USGS Science⌕ Search

SEARCH · USGS Science

Results for “Vegetation Classification and Survey”

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 55 records · Page 3Linked to original sources
USGS2024

Learning from a high-severity fire event—Conditions following the 2018 Carr Fire at Whiskeytown National Recreation Area

The 2018 Carr Fire burned more than 90 percent of Whiskeytown National Recreation Area, with much of the park burning at high severity. California yellow pine and mixed conifer forests are not well adapted to large, high-severity fires, and forest recovery after these events may be problematic. Large, high-severity fire patches pose difficulties for recruitment with interiors that are long distances from potential seed trees and may develop fuel structures that can promote further high-severity fire. This report details patterns of forest structure derived from field plots measured 2–3 years after the Carr Fire, providing a characterization of immediate fire effects. We coupled these observations with remotely sensed information, including data collected from unoccupied aircraft system surveys. The remotely sensed data were used to depict erosion after the Carr Fire as well as to create a high-resolution land cover classification map, a debris flow risk map and hazard assessment, and a post-fire canopy vegetation loss map. Results indicated high levels of tree mortality after the Carr Fire, including high-value old growth forest stands, supporting remotely sensed assessments of fire severity. The high-resolution tree mortality model also aligned well with other remotely sensed estimates of immediate burn severity. Results of the land cover classification illustrated the high percentage of dead vegetation remaining in the understory and canopy 8 months post-fire. Changes in vegetation height identified areas with canopy vegetation loss from 1- to 8-months post-fire. Pairing the post-fire debris accumulation with debris flow probabilities may identify high-risk debris flow areas. The results of this study will help inform future decisions concerning wildland fire and vegetation management strategies at Whiskeytown National Recreation Area and are broadly relevant for management in the aftermath of large, high-severity fires in mixed, dry coniferous forests in the western United States.

California↗
USGS1998

Large-area mapping of biodiversity

The age of discovery, description, and classification of biodiversity is entering a new phase. In responding to the conservation imperative, we can now supplement the essential work of systematics with spatially explicit information on species and assemblages of species. This is possible because of recent conceptual, technical, and organizational progress in generating synoptic views of the earth's surface and a great deal of its biological content, at multiple scales of thematic as well as geographic resolution. The development of extensive spatial data on species distributions and vegetation types provides us with a framework for: (a) assessing what we know and where we know it at meso-scales, and (b) stratifying the biological universe so that higher-resolution surveys can be more efficiently implemented, coveting, for example, geographic adequacy of specimen collections, population abundance, reproductive success, and genetic dynamics. The land areas involved are very large, and the questions, such as resolution, scale, classification, and accuracy, are complex. In this paper, we provide examples from the United States Gap Analysis Program on the advantages and limitations of mapping the occurrence of terrestrial vertebrate species and dominant land-cover types over large areas as joint ventures and in multi-organizational partnerships, and how these cooperative efforts can be designed to implement results from data development and analyses as on-the-ground actions. Clearly, new frameworks for thinking about biogeographic information as well as organizational cooperation are needed if we are to have any hope of documenting the full range of species occurrences and ecological processes in ways meaningful to their management. The Gap Analysis experience provides one model for achieving these new frameworks.

Conference Paper↗
USGS2017

Natural resource inventory and monitoring for Ulaan Taiga Specially Protected Areas—An assessment of needs and opportunities in northern Mongolia

Between 1997 and 2011, Mongolia established three specially protected areas in the north-central part of the country to protect various high-value resources. These areas are jointly referred to as the Ulaan Taiga Specially Protected Areas. In accordance with the goals of the draft general management plan, this report identifies options for initiating an inventory and monitoring program for the three protected areas. Together, the three areas comprise over 1.5 million hectares of mountainous terrain west of Lake Hovsgol and bordering the Darkhad Valley. The area supports numerous rare ungulates, endangered fish, and over 40 species of threatened plants. Illegal mining, illegal logging, and poaching pose the most immediate threats to resources. As a first step, a review of published literature would inform natural resource management at the Ulaan Taiga Specially Protected Areas because it would inform other inventories. Vegetation classification and mapping also would inform other inventory efforts because the process incorporates geographic analysis to identify environmental gradients, fine-scale sampling that captures species composition and structure, and landscape-scale results that represent the variety and extent of habitats for various organisms. Mapping using satellite imagery reduces the cost per hectare. Following a determination of existing knowledge, field surveys of vertebrates and vascular plants would serve to build species lists and fill in gaps in existing knowledge. For abiotic resources, a focus on monitoring air quality, evaluating and monitoring water quality, and assembling and storing weather data would provide information for correlating resource response status with changing environmental conditions. Finally, we identify datasets that, if incorporated into a geographic information system, would inform resource management. They include political boundaries, infrastructure, topography, surficial geology, hydrology, fire history, and soils. In terms of tracking high-value resources, vegetation monitoring at the plot scale would provide a basis for detecting change in such characteristics as plant species composition, vegetation structure, and productivity that are associated with landscape-scale factors such as climate change or biotic interactions. Continued population monitoring of rare ungulates, particularly argali or wild sheep ( Ovis ammon ), would provide information on how populations are responding to natural and anthropogenic stressors. Siberian taimen ( Hucho taimen ) also is an important monitoring target given ongoing threats of poaching and climate change.

Ulaan Taiga Specially Protected Areas↗
USGS2016

Estimated historical distribution of grassland communities of the Southern Great Plains

The purpose of this project was to map the estimated distribution of grassland communities of the Southern Great Plains prior to Euro-American settlement. The Southern Great Plains Rapid Ecoregional Assessment (REA), under the direction of the Bureau of Land Management and the Great Plains Landscape Conservation Cooperative, includes four ecoregions: the High Plains, Central Great Plains, Southwestern Tablelands, and the Nebraska Sand Hills. The REA advisors and stakeholders determined that the mapping accuracy of available national land-cover maps was insufficient in many areas to adequately address management questions for the REA. Based on the recommendation of the REA stakeholders, we estimated the potential historical distribution of 10 grassland communities within the Southern Great Plains project area using data on soils, climate, and vegetation from the Natural Resources Conservation Service (NRCS) including the Soil Survey Geographic Database (SSURGO) and Ecological Site Information System (ESIS). The dominant grassland communities of the Southern Great Plains addressed as conservation elements for the REA area are shortgrass, mixed-grass, and sand prairies. We also mapped tall-grass, mid-grass, northwest mixed-grass, and cool season bunchgrass prairies, saline and foothill grasslands, and semi-desert grassland and steppe. Grassland communities were primarily defined using the annual productivity of dominant species in the ESIS data. The historical grassland community classification was linked to the SSURGO data using vegetation types associated with the predominant component of mapped soil units as defined in the ESIS data. We augmented NRCS data with Landscape Fire and Resource Management Planning Tools (LANDFIRE) Biophysical Settings classifications 1) where NRCS data were unavailable and 2) where fifth-level watersheds intersected the boundary of the High Plains ecoregion in Wyoming. Spatial data representing the estimated historical distribution of grassland communities of the Southern Great Plains are provided as a 30 x 30-meter gridded surface (raster dataset). This information will help to address the priority management questions for grassland communities for the Southern Great Plains REA and can be used to inform other regional-level land management decisions.

Southern Great Plains↗
USGS2020

Final report: Southwestern Willow Flycatcher (Empidonax traillii extimus) and Western Yellow-billed Cuckoo (Coccyzus americanus occidentalis) surveys and habitat availability modeling on the Santa Clara River, California, 26 March 2020

Our project aimed to conduct population surveys for Southwestern Willow Flycatcher (Empidonax trailii extimus; SWFL) and Yellow-billed Cuckoo (Coccyzus americanus; YBCU) in 2018 and 2019, apply existing habitat models to illustrate and predict past, current, and future habitat suitabilities for these two species, and update and standardize classification and mapping of riparian vegetation to reflect recent conditions along the lower 50 miles of the Santa Clara River. Models developed by Hatten and Paradzick (2003), Hatten, et al. (2010), Hatten (2016), and Johnson et al. (2016) were applied to the SCR to provide California Department of Fish and Wildlife (CDFW) and the U.S. Fish and Wildlife Service (USFWS) with useful tools for management of resources for SWFL and YBCU.

California↗
USGS2025

Land cover change within wetland complexes at Dixie Meadows, Churchill County, Nevada: 2015 – 2023

Dixie Meadows, Nevada, is a system of geothermal springs and seeps that feed a complex of marshes and wetland meadows that are located within lands managed by the Bureau of Land Management (BLM) and the Department of Defense (DOD). A previous U.S. Geological Survey report documented variability in satellite imagery-based land cover classifications for seven wetland complexes at near monthly time intervals between October 2015 and January 2022. This report presents additional data, extending analysis to November 2023. Land cover classifications between October 2015 and November 2023 demonstrated an association between vegetation cover characteristics and surface moisture, with Class 1 having dry, bare soil or sparse upland vegetation, Class 2 having moist, bare soil or sparse to small vegetation, Class 3 having dense green vegetation with potentially saturated soil conditions, Class 4 having a mix of shallow surface water, saturated soil, and dense green vegetation, and Class 5 having open surface water. Most of the wetland complexes occur close to spring outflows primarily within land managed by the DOD, though portions are also within BLM lands. The intervening and surrounding landscape outside of the wetland complexes assessed in this study are managed by the BLM. As a result, Class 1 land covers had the largest areal coverage for BLM managed lands. Classes 2 and 3 land covers were primarily mapped inside the wetland complexes and thus had the largest area coverage within DOD managed lands. Class 4 was almost exclusively mapped within the wetland complexes and thus was largely contained within DOD managed lands. Class 5 (open water) was exclusively mapped in and adjacent to a single wetland complex with catchment ponds on land managed by the BLM. The distribution of these land cover classes over the study period was seasonally and annually variable. Land cover areas of Classes 1 and 2 were larger during the spring months. Conversely, land cover areas of Classes 3 and 4 tended to be greatest during the summer or fall. These patterns might be influenced by differences in seasonal water sources and phenology.

Nevada↗
USGS2019

Invasive buffelgrass detection using high-resolution satellite and UAV imagery on Google Earth Engine

Methods to detect and monitor the spread of invasive grasses are critical to avoid ecosystem transformations and large economic costs. The rapid spread of non‐native buffelgrass (Pennisetum ciliare) has intensified fire risk and is replacing fire intolerant native vegetation in the Sonoran Desert of the southwestern US. Coarse‐resolution satellite imagery has had limited success in detecting small patches of buffelgrass, whereas ground‐based and aerial survey methods are often cost prohibitive. To improve detection, we trained 2 m resolution DigitalGlobe WorldView‐2 satellite imagery with 12 cm resolution unmanned aerial vehicle (UAV) imagery and classified buffelgrass on Google Earth Engine, a cloud computing platform, using Random Forest (RF) models in Saguaro National Park, Arizona, USA. Our classification models had an average overall accuracy of 93% and producer's accuracies of 94–96% for buffelgrass, although user's accuracies were low. We detected a 2.92 km 2 area of buffelgrass in the eastern Rincon Mountain District (1.07% of the total area) and a 0.46 km 2 area (0.46% of the total area) in the western Tucson Mountain District of Saguaro National Park. Buffelgrass cover was significantly greater in the Sonoran Paloverde‐Mixed Cacti Desert Scrub vegetation type, on poorly developed Entisols and Inceptisol soils and on south‐facing topographic aspects compared to other areas. Our results demonstrate that high‐resolution imagery improve on previous attempts to detect and classify buffelgrass and indicate potential areas where the invasive grass might spread. The methods demonstrated in this study could be employed by land managers as a low‐cost strategy to identify priority areas for control efforts and continued monitoring.

Arizona↗
USGS2010

Terrestrial Ecosystems of the Conterminous United States

The U.S. Geological Survey (USGS), with support from NatureServe, has modeled the potential distribution of 419 terrestrial ecosystems for the conterminous United States using a comprehensive biophysical stratification approach that identifies distinct biophysical environments and associates them with known vegetation distributions (Sayre and others, 2009). This standardized ecosystem mapping effort used an ecosystems classification developed by NatureServe (Comer and others, 2003). The ecosystem mapping methodology was developed for South America (Sayre and others, 2008) and is now being implemented globally (Sayre and others, 2007). The biophysical stratification approach is based on mapping the major structural components of ecosystems (land surface forms, topographic moisture potential, surficial lithology, isobioclimates and biogeographic regions) and then spatially combining them to produce a set of unique biophysical environments. These physically distinct areas are considered as the fundamental structural units ('building blocks') of ecosystems, and are subsequently aggregated and labeled using the NatureServe classification. The structural footprints were developed from the geospatial union of several base layers including biogeographic regions, isobioclimates (Cress and others, 2009a), land surface forms (Cress and others, 2009b), topographic moisture potential (Cress and others, 2009c), and surficial lithology (Cress and others, in press). Among the 49,168 unique structural footprint classes that resulted from the union, 13,482 classes met a minimum pixel count threshold (20,000 pixels) and were aggregated into 419 NatureServe ecosystems using a semiautomated labeling process based on rule-set formulations for attribution of each ecosystem. The resulting ecosystems are those that are expected to occur based on the combination of the bioclimate, biogeography, and geomorphology. Where land use by humans has not altered land cover, natural vegetation assemblages are expected to occur, and these are described in the ecosystems classification. The map does not show the distribution of urban and agricultural areas - these will be masked out in subsequent analyses to depict the current land cover in addition to the potential distribution of natural ecosystems. This map depicts the smoothed and generalized image of the terrestrial ecosystems dataset. Additional information about this map and any data developed for the ecosystems modeling of the conterminous United States is available online at: http://rmgsc.cr.usgs.gov/ecosystems/.

Scientific Investigations Map↗
USGS2010

Development of a global land cover characteristics database and IGBP DISCover from 1 km AVHRR data

Researchers from the U.S. Geological Survey, University of Nebraska-Lincoln and the European Commission's Joint Research Centre, Ispra, Italy produced a 1 km resolution global land cover characteristics database for use in a wide range of continental-to global-scale environmental studies. This database provides a unique view of the broad patterns of the biogeographical and ecoclimatic diversity of the global land surface, and presents a detailed interpretation of the extent of human development. The project was carried out as an International Geosphere-Biosphere Programme, Data and Information Systems (IGBP-DIS) initiative. The IGBP DISCover global land cover product is an integral component of the global land cover database. DISCover includes 17 general land cover classes defined to meet the needs of IGBP core science projects. A formal accuracy assessment of the DISCover data layer will be completed in 1998. The 1 km global land cover database was developed through a continent-by-continent unsupervised classification of 1 km monthly Advanced Very High Resolution Radiometer (AVHRR) Normalized Difference Vegetation Index (NDVI) composites covering 1992-1993. Extensive post-classification stratification was necessary to resolve spectral/temporal confusion between disparate land cover types. The complete global database consists of 961 seasonal land cover regions that capture patterns of land cover, seasonality and relative primary productivity. The seasonal land cover regions were aggregated to produce seven separate land cover data sets used for global environmental modelling and assessment. The data sets include IGBP DISCover, U.S. Geological Survey Anderson System, Simple Biosphere Model, Simple Biosphere Model 2, Biosphere-Atmosphere Transfer Scheme, Olson Ecosystems and Running Global Remote Sensing Land Cover. The database also includes all digital sources that were used in the classification. The complete database can be sourced from the website: http://edcwww.cr.usgs.gov/landdaac/glcc/glcc.html.

International Journal of Remote Sensing↗
USGS2008

GIS interpolations of witness tree records (1839-1866) for northern Wisconsin at multiple scales

To construct forest landscape of pre-European settlement periods, we developed a GIS interpolation approach to convert witness tree records of the U.S. General Land Office (GLO) survey from point to polygon data, which better described continuously distributed vegetation. The witness tree records (1839-1866) were processed for a 3-million ha landscape in northern Wisconsin, U.S.A. at different scales. We provided implications of processing results at each scale. Compared with traditional GLO mapping that has fixed mapping scales and generalized classifications, our approach allows presettlement forest landscapes to be analysed at the individual species level and reconstructed under various classifications. We calculated vegetation indices including relative density, dominance, and importance value for each species, and quantitatively described the possible outcomes when GLO records are analysed at three different scales (resolution). The 1 x 1-section resolution preserved spatial information but derived the most conservative estimates of species distributions measured in percentage area, which increased at coarser resolutions. Such increases under the 2 x 2-section resolution were in the order of three to four times for the least common species, two to three times for the medium to most common species, and one to two times for the most common or highly contagious species. We marred the distributions of hemlock and sugar maple from the pre-European settlement period based on their witness tree locations and reconstructed presettlement forest landscapes based on species importance values derived for all species. The results provide a unique basis to further study land cover changes occurring after European settlement.

Wisconsin↗
USGS2019

Study design and methods for a wetland condition assessment on U.S. Fish and Wildlife Service fee-title lands in the Prairie Pothole Region of North Dakota, South Dakota, and Montana, USA

The U.S. Fish and Wildlife Service (FWS) manages wetlands and grasslands for wildlife habitat throughout the central North American Prairie Pothole Region (PPR). PPR wetlands, or potholes, are widely recognized as critical habitats for North American migratory waterfowl, waterbirds, and other wildlife. Potholes also provide other ecosystem services such as carbon sequestration, flood mitigation, filtration of pollutants, groundwater recharge, nutrient retention, and recreational opportunities. Wetland condition assessments have been completed nationally at coarse scales, but focused, regionwide assessments of the biological condition of potholes managed by the FWS are lacking. Therefore, FWS personnel require information pertaining to the biological condition and status of wetlands on FWS fee-title lands in the PPR to support management, restoration, and acquisition efforts. The biological condition of wetlands typically is reflected by their plant communities, and these communities correspond to past and current management and anthropogenic disturbances; thus, plant communities are a suitable surrogate of wetland condition. This report describes the study design, selection of sample sites, and field survey methods for a wetland condition assessment for FWS fee-title lands in the PPR of North Dakota, South Dakota, and Montana. Various spatial databases were gathered (for example, National Wetlands Inventory) to identify and assess potholes on FWS fee-title lands and to facilitate the selection of study sites. A spatially balanced, site-selection process resulted in the inclusion of 125 temporarily and 125 seasonally ponded potholes distributed across the area of interest; the first 100 for each classification were considered the primary study sites, whereas the remaining 25 were considered an oversample to replace those deemed not appropriate for sampling by field crews. Study sites were within native prairie and reseeded grasslands on FWS National Wildlife Refuges and Waterfowl Production Areas and are distributed among the primary physiographic subregions of the PPR: the Glaciated Plains, Missouri Coteau, and Prairie Coteau; a small number of sites also are within the Lake Agassiz Plain and Turtle Mountains. Site assessment protocols, vegetation survey methods, data analyses, and condition categories (for example, poor, good, very good) for the wetland assessment are based on the North Dakota Rapid Assessment Method and an Index of Plant Community Integrity developed for potholes. Results of the wetland condition assessment will aid FWS staff in assessing past and current management and help to identify priority areas for future management and acquisition.

North Dakota, South Dakota, Montana↗
USGS1982

Refining Landsat classification results using digital terrain data

Scientists at the U.S. Geological Survey's Earth Resources Observation systems (EROS) Data Center have recently completed two land-cover mapping projects in which digital terrain data were used to refine Landsat classification results. Digital ter rain data were incorporated into the Landsat classification process using two different procedures that required developing decision criteria either subjectively or quantitatively. The subjective procedure was used in a vegetation mapping project in Arizona, and the quantitative procedure was used in a forest-fuels mapping project in Montana. By incorporating digital terrain data into the Landsat classification process, more spatially accurate landcover maps were produced for both projects.

Journal of Applied Photographic Engineering↗
USGS1994

Designing global land cover databases to maximize utility—the US prototype

One of the most pressing problems in global climate and ecosystem studies is a lack of adequate land cover data. Staff from the United States Geological Survey's EROS Data Center and the University of Nebraska have developed a United States prototype for a proposed global land cover characteristics database derived from 1-km Advanced Very High Resolution Radiometer satellite data. A total of 159 seasonally distinct spectral I temporal land cover classes were labelled according to their constituent vegetation types rather than forcing the vegetation complexes into a predefined classification scheme. The database contains attributes that characterize each land cover class, including elevation, climate attributes and biophysical parameters derived from the normalized difference vegetation index (ND VI). The database design permits convenient translation to a variety of land cover classification schemes commonly used in global scale models. This approach allows scientists to continue using classification schemes with which they are comfortable, eliminates duplicating land cover database development, and provides a degree of uniformity in the development of parallel but distinct databases.

Book chapter↗
USGS1993

Hydrology, vegetation, and soils of four north Florida River flood plains with an evaluation of state and federal wetland determinations

A study of hydrologic conditions, vegetation, and soils was made in wetland forests of four north Florida streams from 1987 to 1990. The study was conducted by the U.S. Geological Survey in cooperation with the Florida Department of Environmental Regulation to support State and Federal efforts to improve wetland delineation methodology in flood plains. Plant communities and soils were described and related to topographic position and long-term hydrologic conditions at 10 study plots located on 4 streams. Detailed appendixes give average duration, frequency, and depth of flooding; canopy, subcanopy, and ground-cover vegetation; and taxonomic classification, series, and profile descriptions of soils for each plot. Topographic relief, range in stage, and depth of flooding were greatest on the alluvial flood plain of the Ochlockonee River, the largest of the four streams. Soils were silty in the lower elevations of the flood plain, and tree communities were distinctly different in each topographic zone. The Aucilla River flood plain was dominated by levees and terraces with very few depressions or low backwater areas. Oaks dominated the canopy of both lower and upper terraces of the Aucilla flood plain. Telogia Creek is a blackwater stream that is a major tributary of the Ochlockonee River. Its low, wet flood plain was dominated by Wyssa ogeche (Ogeechee tupelo) trees, had soils with mucky horizons, and was inundated by frequent floods of very short duration. The St. Marks River, a spring-fed stream with high base flow, had the least topographic relief and lowest range in stage of the four streams. St. Marks soils had a higher clay content than the other streams, and limestone bedrock was relatively close to the surface. Wetland determinations of the study plots based on State and Federal regulatory criteria were evaluated. Most State and Federal wetland determinations are based primarily on vegetation and soil characteristics because hydrologic records are usually not available. In this study, plots were located near long-term gaging stations, thus wetland determinations based on plant and soil characteristics could be evaluated at sites where long-term hydrologic conditions were known. Inconsistencies among hydrology, vegetation, and soil determinations were greatest on levee communities of the Ochlockonee and Aucilla River flood plains. Duration of average annual longest flood was almost 2 weeks for both plots. The wetland species list currently used (1991) by the State lacks many ground-cover species common to forested flood plains of north Florida rivers. There were 102 ground-cover species considered upland plants by the State that were present on the nine annually flooded plots of this study. Among them were 34 species that grew in areas continuously flooded for an average of 5 weeks or more each year. Common flood-plain species considered upland plants by the State were: Hypoxis leptocarpa (yellow star-grass), and two woody vines, Brunnichia ovata (ladies' eardrops) and Campsis radican s (trumpet-creeper), which were common in areas flooded continuously for 6 to 9 weeks a year; Sebastiania fruticosa (Sebastian-bush), Chasmanthium laxum (spikegrass), and Panicum dichotomum (panic grass), which typically grew in areas flooded an average of 2 to 3 weeks or more per year; Vitis rotundifolia (muscadine) and Toxicodendron radicans (poison-ivy), usually occurring in areas flooded an average of 1 to 2 weeks a year; and Quercus virginiana (live oak) present most often in areas flooded approximately 1 week a year. Federal wetland regulations (1989) limited wetland jurisdiction to only those areas that are inundated or saturated during the growing season. However, year-round hydrologic records were chosen in this report to describe the influence of hydrology on vegetation, because saturation, inundation, or flowing water can have a variety of both beneficial and adverse effects on flood-plain vegetation at any time of the year. These effects can occur because: (1) Soil temperatures in north Florida are probably high enough in winter for anaerobic conditions to develop in saturated soils. (2) Many plants in the flood plains of north Florida are active in the nongrowing season and might be adapted to anaerobic conditions in winter. (3) Other effects of standing water, such as decreased light penetration, prevention of seed germination, and protection of seeds from herbivores; and effects of flowing water, such as scouring and deposition of sediments, seed dispersal, and mechanical injury, can occur any time of year.

Florida↗
USGS2012

Unsupervised classification of lidar-based vegetation structure metrics at Jean Lafitte National Historical Park and Preserve

Traditional vegetation maps capture the horizontal distribution of various vegetation properties, for example, type, species and age/senescence, across a landscape. Ecologists have long known, however, that many important forest properties, for example, interior microclimate, carbon capacity, biomass and habitat suitability, are also dependent on the vertical arrangement of branches and leaves within tree canopies. The objective of this study was to use a digital elevation model (DEM) along with tree canopy-structure metrics derived from a lidar survey conducted using the Experimental Advanced Airborne Research Lidar (EAARL) to capture a three-dimensional view of vegetation communities in the Barataria Preserve unit of Jean Lafitte National Historical Park and Preserve, Louisiana. The EAARL instrument is a raster-scanning, full waveform-resolving, small-footprint, green-wavelength (532-nanometer) lidar system designed to map coastal bathymetry, topography and vegetation structure simultaneously. An unsupervised clustering procedure was then applied to the 3-dimensional-based metrics and DEM to produce a vegetation map based on the vertical structure of the park's vegetation, which includes a flotant marsh, scrub-shrub wetland, bottomland hardwood forest, and baldcypress-tupelo swamp forest. This study was completed in collaboration with the National Park Service Inventory and Monitoring Program's Gulf Coast Network. The methods presented herein are intended to be used as part of a cost-effective monitoring tool to capture change in park resources.

Louisiana↗
USGS2023

Beyond presence mapping: Predicting fractional cover of non-native vegetation in Sentinel-2 imagery using an ensemble of MaxEnt models

Non-native species maps are important tools for understanding and managing biological invasions. We demonstrate a novel approach to extend presence modeling to map fractional cover (FC) of non-native yellow sweet clover Melilotus officinalis in the Northern Great Plains, USA. We used ensembles of MaxEnt models to map FC across landscapes from satellite imagery trained from regional aerial imagery that was trained by local unmanned aerial vehicle (UAV) imagery. Clover cover from field surveys and classified UAV imagery were nearly identical ( n = 22, R 2 = 0.99). Two classified UAV images provided training data to map clover presence with MaxEnt and National Agricultural Imagery Program (NAIP) aerial imagery. We binned cover predictions from NAIP imagery within each Sentinel-2 pixel into eight cover classes to create pure (100%) and FC (20%–95%) training data and modeled each class separately using MaxEnt and Sentinel-2 imagery. We mapped pure clover with one classification threshold and compared its performance to 15 candidate maps that included FC predictions outside pure predictions. Each FC map represented alternative combinations of five MaxEnt thresholds and three approaches to assign cover to pixels with multiple predictions from the FC ensemble. Evaluations of performance with independent datasets revealed maps including FC corresponded to field ( n = 32, R 2 range: 0.39–0.68) and UAV ( n = 20, R 2 range: 0.61–0.84) data better than pure clover maps ( R 2 = 0.15 and 0.31, respectively). Overall, the pure clover map predicted 3.2% cover, whereas the three best performing FC maps predicted 6.6%–8.0% cover. Including FC predictions increased accuracy and cover predictions which can improve ecological understanding of invasions. Our method allows efficient FC mapping for vegetative species discernible in UAV imagery and may be especially useful for mapping rare, irruptive or patchily distributed species with poor representation in field data, which challenges landscape-level mapping.

Montana, South Dakota↗
USGS2011

Development of a high-resolution binational vegetation map of the Santa Cruz River riparian corridor and surrounding watershed, southern Arizona and northern Sonora, Mexico

This report summarizes the development of a binational vegetation map developed for the Santa Cruz Watershed, which straddles the southern border of Arizona and the northern border of Sonora, Mexico. The map was created as an environmental input to the Santa Cruz Watershed Ecosystem Portfolio Model (SCWEPM) that is being created by the U.S. Geological Survey for the watershed. The SCWEPM is a map-based multicriteria evaluation tool that allows stakeholders to explore tradeoffs between valued ecosystem services at multiple scales within a participatory decision-making process. Maps related to vegetation type and are needed for use in modeling wildlife habitat and other ecosystem services. Although detailed vegetation maps existed for the U.S. side of the border, there was a lack of consistent data for the Santa Cruz Watershed in Mexico. We produced a binational vegetation classification of the Santa Cruz River riparian habitat and watershed vegetation based on NatureServe Terrestrial Ecological Systems (TES) units using Classification And Regression Tree (CART) modeling. Environmental layers used as predictor data were derived from a seasonal set of Landsat Thematic Mapper (TM) images (spring, summer, and fall) and from a 30-meter digital-elevation-model (DEM) grid. Because both sources of environmental data are seamless across the international border, they are particularly suited to this binational modeling effort. Training data were compiled from existing field data for the riparian corridor and data collected by the NM-GAP (New Mexico Gap Analysis Project) team for the original Southwest Regional Gap Analysis Project (SWReGAP) modeling effort. Additional training data were collected from core areas of the SWReGAP classification itself, allowing the extrapolation of the SWReGAP mapping into the Mexican portion of the watershed without collecting additional training data.

Arizona↗
USGS2008

Effect of species rarity on the accuracy of species distribution models for reptiles and amphibians in southern California

Aim: Several studies have found that more accurate predictive models of species' occurrences can be developed for rarer species; however, one recent study found the relationship between range size and model performance to be an artefact of sample prevalence, that is, the proportion of presence versus absence observations in the data used to train the model. We examined the effect of model type, species rarity class, species' survey frequency, detectability and manipulated sample prevalence on the accuracy of distribution models developed for 30 reptile and amphibian species. Location: Coastal southern California, USA. Methods: Classification trees, generalized additive models and generalized linear models were developed using species presence and absence data from 420 locations. Model performance was measured using sensitivity, specificity and the area under the curve (AUC) of the receiver-operating characteristic (ROC) plot based on twofold cross-validation, or on bootstrapping. Predictors included climate, terrain, soil and vegetation variables. Species were assigned to rarity classes by experts. The data were sampled to generate subsets with varying ratios of presences and absences to test for the effect of sample prevalence. Join count statistics were used to characterize spatial dependence in the prediction errors. Results: Species in classes with higher rarity were more accurately predicted than common species, and this effect was independent of sample prevalence. Although positive spatial autocorrelation remained in the prediction errors, it was weaker than was observed in the species occurrence data. The differences in accuracy among model types were slight. Main conclusions: Using a variety of modelling methods, more accurate species distribution models were developed for rarer than for more common species. This was presumably because it is difficult to discriminate suitable from unsuitable habitat for habitat generalists, and not as an artefact of the effect of sample prevalence on model estimation. ?? 2008 The Authors.

Diversity and Distributions↗