Vegetation classification and mapping, Vicksburg National Military Park, Mississippi

Chris Lea; Robert G. Waltermire; Carl Nordman

Home/Archive/USGS

USGS · 70046565

Vegetation classification and mapping, Vicksburg National Military Park, Mississippi

Chris Lea·Robert G. Waltermire·Carl Nordman

View original source ↗Export metadata ↓

Abstract

Executive Summary The National Park Service (NPS) Gulf Coast Inventory and Monitoring Network, with the support of the National Park Service Vegetation Inventory completed vegetation classification and mapping for Vicksburg National Military Park (VICK), in Warren County, Mississippi, from 2004 to 2010. VICK is one of more than 250 NPS units to be covered by the Vegetation Inventory. Methods and procedures follow those of the NPS Vegetation Inventory as of August, 2012 ( http://science.nature.nps.gov/im/inventory/veg/index.cfm ). Ecologists collected floristic and environmental data from 45 vegetation field plots and classified and described 13 plant community types corresponding to US National Vegetation Classification (USNVC) associations from these data. This classification was used to map 712 hectares (1,698 acres) of Vicksburg National Military Park, in ten map classes corresponding to vegetation associations and two non-vegetated land cover classes, from a digital orthomosaic created from 1:24,000 color infrared aerial photographs and digital elevation model (DEM) data. Methods used to delineate stands varied according to vegetation type and were a combination of incorporating stand boundaries from an existing map, new “heads-up” interpretation and modeling of digital environmental spatial data. A thematic accuracy assessment collected vegetation data from 118 field observations stratified across the vegetation classes. The overall accuracy of the vegetation map was estimated to be 60.5% at the thematic level of USNVC association and 92.4% at the thematic level of USNVC group. Estimates of areas of vegetation types, as adjusted for mapping error, were calculated for all types mapped and/or observed during the thematic accuracy assessment. An important finding of this study is that the composition of the forests of VICK is substantially different from that described by earlier, more qualitative, accounts. Implications of these findings and possible applications of the data for resource monitoring and management are presented. This report summarizes the methods and general results of the Vegetation Inventory for Vicksburg National Military Park. Appendices include descriptions of vegetation types, a field key to the types, a list of plant species, and accuracy assessment contingency tables. This report is supplemented by the project data, which include: spatial database containing map polygon attributes. Geospatial products are in Universal Transverse Mercator (UTM) projection, Zone 15, using the North American Datum of 1983. all field data (plot, observation point, and accuracy assessment point) stored in a Microsoft Access database metadata for all digital products The data and reports produced by this investigation reside at: https://irma.nps.gov/App/ and/or at: http://biology.usgs.gov/npsveg/

Explore related subjects

National Park Service Natural Resource Technical Report Report

Source-reported bounding extent: 32.09304606091566° to 32.49586350791503° latitude; -90.889892578125° to -90.615234375° longitude. This indicates report coverage, not an exact sampling location. View area on OpenStreetMap.

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Chris Lea, Robert G. Waltermire, Carl Nordman. 2013. Vegetation classification and mapping, Vicksburg National Military Park, Mississippi. https://pubs.usgs.gov/publication/70046565

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2013

Vegetation projections for Wind Cave National Park with three future climate scenarios: Final report in completion of Task Agreement J8W07100052

Introduction The effects of climate change on the natural resources protected by Parks will likely be substantial, but geographically variable, due to local variation in climate trajectories and differences among ecosystems in their vulnerability to climate change. The projections of general circulation models (GCMs) indicate the possible magnitude and direction of future climate change for a region, but the utility of these projections for more local scales, those of individual National Park Service (NPS) units, are more uncertain because the coarse-scale GCMs lack much of the topographic detail that alters local climates. In addition, complex, interacting effects of temperature, precipitation, atmospheric CO 2 concentrations, fire, and herbivores on the vegetation that is the foundational natural resource of many NPS units present challenges in assessing the effects of projected future climates on plant and animal assemblages managed by the NPS. In spring 2009, Wind Cave National Park (WICA) served as a case study in a workshop assessing the use of scenario planning as a tool for park management planning in the face of rapidly changing climate. One outcome of the workshop was the recognized need for quantitative models to better understand the range of possible vegetation changes under different future climates and management decisions. This report addresses this need; it describes our adaptation of a dynamic global vegetation model (DGVM) to WICA vegetation and the resulting projections of future vegetation under three future climate scenarios and 11 management scenarios determined by Park natural resource managers. Wind Cave National Park lies along a narrow transition zone between the ponderosa pine (Pinus ponderosa) forests of the Black Hills and the mixed grass prairie that once extended with few interruptions over the lower, gentler terrain, subject to warmer, drier climate to the east and south of the Park. The location and character of this transition is strongly influenced by fire frequency and intensity (Brown and Sieg 1999). Furthermore, the mixed grass prairie occupies a broader transition zone between eastern tallgrass prairie and the shortgrass prairie of the western Great Plains. The dominance of species characteristic of these two prairie types varies with soil moisture availability, evaporative demand, and recent grazing history (Cogan et al. 1999). In addition, Wind Cave lies near the midpoint of a long gradient of C 3 (cool season) grass dominance to the north and C 4 (warm season) grass dominance to the south. The ecotonal position of WICA may make it particularly sensitive to climate change. For example, small changes in fire frequency and/or intensity and the vigor of trees vs. grass could dramatically shift the proportions of these two lifeforms. The Park hydrology is also sensitive to changes in the balance between infiltration of precipitation and evapotranspiration, as on average, only a small fraction of annual precipitation reaches the deeper soil layers that feed permanent streamflow. The resources at risk at Wind Cave NP include the Cave itself, as well as small backcountry caves, a genetically important bison herd, and other prairie species including the black-tailed prairie dog and endangered black-footed ferrets. All of these resources will be directly affected by climate change impacts on vegetation and hydrology. Natural resource management challenges at WICA are substantial, diverse, and intertwined. Aboveground, the park has been recognized as exemplary for its high quality vegetation (Marriot et al. 1999), though the park is relatively small for the diversity of vegetation types and species that it supports. Even without a changing climate, maintaining the integrity of the plant communities is complicated by the park’s legislated responsibility to maintain viable populations of bison, elk and pronghorn. In addition, the federally endangered black-footed ferret was recently re-introduced to the park. This species requires large extents of prairie dog towns for prey and habitat. Prairie dogs impact vegetation by constant clipping, grazing and soil disturbance, thereby affecting plant composition and productivity. Moreover, naturally high interannual climate variability and the strong influence of precipitation on grass productivity in this region combine to yield high interannual variability in the amount of forage available for the wildlife that the park is tasked to maintain. Finally, fire, which is now primarily controlled by WICA and NPS Northern Great Plains fire management programs, is intertwined with all other natural resource issues at WICA, as it can impact prairie dog colony and forest expansion, ungulate foraging behavior, invasive plant species, and hydrological processes. Although not capable of capturing all of these complexities, dynamic vegetation models do provide a means for quantitatively projecting vegetation futures in future climates under plausible fire and grazing regimes. Our work uses the DGVM MC1 to simulate the effects of future climate projections and management practices on the vegetation of WICA. MC1 is designed to project potential vegetation as influenced by natural processes and hence is appropriate for national parks, where conservation of native biota and ecosystems is of great importance. Since the initial application of MC1 to a small portion of WICA (Bachelet et al. 2000), the model has been altered to improve model performance with the inclusion of dynamic fire. Applying this improved version to WICA required substantial recalibration, during which we have made a number of improvements to MC1 that will be incorporated as permanent changes. In this report we document these changes and our calibration procedure following a brief overview of the model. We compare the projections of current vegetation to the current state of the park and present projections of vegetation dynamics under future climates downscaled from three GCMs selected to represent the existing range in available GCM projections. In doing so, we examine the consequences of different management options regarding fire and grazing, major aspects of biotic management at Wind Cave.

South Dakota↗

USGS2013

Pacific island landbird monitoring annual report, National Park of American Samoa, Ta‘u and Tutuila units, 2011

The National Park of American Samoa (NPSA) was surveyed for landbirds and habitat characteristics from June through August, 2011. This information provides the first data in the time-series of landbird monitoring for long-term trends in forest bird distribution, density, and abundance within the NPSA. The NPSA survey area was comprised of the terrestrial portions of the Ta‘u and Tutuila Units. Each Unit was surveyed using point-transect distance sampling to estimate bird abundance. Sampling was conducted using a split-panel design where legacy transects are visited during each sampling occasion and newly, randomly located transects are visited only during one sampling occasion. This design optimizes trend detection while allowing for measuring and correcting for estimator bias. A total of 2,516 birds was detected from 13 species in both Units. All species were either endemic or indigenous to the islands of American Samoa. Numbers of detections ranged from 7 to 1,111. Nearly every species detected was broadly distributed in the predominantly native forests of NPSA. Sufficient detections were made of seven species, allowing for density estimation. Densities of species were higher in the Tutuila Unit; with the exception of the Wattled Honeyeater ( Foulehaio carunculata ), which was the most abundant species in both Units. The species occurred at nearly every station sampled and had densities much higher than the Samoan Starling ( Aplonis atrifusca ), Polynesian Starling ( Aplonis tabuensis ), and Collared Kingfisher ( Halcyon chloris ) which occurred in modest densities. The remaining species detected occurred at less than 20% of stations sampled and we were only able to determine the number of birds per station and percent occurrence. The White-rumped Swiftlet ( Aerodramus spodiopygius ) and Cardinal Honeyeater ( Myzomela cardinalis ) were detected in small numbers, but both species can be difficult to detect in closed canopy forests. The Purple Swamphen ( Porphyrio porphyrio ) and Banded Rail ( Gallirallus philippensis ) were most often detected in areas close to villages and agroforestry plantations. The Blue-crowned Lorikeet ( Vini australis ) and Fiji Shrikebill ( Clytorhynchus vitiensis ) only occur in the Manu‘a Island Group. The former was detected in most survey areas and the latter was patchily distributed in the Ta‘u Unit. The Many-colored Fruit-dove ( Ptilinopus perousii ), a species of concern, was detected in very small numbers in both Units. The Spotless Crake ( Porzana tabuensi ), which is extirpated on Tutuila Island, has been incidentally detected in small numbers on Ta‘u Island. However, the species was neither seen nor heard during this survey and remains a species of concern. NPSA canopy and understory composition was predominantly native, and trees formed a dense closed canopy at nearly 90% of the stations sampled. More than half of the tree heights in both units were taller than 5 m and the majority of slopes were steeper than 20 degrees. There were no clear dominant tree species in the mixed native forests. The most common tree species documented included Syzygium spp., Dysoxylum spp., Ficus spp., Hibiscus tiliaceus and Rhus taitensis (among others). There were significant differences in the distribution of bird densities between legacy and random transects. Determining differences in detection probabilities cannot be definitively assessed from a single survey. We recommend both panels be sampled in the future until bias in density and abundance can be evaluated, or if sampling may be reduced.

American Samoa;National Park Of American Samoa↗

USGS2013

Pacific Island landbird monitoring annual report, Haleakalā National Park, 2012

Haleakalā National Park (HALE) was surveyed for landbirds and habitat characteristics from March 20 through July 26, 2012. This information provides data in the time-series of landbird monitoring for long-term trends in forest bird distribution, density, and abundance. The Kīpahulu District of eastern Haleakalā Volcano was surveyed using point-transect distance sampling to estimate bird abundance. We surveyed 160 stations and detected a total of 2,830 birds from 12 species. Half of the species were native and half were non-native. Numbers of detections per species ranged from 1 to 849. There were sufficient detections of seven species to allow density estimation. Āpapane ( Himatione sanguinea ) was the most widely distributed and abundant native species detected in the survey. ‘Alauahio ( Paroreomyza montana newtoni ), Maui ‘Amakihi ( Hemignathus virens wilsoni ), and I‘iwi ( Vestiaria coccinea ) were widespread and occurred in relatively modest densities. Only eight Kiwikiu ( Pseudonestor xanthophrys ) and 20 ‘Ākohekohe ( Palmeria dolei ) were detected and were restricted to high elevation wet forest. We estimated an abundance of 495 ± 261individuals of Kiwikiu in a 2,036 ha inference area which likely includes the entire suitable habitat for this species in HALE. For ‘Ākohekohe, we estimated an abundance of 1,150 ± 389 individuals in the 1,458 ha inference area. There was a strong representation of non-native landbirds in the survey area. The Japanese White-eye ( Zosterops japonicus ), Japanese Bush-warbler ( Cettia diphone ), and Red-billed Leiothrix ( Leiothrix lutea ) accounted for nearly half of all landbird detections. Each species was common in predominantly native forests. Vegetation and topographic characteristics were recorded on 160 landbird monitoring stations. HALE canopy and understory composition was predominantly native, especially at elevations above 1,100 m. Much of the forest canopy was comprised of `ohi`a ( Metrosideros polymorpha ) interspersed with mature olapa ( Cheirodendron platyphyllum ). This canopy class occurred at 92.5% of the stations surveyed. More than three-quarters (77.5%) of the monitoring stations had a dense canopy with most crowns interlocking (> 60% cover). More than half (52%) of the stations surveyed had trees taller than 10 m, while almost a third (31%) had trees 5-10 m. Only 17% of the stations had a canopy shorter than 5 m. The native shrubs Vaccinium calycinum , Broussaisia arguta , and Leptecophylla tameiameae were the most common understory plants recorded, occurring at more than 30% of the stations sampled. Native mosses and ferns were also common at stations, occurring at more than 90% of the stations sampled. The invasive Psidium cattleainum , Clidemia hirta , and Hedychium gardnerianum occurred at approximately 14% of the stations sampled, predominantly at elevations below 1,100 m.

Hawai'i↗