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 559 records · Page 31Linked to original sources
USGS2007

Development and implementation of a Bayesian-based aquifer vulnerability assessment in Florida

The Florida Aquifer Vulnerability Assessment (FAVA) was designed to provide a tool for environmental, regulatory, resource management, and planning professionals to facilitate protection of groundwater resources from surface sources of contamination. The FAVA project implements weights-of-evidence (WofE), a data-driven, Bayesian-probabilistic model to generate a series of maps reflecting relative aquifer vulnerability of Florida's principal aquifer systems. The vulnerability assessment process, from project design to map implementation is described herein in reference to the Floridan aquifer system (FAS). The WofE model calculates weighted relationships between hydrogeologic data layers that influence aquifer vulnerability and ambient groundwater parameters in wells that reflect relative degrees of vulnerability. Statewide model input data layers (evidential themes) include soil hydraulic conductivity, density of karst features, thickness of aquifer confinement, and hydraulic head difference between the FAS and the watertable. Wells with median dissolved nitrogen concentrations exceeding statistically established thresholds serve as training points in the WofE model. The resulting vulnerability map (response theme) reflects classified posterior probabilities based on spatial relationships between the evidential themes and training points. The response theme is subjected to extensive sensitivity and validation testing. Among the model validation techniques is calculation of a response theme based on a different water-quality indicator of relative recharge or vulnerability: dissolved oxygen. Successful implementation of the FAVA maps was facilitated by the overall project design, which included a needs assessment and iterative technical advisory committee input and review. Ongoing programs to protect Florida's springsheds have led to development of larger-scale WofE-based vulnerability assessments. Additional applications of the maps include land-use planning amendments and prioritization of land purchases to protect groundwater resources. ?? International Association for Mathematical Geology 2007.

Natural Resources Research↗
USGS2019

Calibration of Precipitation-Runoff Modeling System (PRMS) to simulate prefire and postfire hydrologic response in the upper Rio Hondo Basin, New Mexico

The Precipitation-Runoff Modeling System (PRMS) is widely used to simulate the effects of climate, topography, land cover, and soils on landscape-level hydrologic responses and streamflow. The U.S. Geological Survey (USGS), in cooperation with the New Mexico Department of Homeland Security and Emergency Management, developed procedures to apply the PRMS model to simulate the effects of fire on hydrologic responses. A PRMS model was built of the upper Rio Hondo Basin from the headwaters to approximately 19 miles downstream from the USGS streamgage Rio Hondo above Chavez Canyon near Hondo, New Mexico, by using 24 hydrologic response units (HRUs), or hydrologically similar subareas, from the National Hydrologic Model. A quasi-graphical user interface was created to easily query and analyze published PRMS sensitivity-analysis data. Simulation of mean daily streamflow was most sensitive to parameters related to snowmelt or infiltration throughout the upper Rio Hondo Basin. In the basin’s eastern and northern HRUs, flashiness and timing of streamflow were most sensitive to interflow; in many western-basin HRUs (higher elevations), flashiness of streamflow was most sensitive to soil moisture parameters, and timing of streamflow was most sensitive to infiltration and evapotranspiration parameters. The PRMS model was calibrated for the fire-affected North Fork Eagle Creek subwatershed by comparing modeled to observed daily streamflow for the nonfrozen (May through October) period for a prefire and postfire time period. The prefire model was calibrated for the period 2007–12 before the 2012 fire, and the postfire model was calibrated for a 2-year (2014–15) period after the fire. Model parameterization combined manual adjustment of 8 parameters on the basis of prior knowledge and automated adjustment of the most sensitive parameters by using the Let Us Calibrate interface. A gridded, daily precipitation dataset that captured the spatial heterogeneity across the study watershed was used as the precipitation input for calibration. Model performance was assessed as satisfactory by using standard statistical measures for prefire and postfire periods. The calibrated model was run by using data from a single precipitation gage to better represent the effect of localized, extreme storms on postfire hydrologic response. The calibrated models for prefire and postfire conditions simulated streamflows with greater consistency than the uncalibrated model for the corresponding (prefire or postfire) period of hydrographic record. The effect of fire on streamflow was found to be primarily a shift from streamflow dominated by base flow prior to fire to streamflow dominated by surface runoff after fire.

New Mexico↗
USGS2017

Lidar aboveground vegetation biomass estimates in shrublands: Prediction, uncertainties and application to coarser scales

Our study objectives were to model the aboveground biomass in a xeric shrub-steppe landscape with airborne light detection and ranging (Lidar) and explore the uncertainty associated with the models we created. We incorporated vegetation vertical structure information obtained from Lidar with ground-measured biomass data, allowing us to scale shrub biomass from small field sites (1 m subplots and 1 ha plots) to a larger landscape. A series of airborne Lidar-derived vegetation metrics were trained and linked with the field-measured biomass in Random Forests (RF) regression models. A Stepwise Multiple Regression (SMR) model was also explored as a comparison. Our results demonstrated that the important predictors from Lidar-derived metrics had a strong correlation with field-measured biomass in the RF regression models with a pseudo R 2 of 0.76 and RMSE of 125 g/m 2 for shrub biomass and a pseudo R 2 of 0.74 and RMSE of 141 g/m 2 for total biomass, and a weak correlation with field-measured herbaceous biomass. The SMR results were similar but slightly better than RF, explaining 77–79% of the variance, with RMSE ranging from 120 to 129 g/m 2 for shrub and total biomass, respectively. We further explored the computational efficiency and relative accuracies of using point cloud and raster Lidar metrics at different resolutions (1 m to 1 ha). Metrics derived from the Lidar point cloud processing led to improved biomass estimates at nearly all resolutions in comparison to raster-derived Lidar metrics. Only at 1 m were the results from the point cloud and raster products nearly equivalent. The best Lidar prediction models of biomass at the plot-level (1 ha) were achieved when Lidar metrics were derived from an average of fine resolution (1 m) metrics to minimize boundary effects and to smooth variability. Overall, both RF and SMR methods explained more than 74% of the variance in biomass, with the most important Lidar variables being associated with vegetation structure and statistical measures of this structure (e.g., standard deviation of height was a strong predictor of biomass). Using our model results, we developed spatially-explicit Lidar estimates of total and shrub biomass across our study site in the Great Basin, U.S.A., for monitoring and planning in this imperiled ecosystem.

Idaho↗
USGS2008

Modeling landslide recurrence in Seattle, Washington, USA

To manage the hazard associated with shallow landslides, decision makers need an understanding of where and when landslides may occur. A variety of approaches have been used to estimate the hazard from shallow, rainfall-triggered landslides, such as empirical rainfall threshold methods or probabilistic methods based on historical records. The wide availability of Geographic Information Systems (GIS) and digital topographic data has led to the development of analytic methods for landslide hazard estimation that couple steady-state hydrological models with slope stability calculations. Because these methods typically neglect the transient effects of infiltration on slope stability, results cannot be linked with historical or forecasted rainfall sequences. Estimates of the frequency of conditions likely to cause landslides are critical for quantitative risk and hazard assessments. We present results to demonstrate how a transient infiltration model coupled with an infinite slope stability calculation may be used to assess shallow landslide frequency in the City of Seattle, Washington, USA. A module called CRF (Critical RainFall) for estimating deterministic rainfall thresholds has been integrated in the TRIGRS (Transient Rainfall Infiltration and Grid-based Slope-Stability) model that combines a transient, one-dimensional analytic solution for pore-pressure response to rainfall infiltration with an infinite slope stability calculation. Input data for the extended model include topographic slope, colluvial thickness, initial water-table depth, material properties, and rainfall durations. This approach is combined with a statistical treatment of rainfall using a GEV (General Extreme Value) probabilistic distribution to produce maps showing the shallow landslide recurrence induced, on a spatially distributed basis, as a function of rainfall duration and hillslope characteristics.

Washington↗
USGS2019

Disease‐structured N‐mixture models: A practical guide to model disease dynamics using count data

Obtaining inferences on disease dynamics (e.g., host population size, pathogen prevalence, transmission rate, host survival probability) typically requires marking and tracking individuals over time. While multistate mark–recapture models can produce high‐quality inference, these techniques are difficult to employ at large spatial and long temporal scales or in small remnant host populations decimated by virulent pathogens, where low recapture rates may preclude the use of mark–recapture techniques. Recently developed N ‐mixture models offer a statistical framework for estimating wildlife disease dynamics from count data. N ‐mixture models are a type of state‐space model in which observation error is attributed to failing to detect some individuals when they are present (i.e., false negatives). The analysis approach uses repeated surveys of sites over a period of population closure to estimate detection probability. We review the challenges of modeling disease dynamics and describe how N ‐mixture models can be used to estimate common metrics, including pathogen prevalence, transmission, and recovery rates while accounting for imperfect host and pathogen detection. We also offer a perspective on future research directions at the intersection of quantitative and disease ecology, including the estimation of false positives in pathogen presence, spatially explicit disease‐structured N ‐mixture models, and the integration of other data types with count data to inform disease dynamics. Managers rely on accurate and precise estimates of disease dynamics to develop strategies to mitigate pathogen impacts on host populations. At a time when pathogens pose one of the greatest threats to biodiversity, statistical methods that lead to robust inferences on host populations are critically needed for rapid, rather than incremental, assessments of the impacts of emerging infectious diseases.

Ecology and Evolution↗
USGS2019

Quantifying trends and uncertainty in prehistoric forest composition

Forest ecosystems in eastern North America were in flux over the last several thousand years, well before Euro-American land clearance and the 20th-century onset of anthropogenic climate change. However, the magnitude and uncertainty of prehistoric vegetation change have been difficult to quantify because of the multiple ecological, dispersal, and sedimentary processes that govern the relationship between forest composition and fossil pollen assemblages. Here we extend STEPPS, a Bayesian hierarchical spatio-temporal pollen-vegetation model, to estimate changes in forest composition in the upper Midwestern United States from about 2000 to 200 years ago. Using this approach, we identify areas of statistically and ecologically significant change. Between 2000 and 200 years ago, forest composition significantly changed across broad regions of north-central Wisconsin and Minnesota. Rates of compositional change varied spatially, and can be linked to previously reported events. The single largest change is the infilling of Tsuga canadensis in northern Wisconsin over the past 2000 years. Despite this range in-filling, the range limit of T. canadensis was largely stable, with modest expansion westward. The regional ecotone between temperate hardwood forests and northern mixed hardwood/conifer forests shifted southwestward by 15-20 km in Minnesota and Northwestern Wisconsin. Fraxinus, Ulmus, and other mesic hardwoods expanded in the Big Woods region of southern Minnesota. However, some areas showed no significant change, suggesting high complexity in the spatiotemporal patterns of past forest dynamics. The increasing density of paleoecological data networks and advances in statistical modeling approaches now enables the confident detection of subtle but significant changes in forest composition over the last 2000 years.

Minnesota, Wisconsin↗
USGS2022

Streambank and floodplain geomorphic change and contribution to watershed material budgets

Stream geomorphic change is highly spatially variable but critical to landform evolution, human infrastructure, habitat, and watershed pollutant transport. However, measurements and process models of streambank erosion and floodplain deposition and resulting sediment fluxes are currently insufficient to predict these rates in all perennial streams over large regions. Here we measured long-term lateral streambank and vertical floodplain change and sediment fluxes using dendrogeomorphology in streams around the U.S. Mid-Atlantic, and then statistically modeled and extrapolated these rates to all 74 133 perennial, nontidal streams in the region using watershed- and reach-scale predictors. Measured long-term rates of streambank erosion and floodplain deposition were highly spatially variable across the landscape from the mountains to the coast. Random Forest regression identified that geomorphic change and resulting fluxes of sediment and nutrients, for both streambank and floodplain, were most influenced by urban and agricultural land use and the drainage area of the upstream watershed. Modeled rates for headwater streams were net erosional whereas downstream reaches were on average net depositional, leading to regional cumulative sediment loads from streambank erosion (−5.1 Tg yr −1 ) being nearly balanced by floodplain deposition (+5.3 Tg yr −1 ). Geomorphic changes in stream valleys had substantial influence on watershed sediment, phosphorus, carbon, and nitrogen budgets in comparison to existing predictions of upland erosion and delivery to streams and of downstream sediment loading. The unprecedented scale of these novel findings provides important insights into the balance of erosion and deposition in streams within disturbed landscapes and the importance of geomorphic change to stream water quality and carbon sequestration, and provides vital understanding for targeting management actions to restore watersheds.

Chesapeake Bay watershed↗
USGS2007

Modeling the spatial distribution of landslide-prone colluvium and shallow groundwater on hillslopes of Seattle, WA

Landslides in partially saturated colluvium on Seattle, WA, hillslopes have resulted in property damage and human casualties. We developed statistical models of colluvium and shallow-groundwater distributions to aid landslide hazard assessments. The models were developed using a geographic information system, digital geologic maps, digital topography, subsurface exploration results, the groundwater flow modeling software VS2DI and regression analyses. Input to the colluvium model includes slope, distance to a hillslope-crest escarpment, and escarpment slope and height. We developed different statistical relations for thickness of colluvium on four landforms. Groundwater model input includes colluvium basal slope and distance from the Fraser aquifer. This distance was used to estimate hydraulic conductivity based on the assumption that addition of finer-grained material from down-section would result in lower conductivity. Colluvial groundwater is perched so we estimated its saturated thickness. We used VS2DI to establish relations between saturated thickness and the hydraulic conductivity and basal slope of the colluvium. We developed different statistical relations for three groundwater flow regimes. All model results were validated using observational data that were excluded from calibration. Eighty percent of colluvium thickness predictions were within 25% of observed values and 88% of saturated thickness predictions were within 20% of observed values. The models are based on conditions common to many areas, so our method can provide accurate results for similar regions; relations in our statistical models require calibration for new regions. Our results suggest that Seattle landslides occur in native deposits and colluvium, ultimately in response to surface-water erosion of hillstope toes. Regional groundwater conditions do not appear to strongly affect the general distribution of Seattle landslides; historical landslides were equally dispersed within and outside of the area potentially affected by regional groundwater conditions.

Earth Surface Processes and Landforms↗
USGS2025

Peak streamflow trends in Montana and northern Wyoming and their relation to changes in climate, water years 1921–2020

Frequency analysis on annual peak streamflow (hereinafter, peak flow) is essential to water-resources management applications, including critical structure design (for example, bridges and culverts) and floodplain mapping. Nonstationarity is a statistical property of a peak-flow series such that the distributional properties (the mean, variance, or skew) change either gradually (monotonic trend) or abruptly (shift, step change or change point) through time. Not incorporating or accounting for observed nonstationarity into peak-flow frequency analysis might result in a poor representation of the true probability of large floods and thus misrepresent the actual flood risks to life and property. This report summarizes how hydroclimatic variability might affect the temporal and spatial distributions of peak-flow data in the State of Montana (and northern Wyoming) and is part of a larger study to document peak-flow nonstationarity and hydroclimatic changes across a nine-State region consisting of Illinois, Iowa, Michigan, Minnesota, Missouri, Montana, North Dakota, South Dakota, and Wisconsin. A wide range of analyses and statistical approaches are applied to document the primary mechanisms controlling floods and characterize temporal changes in hydroclimatic variables and peak flows. This study was completed in cooperation with the Montana Department of Natural Resources and Conservation. The purpose of this report is to characterize temporal and spatial patterns of nonstationarity in peak flows and hydroclimatology in Montana and northern Wyoming. In this evaluation, peak-flow, daily streamflow, and model-simulated gridded climatic data were examined for monotonic trends, change points, and other statistical properties that might indicate changing climatic and environmental conditions. This report includes background information on the study area, the history of U.S. Geological Survey peak-flow data collection and frequency analysis in Montana, and the review of research relating to hydroclimatic variability and change in Montana. This study might help provide a framework for addressing potential nonstationarity issues in peak-flow frequency updates that commonly are completed by the U.S. Geological Survey in cooperation with other agencies throughout the Nation. The analytical structure of this study includes analyses of monotonic trends and change points in numerous hydroclimatic variables in assigned 30-, 50-, 75-, and 100-year analysis periods. For Montana and part of Wyoming, the 30-, 50-, 75, and 100-year analyses included 157, 70, 48, and 12 streamgages, respectively. For those streamgages, nonstationarities were analyzed in the following variables: (1) climatic variables, including annual and seasonal (winter, spring, summer, and fall) temperature and precipitation; (2) daily streamflow variables, including the annual center of volume duration, annual center of volume median, and peaks over threshold with a mean of four events per year; and (3) annual peak-flow variables, including peak-flow timing and magnitude. A likelihood approach was used to express statistical confidence and assign the nonstationarity results as likely upward or downward (highest statistical confidence), somewhat likely upward or downward (less statistical confidence), or about as likely as not (little statistical confidence; hereinafter, neutral). For the nonstationarity analyses of the climatic, daily streamflow, and peak-flow variables, the results are presented in detail and discussed with respect to statewide patterns and geographic variability. For each of the 30-, 50-, and 75-year analyses, peak-flow change-point and monotonic trend analyses were compiled for streamgages classified with likely downward or likely upward trends. For those streamgages, the associated basin characteristics and nonstationarity results for peak-flow timing, daily streamflow, and climatic variables were investigated and statistically compared to discern associations among other variables that might contribute to the peak-flow nonstationarity results. The 50- and 75-year peak-flow nonstationarities identified in this study are mostly downward, in association with mostly upward temperature and potential evapotranspiration:precipitation monotonic trends. For the 50-, 75-, and 100-year analyses, the peak-flow change points are predominantly downward and are concentrated in the 1970s and 1980s, which indicates general consistency among the longer trend periods. These findings are in association with substantial research documenting globally rising temperature and atmospheric greenhouse gas concentrations that might be largely attributed to anthropogenic activities. Anthropogenic effects might represent long-term (on the order of several decades to more than a century) climate changes that might happen within highly variable natural climate fluctuations. Several paleo studies in the north-central United States have indicated that hydroclimatic extremes (that is, low- and high-streamflow conditions) before European settlement have been outside of extremes since the 1900s. Depending on the interactions of anthropogenic effects and natural climate variability, extreme high-streamflow conditions might occur in the future, even in the presence of long-term downward peak-flow trends.

Montana, Wyoming↗
USGS2026

Survey protocols drive heterogeneity in occupancy estimates that cannot be standardized using post hoc analyses

Site occupancy models are widely used to estimate species distributions from presence–nondetection data, yet for mobile animals, the true proportion of sites occupied varies in time. Thus, true occupancy and estimates generated from different protocols are affected by temporal characteristics of survey design. Here we tested whether we could standardize occupancy estimates across different sampling protocols with a post hoc statistical test. Using an individual-based movement model of Wood Thrush ( Hylocichla mustelina ), we simulated point count surveys across 3000 populations under 90 spatially and temporally variable sampling protocols. For each simulation, we fit a standard occupancy model, then used minute-level detections to truncate surveys and generate occupancy estimates across increasing survey lengths. We fit asymptotic models to these series to estimate intercepts (instantaneous occupancy) and asymptotes (longer-term occupancy) and evaluated how well they approximated true instantaneous, daily and seasonal occupancy. We were unable to recover instantaneous or seasonal occupancy from any approach. Both standard occupancy models and asymptotic regression (the top-performing asymptotic model) most closely approximated daily occupancy; although asymptotic regression produced slightly less biased estimates, it showed substantially greater variability, resulting in greater overall error. Our results show that post hoc analytical approaches cannot, at present, reliably standardize occupancy estimates across heterogeneous survey designs. Instead, ensuring biologically informative and comparable estimates will depend on standardized survey protocols. Practical implication : Monitoring programs using presence–nondetection data should prioritize consistent survey protocols when estimating occupancy for mobile species, with a focus on daily occupancy as the most attainable and useful metric. Although the highest accuracy and precision were achieved under the most intensive survey protocols, for point counts of Wood Thrush and similar species, four consecutive 15–20 min surveys provide near-optimal precision and can be completed in a single visit, making them well suited for estimating relative occupancy under logistical constraints. Three 20-min surveys spaced ~24 h apart closely approximate daily occupancy while maintaining reasonable precision. Developing similar protocols for other species and survey methods will be critical to ensuring comparable occupancy estimates across space and time.

Ecological Solutions and Evidence↗
USGS1997

Application of the LI-COR CO 2 analyzer to volcanic plumes: a case study, volcán Popocatépetl, Mexico, June 7 and 10, 1995

Volcanic CO 2 emission rate data are sparse despite their potential importance for constraining the role of magma degassing in the biogeochemical cycle of carbon and for assessing volcanic hazards. We used a LI-COR CO 2 analyzer to determine volcanic CO 2 emission rates by airborne measurements in volcanic plumes at Popocatépetl volcano on June 7 and 10, 1995. LI-COR sample paths of ∼72 m, compared with ∼1 km for the analyzer customarily used, together with fast Fourier transforms to remove instrument noise from raw data greatly improve resolution of volcanic CO 2 anomalies. Parametric models fit to background CO 2 provide a statistical tool for distinguishing volcanic from ambient CO 2 . Global Positioning System referenced flight traverses provide vastly improved data on the shape, coherence, and spatial distribution of volcanic CO 2 in plume cross sections and contrast markedly with previous results based on traverse stacking. The continuous escape of CO 2 and SO 2 from Popocatépetl was fundamentally noneruptive and represented quiescent magma degassing from the top of a magma chamber ∼5 km deep. The average CO 2 emission rate for January-June 1995 is estimated to be at least 6400 t d −1 , one of the highest determined for a quiescently degassing volcano, although correction for downwind dispersion effects on volcanic CO 2 indicates a higher rate of ∼9000 t d −1 . Analysis of random errors indicates emission rates have 95% confidence intervals of ∼±20%, with uncertainty contributed mostly by wind speed variance, although the variance of plume cross-sectional areas during traversing is poorly constrained and possibly significant.

Journal of Geophysical Research B: Solid Earth↗
USGS2026

Variability and consistency in wildfire susceptibility: Insights from a national compilation

Background Wildfire risk in the United States is rising and remains a land management priority. The quantitative wildfire risk assessment (QWRA) framework integrates fuels, topography, weather and values at risk to estimate the potential change in value from wildfire. Within this, response functions (RFs) represent how values respond to fire intensity. These are often based on expert judgment, but variation across assessments is unclear. Aims This study uses data from the US Geological Survey (USGS) Wildfire Hazard and Risk Assessment Clearinghouse to characterize consistency and variation across categories and contexts. Methods We applied descriptive statistics to summarize RFs, using tables, box-and-whisker plots and heat maps stratified by highly valued resource or asset (HVRA) category and spatial scale. Key results RFs and value definitions vary, especially for ecosystem-related resources. Some functions, such as for buildings in the wildland–urban interface (WUI), translate well across contexts, while others require more input. Conclusions Some functions are broadly transferable, while others need customization. This analysis provides references and starting points for improvement to RFs in QWRAs. Interpretations Expanding the clearinghouse and dataset and building more transparency in expert elicitation can build trust among communities, agencies and end-users, and can support efficient use of limited resources to mitigate wildfire risk.

International Journal of Wildland Fire↗
USGS2020

Capturing spatiotemporal patterns in presence-absence data to inform monitoring and sampling designs for the threatened Dakota skipper (Lepidoptera: Hesperiidae) in the Great Plains of the United States

Declines among species of insect pollinators, especially butterflies, has garnered attention from scientists and managers. Often these declines have spurred governments to declare some species as threatened or endangered. We used existing presence–absence data from surveys for the threatened Dakota skipper Hesperia dacotae (Skinner) to build statistical maps of species presence that could be used to inform future monitoring designs. We developed a hierarchical Bayesian modeling approach to estimate the spatial distribution and temporal trend in Dakota skipper probability of presence. Our model included a spatial random effect and fixed effects for the proportion of two grassland habitat types: those on well-drained soils and those on poorly drained soils; as well as the topographic slope. The results from this model were then used to assess sampling strategies with two different monitoring objectives: locating new Dakota skipper colonies or monitoring the proportion of historically (pre-2000) extant colonies. Our modeling results suggested that the distribution of Dakota skippers followed the distribution of remnant grasslands and that probabilities of presence tended to be higher in topographically diverse grasslands with well-drained soils. Our analysis also showed that the probability of presence declined throughout the northern Great Plains range. Our simulations of the different sampling designs suggested that new detections were expected when sampling where Dakota skippers likely occurred historically, but this may lead to a tradeoff with monitoring existing sites. Prior information about the extant sites may help to ameliorate this tradeoff.

Iowa, Minnesota, North Dakota, South Dakota↗
USGS2022

Estimating species misclassification with occupancy dynamics and encounter rates: A semi-supervised, individual-level approach

1. Large-scale, long-term biodiversity monitoring is essential to conservation, land management, and identifying threats to biodiversity. However, multispecies surveys are prone to various types of observation error, including false positive/negative detection, and misclassification, where a species is thought to have been encountered but not correctly identified. Previous methods assume an imperfect classifier produces species-level classifications, but in practice, particularly with human observers, we may end up with extraspecific classifications including `unknown', morphospecies designations, and taxonomic identifications coarser than species. Disregarding these types of species misclassification in biodiversity monitoring datasets can bias estimates of ecologically important quantities such as demographic ratess, occurrence, and species richness. 2. Here we present a joint classification-occupancy model that accounts for species non-detection and misclassification. Our framework accommodates extinction and colonization dynamics, allows for additional uncertain `morphospecies' designations, and makes use of individual specimens with known species identities in a semi-supervised setting. We compare the performance of our model to a classification-only model that discards information about occupancy and encounter rate. We illustrate our model with an empirical case study of the carabid beetle (Carabidae) community at the National Ecological Observatory Network Niwot Ridge Mountain Research Station, near Boulder, CO, USA. We also use simulations to evaluate model performance through validation metrics where varying fractions of the data are confirmed. 3. The model supported imperfect classifier accuracy and favored certain true species classifications strongly for some morphospecies. The model outperformed (e.g., precision) the reduced model that discarded occupancy information, and these differences were most pronounced for abundant species. 4. Spatial and temporal dynamics from modeled occupancy and encounter rates may inform species misclassification probability, but this idea has not yet been tested. Our statistical framework explores this opportunity, and can be applied to datasets with imperfect species detection and classification, limited verification data, and non-species classifications.

Methods in Ecology and Evolution↗
USGS2020

Disentangling the effects of multiple fires on spatially interspersed sagebrush (Artemisia spp.) communities

Questions Relative to a landscape with a mosaic of two sagebrush community types and increasing fire frequency, we asked: (a) do vegetation characteristics vary significantly with number of times burned for each sagebrush community; (b) how do vegetation responses to different fire frequencies compare between the two sagebrush communities? Location Columbia Plateau Ecoregion, Washington, USA. Methods We sampled vegetation across a landscape that burned three times over a 10‐year period in two sagebrush community types that are interspersed on unique land forms: big sagebrush ( Artemisia tridentata ) communities that occur on small “mounds” and scabland sagebrush ( A. rigida ) communities that occur on surrounding “flats.” Spatially overlapping fires permitted a balanced sampling design to assess unburned and once‐, twice‐, and thrice‐burned locations for each land form/community type. We utilized a suite of statistical analyses to determine differences among plant functional groups and biomass among unburned/burned strata by land form and compared results between land forms. Results Big sagebrush and scabland sagebrush communities responded uniquely to multiple fires, due to different fuel loadings, fire severities, succession and invasion dynamics. Big sagebrush experienced nearly complete shrub loss and conversion from exotic‐invaded shrubland to exotic annual grassland after only one fire. In contrast, scabland sagebrush retained a minor shrub component and higher relative cover of native herbaceous species, even after three fires. Both communities retained cover of native perennial grasses, including shallow‐ and deep‐rooted species, likely reflecting decreasing fire intensity with number of times burned. Conclusions Despite different community‐level responses, increasing fire frequency is transforming the entire landscape to a non‐native/native grassland mix. Quantifying unique ecosystem responses to altered wildfire regimes is critical to understanding the relative resilience of communities to disturbance and their resistance to exotic species invasion (and community type conversion). Management actions may help to maintain spatial heterogeneity of ecosystems and fire‐tolerant native species.

Washington↗
USGS1988

Thermochronology of economic mineral deposits: Dating the stages of mineralization at Panasqueira, Portugal, by high-precision 40Ar/ 39Ar age spectrum techniques on muscovite

40 Ar/ 39 Ar age spectrum dates for 13 muscovites have been used to reconstruct the thermal history (thermochronology) of the Panasqueira, Portugal, tin-tungsten deposit, a deposit spatially associated with a belt of Hercynian plutons. Muscovite samples with an age difference as small as 2.2 m.y. (0.7% of the age) are statistically distinct. Statistics are even better for comparison of multiple samples from separate events; that is, a difference of 0.9 m.y. (0.3%) can be resolved in this approximately 300-m.y.-old deposit. The major tin and tungsten ore-forming stages, which are the oxide-silicate stage, the main sulfide stage, and greisenization, occurred between 296.3 + or - 0.8 (1 Sigma ) and 291.6 + or - 0.8 m.y. (1 Sigma ). The first substage of the oxide-silicate stage was a short-lived thermal pulse at 296.3 + or - 0.6 m.y.; the fluids responsible may have emanated from the known granite cupola. The main sulfide stage was active at 294.5 + or - 0.9 m.y. as a slightly longer lived pulse with oldest evidence for this stage (295.8 + or - 0.6 m.y.) coming from areas farthest away from the known cupola and youngest evidence (293.5 + or - 0.8 m.y.) closest to the cupola. A second substage of the oxide-silicate stage occurred as a short-lived thermal pulse at 292.9 + or - 0.7 m.y., synchronous with greisenization of the cupola and alteration of the silica cap at 292.1 + or - 0.4 m.y. The duration of activity of the oxide-silicate stage, the main sulfide stage, greisenization, and alteration of the silica cap based on the ages of all 13 muscovites was greater than 4.2 + or - 0.5 m.y. (1 Sigma ). Minor argon loss from all dated muscovites occurred during later reheating, probably during the longer lived pyrrhotite alteration stage. A single center, the known cupola, had a prolonged role and was the source for main sulfide stage, oxide-silicate stage II, greisenization, and alteration of the silica cap and possibly oxide-silicate stage I and the pyrrhotite alteration stage; however, a separate source for these latter two stages cannot be ruled out.This study is an example of a new and powerful application of 40 Ar/ 39 Ar age spectrum dating of muscovite. Because of the high precision demonstrated in this study, it is now possible to establish time constraints necessary for solving some of the long-standing problems in economic geology. Beyond this, the unique geologic situation of Panasqueira has allowed us to quantify the thermal characteristics of muscovite. Published fluid inclusion data have been used to estimate a muscovite argon closure temperature of approximately 325 degrees C during rapid cooling or short reheating and a temperature of approximately 270 degrees C during slow cooling or extended reheating. Argon-loss patterns displayed by all dated muscovites resulted from reheating after original closure; the mechanism for this argon loss appears to have been argon transport by volume diffusion. Thus, 40 Ar/ 39 Ar age spectrum dating of muscovite can be used to evaluate thermal conditions controlling argon diffusion as well as age, duration, and number of episodes of mineralization.

Economic Geology↗
USGS2013

Salmonids, stream temperatures, and solar loading--modeling the shade provided to the Klamath River by vegetation and geomorphology

The U.S. Geological Survey is studying approaches to characterize the thermal regulation of water and the dynamics of cold water refugia. High temperatures have physiological impacts on anadromous fish species. Factors affecting the presence, variability, and quality of thermal refugia are known, such as riverine and watershed processes, hyporheic flows, deep pools and bathymetric factors, thermal stratification of reservoirs, and other broader climatic considerations. This research develops a conceptual model and methodological techniques to quantify the change in solar insolation load to the Klamath River caused by riparian and floodplain vegetation, the morphology of the river, and the orientation and topographic characteristics of its watersheds. Using multiple scales of input data from digital elevation models and airborne light detection and ranging (LiDAR) derivatives, different analysis methods yielded three different model results. These models are correlated with thermal infrared imagery for ground-truth information at the focal confluence with the Scott River. Results from nonparametric correlation tests, geostatistical cross-covariograms, and cross-correlograms indicate that statistical relationships between the insolation models and the thermal infrared imagery exist and are significant. Furthermore, the use of geostatistics provides insights to the spatial structure of the relationships that would not be apparent otherwise. To incorporate a more complete representation of the temperature dynamics in the river system, other variables including the factors mentioned above, and their influence on solar loading, are discussed. With similar datasets, these methods could be applied to any river in the United States—especially those listed as temperature impaired under Section 303(d) of the Clean Water Act—or international riverine systems. Considering the importance of thermal refugia for aquatic species, these methods can help investigate opportunities for riparian restoration, identify problematic reaches unlikely to provide good habitat, and simulate changes to solar loading estimates from alternative landscape configurations.

California↗
USGS2010

Use of spatial capture-recapture modeling and DNA data to estimate densities of elusive animals

Assessment of abundance, survival, recruitment rates, and density (i.e., population assessment) is especially challenging for elusive species most in need of protection (e.g., rare carnivores). Individual identification methods, such as DNA sampling, provide ways of studying such species efficiently and noninvasively. Additionally, statistical methods that correct for undetected animals and account for locations where animals are captured are available to efficiently estimate density and other demographic parameters. We collected hair samples of European wildcat ( Felis silvestris ) from cheek-rub lure sticks, extracted DNA from the samples, and identified each animals' genotype. To estimate the density of wildcats, we used Bayesian inference in a spatial capture-recapture model. We used WinBUGS to fit a model that accounted for differences in detection probability among individuals and seasons and between two lure arrays. We detected 21 individual wildcats (including possible hybrids) 47 times. Wildcat density was estimated at 0.29/km 2 (SE 0.06), and 95% of the activity of wildcats was estimated to occur within 1.83 km from their home-range center. Lures located systematically were associated with a greater number of detections than lures placed in a cell on the basis of expert opinion. Detection probability of individual cats was greatest in late March. Our model is a generalized linear mixed model; hence, it can be easily extended, for instance, to incorporate trap- and individual-level covariates. We believe that the combined use of noninvasive sampling techniques and spatial capture-recapture models will improve population assessments, especially for rare and elusive animals.

Conservation Biology↗