of the lakes had extensive vertical mixing that resulted in their water quality deteriorating throughout summer. Each of the lakes had mean summer total phosphorus concentrations exceeding 0.088 milligram per liter (mg/L), well above the 0.040 mg/L criterion for the lakes. Detailed phosphorus budgets for the lakes indicated that the primary sources of phosphorus were from their tributaries (for the most upstream lake in the Winnebago Pool–Lake Poygan) or from a combination of input from the upstream lakes and phosphorus release from the bottom sediment when only the summer months were considered (for the other three lakes). Model simulations with the BATHTUB and Jensen models indicated that (1) the lakes should have almost linear response in their total phosphorus concentrations to changes in their phosphorus inputs; (2) phosphorus inputs need to be reduced by about 60 percent to the Upper Pool Lakes and 69–73 percent to Lake Winnebago to reduce their mean summer total phosphorus concentrations to 0.040 mg/L; and (3) if all the anthropogenic phosphorus inputs to the lakes could be eliminated, their best possible mean summer total phosphorus concentrations should decrease to about 0.022–0.028 mg/L in the Upper Pool Lakes and to 0.032–0.033 mg/L in Lake Winnebago. The effects of any reduction in phosphorus loading will take many years (50 to more than 75 years) to be fully realized in lake water quality because of phosphorus release from the lake sediments. The effects of nutrient reductions in the watershed of a chain of lakes, such as the Winnebago Pool, gradually cascades down the chain, which has beneficial and detrimental effects. Any action made in the watershed of upstream lakes to reduce phosphorus inputs should improve the water quality of all downstream lakes; however, the upstream lakes delay the response in the downstream lakes, especially in lakes where internal phosphorus loading is important.

Wisconsin↗

USGS2023

Evaluating drivers of hydrology, water quality, and benthic macroinvertebrates in streams of Fairfax County, Virginia, 2007–18

In 2007, the U.S. Geological Survey partnered with Fairfax County, Virginia, to establish a long-term water-resources monitoring program to evaluate the hydrology, water quality, and ecology of Fairfax County streams and the watershed-scale effects of management practices. Fairfax County uses a variety of management practices, policies, and programs to protect and restore its water resources, but the effects of such strategies are not well understood. This report used streamflow, water-quality, and ecological monitoring data collected from 20 Fairfax County watersheds from 2007 through 2018 to assess the effects of management practices, landscape factors, and climatic conditions on observed nutrient, sediment, salinity, and benthic-macroinvertebrate community responses. Urbanization, climatic variability, and an increase in management practices occurred within Fairfax County during the study period. Impervious cover, housing units, wastewater infrastructure, and (or) stormwater infrastructure increased in most study watersheds. Climatic conditions varied among study years; countywide estimates of average-annual air temperature differed by about 3 degrees Celsius, and total precipitation ranged from about 34 to 63 inches per year. The effects of the management practices, implemented to reduce nitrogen, phosphorus, and (or) sediment loads, are considered in this study. These management practices primarily consist of stormwater retrofits and stream restorations; however, stream restorations account for most of the financial investment and expected load reductions. Management practices were implemented in half of the study watersheds, and most practices were installed and reductions credited late in the study period. Changes in hydrologic response during storm events were evaluated over the study period because many management practices that were implemented were designed to achieve nutrient and sediment reductions by slowing or intercepting runoff. The average number and length of storm events was mostly unchanged throughout the monitoring network. Four watersheds with 10 years of streamflow data showed a mixture of trends in stormflow peak, volume, and rate-of-change. Event-mean nutrient and sediment concentrations from these watersheds were evaluated during storm events and generally showed increases in total phosphorus (TP) and suspended sediment and reductions or no changes in total nitrogen (TN). Landscape inputs of nitrogen and phosphorus and the percentage of inputs delivered to streams were estimated for the study watersheds. Estimated phosphorus from fertilizer and nitrogen from atmospheric deposition represented large nutrient inputs in most watersheds; amounts of other nonpoint sources varied based on land use. Estimated nitrogen inputs declined throughout Fairfax County and in most study watersheds from 2008 through 2018; in comparison, phosphorus input changes were relatively small. Most nonpoint-nutrient inputs were retained on the landscape and did not reach streams, with slightly more nitrogen retention than phosphorus, on average. Retention rates were lower for years with more precipitation and streamflow. After adjusting for streamflow, TN and TP loads were generally higher for years with more nutrient inputs. Calculated as a function of flow-adjusted loads, TP retention declined at most stations from 2009 through 2018, in comparison, TN retention was relatively unchanged. Landscape and climatic conditions affected spatial differences and changes in Fairfax County stream conditions from 2009 through 2018. TN concentrations were higher and increases over time were larger in watersheds with elevated septic-system density. TP concentrations were higher in watersheds with more turfgrass; concentrations were lower, but had larger increases over time, in watersheds with deeper soils. Suspended-sediment concentrations were higher in watersheds with greater stream densities. Specific conductance was higher in watersheds with more developed land use and shallower soils. Benthic-macroinvertebrate index of biotic integrity (IBI) scores were lower in watersheds with high road density and had larger increases over time in bigger, more developed watersheds. Annual variability in TN and TP concentrations and benthic-macroinvertebrate IBI scores was affected by precipitation; annual variability in suspended sediment concentrations and specific conductance was affected by air temperature. After accounting for influences from landscape and climatic conditions, expected management-practice effects were not consistently observed in monitored stream responses. These effects were assessed by comparing expected management-practice load reductions with the timing, direction, and magnitude of changes in storm-event hydrology, nutrient and sediment loads, median-annual water-quality conditions, and benthic-macroinvertebrate IBI scores. An important consideration for future investigations of management-practice effects is how to control for water-quality and ecological variability caused by geologic properties, the urban environment, precipitation, and (or) air temperature. The interpretation of management-practice effects in this report was likely influenced by a combination of factors, including (1) the amount, timing, and location of management-practice implementation; (2) unmeasured landscape and climatic factors; (3) uncertain management-practice expectations; (4) hydrologic variability; and (5) analytical assumptions. Through continued data-collection efforts, particularly after management practices have been completed, many of these factors may become less influential in the future.

Virginia↗

USGS2013

Stochastic empirical loading and dilution model (SELDM) version 1.0.0

The Stochastic Empirical Loading and Dilution Model (SELDM) is designed to transform complex scientific data into meaningful information about the risk of adverse effects of runoff on receiving waters, the potential need for mitigation measures, and the potential effectiveness of such management measures for reducing these risks. The U.S. Geological Survey developed SELDM in cooperation with the Federal Highway Administration to help develop planning-level estimates of event mean concentrations, flows, and loads in stormwater from a site of interest and from an upstream basin. Planning-level estimates are defined as the results of analyses used to evaluate alternative management measures; planning-level estimates are recognized to include substantial uncertainties (commonly orders of magnitude). SELDM uses information about a highway site, the associated receiving-water basin, precipitation events, stormflow, water quality, and the performance of mitigation measures to produce a stochastic population of runoff-quality variables. SELDM provides input statistics for precipitation, prestorm flow, runoff coefficients, and concentrations of selected water-quality constituents from National datasets. Input statistics may be selected on the basis of the latitude, longitude, and physical characteristics of the site of interest and the upstream basin. The user also may derive and input statistics for each variable that are specific to a given site of interest or a given area. SELDM is a stochastic model because it uses Monte Carlo methods to produce the random combinations of input variable values needed to generate the stochastic population of values for each component variable. SELDM calculates the dilution of runoff in the receiving waters and the resulting downstream event mean concentrations and annual average lake concentrations. Results are ranked, and plotting positions are calculated, to indicate the level of risk of adverse effects caused by runoff concentrations, flows, and loads on receiving waters by storm and by year. Unlike deterministic hydrologic models, SELDM is not calibrated by changing values of input variables to match a historical record of values. Instead, input values for SELDM are based on site characteristics and representative statistics for each hydrologic variable. Thus, SELDM is an empirical model based on data and statistics rather than theoretical physiochemical equations. SELDM is a lumped parameter model because the highway site, the upstream basin, and the lake basin each are represented as a single homogeneous unit. Each of these source areas is represented by average basin properties, and results from SELDM are calculated as point estimates for the site of interest. Use of the lumped parameter approach facilitates rapid specification of model parameters to develop planning-level estimates with available data. The approach allows for parsimony in the required inputs to and outputs from the model and flexibility in the use of the model. For example, SELDM can be used to model runoff from various land covers or land uses by using the highway-site definition as long as representative water quality and impervious-fraction data are available.

Techniques and Methods↗

USGS2001

Geohydrology and limnology of Walden Pond, Concord, Massachusetts

The trophic ecology and ground-water contributing area of Walden Pond, in Concord and Lincoln, Mass., were investigated by the U.S. Geological Survey in cooperation with the Massachusetts Department of Environmental Management from April 1997 to July 2000. Bathymetric investigation indicated that Walden Pond (24.88 hectares), a glacial kettle-hole lake with no surface inlet or outlet, has three deep areas. The maximum depth (30.5 meters) essentially was unchanged from measurements made by Henry David Thoreau in 1846. The groundwater contributing area (621,000 square meters) to Walden Pond was determined from water-table contours in areas of stratified glacial deposits and from land-surface contours in areas of bedrock highs. Walden Pond is a flow-through lake: Walden Pond gains water from the aquifer along its eastern perimeter and loses water to the aquifer along its western perimeter. Walden Pond contributing area also includes Goose Pond and its contributing area. A water budget calculated for Walden Pond, expressed as depth of water over the lake surface, indicated that 45 percent of the inflow to the lake was from precipitation (1.215 meters per year) and 55 percent from ground water (1.47 meters per year). The groundwater inflow estimate was based on the average of two different approaches including an isotope mass-balance approach. Evaporation accounted for 26 percent of the outflow from the lake (0.71 meters per year) whereas lake-water seepage to the groundwater system contributed 74 percent of the outflow (1.97 meters per year). The water-residence time of Walden Pond is approximately 5 years. Potential point sources of nutrients to ground water, the Concord municipal landfill and a trailer park, were determined to be outside the Walden Pond groundwater contributing area. A third source, the septic leach field for the Walden Pond State Reservation facilities, was within the groundwater contributing area. Nutrient budgets for the lake indicated that nitrogen inputs (858 kilograms per year) were dominated (30 percent) by plume water from the septic leach field and, possibly, by swimmers (34 percent). Phosphorus inputs (32 kilograms per year) were dominated by atmospheric dry deposition, background ground water, and estimated swimmer inputs. Swimmer inputs may represent more than 50 percent of the phosphorus load during the summer. The septic-system plume did not contribute phosphorus, but increased the nitrogen to phosphorus ratio for inputs from 41 to 59, on an atom-to-atom basis. The ratio of nitrogen to phosphorus in input loads and within the lake indicated algal growth would be strongly phosphorus limited. Nitrogen supply in excess of plant requirements may mitigate against nitrogen fixing organisms including undesirable blooms of cyanobacteria. Based on areal nutrient loading, Walden Pond is a mesotrophic lake. Hypolimnetic oxygen demand of Walden Pond has increased since a profile was measured in 1939. Currently (1999), the entire hypolimnion of Walden Pond becomes devoid of dissolved oxygen before fall turnover in late November; whereas historical data indicated dissolved oxygen likely remained in the hypolimnion during 1939. The complete depletion of dissolved oxygen likely causes release of phosphorus from the sediments. Walden Pond contains a large population of the deep-growing benthic macro alga Nitella , which has been hypothesized to promote water clarity in other clear-water lakes by sequestering nutrients and keeping large areas of the sediment surface oxygenated. Loss of Nitella populations in other lakes has correlated with a decline in water quality. Although the Nitella standing crop is large in Walden Pond, Nitella still appears to be controlled by nutrient availability. Decreasing phosphorus inputs to Walden Pond, by amounts under anthropogenic control would likely contribute to the stability of the Nitella population in the metalimnion, may reverse oxygen depletion in the hypolimnion, and decrease recycling of phosphorus from the sediments.

Massachusetts↗

USGS1996

Uncertainty analysis of the simulations of effects of discharging treated wastewater to the Red River of the North at Fargo, North Dakota, and Moorhead, Minnesota

Two separate studies to simulate the effects of discharging treated wastewater to the Red River of the North at Fargo, North Dakota, and Moorhead, Minnesota, have been completed. In the first study, the Red River at Fargo Water-Quality Model was calibrated and verified for icefree conditions. In the second study, the Red River at Fargo Ice-Cover Water-Quality Model was verified for ice-cover conditions. To better understand and apply the Red River at Fargo Water-Quality Model and the Red River at Fargo Ice-Cover Water-Quality Model, the uncertainty associated with simulated constituent concentrations and property values was analyzed and quantified using the Enhanced Stream Water Quality Model-Uncertainty Analysis. The Monte Carlo simulation and first-order error analysis methods were used to analyze the uncertainty in simulated values for six constituents and properties at sites 5, 10, and 14 (upstream to downstream order). The constituents and properties analyzed for uncertainty are specific conductance, total organic nitrogen (reported as nitrogen), total ammonia (reported as nitrogen), total nitrite plus nitrate (reported as nitrogen), 5-day carbonaceous biochemical oxygen demand for ice-cover conditions and ultimate carbonaceous biochemical oxygen demand for ice-free conditions, and dissolved oxygen. Results are given in detail for both the ice-cover and ice-free conditions for specific conductance, total ammonia, and dissolved oxygen. The sensitivity and uncertainty of the simulated constituent concentrations and property values to input variables differ substantially between ice-cover and ice-free conditions. During ice-cover conditions, simulated specific-conductance values are most sensitive to the headwatersource specific-conductance values upstream of site 10 and the point-source specific-conductance values downstream of site 10. These headwater-source and point-source specific-conductance values also are the key sources of uncertainty. Simulated total ammonia concentrations are most sensitive to the point-source total ammonia concentrations at all three sites. Other input variables that contribute substantially to the variability of simulated total ammonia concentrations are the headwater-source total ammonia and the instream reaction coefficient for biological decay of total ammonia to total nitrite. Simulated dissolved-oxygen concentrations at all three sites are most sensitive to headwater-source dissolved-oxygen concentration. This input variable is the key source of variability for simulated dissolved-oxygen concentrations at sites 5 and 10. Headwatersource and point-source dissolved-oxygen concentrations are the key sources of variability for simulated dissolved-oxygen concentrations at site 14. During ice-free conditions, simulated specific-conductance values at all three sites are most sensitive to the headwater-source specific-conductance values. Headwater-source specificconductance values also are the key source of uncertainty. The input variables to which total ammonia and dissolved oxygen are most sensitive vary from site to site and may or may not correspond to the input variables that contribute the most to the variability. The input variables that contribute the most to the variability of simulated total ammonia concentrations are pointsource total ammonia, instream reaction coefficient for biological decay of total ammonia to total nitrite, and Manning's roughness coefficient. The input variables that contribute the most to the variability of simulated dissolved-oxygen concentrations are reaeration rate, sediment oxygen demand rate, and headwater-source algae as chlorophyll a.

Water-Resources Investigations Report↗

USGS2017

The relative effectiveness of empirical and physical models for simulating the dense undercurrent of pyroclastic flows under different emplacement conditions

High concentration pyroclastic density currents (PDCs) are hot avalanches of volcanic rock and gas and are among the most destructive volcanic hazards due to their speed and mobility. Mitigating the risk associated with these flows depends upon accurate forecasting of possible impacted areas, often using empirical or physical models. TITAN2D, VolcFlow, LAHARZ, and Δ H/L or energy cone models each employ different rheologies or empirical relationships and therefore differ in appropriateness of application for different types of mass flows and topographic environments. This work seeks to test different statistically- and physically-based models against a range of PDCs of different volumes, emplaced under different conditions, over different topography in order to test the relative effectiveness, operational aspects, and ultimately, the utility of each model for use in hazard assessments. The purpose of this work is not to rank models, but rather to understand the extent to which the different modeling approaches can replicate reality in certain conditions, and to explore the dynamics of PDCs themselves. In this work, these models are used to recreate the inundation areas of the dense-basal undercurrent of all 13 mapped, land-confined, Soufrière Hills Volcano dome-collapse PDCs emplaced from 1996 to 2010 to test the relative effectiveness of different computational models. Best-fit model results and their input parameters are compared with results using observation- and deposit-derived input parameters. Additional comparison is made between best-fit model results and those using empirically-derived input parameters from the FlowDat global database, which represent “forward” modeling simulations as would be completed for hazard assessment purposes. Results indicate that TITAN2D is able to reproduce inundated areas well using flux sources, although velocities are often unrealistically high. VolcFlow is also able to replicate flow runout well, but does not capture the lateral spreading in distal regions of larger-volume flows. Both models are better at reproducing the inundated area of single-pulse, valley-confined, smaller-volume flows than sustained, highly unsteady, larger-volume flows, which are often partially unchannelized. The simple rheological models of TITAN2D and VolcFlow are not able to recreate all features of these more complex flows. LAHARZ is fast to run and can give a rough approximation of inundation, but may not be appropriate for all PDCs and the designation of starting locations is difficult. The Δ H/L cone model is also very quick to run and gives reasonable approximations of runout distance, but does not inherently model flow channelization or directionality and thus unrealistically covers all interfluves. Empirically-based models like LAHARZ and Δ H/L cones can be quick, first-approximations of flow runout, provided a database of similar flows, e.g., FlowDat, is available to properly calculate coefficients or Δ H/L . For hazard assessment purposes, geophysical models like TITAN2D and VolcFlow can be useful for producing both scenario-based or probabilistic hazard maps, but must be run many times with varying input parameters. LAHARZ and Δ H/L cones can be used to produce simple modeling-based hazard maps when run with a variety of input volumes, but do not explicitly consider the probability of occurrence of different volumes. For forward modeling purposes, the ability to derive potential input parameters from global or local databases is crucial, though important input parameters for VolcFlow cannot be empirically estimated. Not only does this work provide a useful comparison of the operational aspects and behavior of various models for hazard assessment, but it also enriches conceptual understanding of the dynamics of the PDCs themselves.

Frontiers in Earth Science↗

USGS1989

Basin-scale relations via conditioning

A rainfall-runoff model is used in conjunction with a probabilistic description of the input to this model to obtain simple regression-like relations for basin runoff in terms of basin and storm characteristics. These relations, similar to those sought in regionalization studies, are computed by evaluating the conditional distribution of model output given basin and storm characteristics. This method of conditioning provides a general way of examining model sensitivity to various components of model input. The resulting relations may be expected to resemble corresponding relations obtained by regionalization using actual runoff to the extent that the rainfall-runoff model and the model input specification are physically realistic. The probabilistic description of model input is an extension of so-called "random-model" of channel networks and involves postulating an ensemble of basins and associated probability distributions that mimic the variability of basin characteristics seen in nature. Application is made to small basins in the State of Wyoming. Parameters of the input variable distribution are estimated using data from Wyoming, and basin-scale relations are estimated both, parametrically and nonparametrically using model-generated runoff from simulated basins. Resulting basin-scale relations involving annual flood quantiles are in reasonable agreement with those presented in a previous regionalization study, but error estimates are smaller than those in the previous study, an artifact of the simplicity of the rainfall-runoff model used in this paper. We also obtain relations for peak of the instantaneous unit hydrograph which agree fairly well with theoretical relations given in the literature. Finally, we explore the issues of sensitivity of basin-scale, relations and error estimates to parameterization of the model input probability distribution and of how this sensitivity is related to making inferences about a particular ungaged basin. ?? 1989 Springer-Verlag.

Stochastic Hydrology and Hydraulics↗

USGS2012

The impact of biotic/abiotic interfaces in mineral nutrient cycling: A study of soils of the Santa Cruz chronosequence, California

Biotic/abiotic interactions between soil mineral nutrients and annual grassland vegetation are characterized for five soils in a marine terrace chronosequence near Santa Cruz, California. A Mediterranean climate, with wet winters and dry summers, controls the annual cycle of plant growth and litter decomposition, resulting in net above-ground productivities of 280–600 g m −2 yr −1 . The biotic/abiotic (A/B) interface separates seasonally reversible nutrient gradients, reflecting biological cycling in the shallower soils, from downward chemical weathering gradients in the deeper soils. The A/B interface is pedologically defined by argillic clay horizons centered at soil depths of about one meter which intensify with soil age. Below these horizons, elevated solute Na/Ca, Mg/Ca and Sr/Ca ratios reflect plagioclase and smectite weathering along pore water flow paths. Above the A/B interface, lower cation ratios denote temporal variability due to seasonal plant nutrient uptake and litter leaching. Potassium and Ca exhibit no seasonal variability beneath the A/B interface, indicating closed nutrient cycling within the root zone, whereas Mg variability below the A/B interface denotes downward leakage resulting from higher inputs of marine aerosols and lower plant nutrient requirements. The fraction of a mineral nutrient annually cycled through the plants, compared to that lost from pore water discharge, is defined their respective fluxes F j,plants = q j,plants /( q j,plants + q j,discharge ) with average values for K and Ca ( F K,plants = 0.99; F Ca,plants = 0.93) much higher than for Mg and Na ( F Mg,plants 0.64; F Na,plants = 0.28). The discrimination against Rb and Sr by plants is described by fractionation factors ( K Sr/Ca = 0.86; K Rb/K = 0.83) which are used in Rayleigh fractionation-mixing calculations to fit seasonal patterns in solute K and Ca cycling. K Rb/K and K 24 Mg / 22 Mg "> K24Mg/22Mg values (derived from isotope data in the literature) fall within fractionation envelopes bounded by inputs from rainfall and mineral weathering. K Sr/Ca and K 44 Ca / 40 Ca "> K44Ca/40Ca fractionation factors fall outside these envelopes indicating that Ca nutrient cycling is closed to these external inputs. Small net positive K and Ca fluxes (6–14 mol m −2 yr −1 ), based on annual mass balances, indicate that the soils are accumulating mineral nutrients, probably as a result of long-term environmental disequilibrium.

California↗

USGS2013

Aeolian controls of soil geochemistry and weathering fluxes in high-elevation ecosystems of the Rocky Mountains, Colorado

When dust inputs are large or have persisted for long periods of time, the signature of dust additions are often apparent in soils. The of dust will be greatest where the geochemical composition of dust is distinct from local sources of soil parent material. In this study the influence of dust accretion on soil geochemistry is quantified for two different soils from the San Juan Mountains of southwestern Colorado, USA. At both study sites, dust is enriched in several trace elements relative to local rock, especially Cd, Cu, Pb, and Zn. Mass-balance calculations that do not explicitly account for dust inputs indicate the accumulation of some elements in soil beyond what can be explained by weathering of local rock. Most observed elemental enrichments are explained by accounting for the long-term accretion of dust, based on modern isotopic and geochemical estimates. One notable exception is Pb, which based on mass-balance calculations and isotopic measurements may have an additional source at one of the study sites. These results suggest that dust is a major factor influencing the development of soil in these settings and is also an important control of soil weathering fluxes. After accounting for dust inputs in mass-balance calculations, Si weathering fluxes from San Juan Mountain soils are within the range observed for other temperate systems. Comparing dust inputs with mass-balanced based flux estimates suggests dust could account for as much as 50–80% of total long-term chemical weathering fluxes. These results support the notion that dust inputs may sustain chemical weathering fluxes even in relatively young continental settings. Given the widespread input of far-traveled dust, the weathering of dust is likely and important and underappreciated aspect of the global weathering engine.

Colorado↗

USGS2016

Uncertainty analysis of the Operational Simplified Surface Energy Balance (SSEBop) model at multiple flux tower sites

Evapotranspiration (ET) is an important component of the water cycle – ET from the land surface returns approximately 60% of the global precipitation back to the atmosphere. ET also plays an important role in energy transport among the biosphere, atmosphere, and hydrosphere. Current regional to global and daily to annual ET estimation relies mainly on surface energy balance (SEB) ET models or statistical and empirical methods driven by remote sensing data and various climatological databases. These models have uncertainties due to inevitable input errors, poorly defined parameters, and inadequate model structures. The eddy covariance measurements on water, energy, and carbon fluxes at the AmeriFlux tower sites provide an opportunity to assess the ET modeling uncertainties. In this study, we focused on uncertainty analysis of the Operational Simplified Surface Energy Balance (SSEBop) model for ET estimation at multiple AmeriFlux tower sites with diverse land cover characteristics and climatic conditions. The 8-day composite 1-km MODerate resolution Imaging Spectroradiometer (MODIS) land surface temperature (LST) was used as input land surface temperature for the SSEBop algorithms. The other input data were taken from the AmeriFlux database. Results of statistical analysis indicated that the SSEBop model performed well in estimating ET with an R2 of 0.86 between estimated ET and eddy covariance measurements at 42 AmeriFlux tower sites during 2001–2007. It was encouraging to see that the best performance was observed for croplands, where R2 was 0.92 with a root mean square error of 13 mm/month. The uncertainties or random errors from input variables and parameters of the SSEBop model led to monthly ET estimates with relative errors less than 20% across multiple flux tower sites distributed across different biomes. This uncertainty of the SSEBop model lies within the error range of other SEB models, suggesting systematic error or bias of the SSEBop model is within the normal range. This finding implies that the simplified parameterization of the SSEBop model did not significantly affect the accuracy of the ET estimate while increasing the ease of model setup for operational applications. The sensitivity analysis indicated that the SSEBop model is most sensitive to input variables, land surface temperature (LST) and reference ET ( ET o ); and parameters, differential temperature ( dT ), and maximum ET scalar ( K max ), particularly during the non-growing season and in dry areas. In summary, the uncertainty assessment verifies that the SSEBop model is a reliable and robust method for large-area ET estimation. The SSEBop model estimates can be further improved by reducing errors in two input variables ( ET o and LST) and two key parameters ( K max and dT ).

Journal of Hydrology↗

USGS2018

Metamodeling and mapping of nitrate flux in the unsaturated zone and groundwater, Wisconsin, USA

Nitrate contamination of groundwater in agricultural areas poses a major challenge to the sustainability of water resources. Aquifer vulnerability models are useful tools that can help resource managers identify areas of concern, but quantifying nitrogen (N) inputs in such models is challenging, especially at large spatial scales. We sought to improve regional nitrate (NO 3 − ) input functions by characterizing unsaturated zone NO 3 − transport to groundwater through use of surrogate, machine-learning metamodels of a process-based N flux model. The metamodels used boosted regression trees (BRTs) to relate mappable landscape variables to parameters and outputs of a previous “vertical flux method” (VFM) applied at sampled wells in the Fox, Wolf, and Peshtigo (FWP) river basins in northeastern Wisconsin. In this context, the metamodels upscaled the VFM results throughout the region, and the VFM parameters and outputs are the metamodel response variables. The study area encompassed the domain of a detailed numerical model that provided additional predictor variables, including groundwater recharge, to the metamodels. We used a statistical learning framework to test a range of model complexities to identify suitable hyperparameters of the six BRT metamodels corresponding to each response variable of interest: NO 3 − source concentration factor (which determines the local NO 3 − input concentration); unsaturated zone travel time; NO 3 − concentration at the water table in 1980, 2000, and 2020 (three separate metamodels); and NO 3 − “extinction depth”, the eventual steady state depth of the NO 3 − front. The final metamodels were trained to 129 wells within the active numerical flow model area, and considered 58 mappable predictor variables compiled in a geographic information system (GIS). These metamodels had training and cross-validation testing R 2 values of 0.52 – 0.86 and 0.22 – 0.38, respectively, and predictions were compiled as maps of the above response variables. Testing performance was reasonable, considering that we limited the metamodel predictor variables to mappable factors as opposed to using all available VFM input variables. Relationships between metamodel predictor variables and mapped outputs were generally consistent with expectations, e.g. with greater source concentrations and NO 3 − at the groundwater table in areas of intensive crop use and well drained soils. Shorter unsaturated zone travel times in poorly drained areas likely indicated preferential flow through clay soils, and a tendency for fine grained deposits to collocate with areas of shallower water table. Numerical estimates of groundwater recharge were important in the metamodels and may have been a proxy for N input and redox conditions in the northern FWP, which had shallow predicted NO 3 − extinction depth. The metamodel results provide proof-of-concept for regional characterization of unsaturated zone NO 3 − transport processes in a statistical framework based on readily mappable GIS input variables.

Wisconsin↗