USGS Science⌕ Search

SEARCH · USGS Science

Results for “Remote Sensing Information”

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 505 records · Page 28Linked to original sources
USGS2013

An approach for characterizing the distribution of shrubland ecosystem components as continuous fields as part of NLCD

Characterizing and quantifying distributions of shrubland ecosystem components is one of the major challenges for monitoring shrubland vegetation cover change across the United States. A new approach has been developed to quantify shrubland components as fractional products within National Land Cover Database (NLCD). This approach uses remote sensing data and regression tree models to estimate the fractional cover of shrubland ecosystem components. The approach consists of three major steps: field data collection, high resolution estimates of shrubland ecosystem components using WorldView-2 imagery, and coarse resolution estimates of these components across larger areas using Landsat imagery. This research seeks to explore this method to quantify shrubland ecosystem components as continuous fields in regions that contain wide-ranging shrubland ecosystems. Fractional cover of four shrubland ecosystem components, including bare ground, herbaceous, litter, and shrub, as well as shrub heights, were delineated in three ecological regions in Arizona, Florida, and Texas. Results show that estimates for most components have relatively small normalized root mean square errors and significant correlations with validation data in both Arizona and Texas. The distribution patterns of shrub height also show relatively high accuracies in these two areas. The fractional cover estimates of shrubland components, except for litter, are not well represented in the Florida site. The research results suggest that this method provides good potential to effectively characterize shrubland ecosystem conditions over perennial shrubland although it is less effective in transitional shrubland. The fractional cover of shrub components as continuous elements could offer valuable information to quantify biomass and help improve thematic land cover classification in arid and semiarid areas.

Arizona, Florida, Texas↗
USGS2009

An integrated approach to benthic habitat mapping using remote sensing and GIS: An example from the Hawaiian Islands

This chapter documents our effort to map benthic habitats within the KalokoHonokohau National Historic Park, Hawai`i, USA. We produce detailed benthichabitat maps by using a combination of color aerial photography, high-resolution bathymetry, and georeferenced underwater video and still photography. We classify individual habitat polygons using five basic attributes and additional information regarding geology, morphology, and coral species. Derivative data sets including isobaths, hillshades, and slope maps are also generated. The mapping shows that benthic habitat and seafloor morphology varies greatly throughout the study area. Nearly 73% of the study area consists of a hardbottom structure that is potentially available for coral habitation; the remaining 27% includes unconsolidated sediment and artificial or historical features. Coral cover is generally low and increases with water depth. The offshore geology is predominantly composed of smooth to undulating pahoehoe-type basalt flows that form flat to gently sloping benches, vertical walls, and steep escarpments. In some locations the basalt surface is irregular and mounded into ridges, pinnacles, and arches. Large rounded basaltic boulders and smaller scattered rocks are common throughout the marine portions of the park. Coral or accreted carbonate reef obscures the underlying volcanic surface in only a few areas. The underlying geologic framework and morphology of the submerged volcanic flows within Kaloko-Honokohau National Historical Park provide the primary control on benthic habitats within the park. The habitat maps and associated data can be used as a stand-alone product or in a GIS to provide useful baseline information to scientists, managers and the general public.

Hawaii↗
USGS2005

Yellowstone grizzly bear investigations: Annual report of the Interagency Grizzly Bear Study Team, 2004

The contents of this Annual Report summarize results of monitoring and research from the 2004 field season. The report also contains a summary of nuisance grizzly bear ( Ursus arctos horribilis ) management actions. The study team continues to work on issues associated with counts of unduplicated females with cubs-of-the-year (COY). These counts are used to establish a minimum population size, which is then used to establish mortality thresholds for the Recovery Plan (U.S. Fish and Wildlife Service [USFWS] 1993). A computer program that defines the rule set used by Knight et al. (1995) to differentiate unique family groups was completed in spring 2005. We will use an improved version of this model to verify the accuracy of the rules using known bears and their telemetry locations in test runs. We hope to have this work complete by the end of 2005. The grizzly bear recovery plan (USFWS 1993) established mortality quotas at 4% of the minimum population estimate derived from female with COY data and no more than 30% of the 4% (1.2%) could be female bears. Simulation modeling (Harris 1984) established sustainable mortality at around 6% of the population. We used the latest information on reproduction and survival to estimate population trajectory in the same simulation model originally used by Harris. A Wildlife Monograph has been accepted for publication and should be available by summer 2005. Our project addressing the potential application of stable isotopes and trace elements to quantify consumption rates of whitebark pine ( Pinus albicaulis ) and cutthroat trout ( Oncorhynchus clarki ) by grizzly bears was completed. Our manuscript on consumption rates of whitebark pine has been published (Canadian Journal of Zoology 81:763-770). The manuscript on fish consumption rates was also accepted and is published in the Canadian Journal of Zoology 82:493-501. Both can be found on the Interagency Grizzly Bear Study Team (IGBST) website http://www.nrmsc.usgs.gov/research/igbst-home.htm. We began a new study in Grand Teton National Park evaluating habitat use both temporally and spatially between grizzly and black ( Ursus americanus ) bears. We will employ a new form of Global Positioning System (GPS) technology that incorporates a spread spectrum communication system. Spread spectrum allows for transfer of stored GPS locations from the collar to a remote receiving station. Results of our first yea r’s field season are summarized in this report. Whitebark pine (WBP) has been identified as one of the import ant fall foods of the Yellowstone grizzly bear. Previous efforts to map the distribution of WBP were for the Cumulative Effects Model. Consequently the only coverage of WBP distribution was for the grizzly bear Recovery Zone. We were successful in getting financial support through the U.S. Geological Survey Land Remote Sensing Program and Interdisciplinary Science Support Activities Project to create an ecosystem-wide map of the distribution of WBP. The results of that project are reported in Appendix A. The study team annually estimates WBP cone production on a series of transects. That information is reported annually in our reports. Concern over the long-term health of WBP prompted us to investigate the usefulness of cone counts as an indirect index of WBP health. Results of this analysis (Appendix B) indicated that cone production is too variable to serve this purpose. Consequently, we partnered with several 2 other agencies and embarked on a program to develop a long-term monitoring program directed specifically at WBP health in the Greater Yellowstone Ecosystem (GYE). Our team (Greater Yellowstone Whitebark Pine Monitoring Working Group) was successful in obtaining funds to develop and implement a WBP health monitoring program. Results of our first years work are presented in Appendix C. We also successfully competed for funds in 2005 and will continue to collect information on WBP health. Army cutworm moths ( Euxoa auxiliaris ) are also a very important food for a segment of the GYE grizzly bear population. Hillary Robison, graduate student at University of Nevada, Reno, is nearing completion of her program. In this report, we post her annual work summary, and abstracts of her most recently submitted publications. These include one on the levels of pesticides in cutworm moths and their potential affect on grizzly bears (Appendix D), a spatial analysis to identify army cutworm moth habitat (Appendix E), and the results of a preliminary analysis of pollen grains on the mouth parts of moths (Appendix F) to help identify which plant species are commonly fed upon. Other study team members have also been working on various aspects of grizzly bear science. Study team member Kerry Gunther hosted a workshop on habituated grizzly bears in North America. A copy of the abstract of that report can be found in Appendix G. Additionally, Kerry Gunther and Doug Smith, wolf researcher in Yellowstone National Park (YNP), reported on the interactions between gray wolves ( Canis lupus ) and female grizzly bears with young. They report that of 15 interactions between these 2 carnivores, 8 involve d females with COY. They observed 2 incidents where cubs were killed by wolves at ungulate carcasses (Appendix H). The annual reports of the IGBST summarize annual data collection. Because additional information can be obtained after publication, data summaries are subject to change. For that reason, data analyses and summaries presented in this report supersede all previously published data. The study area and sampling techniques are reported by Blanchard (1985), Mattson et al. (1991 a ), and Haroldson et al. (1998).

Idaho, Montana, Wyoming↗
USGS2006

An analysis of urban thermal characteristics and associated land cover in Tampa Bay and Las Vegas using Landsat satellite data

Remote sensing data from both Landsat 5 and Landsat 7 systems were utilized to assess urban area thermal characteristics in Tampa Bay watershed of west-central Florida, and the Las Vegas valley of southern Nevada. To quantitatively determine urban land use extents and development densities, sub-pixel impervious surface areas were mapped for both areas. The urban–rural boundaries and urban development densities were defined by selecting certain imperviousness threshold values and Landsat thermal bands were used to investigate urban surface thermal patterns. Analysis results suggest that urban surface thermal characteristics and patterns can be identified through qualitatively based urban land use and development density data. Results show the urban area of the Tampa Bay watershed has a daytime heating effect (heat-source), whereas the urban surface in Las Vegas has a daytime cooling effect (heat-sink). These thermal effects strongly correlated with urban development densities where higher percent imperviousness is usually associated with higher surface temperature. Using vegetation canopy coverage information, the spatial and temporal distributions of urban impervious surface and associated thermal characteristics are demonstrated to be very useful sources in quantifying urban land use, development intensity, and urban thermal patterns.

Remote Sensing of Environment↗
USGS2020

Oases of the future? Evaluating springs as potential hydrologic refugia in drying climates

Springs in water-limited landscapes are biodiversity hotspots and keystone ecosystems, disproportionately influencing surrounding landscapes despite their often small areas. Some springs served as evolutionary refugia during previous climate drying, supporting relict species in isolated habitats. Understanding whether springs will provide hydrologic refugia from future climate change is important to biodiversity conservation but complicated by hydrologic variability among springs, data limitations, and multiple non-climate threats to groundwater-dependent ecosystems. Here, we present a conceptual framework for categorizing springs as potentially stable, relative, or transient hydrologic refugia in a drying climate. Clues about refugial capacity of springs can be assembled from diverse approaches, including citizen-science-powered ecohydrologic monitoring, remote sensing, landowner interviews, and environmental tracer analysis. Managers can integrate multiple lines of evidence to predict which springs may become future refugia for species of concern, strengthening the long-term effectiveness of springs conservation and restoration and informing climate adaptation for terrestrial and freshwater species.

Frontiers in Ecology and the Environment↗
USGS2006

Predicting tree species presence and basal area in Utah: A comparison of stochastic gradient boosting, generalized additive models, and tree-based methods

Many efforts are underway to produce broad-scale forest attribute maps by modelling forest class and structure variables collected in forest inventories as functions of satellite-based and biophysical information. Typically, variants of classification and regression trees implemented in Rulequest's?? See5 and Cubist (for binary and continuous responses, respectively) are the tools of choice in many of these applications. These tools are widely used in large remote sensing applications, but are not easily interpretable, do not have ties with survey estimation methods, and use proprietary unpublished algorithms. Consequently, three alternative modelling techniques were compared for mapping presence and basal area of 13 species located in the mountain ranges of Utah, USA. The modelling techniques compared included the widely used See5/Cubist, generalized additive models (GAMs), and stochastic gradient boosting (SGB). Model performance was evaluated using independent test data sets. Evaluation criteria for mapping species presence included specificity, sensitivity, Kappa, and area under the curve (AUC). Evaluation criteria for the continuous basal area variables included correlation and relative mean squared error. For predicting species presence (setting thresholds to maximize Kappa), SGB had higher values for the majority of the species for specificity and Kappa, while GAMs had higher values for the majority of the species for sensitivity. In evaluating resultant AUC values, GAM and/or SGB models had significantly better results than the See5 models where significant differences could be detected between models. For nine out of 13 species, basal area prediction results for all modelling techniques were poor (correlations less than 0.5 and relative mean squared errors greater than 0.8), but SGB provided the most stable predictions in these instances. SGB and Cubist performed equally well for modelling basal area for three species with moderate prediction success, while all three modelling tools produced comparably good predictions (correlation of 0.68 and relative mean squared error of 0.56) for one species. ?? 2006 Elsevier B.V. All rights reserved.

Ecological Modelling↗
USGS1993

Surficial geology of the Safsaf region, south-central Egypt, derived from remote-sensing and field data

We used a decorrelation-stretched image of Landsat Thematic Mapper (TM) Bands 1, 4, and 7 and field data to map and describe the main surficial units in the hyperarid Safsaf region in south-central Egypt. We show that the near-infrared bands on Landsat TM, which are sensitive to very subtle changes in mineralogy common to arid regions, significantly improve the geologist's capability to discriminate geologic units in desert regions. These data also provide the spatial and spectral information necessary to determine the migration patterns and provenance of eolian materials. The Safsaf area was the focus of our post flight field studies using Shuttle Imaging Radar (SIR) data following the discovery of buried paleochannels in North Africa. Most of the channels discernible on SIR images are not expressed in TM data, but traces of a few channels are present in both the SIR and the TM data within the Wadi Safsaf area. Here we present a detailed digital examination of the SIR and the TM-band reflectance and reflectance-ratio data at three locations of the more obvious surface expressions of the buried channels. Our results indicate that the TM expressions of the channels are not purely topographic but are more compositional in nature. Two possibilities may account for the TM expressions of the buried channels: 1) concentrations of windblown, iron-rich materials that accumulated along subtle curvilinear topograpohic traps, or 2) curvilinear exposures of an iron-rich underlying unit of the flat sand sheet.

Safsaf region↗
USGS2024

Quantifying the relative importance of biotic and abiotic factors in landscape-based models of stream fish distributions

Lotic fish species distributions are frequently predicted using remotely sensed habitat variables that characterize the adjacent landscape and serve as proxies for instream habitat. Recent advancements in statistical methodology, however, allow for leveraging fish assemblage data when predicting distributions. This is important because assemblage composition likely provides better information about instream habitat compared to landscape-derived metrics and therefore may improve predictions. To better understand the value of using multi-species fish data in species distribution modeling, we fit two conditional random fields (CRF) models to quantify the relative importance of fish assemblage co-occurrence, landscape-derived habitat variables, and interactions between these two predictor groups (i.e., effects of co-occurrence could be context-dependent) at over 1200 stream catchments in Pennsylvania, USA. We first compared predictive performance of CRF models against traditionally used single-species logistic regressions (generalized linear models; GLMs) and found that inclusion of fish assemblage data often improved predictive performance. The multi-species CRF models performed significantly better at predicting occurrence for 63% of species with an average percent increase in AUC of 25% compared to GLMs. Furthermore, the CRF identified species co-occurrences as more informative, and thus relatively more important, at predicting occurrence than the other effect types. The CRF also suggested that allowing these biotic effects to be context-dependent was important for predicting occurrence of many species. These findings illustrate the value of fish assemblage data for landscape-scale species distribution modeling and leveraging this information can improve predictions and inferences to help inform the management and conservation of freshwater fishes.

Pennsylvania↗
USGS1975

Thermal-infrared spectra and chemical analyses of twenty-six igneous rock samples

Emittance spectra in the 7.5 μm to 14 μm wavelength region and chemical compositions of 26 igneous rocks are reported. Experimental measurements on the rocks were made under simulated daytime field conditions. Some surface silicate contaminants, such as clayey silt, significantly altered the spectral emittance of a fresh sample, whereas, for these samples, hydrous and anhydrous ferric oxide weathering products did not mask important silicate spectral information. In the 11.75 μm to 13.75 μm wavelength region, the mean emittance of all the silicate samples was0.956 ± 0.008, except for periodtite, which had an average emittance of 0.895. This region of uniform emittance should be useful in remote sensing experiments for the separation of the effects of temperature and chemical composition on the spectral emittance of silicate rocks.

Remote Sensing of Environment↗
USGS2013

Summary of the Third International Planetary Dunes Workshop: remote sensing and image analysis of planetary dunes

The Third International Planetary Dunes Workshop took place in Flagstaff, AZ, USA during June 12–15, 2012. This meeting brought together a diverse group of researchers to discuss recent advances in terrestrial and planetary research on aeolian bedforms. The workshop included two and a half days of oral and poster presentations, as well as one formal (and one informal) full-day field trip. Similar to its predecessors, the presented work provided new insight on the morphology, dynamics, composition, and origin of aeolian bedforms on Venus, Earth, Mars, and Titan, with some intriguing speculation about potential aeolian processes on Triton (a satellite of Neptune) and Pluto. Major advancements since the previous International Planetary Dunes Workshop include the introduction of several new data analysis and numerical tools and utilization of low-cost field instruments (most notably the time-lapse camera). Most presentations represented advancement towards research priorities identified in both of the prior two workshops, although some previously recommended research approaches were not discussed. In addition, this workshop provided a forum for participants to discuss the uncertain future of the Planetary Aeolian Laboratory; subsequent actions taken as a result of the decisions made during the workshop may lead to an expansion of funding opportunities to use the facilities, as well as other improvements. The interactions during this workshop contributed to the success of the Third International Planetary Dunes Workshop, further developing our understanding of aeolian processes on the aeolian worlds of the Solar System.

Aeolian Research↗
USGS2006

Remote sensing and geophysical investigations of glacial buried valleys in northeastern Kansas

Aquifers found in glacial buried valleys are a major source of good-quality ground water in northeastern Kansas. The extent and character of many of these deposits are not precisely known, so a detailed study of the buried valleys was undertaken. Test drilling, Landsat imagery, shallow-earth temperature measurements, seismic refraction, surface electrical resistivity, and gravity data were used to evaluate two sites in Nemaha and Jefferson Counties. Tonal patterns on springtime Landsat imagery and winter/summer anomalies in shallow-earth temperatures were quick and inexpensive methods for locating some glacial buried aquifers and suggested areas for more intensive field studies. Reversed seismic refraction and resistivity surveys were generally reliable indicators of the presence or absence of glacial buried valleys, with most depth determinations being within 25% of test-drilling results. The effectiveness of expensive test-hole drilling was greatly increased by integrating remote sensing, shallow-earth temperature, seismic, and resistivity techniques in the two buried valley test areas. A gravity profile allowed precise definition of the extent of one of the channels after the other techniques had been used for general information.

Kansas↗
USGS1997

Use of remote-sensing techniques to survey the physical habitat of large rivers

Remote-sensing techniques that can be used to quantitatively characterize the physical habitat in large rivers in the United States where traditional survey approaches typically used in small- and medium-sized streams and rivers would be ineffective or impossible to apply. The state-of-the-art remote-sensing technologies that we discuss here include side-scan sonar, RoxAnn, acoustic Doppler current profiler, remotely operated vehicles and camera systems, global positioning systems, and laser level survey systems. The use of these technologies will permit the collection of information needed to create computer visualizations and hard copy maps and generate quantitative databases that can be used in real-time mode in the field to characterize the physical habitat at a study location of interest and to guide the distribution of sampling effort needed to address other habitat-related study objectives. This report augments habitat sampling and characterization guidance provided by Meador et al. (1993) and is intended for use primarily by U.S. Geological Survey National Water Quality Assessment program managers and scientists who are documenting water quality in streams and rivers of the United States.

Report↗
USGS2007

Hurricane Katrina flooding and oil slicks mapped with satellite imagery

A multiple-database approach that combined remotely sensed data from Radarsat-1 and Landsat Thematic Mapper Plus (ETM+) imagery was used to map Hurricane Katrinainduced flooding and to identify offshore oil slicks. Maps depicting the areal extent of flooding, oil slicks, and floating debris provide vital information to emergency managers for directing floodrelief efforts and the clean-up of polluted waters.

Gulf Of Mexico↗
USGS2002

USGS environmental studies of the World Trade Center area, New York City, after September 11, 2001

Two days after the September 11, 2001, attack on World Trade Center (WTC), the U.S. Geological Survey (USGS) was asked by the U.S. Environmental Protection Agency (EPA) and the U.S. Public Health Service to conduct a remote sensing and mineralogical characterization study of lower Manhattan around the WTC. This study, conducted in cooperation with the National Aeronautics and Space Administration (NASA) and the Jet Propulsion Laboratory (JPL), was requested to rapidly provide emergency response teams with information on the concentrations and distribution of asbestos and other materials in the dusts deposited around lower Manhattan after the September 11 WTC building collapse in New York City. Preliminary results of the study were released via the internet to emergency response teams on September 18 and September 27, 2001. After September 27, additional work was done to fill remaining data gaps, and the study report underwent further detailed peer review. The report was released to the general public via the internet on November 27, 2001. This fact sheet summarizes the results of the interdisciplinary study; the full report can be viewed at http://geology.cr.usgs.gov/pub/open-file-reports/ofr-01-0429/ .

New York↗
USGS2024

Visual interpretation of high-resolution aerial imagery: A tool for land managers

Remotely sensed imagery from various collection platforms (e.g., satellites, crewed and uncrewed aircraft) are used by biologists and other conservation personnel to support management activities ranging from monitoring invasive species to assessing land cover and vegetation characteristics. Although remote sensing–based vegetation indices and models have been developed and used for some management applications, straightforward visual interpretation of imagery by on-the-ground personnel may be a pragmatic approach for obtaining time-sensitive and spatially relevant information to support and guide local management activities. Our primary objective was to qualitatively assess our ability to identify patches of target invasive plant species based on simple visual interpretation of high-resolution aerial imagery. We also sought to compare the high-resolution imagery to widely available imagery (e.g., National Agriculture Imagery Program) to determine the efficacy of each for assessing vegetation communities and land-cover features in support of management activities. To accomplish these objectives, we obtained high-resolution imagery and visually scanned and assessed the imagery by using standard geographic information system software. We were able to differentiate patches of crownvetch Securigera varia (L.) Lassen and wild parsnip Pastinaca sativa L., but not spotted knapweed Centaurea stoebe L. or leafy spurge Euphorbia esula L. The relative success in identifying these species had a relationship to plant characteristics (e.g., flower color and morphology, height), time of year (phenology), patch size and density, and potentially site characteristics such density of the underlying vegetation (e.g., grasses), substrate color characteristics (i.e., color contrast with flowers), and physical disturbance. Our straightforward, qualitative assessment suggests that visual interpretation of high-resolution imagery, but not some lower-resolution imagery, may be an efficient and effective tool for supporting local invasive species management through activities such as monitoring known patches, identifying undetected infestations, assessing management actions, guiding field work, or prioritizing on-the-ground monitoring activities.

Journal of Fish and Wildlife Management↗
USGS1999

The way forward

This paper focuses on the lessons hearned in the conduct of the lnternational Geosphere Biosphere Program's Data and Information System (rcnr-nts), global 1-km Land-Cover Mapping Project (n$cover). There is stiLL considerable fundamental research to be conducted dealing with the development and validation of thematic geospatial products derived from a combination of remotely sensed and ancillary data. Issues include database and data product development, classification legend definitions, processing and analysis techniques, and sampling strategies. A significant infrastructure is required to support an effort such as DISCover. The infrastructure put in place under the auspices of the IGBP-DIS serves as a model, and must be put in place to enable replication and development of projects such as Discover.

Photogrammetric Engineering and Remote Sensing↗
USGS2002

A flood early warning system for southern Africa

Sizeable areas of the Southern African Region experienced widespread flooding in 2000. Deployment of hydrologic models can help reduce the human and economic losses in the regions by providing improved monitoring and forecast information to guide relief activities. In this study, we describe a hydrologic model developed for wide-area flood risk monitoring for the Southern African region. The model is forced by daily estimates of rainfall and evapotranspiration derived from remotely sensed data and assimilation fields. Model predictive skills were verified with data observed stream flow data from locations within the Limpopo basin. The model performed well in simulating the timing and magnitude of the stream flow during a recent episode of flooding in Mozambique in 2000.

Conference Paper↗
USGS2007

Improved wetland remote sensing in Yellowstone National Park using classification trees to combine TM imagery and ancillary environmental data

The U.S. Fish and Wildlife Service uses the term palustrine wetland to describe vegetated wetlands traditionally identified as marsh, bog, fen, swamp, or wet meadow. Landsat TM imagery was combined with image texture and ancillary environmental data to model probabilities of palustrine wetland occurrence in Yellowstone National Park using classification trees. Model training and test locations were identified from National Wetlands Inventory maps, and classification trees were built for seven years spanning a range of annual precipitation. At a coarse level, palustrine wetland was separated from upland. At a finer level, five palustrine wetland types were discriminated: aquatic bed (PAB), emergent (PEM), forested (PFO), scrub–shrub (PSS), and unconsolidated shore (PUS). TM-derived variables alone were relatively accurate at separating wetland from upland, but model error rates dropped incrementally as image texture, DEM-derived terrain variables, and other ancillary GIS layers were added. For classification trees making use of all available predictors, average overall test error rates were 7.8% for palustrine wetland/upland models and 17.0% for palustrine wetland type models, with consistent accuracies across years. However, models were prone to wetland over-prediction. While the predominant PEM class was classified with omission and commission error rates less than 14%, we had difficulty identifying the PAB and PSS classes. Ancillary vegetation information greatly improved PSS classification and moderately improved PFO discrimination. Association with geothermal areas distinguished PUS wetlands. Wetland over-prediction was exacerbated by class imbalance in likely combination with spatial and spectral limitations of the TM sensor. Wetland probability surfaces may be more informative than hard classification, and appear to respond to climate-driven wetland variability. The developed method is portable, relatively easy to implement, and should be applicable in other settings and over larger extents.

Remote Sensing of Environment↗