USGS Science⌕ Search

SEARCH · USGS Science

Results for “Spatial Statistics”

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

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

At least 721 records · Page 40Linked to original sources

Urbanization impacts on evapotranspiration across various spatio-temporal scales

Urbanization has been shown to locally increase the nighttime temperatures creating urban heat islands, which partly arise due to evapotranspiration (ET) reduction. It is unclear how the direction and magnitude of the change in local ET due to urbanization varies globally across different climatic regimes. This knowledge gap is critical, both for the key role of ET in the energy and water balance accounting for the majority of local precipitation, and for reducing the urban heat island effect. We explore and assess the impacts of urbanization on monthly and mean annual ET across a range of landscapes from local to global spatial scales. Remotely sensed land cover and ET available at 1 km resolution are used to quantify the differences in ET between urban and surrounding non-urban areas across the globe. The observed patterns show that the statistically significant difference between urban and non-urban ET can be estimated to first order as a function of local hydroclimate, with arid regions seeing increased ET, and humid regions showing decreased ET. Cities under cold climates also evaporate more than their non-urban surroundings during the winter, as the urban micro-climate has increased energy availability resulting from human activities. Increased ET in arid cities arises from municipal water withdrawals and increased irrigation during drought conditions. These results can help inform planners to improve the integration of environmental conditions into the design and management of urban landscapes.

Earth's Future↗

Monitoring on-orbit stability of Terra MODIS and Landsat 7 ETM+ reflective solar bands using Railroad Valley Playa, Nevada (RVPN) test site

The Moderate Resolution Imaging Spectroradiometer (MODIS) Proto-Flight Model (PFM), launched on December 18, 1999, aboard NASA's Earth Observing System (EOS) Terra satellite has 20 reflective solar bands (RSB) with wavelengths ranging from 0.41 to 2.1 mum over a wide field of view (plusmn55deg). The Landsat 7 (L7) Enhanced Thematic Mapper Plus (ETM+) sensor was launched on April 15, 1999, and has six spectral bands located in the visible and shortwave infrared (SWIR) part of the electromagnetic spectrum (0.4 - 2.5 mum). The ETM+ belongs to the family of Thematic Mapper sensors flown on previous Landsat missions. In this study, over 75 cloud-free nadir near-simultaneous images over Railroad Valley Playa, Nevada (RVPN) were chosen covering entire missions of both sensors. RVPN (38.5degN and 115.7degW), located between the cities of Ely and Tonopah, Nevada, USA, is a high reflectance site with very high spatial, spectral, and temporal uniformity. It is referenced to the Worldwide Reference System-2 (WRS-2) with path 40 and row 33. Homogeneous regions of interest (ROI) were chosen and cross-calibration was performed using an image statistics approach to monitor the long-term stability of the two sensors.

Nevada↗

Digital spatial data for observed, predicted, and misclassification errors for observations in the training dataset for nitrate and arsenic concentrations in basin-fill aquifers in the Southwest Principal Aquifers study area

This product "Digital spatial data for observed, predicted, and misclassification errors for observations in the training dataset for nitrate and arsenic concentrations in basin-fill aquifers in the Southwest Principal Aquifers study area" is a 1:250,000-scale point spatial dataset developed as part of a regional Southwest Principal Aquifers (SWPA) study (Anning and others, 2012). The study examined the vulnerability of basin-fill aquifers in the southwestern United States to nitrate contamination and arsenic enrichment. Statistical models were developed by using the random forest classifier algorithm to predict concentrations of nitrate and arsenic across a model grid that represents local- and basin-scale measures of source, aquifer susceptibility, and geochemical conditions.

Arizona, California, Colorado, Idaho, Nevada, New ↗

Digital spatial data for predicted nitrate and arsenic concentrations in basin-fill aquifers of the Southwest Principal Aquifers study area

This product "Digital spatial data for predicted nitrate and arsenic concentrations in basin-fill aquifers of the Southwest Principal Aquifers study area" is a 1:250,000-scale vector spatial dataset developed as part of a regional Southwest Principal Aquifers (SWPA) study (Anning and others, 2012). The study examined the vulnerability of basin-fill aquifers in the southwestern United States to nitrate contamination and arsenic enrichment. Statistical models were developed by using the random forest classifier algorithm to predict concentrations of nitrate and arsenic across a model grid that represents local- and basin-scale measures of source, aquifer susceptibility, and geochemical conditions.

Arizona, California, Colorado, Idaho, Nevada, New ↗

Greater sage-grouse ( Centrocercus urophasianus ) nesting and brood-rearing microhabitat in Nevada and California—Spatial variation in selection and survival patterns

Greater sage-grouse ( Centrocercus urophasianus ; hereinafter, "sage-grouse") are highly dependent on sagebrush ( Artemisia spp.) dominated vegetation communities for food and cover from predators. Although this species requires the presence of sagebrush shrubs in the overstory, it also inhabits a broad geographic distribution with significant gradients in precipitation and temperature that drive variation in sagebrush ecosystem structure and concomitant shrub understory conditions. Variability in understory conditions across the species’ range may be responsible for the sometimes contradictory findings in the scientific literature describing sage-grouse habitat use and selection during important life history stages, such as nesting. To help understand the importance of this variability and to help guide management actions, we evaluated the nesting and brood-rearing microhabitat factors that influence selection and survival patterns in the Great Basin using a large dataset of microhabitat characteristics from study areas spanning northern Nevada and a portion of northeastern California from 2009 to 2016. The spatial and temporal coverage of the dataset provided a powerful opportunity to evaluate microhabitat factors important to sage-grouse reproduction, while also considering habitat variation associated with different climatic conditions and areas affected by wildfire. The summary statistics for numerous microhabitat factors, and the strength of their association with sage-grouse habitat selection and survival, are provided in this report to support decisions by land managers, policy-makers, and others with the best-available science in a timely manner.

California, Nevada↗

Earthquake likelihood model testing

INTRODUCTION The Regional Earthquake Likelihood Models (RELM) project aims to produce and evaluate alternate models of earthquake potential (probability per unit volume, magnitude, and time) for California. Based on differing assumptions, these models are produced to test the validity of their assumptions and to explore which models should be incorporated in seismic hazard and risk evaluation. Tests based on physical and geological criteria are useful but we focus on statistical methods using future earthquake catalog data only. We envision two evaluations: a test of consistency with observed data and a comparison of all pairs of models for relative consistency. Both tests are based on the likelihood method, and both are fully prospective ( i.e. , the models are not adjusted to fit the test data). To be tested, each model must assign a probability to any possible event within a specified region of space, time, and magnitude. For our tests the models must use a common format: earthquake rates in specified “bins” with location, magnitude, time, and focal mechanism limits. Seismology cannot yet deterministically predict individual earthquakes; however, it should seek the best possible models for forecasting earthquake occurrence. This paper describes the statistical rules of an experiment to examine and test earthquake forecasts. The primary purposes of the tests described below are to evaluate physical models for earthquakes, assure that source models used in seismic hazard and risk studies are consistent with earthquake data, and provide quantitative measures by which models can be assigned weights in a consensus model or be judged as suitable for particular regions. In this paper we develop a statistical method for testing earthquake likelihood models. A companion paper ( Schorlemmer and Gerstenberger 2007 , this issue) discusses the actual implementation of these tests in the framework of the RELM initiative. Statistical testing of hypotheses is a common task and a wide range of possible testing procedures exist. Jolliffe and Stephenson ( 2003 ) present different forecast verifications from atmospheric science, among them likelihood testing of probability forecasts and testing the occurrence of binary events. Testing binary events requires that for each forecasted event, the spatial, temporal and magnitude limits be given. Although major earthquakes can be considered binary events, the models within the RELM project express their forecasts on a spatial grid and in 0.1 magnitude units; thus the results are a distribution of rates over space and magnitude. These forecasts can be tested with likelihood tests. In general, likelihood tests assume a valid null hypothesis against which a given hypothesis is tested. The outcome is either a rejection of the null hypothesis in favor of the test hypothesis or a nonrejection, meaning the test hypothesis cannot outperform the null hypothesis at a given significance level. Within RELM, there is no accepted null hypothesis and thus the likelihood test needs to be expanded to allow comparable testing of equipollent hypotheses. To test models against one another, we require that forecasts are expressed in a standard format: the average rate of earthquake occurrence within pre-specified limits of hypocentral latitude, longitude, depth, magnitude, time period, and focal mechanisms. Focal mechanisms should either be described as the inclination of P -axis, declination of P -axis, and inclination of the T -axis, or as strike, dip, and rake angles. Schorlemmer and Gerstenberger ( 2007 , this issue) designed classes of these parameters such that similar models will be tested against each other. These classes make the forecasts comparable between models. Additionally, we are limited to testing only what is precisely defined and consistently reported in earthquake catalogs. Therefore it is currently not possible to test such information as fault rupture length or area, asperity location, etc. Also, to account for data quality issues, we allow for location and magnitude uncertainties as well as the probability that an event is dependent on another event. As we mentioned above, only models with comparable forecasts can be tested against each other. Our current tests are designed to examine grid-based models. This requires that any fault-based model be adapted to a grid before testing is possible. While this is a limitation of the testing, it is an inherent difficulty in any such comparative testing. Please refer to appendix B for a statistical evaluation of the application of the Poisson hypothesis to fault-based models. The testing suite we present consists of three different tests: L-Test, N-Test, and R-Test. These tests are defined similarily to Kagan and Jackson ( 1995 ). The first two tests examine the consistency of the hypotheses with the observations while the last test compares the spatial performances of the models.

Seismological Research Letters↗

Chapter 5: Greater sage-grouse: General use and roost site occurrence with pellet counts as a measure of relative abundance

Greater sage-grouse ( Centrocercus urophasianus ) have been declining both spatially and numerically throughout their range because of anthropogenic disturbance and loss and fragmentation of sagebrush ( Artemisia spp.) habitats. Understanding how sage-grouse respond to these habitat alterations and disturbances, particularly the types of disturbances and extent at which they respond, is critical to designing management actions and prioritizing areas of conservation. To address these needs, we developed statistical models of the relationships between occurrence and abundance of greater sage-grouse and multi-scaled measures of vegetation, abiotic, and disturbance in the Wyoming Basins Ecoregional Assessment (WBEA) area. Sage-grouse occurrence was strongly related to the amount of sagebrush within 1 km for both roost site and general use locations. Roost sites were identified by presence of sage-grouse fecal pellet groups whereas general use locations had single pellets. Proximity to anthropogenic disturbance including energy development, power lines, and major roads was negatively associated with sage-grouse occurrence. Models of sage-grouse occurrence correctly predicted active lek locations with >75% accuracy. Our spatially explicit models identified areas of high occurrence probability in the WBEA area that can be used to delineate areas for conservation and refine existing conservation plans. These models can also facilitate identification of pathways and corridors important for maintenance of sage-grouse population connectivity.

Wyoming↗

Description of the National Hydrologic Model for use with the Precipitation-Runoff Modeling System (PRMS)

This report documents several components of the U.S. Geological Survey National Hydrologic Model of the conterminous United States for use with the Precipitation-Runoff Modeling System (PRMS). It provides descriptions of the (1) National Hydrologic Model, (2) Geospatial Fabric for National Hydrologic Modeling, (3) PRMS hydrologic simulation code, (4) parameters and estimation methods used to compute spatially and temporally distributed default values as required by PRMS, (5) National Hydrologic Model Parameter Database, and (6) model extraction tool named Bandit. The National Hydrologic Model Parameter Database contains values for all PRMS parameters used in the National Hydrologic Model. The methods and national datasets used to estimate all the PRMS parameters are described. Some parameter values are derived from characteristics of topography, land cover, soils, geology, and hydrography using traditional Geographic Information System methods. Other parameters are set to long-established default values and computation of initial values. Additionally, methods (statistical, sensitivity, calibration, and algebraic) were developed to compute parameter values on the basis of a variety of nationally-consistent datasets. Values in the National Hydrologic Model Parameter Database can periodically be updated on the basis of new parameter estimation methods and as additional national datasets become available. A companion ScienceBase resource provides a set of static parameter values as well as images of spatially-distributed parameters associated with PRMS states and fluxes for each Hydrologic Response Unit across the conterminuous United States.

Techniques and Methods↗

Development of spatial data guidelines and standards: spatial data set documentation to support hydrologic analysis in the U.S. Geological Survey

Spatial data analysis has become an integral component in many surface and sub-surface hydrologic investigations within the U.S. Geological Survey (USGS). Currently, one of the largest costs in applying spatial data analysis is the cost of developing the needed spatial data. Therefore, guidelines and standards are required for the development of spatial data in order to allow for data sharing and reuse; this eliminates costly redevelopment. In order to attain this goal, the USGS is expanding efforts to identify guidelines and standards for the development of spatial data for hydrologic analysis. Because of the variety of project and database needs, the USGS has concentrated on developing standards for documenting spatial sets to aid in the assessment of data set quality and compatibility of different data sets. An interim data set documentation standard (1990) has been developed that provides a mechanism for associating a wide variety of information with a data set, including data about source material, data automation and editing procedures used, projection parameters, data statistics, descriptions of features and feature attributes, information on organizational contacts lists of operations performed on the data, and free-form comments and notes about the data, made at various times in the evolution of the data set. The interim data set documentation standard has been automated using a commercial geographic information system (GIS) and data set documentation software developed by the USGS. Where possible, USGS developed software is used to enter data into the data set documentation file automatically. The GIS software closely associates a data set with its data set documentation file; the documentation file is retained with the data set whenever it is modified, copied, or transferred to another computer system. The Water Resources Division of the USGS is continuing to develop spatial data and data processing standards, with emphasis on standards needed to support hydrologic analysis, hydrologic data processing, and publication of hydrologic thermatic maps. There is a need for the GIS vendor community to develop data set documentation tools similar to those developed by the USGS, or to incorporate USGS developed tools in their software.

Conference Paper↗

Characterization of water use and water balance for the croplands of Kansas using satellite, climate, and irrigation data

Kansas is one of the most productive agricultural states in the United States, where agricultural irrigation is a primary user of underground and surface water. Because of low precipitation and declining groundwater levels in western and central Kansas, sustainable management of irrigation water resources is a critical issue in the agricultural productivity of the state. The objective of this study is to analyze and characterize the water use and water balance in the croplands of Kansas using satellite observations, meteorological data, and in situ irrigation water use records. We used actual evapotranspiration ( ETa ), precipitation, soil moisture, and irrigation water use to calculate water balance for Kansas in 2015 at scales of counties, climatic divisions, and groundwater management districts (GMD). The Operational Simplified Surface Energy Balance model was implemented to estimate 30-m resolution ETa . Results showed that the seasonal (May – September) precipitation, soil water storage change, and ETa are 528 mm, 80 mm, and 555 mm, respectively, on average of all croplands in the state. The annual net irrigation water consumption was 293 mm for irrigated croplands, indicating that irrigation water constitutes an substantial portion of the water supply in the state. The total volumetric irrigation water use was 3.24 km 3 for all croplands within five GMDs in western and south-central Kansas, while only 0.38 km 3 was outside of GMDs. The multiple regression models of ETa against precipitation and irrigation water use were statistically significant with R 2 values of 0.71 and 0.87, respectively, at county and climate division scales. Regression models also indicated a higher rate of ETa response to irrigation water use than that to precipitation. Our study demonstrated the spatial patterns of crop water use and water balance in Kansas, which could provide useful information for management of irrigation agriculture and water resources for the state.

Kansas↗

Individual encounter data of six African carnivore species optimized for multi-species density estimation

The ability to estimate abundances of multiple wildlife species within an area is valuable for both conservation and ecological inquiry. Spatially explicit capture–recapture (SCR) methods are commonly used to obtain reliable population size estimates, particularly for low-density and individually identifiable carnivore species. However, estimating abundance within multi-species communities poses a methodological challenge as survey designs and analytical tools are primarily tailored for single target species. Here, we present a dataset of spatially referenced individual encounter histories of six carnivore species with varying space requirements (lion, Panthera leo ; leopard, Panthera pardus ; spotted hyena, Crocuta crocuta ; cheetah, Acinonyx jubatus ; serval, Leptailurus serval ; large-spotted genet, Genetta tigrina ). These data were collected in a South African game reserve using a camera trap array optimized for multi-species density estimation using SCR methods. This dataset will be a valuable resource for studying spatial processes among potentially interacting carnivores without the common pitfalls that come with by-catch data of non-target species, and will provide a much-needed case study for the further development of multi-species statistical method development.

Munywana Conservancy↗

Bed-sediment grain-size and morphologic data from Suisun, Grizzly, and Honker Bays, CA, 1998-2002

The USGS Place Based Studies Program for San Francisco Bay investigates this sensitive estuarine system to aid in resource management. As part of the inter-disciplinary research program, the USGS collected side-scan sonar data and bed-sediment samples from north San Francisco Bay to characterize bed-sediment texture and investigate temporal trends in sedimentation. The study area is located in central California and consists of Suisun Bay, and Grizzly and Honker Bays, sub-embayments of Suisun Bay. During the study (1998-2002), the USGS collected three side-scan sonar data sets and approximately 300 sediment samples. The side-scan data revealed predominantly fine-grained material on the bayfloor. We also mapped five different bottom types from the data set, categorized as featureless, furrows, sand waves, machine-made, and miscellaneous. We performed detailed grain-size and statistical analyses on the sediment samples. Overall, we found that grain size ranged from clay to fine sand, with the coarsest material in the channels and finer material located in the shallow bays. Grain-size analyses revealed high spatial variability in size distributions in the channel areas. In contrast, the shallow regions exhibited low spatial variability and consistent sediment size over time.

California↗

Areal distribution and concentration of contaminants of concern in surficial streambed and lakebed sediments, Lake St. Clair and tributaries, Michigan, 1990-2003

As part of the Lake St. Clair Regional Monitoring Project, the U.S. Geological Survey evaluated data collected from surficial streambed and lakebed sediments in the Lake Erie-Lake St. Clair drainages. This study incorporates data collected from 1990 through 2003 and focuses primarily on the U.S. part of the Lake St. Clair Basin, including Lake St. Clair, the St. Clair River, and tributaries to Lake St. Clair. Comparable data from the Canadian part of the study area are included where available. The data are compiled into 4 chemical classes and consist of 21 compounds. The data are compared to effects-based sediment-quality guidelines, where the Threshold Effect Level and Lowest Effect Level represent concentrations below which adverse effects on biota are not expected and the Probable Effect Level and Severe Effect Level represent concentrations above which adverse effects on biota are expected to be frequent. Maps in the report show the spatial distribution of the sampling locations and illustrate the concentrations relative to the selected sediment-quality guidelines. These maps indicate that sediment samples from certain areas routinely had contaminant concentrations greater than the Threshold Effect Concentration or Lowest Effect Level. These locations are the upper reach of the St. Clair River, the main stem and mouth of the Clinton River, Big Beaver Creek, Red Run, and Paint Creek. Maps also indicated areas that routinely contained sediment contaminant concentrations that were greater than the Probable Effect Concentration or Severe Effect Level. These locations include the upper reach of the St. Clair River, the main stem and mouth of the Clinton River, Red Run, within direct tributaries along Lake St. Clair and in marinas within the lake, and within the Clinton River headwaters in Oakland County. Although most samples collected within Lake St. Clair were from sites adjacent to the mouths of its tributaries, samples analyzed for trace-element concentrations were collected throughout the lake. The distribution of trace-element concentrations corresponded well with the results of a two-dimensional hydrodynamic model of flow patterns from the Clinton River into Lake St. Clair. The model was developed independent from the bed sediment analysis described in this report; yet it showed a zone of deposition for outflow from the Clinton River into Lake St. Clair that corresponded well with the spatial distribution of trace-element concentrations. This zone runs along the western shoreline of Lake St. Clair from L'Anse Creuse Bay to St. Clair Shores, Michigan and is reflected in the samples analyzed for mercury and cadmium. Statistical summaries of the concentration data are presented for most contaminants, and selected statistics are compared to effects-based sediment-quality guidelines. Summaries were not computed for dieldrin, chlordane, hexachlorocyclohexane, lindane, and mirex because insufficient data are available for these contaminants. A statistical comparison showed that the median concentration for hexachlorobenzene, anthracene, benz[ a ]anthracene, chrysene, and pyrene are greater than the Threshold Effect Concentration or Lowest Effect Level. Probable Effect Concentration Quotients provide a mechanism for comparing the concentrations of contaminant mixtures against effects-based biota data. Probable Effect Concentration Quotients were calculated for individual samples and compared to effects-based toxicity ranges. The toxicity-range categories used in this study were nontoxic (quotients < 0.5) and toxic (quotients > 0.5). Of the 546 individual samples for which Probable Effect Concentration Quotients were calculated, 469 (86 percent) were categorized as being nontoxic and 77 (14 percent) were categorized as being toxic. Bed-sediment samples with toxic Probable Effect Concentration Quotients were collected from Paint Creek, Galloway Creek, the main stem of the Clinton River, Big Beaver Creek, Red Run, Clinton River towards the mouth, Lake St. Clair along the western shore, and the St. Clair River near Sarnia.

Michigan↗

A regional classification of the effectiveness of depressional wetlands at mitigating nitrogen transport to surface waters in the Northern Atlantic Coastal Plain

Nitrogen from nonpoint sources contributes to eutrophication, hypoxia, and related ecological degradation in Atlantic Coastal Plain streams and adjacent coastal estuaries such as Chesapeake Bay and Pamlico Sound. Although denitrification in depressional (non-riparian) wetlands common to the Coastal Plain can be a significant landscape sink for nitrogen, the effectiveness of individual wetlands at removing nitrogen varies substantially due to varying hydrogeologic, geochemical, and other landscape conditions, which are often poorly or inconsistently mapped over large areas. A geographic model describing the spatial variability in the likely effectiveness of depressional wetlands in watershed uplands at mitigating nitrogen transport from nonpoint sources to surface waters was constructed for the Northern Atlantic Coastal Plain (NACP), from North Carolina through New Jersey. Geographic and statistical techniques were used to develop the model. Available medium-resolution (1:100,000-scale) stream hydrography was used to define 33,799 individual watershed catchments in the study area. Sixteen landscape metrics relevant to the occurrence of depressional wetlands and their effectiveness as nitrogen sinks were defined for each catchment, based primarily on available topographic and soils data. Cluster analysis was used to aggregate the 33,799 catchments into eight wetland landscape regions (WLRs) based on the value of three principal components computed for the 16 original landscape metrics. Significant differences in topography, soil, and land cover among the eight WLRs demonstrate the effectiveness of the clustering technique. Results were used to interpret the relative likelihood of depressional wetlands in each WLR and their likely effectiveness at mitigating nitrogen transport from upland source areas to surface waters. The potential effectiveness of depressional wetlands at mitigating nitrogen transport varies substantially over different parts of the NACP. Depressional wetlands are common in three WLRs covering 32 percent of the area, and have a relatively high potential to mitigate nitrogen transport from nonpoint sources. Conversely, 37 percent of the study area includes rolling hills with relatively high slope and relief, and little likelihood of depressional wetlands. The remainder of the Coastal Plain includes relatively flat watersheds with moderate to low relative likelihood of nitrogen mitigation. The delineation of WLRs in this model should be useful for targeting wetland conservation or restoration efforts, and for estimating the effects of depressional wetlands on the regional nitrogen budget, but should be considered in light of limitations and assumptions inherent in the model.

Scientific Investigations Report↗

Comparison of nitrate, pesticides, and volatile organic compounds in samples from monitoring and public-supply wells, Kirkwood-Cohansey aquifer system, southern New Jersey

The number and total concentration of volatile organic compounds (VOCs) per sample were significantly greater in water from public-supply wells than in water from shallow and moderate-depth monitoring wells in the surficial Kirkwood-Cohansey aquifer system in the Glassboro area of southern New Jersey. In contrast, concentrations of nitrate (as nitrogen) and the number and total concentration of pesticides per sample were statistically similar in samples from shallow and moderate-depth monitoring wells and those from public-supply wells. VOCs in ground water typically are derived from point sources, which commonly exist in urban areas and which result in spatially variable contaminant concentrations near the water table. Because larger volumes of water are withdrawn from public-supply wells than from monitoring wells, their contributing areas are larger and, therefore, they are more likely to intercept water flowing from VOC point sources. Additionally, public-supply wells intercept flow paths that span a large temporal interval. Public-supply wells in the Glassboro study area withdraw water flowing along short paths, which contains VOCs that recently entered the aquifer system, and water flowing along relatively long paths, which contains VOCs that originated from the degradation of parent compounds or that are associated with past land uses. Because the volume of water withdrawn from monitoring wells is small and because shallow monitoring wells are screened near the water table, they generally intercept only relatively short flow paths. Therefore, samples from these wells represent relatively recent, discrete time intervals and contain both fewer VOCs and a lower total VOC concentration than samples from public-supply wells. Nitrate and pesticides in ground water typically are derived from nonpoint sources, which commonly are found in both agricultural and urban areas and typically result in lowlevel, relatively uniform concentrations near the water table. Because nonpoint sources are diffuse and because processes such as degradation or sorption/dispersion do not occur at rates sufficient to prevent detection of these constituents in parts of the aquifer used for domestic and public supply in the study area, concentrations of nitrate and pesticides and numbers of pesticide compounds are likely to be similar in samples from shallow monitoring wells and samples from public-supply wells. Results of a comparison of (1) the general characteristics of, and water-quality data from, public-supply wells in the Glassboro study area to available data from public-supply wells screened in the Kirkwood-Cohansey aquifer system outside the study area, and (2) land-use settings, soil characteristics, and aquifer properties in and outside the study area indicate that the findings of this study likely are applicable to the entire extent of the Kirkwood- Cohansey aquifer system in southern New Jersey.

New Jersey↗

Major and trace element geochemistry and background concentrations for soils in Connecticut

Soil samples were collected throughout Connecticut (CT) to determine the relationship of soil chemistry with the underlying geology and to better understand background concentrations of major and trace elements in soils. Soil samples were collected (1) from the upper 5 cm of surficial soil at 100 sites, (2) from the A horizon at 86 of these sites, and (3) from the deeper horizon, typically the C horizon, at 79 of these sites. The <2-millimeter fraction of each sample was analyzed for 44 elements by methods that yield the total or near-total elemental content. Sample sites were characterized by glacial setting, underlying bedrock geology, and soil type. These spatial data were used with element concentrations in the C-horizon to relate geologic factors to soil chemistry. Concentrations of elements in C-horizon soils varied with grain size in surficial glacial materials and with underlying rock types, as determined using nonparametric statistical procedures. Concentrations of most elements in C-horizon soils showed a positive correlation with silt and (or) clay content and were higher in surficial materials mapped as till, thick till, and (or) fines. Element concentrations in C-horizon soils showed significant differences among the underlying geologic provinces and were highest overlying the Grenville Belt and (or) the Grenville Shelf Sequence Provinces in western CT. These rocks consist mainly of carbonates and the relatively high element concentrations in overlying soils likely result from less influence of dilution by quartz compared to other provinces. Element concentrations in C-horizon soils in CT were compared with those in samples from other New England states overlying similar lithologic bedrock types. The upper range of As concentrations in C-horizon soils overlying the New Hampshire-Maine (NH-ME) Sequence in CT was 15 mg/kg, lower than the upper range of 24 mg/kg in C-horizon soils overlying the same sequence in ME. In CT, U concentration means were significantly higher in C-horizon soils overlying Avalonian granites, and U concentrations ranged as high as 14 mg/kg, compared to those in C-horizon soil samples collected from other New England states, which ranged as high as 6.1 mg/kg in a sample in NH overlying the NH-ME Sequence. Element concentrations in C-horizon soils in CT were compared with those in samples collected from shallower depths. Concentrations of most major elements were highest in C-horizon soil samples, including Al, Ca, Fe, K, Na, and Ti, but element concentrations showed a relatively similar pattern in A-horizon and surficial soil samples among the underlying geologic provinces. Trace element concentrations, including Ba, W, Ga, Ni, Cs, Rb, Sr, Th, Sc, and U, also were higher in C-horizon soil samples than in overlying soil samples. Concentrations of Mg, and several trace elements, including Mn, P, As, Nb, Sn, Be, Bi, Hg, Se, Sb, La, Co, Cr, Pb, V, Y, Cu, Pb, and Zn were highest in some A-horizon or surficial soils, and indicate possible contributions from anthropogenic sources. Because element concentrations in soils above the C horizon are more likely to be affected by anthropogenic factors, concentration ranges in C-horizon soils and their spatially varying geologic associations should be considered when estimating background concentrations of elements in CT soils.

Connecticut↗

A probabilistic method for constructing wave time-series at inshore locations using model scenarios

Continuous time-series of wave characteristics (height, period, and direction) are constructed using a base set of model scenarios and simple probabilistic methods. This approach utilizes an archive of computationally intensive, highly spatially resolved numerical wave model output to develop time-series of historical or future wave conditions without performing additional, continuous numerical simulations. The archive of model output contains wave simulations from a set of model scenarios derived from an offshore wave climatology. Time-series of wave height, period, direction, and associated uncertainties are constructed at locations included in the numerical model domain. The confidence limits are derived using statistical variability of oceanographic parameters contained in the wave model scenarios. The method was applied to a region in the northern Gulf of Mexico and assessed using wave observations at 12 m and 30 m water depths. Prediction skill for significant wave height is 0.58 and 0.67 at the 12 m and 30 m locations, respectively, with similar performance for wave period and direction. The skill of this simplified, probabilistic time-series construction method is comparable to existing large-scale, high-fidelity operational wave models but provides higher spatial resolution output at low computational expense. The constructed time-series can be developed to support a variety of applications including climate studies and other situations where a comprehensive survey of wave impacts on the coastal area is of interest.

Coastal Engineering↗

Predicting the potential distribution of invasive exotic species using GIS and information-theoretic approaches: A case of ragweed (Ambrosia artemisiifolia L.) distribution in China

Invasive exotic species pose a growing threat to the economy, public health, and ecological integrity of nations worldwide. Explaining and predicting the spatial distribution of invasive exotic species is of great importance to prevention and early warning efforts. We are investigating the potential distribution of invasive exotic species, the environmental factors that influence these distributions, and the ability to predict them using statistical and information-theoretic approaches. For some species, detailed presence/absence occurrence data are available, allowing the use of a variety of standard statistical techniques. However, for most species, absence data are not available. Presented with the challenge of developing a model based on presence-only information, we developed an improved logistic regression approach using Information Theory and Frequency Statistics to produce a relative suitability map. This paper generated a variety of distributions of ragweed (Ambrosia artemisiifolia L.) from logistic regression models applied to herbarium specimen location data and a suite of GIS layers including climatic, topographic, and land cover information. Our logistic regression model was based on Akaike's Information Criterion (AIC) from a suite of ecologically reasonable predictor variables. Based on the results we provided a new Frequency Statistical method to compartmentalize habitat-suitability in the native range. Finally, we used the model and the compartmentalized criterion developed in native ranges to "project" a potential distribution onto the exotic ranges to build habitat-suitability maps. ?? Science in China Press 2007.

Chinese Science Bulletin↗