National Vegetation Classification Standard
Mapping Guidelines

Choice of Methods: The first step in any project is to determine the methods to be used. The project manager should consider the following before determining a mapping method:

1. Project objectives and any objectives for affected refuges that may be different. (The project area may be a refuge in some cases)
2. Needs of the manager and staff for the project area.
3. Geographic extent of the project area.
4. The Service standard of classifying vegetation to the "Formation" level of NVCS hierarchy.
5. The use of common Minimum Mapping Units (MMU) for large and small scale maps at .25 hectares and 5 acres respectively.
6. Homogeneous -vs- Heterogeneous vegetation communities.
7. Whether or not multiple scales are needed for the same location. The manager may need to consider two separate mapping methods if that is the case.
8. Resolution -vs- Scale (topic needs more information)

Depending on the answers to questions raised by the items considered above, there is additional information needed, as well as materials and staff resources. The following items discuss the information needed in greater detail.

Source Materials: The following are recommendations for large scale and small scale vegetation mapping projects. The items with an asterisk (*) are ideal for further detail at a large scale. Data may or may not be available for the specified project locations.

Take care to ensure that the timing and temporal selection of photos and images meet your needs for the project you are undertaking. For example, leaf-on versus leaf-off aerial photos impact the ability to even see under story vegetation.

Cartographic Base Materials: Cartographic base materials are similar for both large and small scale projects and can include:

DOQQ's*
Tiger Data
DLG's
DEM's
DRG's
The base materials will be used for general location information only and should not be depended upon for positional accuracy. (Topic needs more detail)

Minimum Mapping Units: To classify vegetation to the Formation Level these are acceptable MMU's. (this topic needs more research)

LARGE SCALE: 0.25 hectares
SMALL SCALE: 2 hectares (18 pixels)

Expertise and Contacts Please contact the Regional GIS Coordinator for a list of botanical, remote sensing, and cartographic experts.

Accuracy Requirements: When classifying vegetation to the formation level of the NVCS an accuracy of 80% is required. When mapping to a larger scale beyond the formation level of the hierarchy, a lower percentage level of accuracy will be required. (we recommend waiting for Paul Steblein to consult on the Chincoteague Project)

Accuracy Assessment: This section refers to field verification of the classification as well as photo interpretation. More information is needed, but the following items should be included.

1. Independent data set from those used in training.
2. 30 Samples per class is ideal. Based on a normal distribution. (This idea is taken from the basics of population/distribution statistics. More detail needed)
3. Collection of data in the field prior to initial project start.
4. Final report on the methods and ID's of the site visited for accuracy assessment as well as the initial field reconnaissance.
5. Determination of omission and co-mission errors.
6. Supporting evidence on field sites visited. (i.e. photographs, 35mm and digital)
7. Kappa statistic. This number reports the accuracy compared to an expected accuracy level using an random sample. (Needs information on how to compute)

QA/QC Intermediate and Final: This section needs more detail, but at a minimum the following steps should be included.

1. Draft products should be delivered to the project manager for quality checks at established intermediate points within the project.
2. Continued communication between "mappers", ecologists, and the manager is essential.

The end result of any project should be to obtain the needed products to meet the objectives of the projects.. There may be a variety of types of products ranging from hard copy to digital to verbal information. The following list covers many common products, but is not all-inclusive.

End Products (mapping and reports): These items can be produced for large or small scale projects.

1. Hard Copy of Maps
2. Digital Maps
3. All data layers or covers used to produce map
4. Processing Methods
5. Programming i.e. AMLs, menus, scripts, etc
6. Accuracy Assessments
7. Complete Metadata for all products. Complies with FGDC standards.
8. Finished digital products sent to a central location to serve on web.
9. Second copy to Regional Repository (data steward)
10. Time line for remapping. If appropriate for follow up change analysis and monitoring.
11. Field data sheets.
12. Vegetation classification key and/or crosswalk to NVCS.
13. Training for field station: in house capability to use the data
14. Field photographs.
15. All reports, correspondence, etc.

Contract Guidance: There are some sample guidelines in the National Park Service Documentation. At a minimum contract guidance needs to include:

1. Accuracy documented and Assessment.
2. Minimum Mapping Unit clearly defined and documented.
3. Mapping Scale clearly defined and documented.
4. NVCS followed to the appropriate level of the hierarchy.
5. Deliverables. Intermediate and Final Products.

Cost Comparisons: information for high and low end projects is given below. (Kevin Hop may be a good source of information for the large scale information )

Examples of project costs were given ranging from $35,000 to $480,000 for Agassiz NWR and Acadia NP respectively. Both projects covered approximately 60,000 - 70,000 acres. In many cases FWS would not need to bear all of the costs. Partnerships to jointly purchases aerial photography or TM scenes with other federal or state agencies are common and effective in reducing individual costs.

Other Cost Considerations:

1. Digitization of base data layers which are not available.
2. The cost of technical experts.
3. The cost of meetings and travel to the field.
4. Accuracy Assessments by external means. (This may raise the cost by 33% per BRD- Maury)
5. Contracting is expensive. Costly in lost experience and dollars.

EXAMPLES: Examples were given to walk through the process. The idea is to aid managers in determining vegetation mapping needs/objectives and to highlight the types of decisions that are necessary when planning a vegetation mapping project.

Small Scale:

Large Scale:

For additional information regarding this Web page, contact Deb Southworth Green, in the Division of Information Resources and Technology Management, at Deb_Green@fws.gov

Return to the U.S. Fish and Wildlife Service GIS Home Page
Visit the U.S. Fish and Wildlife Service Home Page

Keywords=GIS, data, metadata, steering, veg, vegetation
Last Modified January 02, 2001 01:03 PM