SEARCH · Search USGS

Results for “Spatial Statistics”

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 217 records · Page 12Linked to original sources

USGS2015

Soil ecology of a rock outcrop ecosystem: Abiotic stresses, soil respiration, and microbial community profiles in limestone cedar glades

Limestone cedar glades are a type of rock outcrop ecosystem characterized by shallow soil and extreme hydrologic conditions—seasonally ranging from xeric to saturated—that support a number of plant species of conservation concern. Although a rich botanical literature exists on cedar glades, soil biochemical processes and the ecology of soil microbial communities in limestone cedar glades have largely been ignored. This investigation documents the abiotic stress regime of this ecosystem (shallow soil, extreme hydrologic fluctuations and seasonally high soil surface temperatures) as well as soil physical and chemical characteristics, and relates both types of information to ecological structures and functions including vegetation, soil respiration, and soil microbial community metabolic profiles and diversity. Methods used in this investigation include field observations and measurements of soil physical and chemical properties and processes, laboratory analyses, and microbiological assays of soil samples. Stress factors quantified by this research include shallow soil (depth to bedrock ranging from 2.4 to 22.6 cm), volumetric soil water content levels seasonally ranging from xeric (below 5%) to saturated (above 50%), and seasonally extreme ground-surface temperatures (above 48°C). Findings from this research indicate that spatial and temporal heterogeneity exists in limestone cedar glades in terms of abiotic stress factors and soil physical and chemical properties. Several such soil properties (e.g. soil depth, organic matter levels, pH, and particle size distribution) are spatially correlated. These soil properties were statistically related to ecological structures and functions such as vegetation patterns, soil respiration, the density of culturable heterotrophic microbes in soil and metabolic diversity of soil microbial community profiles. In general, zones within limestone cedar glades that had relatively shallow soil, alkaline pH, low levels of organic matter and high levels of silt also tended to have depressed rates of soil respiration and reduced densities and metabolic diversity of culturable heterotrophic soil microbes. Additionally, seasonally-relevant stress factors including soil water content and temperatures at or near the soil surface were related to the same set of ecological structures and functions.

Tennessee↗

USGS2010

Status and understanding of groundwater quality in the central-eastside San Joaquin Basin, 2006: California GAMA Priority Basin Project

Groundwater quality in the approximately 1,695-square-mile Central Eastside San Joaquin Basin (Central Eastside) study unit was investigated as part of the Priority Basin Project (PBP) of the Groundwater Ambient Monitoring and Assessment (GAMA) Program. The GAMA PBP was developed in response to the California Groundwater Quality Monitoring Act of 2001, and is being conducted by the California State Water Resources Control Board in collaboration with the U.S. Geological Survey and the Lawrence Livermore National Laboratory. The GAMA Central Eastside study unit was designed to provide a spatially unbiased assessment of untreated-groundwater quality, as well as a statistically consistent basis for comparing water quality throughout California. During March through June 2006, samples were collected from 78 wells in Stanislaus and Merced Counties, 58 of which were selected using a spatially distributed, randomized grid-based method to provide statistical representation of the study unit (grid wells), and 20 of which were sampled to evaluate changes in water chemistry along groundwater-flow paths (understanding wells). Water-quality data from the California Department of Public Health (CDPH) database also were used for the assessment. An assessment of the current status of the groundwater quality included collecting samples from wells for analysis of anthropogenic constituents such as volatile organic compounds (VOCs) and pesticides, as well as naturally occurring constituents such as major ions and trace elements. The assessment of status is intended to characterize the quality of untreated-groundwater resources within the primary aquifer system, not the treated drinking water delivered to consumers by water purveyors. The primary aquifer system (hereinafter, primary aquifer) is defined as that part of the aquifer corresponding to the perforation interval of wells listed in the CDPH database for the Central Eastside study unit. The quality of groundwater in shallower or deeper water-bearing zones may differ from that in the primary aquifer; shallower groundwater may be more vulnerable to surficial contamination. The primary aquifer is represented by the grid wells, of which 90 percent had depths to the tops of their perforations of about 80 to 330 feet and depths to bottom of about 100 to 670 feet. Relative-concentrations (sample concentration divided by benchmark concentration) were used as the primary metric for assessing the status of water quality for those constituents that have Federal and (or) California human health or aesthetic benchmarks. A relative-concentration greater than (>) 1.0 indicates a concentration above a benchmark, and less than or equal to (≤) 1.0 indicates a concentration equal to or below a benchmark. For organic and special interest constituents, relative-concentrations were classified as high (>1.0), moderate (≤1.0 and >0.1), or low (≤0.1). For inorganic constituents, relative-concentrations were classified as high (>1.0), moderate (≤1.0 and >0.5), or low (≤0.5). The threshold between low and moderate classifications was lower for organic and special interest constituents than for inorganic constituents because organic constituents generally are less prevalent and have smaller relative-concentrations than inorganic constituents. Grid-based and spatially-weighted approaches, the latter incorporating data from all CDPH wells, were used to evaluate the proportion of the primary aquifer (aquifer-scale proportions) with high, moderate, or low relative-concentrations. For individual constituents or classes of constituents, the aquifer-scale high proportion is the percentage of the area of the study unit having high relative-concentrations within the depth-zones of the primary aquifer. Aquifer-scale moderate and low proportions are defined similarly. Spatially-weighted aquifer-scale high proportions nearly always fell within the 90-percent confidence interval of grid-based aquifer-scale high proportions, indicating that the grid-based approach yielded statistically equivalent results to the spatially-weighted approach incorporating CDPH data. The status assessment for inorganic constituents showed that inorganic constituents (one or more) were high, relative to human-health benchmarks, in 18.0 percent of the primary aquifer, moderate in 44.0 percent, and low in 38.0 percent. Of inorganic constituents with human-health benchmarks, arsenic, vanadium, and nitrate were detected at high relative-concentrations in 15.6 percent, 3.6 percent, and 2.1 percent, respectively, of the primary aquifer. Of inorganic constituents with secondary maximum contaminant levels (SMCL), manganese, iron, and TDS were detected at high relative-concentrations in 4.5 percent, 2.2 percent, and 1.7 percent, respectively, of the primary aquifer. The status assessment for organic constituents showed that organic constituents (one or more) were high, relative to human-health benchmarks, in a smaller proportion of the primary aquifer (1.2 percent) than inorganic constituents (18.0 percent). Organic constituents had moderate relative-concentrations in 14.3 percent, and had low relative-concentrations or were not detected in 84.5 percent, of the primary aquifer. The proportion of the primary aquifer with high relative-concentrations of organic constituents reflected high proportions of the discontinued soil fumigant 1,2-dibromo-3-chlororopane (DBCP; 1.0 percent) and the solvent tetrachloroethene (PCE; 0.2 percent). Most of the organic and special interest constituents detected in groundwater in the Central Eastside study unit have human-health benchmarks. Of the 205 organic and special interest constituents analyzed for, 36 constituents were detected. Of these constituents, 32 were detected only at low relative-concentrations. Four constituents, chloroform, carbon tetrachloride, DBCP, and perchlorate, were detected at moderate relative-concentrations in grid wells. Nine organic and special-interest constituents were detected frequently (detected in greater than 10 percent of samples): the trihalomethanes chloroform, bromoform, bromodichloromethane, and dibromochloromethane; the solvent PCE; the herbicides atrazine, simazine, and metolachlor, and special-interest constituent perchlorate. An assessment of understanding of the groundwater quality included sampling of understanding wells, some of which were perforated in shallower or deeper portions of the aquifer system than the primary aquifer, and analysis of correlations of groundwater quality with land use, depth, age classification, and other potential explanatory factors. The understanding assessment indicated that the concentrations of many constituents were related to depth and groundwater age. However, concentrations of individual constituents or constituent classes also were sometimes related to geochemical conditions, lateral position in the flow system, or land use. High and moderate relative-concentrations of uranium, nitrate, and total dissolved solids (TDS) were detected in some wells where the tops of perforations are within the upper 200 feet of the aquifer system. In wells with the depth to the top of perforations below this depth, concentrations were low. A similar pattern occurred for the sum of herbicide concentrations. These vertical water-chemistry patterns are consistent with the hydrogeologic setting, in which return flows from agricultural and urban land use are the major source of recharge, and withdrawals for irrigation and urban supply are the major source of discharge, resulting in substantial vertical components of groundwater flow. The decrease in concentrations of many constituents with depth reflects in part that groundwater gets older with depth. Tritium, helium-isotopes, and carbon-14 data were used to classify the predominant age of groundwater samples into three categories: modern (water that has entered the aquifer in the last 50 years), pre-modern (water that entered the aquifer more than 50 years, up to tens of thousands of years, ago), and mixed (mixtures of waters with modern and pre-modern ages). Uranium, nitrate, and herbicide concentrations were significantly higher in groundwater having modern- and mixed-ages than pre-modern ages, indicating that these constituents may be affected by anthropogenic activities in the last 50 years. Other patterns in the distribution of nitrate, uranium, and TDS are evident. Isotopic and geochemical data are consistent with partial denitrification of nitrate in some reducing groundwaters in the western and deeper parts of the flow system. Uranium and TDS concentrations increase from east to west across the valley, along the direction of regional lateral groundwater flow. High and moderate relative-concentrations of arsenic can be attributed to reductive dissolution of manganese or iron oxides, or to desorption by high pH waters. Arsenic concentrations also increased with increasing depth and groundwater age. High to moderate relative-concentrations of vanadium primarily are related to high pH under oxic conditions. The frequency of detections of DBCP was greater in areas with orchard-vineyard land use >40 percent and at depths <200 feet. THMs and solvents were correlated positively with percent urban land use. Herbicide concentrations were correlated negatively with percent natural land use. Perchlorate concentrations were significantly greater in waters having modern and mixed ages than waters having pre-modern ages and were significantly and positively correlated with two land uses—percent orchard/vineyard land use and percent urban land use.

California↗

USGS2025

Estimating spatially explicit survival and mortality risk from telemetry data with thinned point process models

Mortality risk for animals often varies spatially and can be linked to how animals use landscapes. While numerous studies collect telemetry data on animals, the focus is typically on the period when animals are alive, even though there is important information that could be gleaned about mortality risk. We introduce a thinned spatial point process (SPP) modelling framework that couples relative abundance and space use with a mortality process to formally treat the occurrence of mortality events across the landscape as a spatial process. We show how this model can be embedded in a hierarchical statistical framework and fit to telemetry data to make inferences about how spatial covariates drive both space use and mortality risk. We apply the method to two data sets to study the effects of roads and habitat on spatially explicit mortality risk: (1) VHF telemetry data collected for willow ptarmigan in Alaska, and (2) hourly GPS telemetry data collected for black bears in Colorado. These case studies demonstrate the applicability of this method for different species and data types, making it broadly useful in enabling inferences about the mechanisms influencing animal survival and spatial population processes while formally treating survival as a spatial process, especially as the development and implementation of joint analyses continue to progress.

Ecology Letters↗

USGS2020

An enhanced hydrologic stream network based on the NHDPlus medium resolution dataset

The National Hydrography Dataset Plus, Version 2.1 (NHDPlusV2.1) is an attribute-rich digital stream network for the conterminous United States, serving as a foundational infrastructure for reporting hydrologic information at both regional and national scales. SPAtially Referenced Regressions On Watershed attributes (SPARROW) is a process-based statistical model that relies on a digital hydrologic network like NHDPlusV2.1 to establish spatial relations between quantities of monitored contaminant loads and contaminant sources, accounting for the physical characteristics along flow paths affecting contaminant transport. The U.S. Geological Survey National Water Quality Assessment project adopted and modified the medium-resolution NHDPlusV2.1 network for use as the primary framework supporting SPARROW modeling. This report describes the enhancements made to improve the routing capabilities and the value-added attributes of NHDPlusV2.1 to support modeling and other hydrologic analyses. These enhancements include corrections to inconsistencies in network/routing information, filling in missing attribute values of associated characteristics, accounting of water use affecting flow, new variables useful for interpreting network data, revised flowline attributes such as slope and flow, and incorporation of ancillary spatial data into the network. The resulting dataset containing the enhancements to the network is named E2NHDPlusV2_US. Although the enhancements described in the report were developed for use in SPARROW modeling, the enhancements are expected to be useful for a wide variety of hydrologic studies within the United States.

Conterminous United States↗

USGS2005

Mining Claim Activity on Federal Land for the Period 1976 through 2003

Previous reports on mining claim records provided information and statistics (number of claims) using data from the U.S. Bureau of Land Management's (BLM) Mining Claim Recordation System. Since that time, BLM converted their mining claim data to the Legacy Repost 2000 system (LR2000). This report describes a process to extract similar statistical data about mining claims from LR2000 data using different software and procedures than were used in the earlier work. A major difference between this process and the previous work is that every section that has a mining claim record is assigned a value. This is done by proportioning a claim between each section in which it is recorded. Also, the mining claim data in this report includes all BLM records, not just the western states. LR2000 mining claim database tables for the United States were provided by BLM in text format and imported into a Microsoft? Access2000 database in January, 2004. Data from two tables in the BLM LR2000 database were summarized through a series of database queries to determine a number that represents active mining claims in each Public Land Survey (PLS) section for each of the years from 1976 to 2002. For most of the area, spatial databases are also provided. The spatial databases are only configured to work with the statistics provided in the non-spatial data files. They are suitable for geographic information system (GIS)-based regional assessments at a scale of 1:100,000 or smaller (for example, 1:250,000).

Data Series↗

USGS2002

Extracting temporal and spatial information from remotely sensed data for mapping wildlife habitat: Tucson

The research accomplished in this dissertation used both mathematical and statistical techniques to extract and evaluate measures of landscape temporal dynamics and spatial structure from remotely sensed data for the purpose of mapping wildlife habitat . By coupling the landscape measures gleaned from the remotely sensed data with various sets of animal sightings and population data , effective models of habitat preference were created. Measures of temporal dynamics of vegetation greenness as measured by National Oceanographic and Atmospheric Administration’s Advanced Very High Resolution Radiometer (AVHRR) satellite were used to effectively characterize and map season specific habitat of the Sonoran pronghorn antelope, as well as produce preliminary models of potential yellow-billed cuckoo habitat in Arizona. Various measures that capture different aspects of the temporal dynamics of the landscape were derived from AVHRR Normalized Difference Vegetation Index composite data using three main classes of calculations: basic statistics, standardized principal components analysis, and Fourier analysis. Pronghorn habitat models based on the AVHRR measures correspond visually and statistically to GIS-based models produced using data that represent detailed knowledge of ground-condition. Measures of temporal dynamics also revealed statistically significant correlations with annual estimates of elk population in selected Arizona Game Management Units, suggesting elk respond to regional environmental changes that can be measured using satellite data . Such relationships, once verified and established, can be used to help indirectly monitor the population. Measures of landscape spatial structure derived from IKONOS high spatial resolution (1-m) satellite data using geostatistics effectively map details of Sonoran pronghorn antelope habitat . Local estimates of the nugget, sill, and range variogram parameters calculated within 25 x 25-meter image windows describe the spatial autocorrelation of the image, permitting classification of all pixels into coherent units whose signature graphs exhibit a classic variogram shape. The variogram parameters captured in these signatures have been shown in previous studies to discriminate between different species-specific vegetation associations. The synoptic view of the landscape provided by satellite data can inform resource management efforts. The ability to characterize the spatial structure and temporal dynamics of habitat using repeatable remote sensing data allows closer monitoring of the relationship between a species and its landscape.

Arizona↗

USGS2015

Linear models for airborne-laser-scanning-based operational forest inventory with small field sample size and highly correlated LiDAR data

Modern operational forest inventory often uses remotely sensed data that cover the whole inventory area to produce spatially explicit estimates of forest properties through statistical models. The data obtained by airborne light detection and ranging (LiDAR) correlate well with many forest inventory variables, such as the tree height, the timber volume, and the biomass. To construct an accurate model over thousands of hectares, LiDAR data must be supplemented with several hundred field sample measurements of forest inventory variables. This can be costly and time consuming. Different LiDAR-data-based and spatial-data-based sampling designs can reduce the number of field sample plots needed. However, problems arising from the features of the LiDAR data, such as a large number of predictors compared with the sample size (overfitting) or a strong correlation among predictors (multicollinearity), may decrease the accuracy and precision of the estimates and predictions. To overcome these problems, a Bayesian linear model with the singular value decomposition of predictors, combined with regularization, is proposed. The model performance in predicting different forest inventory variables is verified in ten inventory areas from two continents, where the number of field sample plots is reduced using different sampling designs. The results show that, with an appropriate field plot selection strategy and the proposed linear model, the total relative error of the predicted forest inventory variables is only 5%–15% larger using 50 field sample plots than the error of a linear model estimated with several hundred field sample plots when we sum up the error due to both the model noise variance and the model’s lack of fit.

IEEE Transactions on Geoscience and Remote Sensing↗

USGS2017

Computing spatial correlation of ground motion intensities for ShakeMap

Modeling the spatial correlation of ground motion residuals, caused by coherent contributions from source, path, and site, can provide valuable loss and hazard information, as well as a more realistic depiction of ground motion intensities. The U.S. Geological Survey (USGS) software package, ShakeMap, utilizes a deterministic empirical approach to estimate median ground shaking in conjunction with observed seismic data. ShakeMap-based shaking estimates are used in concert with loss estimation algorithms to estimate fatalities and economic losses after significant seismic events around the globe. Incorporating the spatial correlation of ground motion residuals has been shown to improve seismic loss estimates. In particular, Park, Bazzuro, and Baker (Applications of Statistics and Probability in Civil Engineering, 2007) investigated computing spatially correlated random fields of residuals. However, for large scale ShakeMap grids, computational requirements of the method are prohibitive. In this work, a memory efficient algorithm is developed to compute the random fields and implemented using the ShakeMap framework. This new, iterative parallel algorithm is based on decay properties of an associated ground motion correlation function and is shown to significantly reduce computational requirements associated with adding spatial variability to the ShakeMap g round motion estimates. Further, we demonstrate and quantify the impact of adding peak ground motion spatial variability on resulting earthquake loss estimates.

Computers & Geosciences↗

USGS2012

Hierarchical distance-sampling models to estimate population size and habitat-specific abundance of an island endemic

Population size and habitat-specific abundance estimates are essential for conservation management. A major impediment to obtaining such estimates is that few statistical models are able to simultaneously account for both spatial variation in abundance and heterogeneity in detection probability, and still be amenable to large-scale applications. The hierarchical distance-sampling model of J. A. Royle, D. K. Dawson, and S. Bates provides a practical solution. Here, we extend this model to estimate habitat-specific abundance and rangewide population size of a bird species of management concern, the Island Scrub-Jay ( Aphelocoma insularis ), which occurs solely on Santa Cruz Island, California, USA. We surveyed 307 randomly selected, 300 m diameter, point locations throughout the 250-km 2 island during October 2008 and April 2009. Population size was estimated to be 2267 (95% CI 1613-3007) and 1705 (1212-2369) during the fall and spring respectively, considerably lower than a previously published but statistically problematic estimate of 12 500. This large discrepancy emphasizes the importance of proper survey design and analysis for obtaining reliable information for management decisions. Jays were most abundant in low-elevation chaparral habitat; the detection function depended primarily on the percent cover of chaparral and forest within count circles. Vegetation change on the island has been dramatic in recent decades, due to release from herbivory following the eradication of feral sheep ( Ovis aries ) from the majority of the island in the mid-1980s. We applied best-fit fall and spring models of habitat-specific jay abundance to a vegetation map from 1985, and estimated the population size of A. insularis was 1400-1500 at that time. The 20-30% increase in the jay population suggests that the species has benefited from the recovery of native vegetation since sheep removal. Nevertheless, this jay's tiny range and small population size make it vulnerable to natural disasters and to habitat alteration related to climate change. Our results demonstrate that hierarchical distance-sampling models hold promise for estimating population size and spatial density variation at large scales. Our statistical methods have been incorporated into the R package unmarked to facilitate their use by animal ecologists, and we provide annotated code in the Supplement.

California↗

USGS2010

Status and understanding of groundwater quality in the northern San Joaquin Basin, 2005: California GAMA Priority Basin Project

Groundwater quality in the 2,079 square mile Northern San Joaquin Basin (Northern San Joaquin) study unit was investigated from December 2004 through February 2005 as part of the Priority Basin Project of the Groundwater Ambient Monitoring and Assessment (GAMA) Program. The GAMA Priority Basin Project was developed in response to the Groundwater Quality Monitoring Act of 2001 that was passed by the State of California and is being conducted by the California State Water Resources Control Board in collaboration with the U.S. Geological Survey and the Lawrence Livermore National Laboratory. The Northern San Joaquin study unit was the third study unit to be designed and sampled as part of the Priority Basin Project. Results of the study provide a spatially unbiased assessment of the quality of raw (untreated) groundwater, as well as a statistically consistent basis for comparing water quality throughout California. Samples were collected from 61 wells in parts of Alameda, Amador, Calaveras, Contra Costa, San Joaquin, and Stanislaus Counties; 51 of the wells were selected using a spatially distributed, randomized grid-based approach to provide statistical representation of the study area (grid wells), and 10 of the wells were sampled to increase spatial density and provide additional information for the evaluation of water chemistry in the study unit (understanding/flowpath wells). The primary aquifer systems (hereinafter, primary aquifers) assessed in this study are defined by the depth intervals of the wells in the California Department of Public Health database for each study unit. The quality of groundwater in shallow or deep water-bearing zones may differ from quality of groundwater in the primary aquifers; shallow groundwater may be more vulnerable to contamination from the surface. Two types of assessments were made: (1) status, assessment of the current quality of the groundwater resource; and (2) understanding, identification of the natural and human factors affecting groundwater quality. Relative-concentrations (sample concentrations divided by benchmark concentrations) were used for evaluating groundwater quality for those constituents that have Federal or California regulatory or non-regulatory benchmarks for drinking-water quality. Benchmarks used in this study were either health-based (regulatory and non-regulatory) or aesthetic based (non-regulatory). For inorganic constituents, relative-concentrations were classified as high (equal to or greater than 1.0), indicating relative-concentrations greater than benchmarks; moderate (equal to or greater than 0.5, and less than 1.0); or, low (less than 0.5). For organic and special- interest constituents [1,2,3-trichloropropane (1,2,3-TCP), N-nitrosodimethylamine (NDMA), and perchlorate], relative- concentrations were classified as high (equal to or greater than 1.0); moderate (equal to or greater than 0.1 and less than 1.0); or, low (less than 0.1). Aquifer-scale proportion was used as the primary metric in the status assessment for groundwater quality. High aquifer- scale proportion is defined as the percentage of the primary aquifer with relative-concentrations greater than 1.0; moderate and low aquifer-scale proportions are defined as the percentage of the primary aquifer with moderate and low relative- concentrations, respectively. The methods used to calculate aquifer-scale proportions are based on an equal-area grid; thus, the proportions are areal rather than volumetric. Two statistical approaches - grid-based, which used one value per grid cell, and spatially weighted, which used the full dataset - were used to calculate aquifer-scale proportions for individual constituents and classes of constituents. The spatially weighted estimates of high aquifer-scale proportions were within the 90-percent confidence intervals of the grid-based estimates in all cases. The understanding assessment used statistical correlations between constituent relative-concentrations and

California↗

USGS2013

Topological and canonical kriging for design flood prediction in ungauged catchments: an improvement over a traditional regional regression approach?

In the United States, estimation of flood frequency quantiles at ungauged locations has been largely based on regional regression techniques that relate measurable catchment descriptors to flood quantiles. More recently, spatial interpolation techniques of point data have been shown to be effective for predicting streamflow statistics (i.e., flood flows and low-flow indices) in ungauged catchments. Literature reports successful applications of two techniques, canonical kriging, CK (or physiographical-space-based interpolation, PSBI), and topological kriging, TK (or top-kriging). CK performs the spatial interpolation of the streamflow statistic of interest in the two-dimensional space of catchment descriptors. TK predicts the streamflow statistic along river networks taking both the catchment area and nested nature of catchments into account. It is of interest to understand how these spatial interpolation methods compare with generalized least squares (GLS) regression, one of the most common approaches to estimate flood quantiles at ungauged locations. By means of a leave-one-out cross-validation procedure, the performance of CK and TK was compared to GLS regression equations developed for the prediction of 10, 50, 100 and 500 yr floods for 61 streamgauges in the southeast United States. TK substantially outperforms GLS and CK for the study area, particularly for large catchments. The performance of TK over GLS highlights an important distinction between the treatments of spatial correlation when using regression-based or spatial interpolation methods to estimate flood quantiles at ungauged locations. The analysis also shows that coupling TK with CK slightly improves the performance of TK; however, the improvement is marginal when compared to the improvement in performance over GLS.

Hydrology and Earth System Sciences↗

USGS2003

A stream-gaging network analysis for the 7-day, 10-year annual low flow in New Hampshire streams

The 7-day, 10-year (7Q10) low-flow-frequency statistic is a widely used measure of surface-water availability in New Hampshire. Regression equations and basin-characteristic digital data sets were developed to help water-resource managers determine surface-water resources during periods of low flow in New Hampshire streams. These regression equations and data sets were developed to estimate streamflow statistics for the annual and seasonal low-flow-frequency, and period-of-record and seasonal period-of-record flow durations. generalized-least-squares (GLS) regression methods were used to develop the annual 7Q10 low-flow-frequency regression equation from 60 continuous-record stream-gaging stations in New Hampshire and in neighboring States. In the regression equation, the dependent variables were the annual 7Q10 flows at the 60 stream-gaging stations. The independent (or predictor) variables were objectively selected characteristics of the drainage basins that contribute flow to those stations. In contrast to ordinary-least-squares (OLS) regression analysis, GLS-developed estimating equations account for differences in length of record and spatial correlations among the flow-frequency statistics at the various stations. A total of 93 measurable drainage-basin characteristics were candidate independent variables. On the basis of several statistical parameters that were used to evaluate which combination of basin characteristics contribute the most to the predictive power of the equations, three drainage-basin characteristics were determined to be statistically significant predictors of the annual 7Q10: (1) total drainage area, (2) mean summer stream-gaging station precipitation from 1961 to 90, and (3) average mean annual basinwide temperature from 1961 to 1990. To evaluate the effectiveness of the stream-gaging network in providing regional streamflow data for the annual 7Q10, the computer program GLSNET (generalized-least-squares NETwork) was used to analyze the network by application of GLS regression between streamflow and the climatic and basin characteristics of the drainage basin upstream from each stream-gaging station. Improvement to the predictive ability of the regression equations developed for the network analyses is measured by the reduction in the average sampling-error variance, and can be achieved by collecting additional streamflow data at existing stations. The predictive ability of the regression equations is enhanced even further with the addition of new stations to the network. Continued data collection at unregulated stream-gaging stations with less than 14 years of record resulted in the greatest cost-weighted reduction to the average sampling-error variance of the annual 7Q10 regional regression equation. The addition of new stations in basins with underrepresented values for the independent variables of the total drainage area, average mean annual basinwide temperature, or mean summer stream-gaging station precipitation in the annual 7Q10 regression equation yielded a much greater cost-weighted reduction to the average sampling-error variance than when more data were collected at existing unregulated stations. To maximize the regional information obtained from the stream-gaging network for the annual 7Q10, ranking of the streamflow data can be used to determine whether an active station should be continued or if a new or discontinued station should be activated for streamflow data collection. Thus, this network analysis can help determine the costs and benefits of continuing the operation of a particular station or activating a new station at another location to predict the 7Q10 at ungaged stream reaches. The decision to discontinue an existing station or activate a new station, however, must also consider its contribution to other water-resource analyses such as flood management, water quality, or trends in land use or climatic change.

Water-Resources Investigations Report↗

USGS2011

Estimating temporal trend in the presence of spatial complexity: A Bayesian hierarchical model for a wetland plant population undergoing restoration

Monitoring programs that evaluate restoration and inform adaptive management are important for addressing environmental degradation. These efforts may be well served by spatially explicit hierarchical approaches to modeling because of unavoidable spatial structure inherited from past land use patterns and other factors. We developed Bayesian hierarchical models to estimate trends from annual density counts observed in a spatially structured wetland forb (Camassia quamash [camas]) population following the cessation of grazing and mowing on the study area, and in a separate reference population of camas. The restoration site was bisected by roads and drainage ditches, resulting in distinct subpopulations ("zones") with different land use histories. We modeled this spatial structure by fitting zone-specific intercepts and slopes. We allowed spatial covariance parameters in the model to vary by zone, as in stratified kriging, accommodating anisotropy and improving computation and biological interpretation. Trend estimates provided evidence of a positive effect of passive restoration, and the strength of evidence was influenced by the amount of spatial structure in the model. Allowing trends to vary among zones and accounting for topographic heterogeneity increased precision of trend estimates. Accounting for spatial autocorrelation shifted parameter coefficients in ways that varied among zones depending on strength of statistical shrinkage, autocorrelation and topographic heterogeneity-a phenomenon not widely described. Spatially explicit estimates of trend from hierarchical models will generally be more useful to land managers than pooled regional estimates and provide more realistic assessments of uncertainty. The ability to grapple with historical contingency is an appealing benefit of this approach.

PLoS ONE↗

USGS2025

A framework for analyzing wild turkey summer sighting data.

Wildlife agencies collect data on productivity (e.g., proportion of hens with poults and number of poults per hen) of wild turkey ( Meleagris gallopavo ) to monitor population status and trends. However, sampling protocols to collect productivity data rely on opportunistic observations reported by wildlife agency personnel and the public and have changed over time and differed among agencies. A protocol to standardize data collection was adopted by most state wildlife agencies in 2019, but long-term historical datasets exist that cannot be analyzed readily to make inferences about spatial and temporal patterns in wild turkey productivity. We developed statistical models to allow comparisons and model trends in productivity among and within states even though data collection protocols changed over time and differed among states. We found greater spatial variation in the proportion of hens with poults than the number of poults per brood, which may reflect how environmental factors influence wild turkey productivity. Our models can also provide inferences about productivity when data are limited or temporally discontinuous for some spatial units. Additionally, we found that temporal and spatial variation in data collection, even under the new protocol, can affect inferences about trends in productivity. The statistical models we developed address the uncontrolled nature of when and where data are collected and offer the ability to investigate long-term patterns of productivity in relation to factors such as changing climate or habitat conditions.

Maryland, New Jersey, New York, Ohio, Pennsylvania↗

USGS1994

Characteristics of zircons from volcanic ash-derived tonsteins in Late Permian coal fields of eastern Yunnan, China

Kaolinitic tonsteins of altered synsedimentary volcanic ash-fall origin are well developed in the Late Permian coal-bearing formations of eastern Yunnan Province. Because of their unique origin, wide lateral extent, relatively constant thickness and sharp contacts with enclosing strata, great importance has been attached to these isochronous petrographic markers. In order to compare tonsteins with co-existing, non-cineritic claystones and characterize the individuality of tonsteins from different horizons for coal bed correlation, a semi-quantitative method was developed that is based on statistical analyses of the concentration and morphology of zircons and their spatial distribution patterns. This zircon-based analytical method also serves as a means for reconstructing volcanic ash-fall dispersal patterns. The results demonstrate that zircons from claystones of two different origins (i.e., tonstein and non-cineritic claystone) differ greatly in their relative abundances, crystal morphologies and spatial distribution patterns. Tonsteins from the same area but from different horizons are characterized by their own unique statistical patterns in terms of zircon concentration values and morphologic parameters (crystal length, width and the ratio of these values), thus facilitating stratigraphic correlation. Zircons from the same tonstein horizon also show continuous variation in these statistical patterns as a function of areal distribution, making it possible to identify the main path and direction in which the volcanic source materials were transported by prevailing winds.

International Journal of Coal Geology↗

USGS2010

Status and understanding of groundwater quality in the North San Francisco Bay groundwater basins, 2004: California GAMA Priority Basin Project

Groundwater quality in the approximately 1,000-square-mile (2,590-square-kilometer) North San Francisco Bay study unit was investigated as part of the Priority Basin Project of the Groundwater Ambient Monitoring and Assessment (GAMA) Program. The study unit is located in northern California in Marin, Napa, and Sonoma Counties. The GAMA Priority Basin Project is being conducted by the California State Water Resources Control Board in collaboration with the U.S. Geological Survey (USGS) and the Lawrence Livermore National Laboratory. The GAMA North San Francisco Bay study was designed to provide a spatially unbiased assessment of untreated groundwater quality in the primary aquifer systems. The assessment is based on water-quality and ancillary data collected by the USGS from 89 wells in 2004 and water-quality data from the California Department of Public Health (CDPH) database. The primary aquifer systems (hereinafter referred to as primary aquifers) were defined by the depth interval of the wells listed in the CDPH database for the North San Francisco Bay study unit. The quality of groundwater in shallower or deeper water-bearing zones may differ from that in the primary aquifers; shallower groundwater may be more vulnerable to surficial contamination. The first component of this study, the status of the current quality of the groundwater resource, was assessed by using data from samples analyzed for volatile organic compounds (VOC), pesticides, and naturally occurring inorganic constituents, such as major ions and trace elements. This status assessment is intended to characterize the quality of groundwater resources within the primary aquifers of the North San Francisco Bay study unit, not the treated drinking water delivered to consumers by water purveyors. Relative-concentrations (sample concentration divided by the health- or aesthetic-based benchmark concentration) were used for evaluating groundwater quality for those constituents that have Federal and (or) California benchmarks. A relative-concentration greater than (>) 1.0 indicates a concentration above a benchmark, and less than or equal to (≤) 1.0 indicates a concentration equal to or below a benchmark. Relative-concentrations of organic and special interest constituents were classified as “high” (relative-concentration > 1.0), “moderate” (0.1 < relative-concentration ≤ 1.0), or “low” (relative-concentration ≤ 0.1). Inorganic constituent relative-concentrations were classified as “high” (relative-concentration > 1.0), “moderate” (0.5 < relative-concentration ≤ 1.0), or “low” (relative-concentration ≤ 0.5). Aquifer-scale proportion was used as a metric for evaluating regional-scale groundwater quality. High aquifer-scale proportion is defined as the percentage of the primary aquifers that have a relative-concentration greater than 1.0; proportion is calculated on an areal rather than a volumetric basis. Moderate and low aquifer-scale proportions were defined as the percentage of the primary aquifers that have moderate and low relative-concentrations, respectively. Two statistical approaches—grid-based and spatially-weighted—were used to evaluate aquifer-scale proportion for individual constituents and classes of constituents. Grid-based and spatially-weighted estimates were comparable in the North San Francisco Bay study unit (90-percent confidence intervals). For inorganic constituents with human-health benchmarks, relative-concentrations were high in 14.0 percent of the primary aquifers, moderate in 35.8 percent, and low in 50.2 percent. The high aquifer-scale proportion of inorganic constituents primarily reflected high aquifer-scale proportions of arsenic (10.0 percent), boron (4.1 percent), and lead (1.6 percent). In contrast, relative-concentrations of organic constituents (one or more) were high in 1.4 percent, moderate in 4.9 percent, and low in 93.7 percent (not detected in 64.8 percent) of the primary aquifers. The high aquifer-scale proportion of organic constituents primarily reflected high aquifer-scale proportions of PCE (1.3 percent), TCE (0.1 percent), and 1,1-dichloroethene (0.1 percent). The inorganic constituents with secondary maximum contaminant levels (SMCL), manganese and iron, had relative-concentrations that were high in 40.8 percent and 24.4 percent of the primary aquifers, respectively. Of the 255 organic and special-interest constituents analyzed for, 26 constituents were detected. Two organic constituents were frequently detected (in 10 percent or more of samples), the trihalomethane chloroform and the herbicide simazine, but both were detected at low relative-concentrations. The second component of this study, the understanding assessment, identified the natural and human factors that affect groundwater quality by evaluating land use, physical characteristics of the wells, geochemical conditions of the aquifer, and water temperature. Results from these evaluations were used to explain the occurrence and distribution of constituents in the study unit. The understanding assessment indicated that a majority of the wells that contained nitrate also had an urban or agricultural land-use classification, had a modern or mixed age classification, and had depths to their top perforations <100 ft (30 m). Geochemical data are consistent with partial denitrification of nitrate in some reducing groundwaters in the terminal and deeper parts of the flow system. High and moderate relative-concentrations of arsenic may be attributed to reductive dissolution of manganese or iron oxides, or to desorption or inhibition of arsenic sorption under alkaline conditions. Arsenic concentrations increased with increasing depth and groundwater age in the North San Francisco Bay study unit. High to moderate relative-concentrations of boron were primarily associated with hydrothermal activity or high-salinity waters in the Napa Sonoma lowlands. Simazine was detected in groundwater classified as modern and mixed age more often than in groundwater classified as pre-modern age, while chloroform was detected most often in groundwater classified as mixed age. Simazine and chloroform also were observed in wells that had surrounding land use classified as agricultural or land use classified as urban, and top of perforation depths less than 100 ft (30 m). Together, the occurrence of chloroform and simazine in shallow wells with modern or mixed groundwater located in urban or agricultural areas suggests that these constituents result from anthropogenic activities during the last 50 years. Tritium, helium-isotope, and carbon-14 data were used to classify the predominant age of groundwater samples into three categories: modern (water that has entered the aquifer in the last 50 years), pre-modern (water that entered the aquifer more than 50 years to tens of thousands of years ago), and mixed (mixtures of modern- and pre-modern-age waters). Arsenic, iron, and total dissolved solids (TDS) concentrations were significantly greater in groundwater having pre-modern-age classification than modern, suggesting that these constituents accumulate with groundwater residence time.

California↗

USGS1997

The sandflat habitat: Scaling from experiments to conclusions

Ecological systems are characterised by spatial and temporal variations in the density of organisms and resources, and in the intensity of processes which affect them (e.g. Watt, 1947, Dayton, 1971, Addicot et al., 1987, Kolasa and Pickett, 1991 and Giller et al., 1994). This heterogeneity represents both a difficulty for field study design and statistical testing, and a challenge to describe the spatial structuring of populations, communities and ecosystems (Legendre, 1993). Patterns in ecological and environmental variables are fundamental to developing hypotheses and in designing subsequent field studies, because they determine the spatial and temporal scales of study. Spatial and/or temporal patterns also provide useful stepping stones for addressing issues of scale. Patterns apparent at one scale can collapse to noise when viewed from other scales, indicating that patterns, processes and our perceptions vary in a scale-dependent manner. Perhaps the most important but rarely discussed issue is the question of how to scale-up from small-scale surveys and experiments to conclusions relevant at larger spatial and temporal scales.

Conference Paper↗

USGS1998

Interpreting carnivore scent-station surveys

The scent-station survey method has been widely used to estimate trends in carnivore abundance. However, statistical properties of scent-station data are poorly understood, and the relation between scent-station indices and carnivore abundance has not been adequately evaluated. We assessed properties of scent-station indices by analyzing data collected in Minnesota during 1986-03. Visits to stations separated by <2 km were correlated for all species because individual carnivores sometimes visited several stations in succession. Thus, visits to stations had an intractable statistical distribution. Dichotomizing results for lines of 10 stations (0 or ≥ 1 visits) produced binomially distributed data that were robust to multiple visits by individuals. We abandoned 2-way comparisons among years in favor of tests for population trend, which are less susceptible to bias, and analyzed results separately for biogeographic sections of Minnesota because trends differed among sections. Before drawing inferences about carnivore population trends, we reevaluated published validation experiments. Results implicated low statistical power and confounding as possible explanations for equivocal or conflicting results of validation efforts. Long-term trends in visitation rates probably reflect real changes in populations, but poor spatial and temporal resolution, susceptibility to confounding, and low statistical power limit the usefulness of this survey method.

Minnesota↗