Geology topics

Research about Indiana

Source-linked reports with geographic coverage including Indiana.

At least 55 records · Page 3Linked to original sources

USGS2020

Analysis of nearshore placement of sediments at Ogden Dunes, Indiana

The harbor structures/shoreline armoring on the southern Lake Michigan shoreline interrupt sand migration. Ogden Dunes, Indiana, and the nearby Indiana Dunes National Lakeshore observed shoreline erosion due to engineered structures associated with Burns Waterway Harbor, east of Ogden Dunes, impeding natural east to west sediment migration. To remedy this, USACE placed over 450,000 cubic meters, or m³, of dredged material post 2006 in the nearshore of Ogden Dunes. However, the effectiveness of nearshore placements for shoreline protection and littoral nourishment is not fully established. To improve nearshore placement effectiveness, USACE monitored the June/July 2016 placement and subsequent movement of 107,000 m³ of dredged material in the nearshore region at Ogden Dunes. This involved an extensive monitoring scheme of three bathymetry surveys, and two acoustic Doppler current profiler deployments, a Coastal Modeling System numerical model of the changes following placement, and a prediction of sediment transport direction using the Sediment Mobility Tool. The SMT predicted sediment migration direction was compared to observations. Observations indicated that between 10/11/2016 and 11/15/2016 the centroid of the sediment above the pre-placement survey moved 17 m onshore. These observations agreed with SMT predictions onshore migration under storm and typical wave conditions. CMS accurately reproduced the hydrodynamic features.

Indiana↗

USGS2020

Inﬂuence of remediation on sediment toxicity within the Grand Calumet River, Indiana, USA

The Grand Calumet River (GCR), located in northern Indiana, is contaminated due to a wide range of historical industrial activities. This study was conducted to determine the influence of sediment remediation within the GCR on concentrations of chemical contaminants and toxicity to sediment-dwelling organisms. Between 2005 and 2016, sediments with high concentrations of metals and toxic organic compounds were remediated through a combination of removal, addition of activated carbon and organoclay amendments, and capping with sand or relatively uncontaminated sediment. Short-term and long-term sediment toxicity tests with the amphipod Hyalella azteca, the midge Chironomus dilutus, and the mussel Lampsilis siliquoidea were conducted with samples collected in 2013, 2015, and 2017, from 29 sites, including both remediated and non-remediated sites. Sediment chemistry and toxicity data for three groups of remediated sites (US Steel, West Branch, and East Branch) were compared to samples from contaminated but unremediated sites and to relatively uncontaminated reference sites. In general, remediated sediments had lower levels of PAHs, PCBs and metals, although sediments from the US Steel area still had elevated levels of PAH, PCB and chromium. Sediments from the three remediated sites and from reference sites showed significantly reduced toxic effects in short-term sediment bioassays, compared to unremediated sites. Variation in the long-term success of remediation may reflect site-specific factors such as the type of remediation and the potential for recontamination from uncontrolled sources.

Indiana↗

USGS2020

Monthly suspended-sediment apportionment for a western Lake Erie agricultural tributary

Black Creek, a headwater to the Maumee River and western Lake Erie, is an agricultural basin with a mix of cropland (66%), pasture (19%), and forest (7%) linked by a road network to the rural community. Suspended sediment was collected monthly during the 2018 water year for the main stem and two sub-basins using in-situ, passive samplers that integrated a range of streamflow conditions. Sediment fingerprinting used 44 indicators to apportion samples among five sources: cropland, pasture, forest, road dirt, and streambanks. Cropland, pasture, and streambanks had similar ranges in sediment-bound phosphorus (679-1670 ppm). Cropland contributed 21 ± 15% (monthly mean ± standard deviation; 0-46% among individual months) of suspended sediment during the year. Fall and spring peaks in cropland contribution highlight the ongoing importance of on-field management, but this small contribution of suspended sediment relative to the expanse of cropland may reflect implementation of best-management practices. Pasture contributed 0-66% (16±19%) of suspended-sediment and roads 0-26% (6±6%). Streambanks contributed 12-100% (55±25%) and was the only source identified in all sediment samples. In this basin, most cropland-adjacent streambanks are protected by a riparian setback. However, streams traversing other land-use types are not as consistently protected, and these setbacks do not protect the stream channel from discharge of water from sump pumps, road culverts, or tile drains. The contribution of sediment from other land uses combined with that from the agricultural drainage network (as streambank material) underscores the need to consider water movement in the basin as a whole.

Indiana↗

USGS2020

Quantifying microplastics in Indiana’s Lake Michigan watershed and submerged aquatic vegetation

Environmental breakdown of plastics produces synthetic microfibers, a class of microplastics that are most prevalent in surface waters. A main source of these pollutants is wastewater treatment plants which discharge into surface waters, including those that flow into Lake Michigan. Microplastics can settle into aquatic sediment or exist for lengths of time in the swash zone of the lakes; they can be ingested by organisms or concentrated in algae or other lake plankton, among other possibilities. As part of a larger Great Lakes study of the growth and senescence of submerged aquatic vegetation (SAV), the loads of synthetic microfibers in the SAV have been determined from Lake Michigan samples collected during the summer months of 2018, 2019 and 2020, and from various samples collected in the watershed and along the shoreline in Northwest Indiana. In order to effectively address microplastics pollution, a complete understanding of the distribution and fate of these pollutants is needed; this project is working toward that goal.

Indiana↗

USGS2019

The effect of sediment cover and female characteristics on the hatching success of walleye

Natural and anthropogenic sources of sedimentation have the potential to degrade spawning habitat and negatively affect incubating fish embryos. Walleye Sander vitreus are lithophilic broadcast spawners that use specific spawning habitats that are vulnerable to degradation caused by deposition of suspended sediments. We measured the effect of different types of sediment cover on hatching success of Walleye eggs and assessed whether differences in female Walleye (female length and mean egg size) account for tolerance to sediment cover. Experiments were carried out in 2018 to test the effect of sediment cover on hatching success and in 2019 to test how female identity and female length or mean egg size may interact with sediment cover to influence hatching success. Eggs in both experiments were exposed to instantaneous sediment cover (0–7 mm) of either sand or silt from fertilization until day 15 of incubation. Results indicated that Walleye eggs were sensitive to silt cover (71% mortality with 2 mm of silt cover) but not sand cover (47% mortality with 7 mm of sand cover). Hatching success differed significantly among individual females. Although there was an indication that hatching success was marginally negatively related to female length and positively related to mean egg size, sediment cover seemed to have similar effects on eggs, regardless of female length or egg size. Susceptibility of Walleye eggs to mortality caused by sediment cover further underscores the need to limit large‐scale sediment loading and resuspension in aquatic systems.

Indiana↗

USGS2019

Trends in streamflow and concentrations and flux of nutrients and total suspended solids in the Upper White River at Muncie, near Nora, and near Centerton, Indiana

The U.S. Geological Survey (USGS), in cooperation with The Nature Conservancy, completed a study to estimate and assess trends in streamflow and annual mean concentrations and flux of nutrients (nitrate plus nitrite, total Kjeldahl nitrogen, and total phosphorus) and total suspended solids at three USGS streamgages (hereafter referred to as “study gages”) on the Upper White River at Muncie (USGS station 03347000), near Nora (USGS station 03351000), and near Centerton (USGS station 03354000), Indiana. Water-quality data used in the analyses were collected by several agencies between calendar years 1991 and 2017, and streamflow (discharge) data were collected by the USGS. For most of the water-quality constituents, there were suitable data to facilitate an analysis of the 26-year period extending from calendar years 1991 to 2017 (water years 1992 to 2017); however, shorter analytical periods were necessary for total Kjeldahl nitrogen for the study gages at Muncie and near Centerton and for total suspended solids for the study gage near Centerton. Temporal trends in streamflows at the study gages for the period extending from water years 1978 to 2017 were assessed using Exploration and Graphics for RivEr Trends (EGRET) and Mann-Kendall and Pettitt tests. With just one exception, the annual maximum and mean daily streamflows and the annual minimum 7-day mean streamflows at the study gages demonstrated upward trends (increasing streamflows) in the EGRET analyses. The exception was the annual 7-day minimum streamflow at the study gage near Nora, which indicated no trend. Mann-Kendall tests also indicated that the average trend for the annual maximum daily, annual mean daily, and annual 7-day minimum streamflow statistics between water years 1978 and 2017 was upward at each of the study gages; however, only the trends in the annual mean daily streamflows at the study gage at Muncie and the annual maximum daily streamflows at the study gages near Nora and near Centerton were statistically significant at a 0.05 probability level. The Pettitt tests indicated that a statistically significant step trend (abrupt change) in annual mean daily streamflows occurred at each of the study gages around water year 2001. The seasonal distributions of total suspended solids, total phosphorus, nitrate plus nitrite, and total Kjeldahl nitrogen concentrations at the study gages were evaluated to identify patterns and other distinguishing characteristics by examining boxplots of concentrations as a function of month of the year. Seasonal distributions of nitrate plus nitrite concentrations and total suspended solids concentrations differed from each other but were generally similar among the three study gages for a given constituent. Median concentrations of nitrate plus nitrite were highest during the January–June months, whereas median concentrations of total suspended solids were highest during June and July. Seasonal distributions of total phosphorus concentrations were similar at the study gages near Nora and near Centerton, but the seasonal distribution was noticeably different at the study gage at Muncie, which had monthly median concentrations that were substantially lower than at the two downstream study gages (near Nora and near Centerton). The seasonal distribution of total Kjeldahl nitrogen concentrations differed in pattern among the three study gages; however, in general, some of the higher monthly median total Kjeldahl nitrogen concentrations at each study gage were associated with the late spring and summer periods. The Weighted Regressions on Time, Discharge, and Season (WRTDS) method implemented in EGRET was used to estimate water-year annual mean daily concentrations and flux of nutrients and total suspended solids, as well as estimates of concentrations and flux that were “normalized” to remove the effect of year-to-year variation in streamflow. The approximate coefficients of determination for the WRTDS regression models ranged from a high of 0.82 for total phosphorus for the study gage near Centerton to a low of 0.19 for nitrate plus nitrite for the study gage near Nora. Loads and yields of total suspended solids, total phosphorus, nitrate plus nitrite, and total Kjeldahl nitrogen were estimated for analytical periods consisting of the longest periods of concurrent record at the three study gages. Loads of each of the constituents increased sequentially from the most upstream study gage to the most downstream study gage; however, the same was not true for yields. The highest yields of total suspended solids, total phosphorus, and total Kjeldahl nitrogen occurred at the most upstream study gage (at Muncie); however, the highest yield of nitrate plus nitrite occurred at the most downstream study gage (near Centerton). WRTDS bootstrap tests were used to assess the magnitude, direction, and likelihood of changes in annual flow-normalized mean daily concentrations and flux of total suspended solids, total phosphorus, nitrate plus nitrite, and total Kjeldahl nitrogen at the study gages between water years 1997 and 2017. Changes in flow-normalized concentrations and flux of the constituents between water years 1997 and 2017 were mostly downward (decreasing). The exceptions were likely to highly likely upward (increasing) changes in (1) flow-normalized annual mean daily concentration and annual flux for total suspended solids and total phosphorus at the study gage at Muncie, (2) flow-normalized annual mean daily total phosphorus concentration at the study gage near Centerton, (3) flow-normalized annual flux of total phosphorus at the study gage near Centerton, and (4) flow-normalized annual mean daily nitrate plus nitrite concentration at the study gage near Centerton. Although an upward change in flow-normalized nitrate plus nitrite concentrations was likely at the study gage near Centerton, flow-normalized annual flux of nitrate plus nitrite at that study gage was determined to have a highly likely downward change. EGRET and Exploration and Graphics for RivEr Trends Confidence Intervals (EGRETci) analyses can be used to improve our understanding of how concentrations and flux change as functions of time and streamflow, as well as provide information on how the relations between streamflow and constituent concentrations have changed within the calendar year between any 2 years included in the analyses. Examples of those uses, illustrating changes between calendar years 1992 and 2017, were given for total suspended solids concentrations at the study gage near Nora and for nitrate plus nitrite concentrations at the study gage near Centerton.

Indiana↗

USGS2019

Evidence of an extreme weather‐induced phenological mismatch and a local extirpation of the endangered Karner blue butterfly

In 2011, an experiment was undertaken to examine spring synchrony between the endangered Karner blue butterfly ( Lycaeides melissa samuelis ) (Kbb) and its obligate host plant, wild blue lupine ( Lupinus perennis ) at Indiana Dunes National Lakeshore (INDU), where the southernmost population of Kbb occurred at the time of this study. From 2012 to 2014, field‐placed Kbb eggs were observed for larvae hatching in conjunction with observations of lupine emergence in oak savanna habitat. In 2012, 61% of Kbb hatched when <5% of lupine had emerged due to an extreme early spring event as compared to subsequent years where temporal overlap was >15% between Kbb and lupine. Laboratory experiments testing the sensitivity of Kbb hatching to warm temperatures during the winter of 2011–2012 confirmed that Kbb eggs were susceptible to temperature‐induced hatching. In the summer of 2012, second generation Kbb larvae feeding on sun‐exposed lupine had higher mortality due to the heat and drought conditions that resulted in earlier plant senescence. Following 2012, Kbb were no longer observed at INDU. This observation illustrates the pressing need for adaptive management strategies that account for extreme weather events brought on by climate change.

Indiana↗

USGS2019

Comparison of groundwater-model construction methods, representations of glacial geology, model designs, and groundwater-model flow simulations within Elkhart County, Indiana

Automated data-processing methods allow hydrologists to efficiently incorporate digital well-record datasets into the construction of hydrostratigraphic frameworks for groundwater-flow models. The method selected to construct the hydrostratigraphic framework can affect the extent of geologic heterogeneity that can be included in the model. The detail generated from a hydrostratigraphic framework can affect groundwater simulation results. The effects of detail on model accuracy, groundwater-flow simulations, and particle-tracking simulations are described in this study. This report compares differences in hydrostratigraphic frameworks and results of groundwater models using (1) a method that incorporates more hydrologic judgment at the expense of using limited lithologic data and (2) a method that is more automated and uses all available lithologic data. The study additionally evaluates the effect of model discretization and inclusion of more (or less) geologic detail on simulation results. Two methods were used to create hydrostratigraphic frameworks of glacial deposits in the St. Joseph River Basin. One method, referred to as the subjective method, manually identifies stratigraphic boundaries using a sample of well logs from State databases and uses two-dimensional kriging to create three model layers of the study area. Indicator kriging is used to define aquifer extent in each layer. The second method, referred to as the objective method, uses three-dimensional kriging to automatically create a detailed heterogeneous model of the study area using all wells logs from the State database. The objective method increases detail in the vertical by greatly increasing the number of computer groundwater model layers from 3 to 30. In Elkhart County, Indiana, a previously published model represents the product of the subjective method, and a newly calibrated model of the same area represents the product of the objective method. An automated calibration procedure was used with the objective model (derived from the objective method) for Elkhart County. The two most-sensitive parameters for the Elkhart County objective model are horizontal hydraulic conductivity of the sand and the combined sand and gravel/gravel deposits. Vertical hydraulic conductivity of the fine-grained and intermediate-sized deposits could not be estimated, possibly indicating major flow paths are along a continuously connected series of sand and gravel deposits and not through a confining layer. The statistics measuring model calibration accuracy for the objective model were slightly better than statistics for the subjective model (model derived from the subjective method) of Elkhart County, but the hydraulic conductivities and flow rates for the two models were different. The mean absolute errors between simulated and measured groundwater levels are 2.04 and 2.16 feet for the objective and subjective models, respectively. Simulated seepage losses from and groundwater discharges to measured stream reaches in the objective model were evenly balanced in terms of over and under simulations of measured values; the subjective model tended to overpredict measured groundwater discharge to streams. The overprediction may be related to the 58 percent greater total inflow and outflow through the subjective model. The greater flow rate through the subjective model results from higher horizontal hydraulic conductivities in the subjective model than in the objective model. Horizontal hydraulic conductivity ranged from 23.9 to 111 feet per day in the objective model and generally ranged from 170 to 370 feet per day in the subjective model. The improvement in calibration statistics for the objective model relative to the subjective model may be from increased detail in how the objective model represents the distribution of fine- and coarse-grained deposits. The improvement also could be associated with the difference in methods used to represent the continuity of the confining unit. The effect of differences in horizontal hydraulic conductivity distributions between the two models for Elkhart County is evident in the groundwater-flow paths simulated by the objective and subjective models. At a withdrawal well location, the flow lines produced by the objective model indicate a wider contributing area than that for the subjective model. The discontinuous confining unit represented in the objective model provided the opportunity for groundwater flow to split into an upper and lower path. The split in flow simulated by the objective model at one location was independently supported by bromide concentrations in groundwater; the subjective model did not duplicate the split in flow.

Indiana↗

USGS2019

Regression models for estimating sediment and nutrient concentrations and loads at the Iroquois River near Foresman, Indiana, March 2015 through July 2018

In 2015, the U.S. Geological Survey, in cooperation with the Iroquois River Conservancy District, deployed continuous water-quality monitors and began collecting representative discrete water-quality samples at the Iroquois River near Foresman, Indiana, streamflow-gaging station (U.S. Geological Survey station 05524500). By relating continuously monitored water-quality data and discrete water-quality samples collected from April 2015 through July 2018, regression models that estimate concentrations of suspended sediment, total nitrogen, and total phosphorus were developed. Developed regression models indicated a strong correlation between turbidity and streamflow with suspended-sediment concentration (adjusted coefficient of determination equals 0.84, predicted residual error sum of squares equals 0.493), nitrate plus nitrite and streamflow with total nitrogen (adjusted coefficient of determination equals 0.99, predicted residual error sum of squares equals 0.0202), and specific conductance and turbidity with total phosphorus (adjusted coefficient of determination equals 0.84, predicted residual error sum of squares equals 0.0935). Daily loads of suspended sediment, total nitrogen, and total phosphorus were computed as the product of daily mean regression model concentrations and daily mean streamflow. During periods when regression model concentrations could not be computed, rloadest models, the R programming language version of the LOADEST FORTRAN program, were used to compute daily loads of each constituent. For 2016 and 2017, the estimated annual suspended-sediment loads were 25,000 and 32,100 tons; estimated total nitrogen loads were 4,260 and 5,780 tons; and estimated total phosphorus loads were 104 and 128 tons, respectively.

Indiana↗

USGS2019

Amphibian monitoring in hardwood forests: Optimizing methods for contaminant‐based compensatory restorations

Amphibians such as frogs, toads, and salamanders provide important services in aquatic and terrestrial ecosystems and have been proposed as useful indicators of progress and success for ecological restoration projects. Limited guidance is available, however, on the costs and benefits of different amphibian monitoring techniques that might be applied to sites restored in compensation for contaminant injury. We used a variety of methods to document the amphibian communities present at 4 restored bottomland hardwood sites in Indiana, USA, and to compare the information return and cost of each method. For 1 method—automated recording units—we also modeled the effect of varying levels of sampling effort on the number of species detected, using sample-based rarefaction and Bayesian nonlinear (Michaelis–Menten) mixed effects models. We detected 13 amphibian species across the restored sites, including 2 species of conservation concern in Indiana—northern leopard frogs ( Lithobates pipiens ) and Blanchard's cricket frogs ( Acris blanchardi ). Sites across a range of restoration ages demonstrated encouraging returns of amphibian communities. Although more mature sites showed greater species richness, recently restored sites still provided important habitat for amphibians, including species of conservation concern. Among the 4 methods compared, amphibian rapid assessment yielded the highest number of species detected and the greatest catch per unit effort, with the lowest per-site cost. Our analysis of level-of-effort effects in the rarefied acoustic data found that number of nights sampled was a better predictor of observed species richness than the number of hours sampled within a night or minutes sampled within an hour. These data will assist restoration practitioners in selecting amphibian monitoring methods appropriate for their site characteristics and budget.

Indiana↗

USGS2019

Regression models for estimating sediment and nutrient concentrations and loads at the Kankakee River, Shelby, Indiana, December 2015 through May 2018

The Kankakee River in northern Indiana flows through the area once known as the Grand Marsh. Beginning in the 1860s, anthropogenic changes to the river within Indiana resulted in downstream flooding and additional transport of sediment and nutrients. In 2015, the U.S. Geological Survey, in cooperation with the Indiana Department of Environmental Management, upgraded the gaging station Kankakee River at Shelby, Indiana, to include the collection of water-quality data. By relating continuously monitored water-quality data to discrete data collected from December 2015 through May 2018, linear regression was used to develop models for estimating concentrations of suspended sediment, total nitrogen, and total phosphorus. Developed regression models indicated a strong correlation between turbidity and specific conductance with suspended-sediment concentration (adjusted coefficient of determination equals 0.92, predicted residual error sum of squares equals 0.151), nitrate plus nitrite and specific conductance with total nitrogen (adjusted coefficient of determination equals 0.95, predicted residual error sum of squares equals 0.0248), and turbidity with total phosphorus (adjusted coefficient of determination equals 0.89, predicted residual error sum of squares equals 0.0103). Daily loads of suspended sediment, total nitrogen, and total phosphorus were computed as the product of daily mean regression model concentrations and daily mean streamflow. Rloadest models were used to compute daily loads of each constituent during gaps in regression model loads. For 2016 and 2017, the estimated annual suspended-sediment loads were 105,000 and 91,000 tons; estimated total nitrogen loads were 8,690 and 8,890 tons; and estimated total phosphorus loads were 265 and 236 tons, respectively.

Indiana↗

USGS2019

Sharp savanna-forest transitions in the Midwest followed environmental gradients but are absent from the modern landscape

Historically, closed eastern forests transitioned into open savannas and prairies in the US Midwest, but this transition is poorly understood. To investigate the eastern boundary of the prairie-forest ecotone, we conducted a case study of historic and modern vegetation patterns of the Yellow River watershed in northwest Indiana. Historic vegetation came from the Public Land Survey notes collected in the early 1800s, whereas modern vegetation came from the Forest Inventory Analysis and USGS National Land Cover Database. We mapped historical survey vegetation data using GIS to reconstruct the region’s past and current forest composition and structure. We also mapped climate, topography, and soil composition across the watershed to investigate the relationship between historic vegetation and environmental gradients. We found a sharp transition in the presettlement forest structure and composition, with dense deciduous forests in the eastern portion of our study area and open oak savannas in the west. The savanna ecosystem dominated in sandy well-drained soils and was at a slightly lower elevation than the adjacent closed forest. Modest environmental changes accompanied major vegetation changes in the past, which might suggest fire and hydrological patterns helped maintain the sharp ecotone. By contrast, the modern forest shows no difference in tree density and composition across the watershed, which is consistent with major land use and hydrology changes in the watershed since settlement. On the modern landscape, land that was historically closed forest now has higher agricultural productivity compared to land that was historically savanna, whereas the historic savanna currently supports more mesic forest. These results suggest the environmental gradient continues to subtly shape the landscape. Though land use change has largely removed the closed mixed hardwood forests and oak savannas from this area, a better understanding of the historic vegetation and the conditions that supported it can help inform land management and restoration, as well as reveal ecological processes that drive vegetation transitions.

Indiana↗

USGS2018

Flood-inundation maps for the Salamonie River at Portland, Indiana

Digital flood-inundation maps for a 6.5-mile reach of the Salamonie River at Portland, Indiana, were created by the U.S. Geological Survey (USGS) in cooperation with the Indiana Department of Transportation. The flood-inundation maps, which can be accessed through the USGS Flood Inundation Mapping Science website at https://water.usgs.gov/osw/flood_inundation/ , depict estimates of the areal extent and depth of flooding corresponding to selected water levels (stages) at the USGS streamgage on the Salamonie River at Portland, Ind. (station 03324200). Near-real-time stages at this streamgage may be obtained from the USGS National Water Information System web interface at https://doi.org/10.5066/F7P55KJN or from the National Weather Service Advanced Hydrologic Prediction Service (site PORI3) at https:/water.weather.gov/ahps/ . Flood profiles were computed for the stream reach by means of a one-dimensional step-backwater model. The model was calibrated using the current (2018) stage-discharge relation at the Salamonie River at Portland, Ind., streamgage. The hydraulic model then was used to compute nine water-surface profiles for flood stages at 1-foot (ft) intervals referenced to the streamgage datum and ranging from 10.7 ft or near bankfull to 18.7 ft, which equals the highest point on the streamgage rating curve. The simulated water-surface profiles then were combined with a geographic information system digital elevation model derived from light detection and ranging data having a 0.49-ft root mean square error and 4.9-ft horizontal resolution resampled to a 10-ft grid to delineate the area flooded at each stage. The availability of these maps, along with information regarding current stage from the USGS streamgage, will provide emergency management personnel and residents with information that is critical for flood response activities such as evacuations and road closures, as well as for postflood recovery efforts.

Indiana↗

USGS2018

An exploratory Bayesian network for estimating the magnitudes and uncertainties of selected water-quality parameters at streamgage 03374100 White River at Hazleton, Indiana, from partially observed data

An exploratory discrete Bayesian network (BN) was developed to assess the potential of this type of model for estimating the magnitudes and uncertainties of an arbitrary subset of unmeasured water-quality parameters given the measured complement of parameters historically measured at a U.S. Geological Survey streamgage. Water-quality data for 27 water-quality parameters from 596 discrete measurements at U.S. Geological Survey streamgage 03374100 White River at Hazleton, Indiana, were used to develop this BN. Data for each of the water-quality parameters were discretized into five intervals based on the quintiles of the measured values. The 596 discrete measurements were randomly partitioned into a training set with 80 percent of the data and a testing set with 20 percent of the data to identify, estimate, and assess the training and testing accuracy of the Bayesian network. A BN with 28 nodes was formed from the 27 water-quality parameters and the month of sample collection. Based on data in the training set, a network with 53 directed edges and month as the target node was identified by minimizing the negative log-likelihood function for all nodes treated, in turn, as the target variable. The edge structure determines the number and magnitude of elements in conditional probability tables associated with all nodes. The effectiveness of the BN was assessed on the basis of correct classification rates to one of the five discrete intervals, which were computed separately for the training and testing datasets and for two conditioning variable sets. The selected sets of conditioning variables represent two of many possible sets of measured parameters on which to base estimates of unmeasured parameters. The first set includes only the month of sample collection ( month ), and an expanded set includes month and six other continuously measurable parameters, referred to as the ContMeasSet , all of which were obtained from the discrete data. Results indicated that the training dataset had average correct classification rates of 41.7- and 61.2-percent rates conditioned on the month and ContMeasSet sets, respectively. The testing dataset had somewhat lower average correct classification rates of 40.8 and 56.5 percent for the two conditioning variable sets. When conditioned on month only, the average correct classification rate for the testing dataset was only slightly lower than the average correct classification rate in the training dataset, indicating little model overfitting. When using the ContMeasSet , however, the average decrease in accuracy between training and testing sets was 4.9 percent. The training and testing datasets and both sets of conditioning variables, however, indicate that the BN would substantially outperform a random assignment model, which would be expected to have a 20-percent correct classification rate. In addition, the edge structure of the BN depicts how information can flow through the network, which may help prioritize parameters for measurement to facilitate estimation of unmeasured parameters. Finally, extension of a static BN, like the one developed in this report, to a dynamic BN may provide a basis for using high-frequency or continuous water-quality data to extend information in time between discrete water-quality samples, and this integration could mitigate some of the limitations of high-frequency and discrete water-quality sampling methods.

Indiana↗

USGS2018

Geochemistry and microbiology of groundwater and solids from extraction and monitoring wells and their relation to well efficiency at a Federally operated confined disposal facility, East Chicago, Indiana

In cooperation with the U.S. Army Corps of Engineers, Chicago District, the U.S. Geological Survey investigated the processes affecting water quality, geochemistry, and microbiology in representative extraction and monitoring wells at a confined disposal facility (CDF) in East Chicago, Indiana. The CDF is a 140-acre Federally-managed facility that was the former location of an oil refinery and is now used for the long-term disposal and storage of dredge material from the Indiana Harbor and Indiana Harbor Canal. Residual petroleum hydrocarbons and leachate from the CDF are contained within the facility by use of a groundwater cutoff wall. The wall consists of a soil-bentonite slurry and a gradient control system made up of an automated network of 96 extraction wells, 42 monitoring wells, and 2 ultrasonic sensors that maintain an inward hydraulic gradient at the site. The pumps in the extraction wells require vigilant maintenance and must be replaced when unable to withdraw water at a rate sufficient to maintain the required inward gradient. The wells are screened in the Calumet aquifer, a coarse-grained sand and gravel unit that extends approximately 35 feet below the land surface and is not utilized for drinking-water supply at the CDF or in the surrounding area. This study was initiated to identify the cause of decreased pump discharges and to identify potential mitigation strategies. For this study, the U.S. Geological Survey collected groundwater and solids from monitoring and extraction wells. Groundwater samples were collected during June 2014 for precautionary health screening and on four occasions during September 2014 through November 2014. Groundwater samples collected from two extraction wells during June 2014 were analyzed for concentrations of anthropogenic organic constituents. During September through November 2014, groundwater samples were collected from one additional extraction well, and samples from three monitoring wells were analyzed for concentrations of inorganic and organic constituents, dissolved gases, and bacterial abundance and diversity. Solid samples were collected during April 2014, during September 2014 through November 2014, and during November 2016. Solid samples were collected from the exterior of extraction-well pumps and as flocculent from water samples. Solid samples were collected from 10 wells, including 1 extraction well and 3 monitoring wells sampled for water quality. Solid samples were analyzed for mineralogy, solid-phase habit, geochemistry, and organic composition. The following is a list of observations that were made during this study: (1) the water quality is substantially variable among the six well locations sampled as part of this study—lower (more negative) redox values and higher concentrations of many constituents (including calcium, magnesium, sodium, and sulfate) and properties (including dissolved solids, hardness, and turbidity) were detected in sampled wells located near the extraction wells with the highest frequency of failure; (2) water-level drawdown is variable between extraction wells—wells with the greatest drawdown may pull deeper groundwater into the borehole; (3) dissolved gas results indicate reducing oxidation-reduction processes in the aquifer material that can feasibly contribute iron, carbon dioxide, and other byproducts from hydrocarbon degradation to precipitates and solids that accumulate on and impair pump operation; (4) crystalline and amorphous solid-phase minerals are precipitating in the borehole; (5) several types of bacteria are present in water pumped from extraction wells and are likely responsible for bonding mineral and microbiologic matter to the pump (and other well components); and (6) bacteria may create microenvironments that facilitate precipitation of solids or inhibit dissolution of unstable minerals once the bacteria adhere to biofilm attached to the pump. Results of the study indicate that bacteria may be accumulating and entrapping solid material on the exterior of pumps. This accumulation reduces heat transfer and water discharge from the pump and may lead to decreased efficiency or mechanical failure. Observations could not be made on the well screen, gravel pack, or surrounding geologic formation; therefore, mitigating measures in the borehole may not solve well-productivity issues. Remedies for the pump fouling problems were derived from the review and interpretation of data collected during this study and from information documented in other sources about groundwater well fouling. Potential remedies to problems associated with pump fouling at the CDF may include the following: (1) reducing attractiveness of the extraction wells for microbiological growth by modifying the chemical or physical environment of the well, (2) modifying the pump exterior to decrease microbiological adherence, (3) changing the pumping regime to control the chemistry of water entering the well from the surrounding aquifer material, (4) modifying the pumps to be less physically and thermally attractive, and (5) removing hydrocarbons from groundwater and the aquifer material surrounding the wells or adding surfactants to make them more mobile. Pilot scale testing may be necessary to identify the most effective treatment or combination of treatments.

Indiana↗

USGS2018

Flood-inundation maps for the Wabash River at Lafayette, Indiana

Digital flood-inundation maps for an approximately 4.8-mile reach of the Wabash River at Lafayette, Indiana (Ind.) were created by the U.S. Geological Survey (USGS) in cooperation with the Indiana Office of Community and Rural Affairs. The inundation maps, which can be accessed through the USGS Flood Inundation Mapping Science web site at https://water.usgs.gov/osw/flood_inundation/ , depict estimates of the areal extent and depth of flooding corresponding to selected water levels (stages) at USGS streamgage 03335500, Wabash River at Lafayette, Ind. Current streamflow conditions for estimating near-real-time areas of inundation using USGS streamgage information may be obtained on the internet at https://waterdata.usgs.gov/in/nwis/uv?site_no=03335500 . In addition, information has been provided to the National Weather Service (NWS) for incorporation into their Advanced Hydrologic Prediction Service (AHPS) flood-warning system ( https://water.weather.gov/ahps/ ). The NWS AHPS forecasts flood hydrographs at many places that are often colocated with USGS streamgages, including the Wabash River at Lafayette, Ind. NWS AHPS-forecast peak-stage information may be used with the maps developed in this study to show predicted areas of flood inundation. For this study, flood profiles were computed for the Wabash River reach by means of a one-dimensional step-backwater model. The hydraulic model was calibrated by using the most current stage-discharge relations at USGS streamgage 03335500, Wabash River at Lafayette, Ind., and high-water marks from the flood of July 2003 (U.S. Army Corps of Engineers [USACE], 2007). The calibrated hydraulic model was then used to determine 23 water-surface profiles for flood stages at 1-foot intervals referenced to the streamgage datum and ranging from bankfull to the highest stage of the current stage-discharge rating curve. The simulated water-surface profiles were then combined with a geographic information system digital elevation model derived from light detection and ranging to delineate the area flooded at each water level. The availability of these maps, along with internet information regarding current stage from the USGS streamgage 03335500, Wabash River at Lafayette, Ind., and forecasted high-flow stages from the NWS AHPS, will provide emergency management personnel and residents with information that is critical for flood-response activities such as evacuations and road closures, and for postflood recovery efforts.

Indiana↗

USGS2018

Flood-inundation maps for Cedar Creek at 18th Street at Auburn, Indiana

Digital flood-inundation maps for a 1.9-mile reach of Cedar Creek at Auburn, Indiana (Ind.), from the First Street bridge, downstream to the streamgage at 18th Street, then ending approximately 1,100 feet (ft) downstream of the Baltimore and Ohio railroad, were created by the U.S. Geological Survey (USGS) in cooperation with the Indiana Department of Transportation. The flood-inundation maps, which can be accessed through the USGS Flood Inundation Mapping Science web site at https://water.usgs.gov/osw/flood_inundation/ , depict estimates of the areal extent and depth of flooding corresponding to selected water levels (stages) at the USGS streamgage on Cedar Creek at 18th Street at Auburn, Ind. (station number 04179520). Near-real-time stages at this streamgage may be obtained from the USGS National Water Information System at https://waterdata.usgs.gov/ or the National Weather Service Advanced Hydrologic Prediction Service at http://water.weather.gov/ahps/ , although forecasts of flood hydrographs are not available at this site (ABBI3). Flood profiles were computed for the stream reach by means of a one-dimensional step-backwater model. The hydraulic model was calibrated by using the most current stage-discharge relation at the Cedar Creek at 18th Street at Auburn, Ind. streamgage and the documented high-water marks from the flood of March 11, 2009. The calibrated hydraulic model was then used to compute seven water-surface profiles for flood stages referenced to the streamgage datum and ranging from 7 ft, or near bankfull, to 13 ft, in 1-foot increments. The simulated water-surface profiles were then combined with a geographic information system digital elevation model (derived from light detection and ranging [lidar] data having a 0.98-ft vertical accuracy and 4.9-ft horizontal resolution) to delineate the area flooded at each water level. The availability of these maps, along with internet information regarding current stage from the USGS streamgage at Cedar Creek at 18th Street at Auburn, Ind., and stream information from the National Weather Service, will provide emergency management personnel and residents with information that is critical for flood response activities such as evacuations and road closures as well as for postflood recovery efforts.

Indiana↗

USGS2018

The art, history, and geoscience of Hindustan Whetstone gravestones in Indiana

Cemeteries can be intriguing places to people, in part because of a fascination with death but also because of the quiet solitude and artistic beauty found there. Many grave monuments are really works of art and can be appreciated on that basis alone. Cemeteries can also serve as teaching laboratories for geologists. Monument types, carving styles, ornamentation, and durability are all related, to some extent, to the type of rock used. The older the monument dates the more variability one can see in the character of the stones. Pioneer cemeteries in southern Indiana, some of which date back to the early 1800s, can be used to teach concepts in mineralogy, depositional environments, and paleoastronomy. This can be very useful to someone teaching some of the basic concepts of geology.

Indiana↗