National Park Service


Date of this Version



Natural Resource Report NPS/NIOB/NRR 2019/2040 / NPS 656/165508, November 2019

Published by the United States National Park Service, Natural Resource Stewardship and Science, Fort Collins, Colorado, USA

Also available at:

Please cite this publication as:

Baldvins, T., M. Ley, J. Stevens, D. Jones, and H. Pilkington. 2019. Vegetation classification and mapping project: Niobrara National Scenic River. Natural Resource Report NPS/NIOB/NRR—2019/2040. National Park Service, Fort Collins, Colorado.


United States government work. Public domain material.


Executive Summary

The vegetation inventory project at Niobrara National Scenic River (NIOB) classified and mapped vegetation within the park administrative boundary and estimated thematic map accuracy quantitatively. The project was conducted over a four year period from the summer of 2015 to the winter of 2019. Located in north-central Nebraska, approximately 76 miles of the park is designated as a wild and scenic river. The vast majority of land within the NIOB administrative boundary is private or state-owned.

The project follows guidance provided by the National Park Service (NPS) Vegetation Mapping Inventory (VMI) Program. The overall process includes initial planning and scoping, imagery procurement, field data collection, data analysis, imagery interpretation and classification, and accuracy assessment (AA). The initial planning and scoping meetings to develop the study plan took place in December 2009 in Valentine, Nebraska, and included representation by Niobrara National Scenic River (NIOB), NPS Northern Great Plains Network, Colorado State University, and other interested agencies and organizations.

New imagery was acquired for the NIOB mapping project in early October 2015. This imagery was delivered as 30 cm 4-band (RGB and CIR) high-resolution orthoimages. Additional imagery supplementing the interpretation phase included current and historic true-color Google Earth and Bing Maps imagery.

Prior to field work, the preliminary classification of the vegetation associations included 82 United States National Vegetation Classification (USNVC) associations; 36 were known to occur in the project area while the other 46 associations were potentially located in the area. Existing vegetation and mapping data combined with field-collected vegetation plot data contributed to the final vegetation classification. Vegetation data collections by field crews at 156 NIOB subjective plot locations over the summer of 2015 supported vegetation classification using hierarchical clustering and professional expertise. Other types were identified in the course of additional field work and photointerpretation reconnaissance. The final vegetation classification includes 46 USNVC associations and 13 park special types. Types include 24 forest and woodland types, 8 shrubland types, 21 herbaceous types, and 6 sparse vegetation types.

The final mapping classification, which cross walks vegetation type(s) present with what can be consistently classified and mapped (i.e., interpreted), consists of 35 map classes within NIOB. Of these 35 map classes, 6 represent land use cover classes (cultivated crops, pasture/hay ground, non-vegetated barren land or borrow pit, developed open space, developed low, medium, or high intensity, and water). Of the 29 non-land use map classes, 17 map classes represent a single USNVC association or park special, five map classes represent two USNVC associations or park specials, three map classes represent three USNVC associations or park specials, two map classes represent four USNVC associations or park specials, one map class represents six USNVC associations or park specials, and one map class represents nine USNVC associations or park specials.

Species dominating in forest and woodland types include eastern redcedar (Juniperus virginiana), bur oak (Quercus macrocarpa), green ash (Fraxinus pennsylvanica), hophornbeam (Ostrya virginiana), Virginia wildrye (Elymus virginicus), and Kentucky bluegrass (Poa pratensis). Dominant species in shrubland types found in NIOB are false indigo bush (Amorpha fruticosa), Kentucky bluegrass (Poa pratensis), cuman ragweed (Ambrosia psilostachya), eastern redcedar (Juniperus virginiana), and smooth sumac (Rhus glabra). Other, less extensive shrubland types are dominated by western snowberry (Symphoricarpos occidentalis). The most common species within herbaceous types include cuman ragweed (Ambrosia psilostachya), Heller’s rosette grass (Dichanthelium oligosanthes), little bluestem (Schizachyrium scoparium), white sagebrush (Artemisia ludoviciana), switchgrass (Panicum virgatum), and blue grama (Bouteloua gracilis).

The final NIOB vegetation map consists of 2,762 polygons totaling 29,081 acres (11,768 ha). Mean polygon size for vegetated types is 8.2 acres (3.32 ha). Of the total area, 20,878 acres (8,449 ha) or 72% represent natural or ruderal vegetation map classes. Agricultural vegetation, such as cultivated crops and pasture, account for approximately 3,267 acres (1,322 ha) or 11% of the total mapped area. Non-vegetated barren land is rare and only accounts for 520 acres (210 ha) or 1.8%. Open water is the most widespread land cover class with an area of approximately 4,415 acres (1,786 ha) or 15% of the total mapped area. Within the total area occupied by vegetation map classes, forest and woodland types were the most extensive (12,278 acres (4,969 ha) or 42%), followed by herbaceous types (7,021 acres (2,841 ha) or 24%), shrubland types (1,134 acres (532 ha) or 4.5%), and sparse types (265 acres (107 ha) or 0.9%).

A total of 755 accuracy assessment (AA) samples were collected to evaluate the thematic accuracy of the vegetation polygon data. As a simple proportion, the final thematic accuracy was 84.5%. When map class accuracies were weighted in proportion to the area they occupy within NIOB, the overall accuracy was 87.8%. Two map classes were retained that still fell below the 80% thematic accuracy threshold. Of these two retained map classes, one had accuracy close to 80%, and the other was a map class represented by only a few polygons (i.e. a map class with three polygons in which only one was mapped incorrectly).

In addition to the vegetation polygon database, the project delivered several other products to support park resource management. A geodatabase links the vegetation data layer to other feature classes, such as vegetation classification and AA plots and associated sampling data from the PLOTS database, plot photos, and project boundary extent. The database includes tables documenting the USNVC hierarchy and allows for spatial queries of data associated with a vegetation polygon or sample point. All geospatial products are projected using North American Datum 1983 (NAD83) in Universal Transverse Mercator (UTM) Zone 14 North. Other products include ArcGIS .mxd files for each of the maps along with the aerial imagery acquired for the project. The final report includes methods and results, descriptions for each vegetation type, a dichotomous field key to vegetation types, contingency tables showing AA results, field forms, a species list, and a guide to imagery interpretation. These products provided NIOB with an array of tools to assist in managing park resources and making informed resource management decisions.

The use of standard national vegetation classification and mapping protocols facilitates effective resource stewardship by ensuring compatibility and widespread use of the information throughout the NPS as well as by other federal and state agencies. This comprehensive geospatial database and associated information support a wide variety of resource assessment, park management, and planning needs. In addition, the associated information provides a structure for framing and answering critical scientific questions about vegetation communities and their relationship to environmental processes across the landscape.