Map Development

The Oklahoma Botanical Garden and Arboretum, or OBGA, is a statewide organization comprised of gardens, parks, zoos, and arboreta. The organization was created in 1991 as a support network for member gardens. The OBGA is under the direction and supervision of the Board of Regents of Oklahoma Agricultural and Mechanical Colleges and is coordinated by the Oklahoma State University Department of Horticulture and Landscape Architecture. The headquarters’ garden and arboretum is a 60 acre plot west of the OSU campus in Stillwater, Oklahoma.  The headquarter's garden was renamed Oklahoma State University Botanical Garden in the Spring of 2003.

In 2000, curators of the OSU Botanical Garden began researching new ways to manage their specimens. Dr. Douglas Needham of the OSU Department of Horticulture and Landscape Architecture had already developed Plant Information Manager, a Filemaker Pro relational database which contains not only general biological data, but also information on individual specimens. PIM is an extensive database, but it contains no references to specimen locations within the arboretum. Dr. Lou Anella, of the same department, was interested in finding a way to keep track of where each specimen was located.

The Urban and Community Forest Council awarded Dr. Anella a grant to map the OSU Botanical Garden using Global Positioning System, or GPS, technology. The resulting map would be utilized for garden management and as an educational tool for teaching other gardens the value of using this technology.

This project focuses on the development of an on-line GPS/GIS map. The researcher collected GPS data from the OSUBG using a Trimble GPS receiver, manipulated the data in Pathfinder Office and ArcView 3.3, and displayed the resulting map on-line using ArcIMS. The goal of this project is three fold. The map will be used by the curators of the OSUBG as a management tool. The map will also serve as an educational tool not only for Horticulture and Landscape Architecture students, but also for the public. Finally, completion of the map will satisfy the requirements of Dr. Anella’s Urban and Community Forest Council grant.

Collection of GPS field data began during the summer of 2001. The data dictionary was built by Bruce Battles, a staff member from the OSU Department of Geography. The data dictionary contained nine different data types: tree, tree area, shrub, shrub area, herbaceous, herbaceous area, generic point, generic line, and generic area.

Tree, shrub, and herbaceous attribute data included the accession number, height, trunk diameter, canopy diameter, conditions, and notes. The accession number is an identification number given by the PIM database which tells when the specimen was cataloged and which garden it is a part of. Specimens without accession numbers were given a temporary identification number. Tree, shrub, and herbaceous area attribute data included accession number, height, trunk diameter, canopy diameter, conditions, and notes.

Generic points, lines, and areas were reserved for human structures and infrastructure. The researcher did consider using pre-existing AutoCad data from the Oklahoma State University Physical Plant for streets and buildings. This data still might be utilized, but it will have to be geo-referenced first. Instead, the researcher chose to map these features using GPS. An example of a generic point would be a water hydrant and they had no attributes. Generic lines consisted of roads, paths, and boundaries. They had no attribute data. Generic areas were buildings, test plots, and brush piles. They also had no attribute data.

All GPS data entry was performed by the researcher and a Horticulture and Landscape Architecture Department undergraduate, Cheryl Boyer. The GPS receiver was a Trimble Pro XRS with a TDCI datalogger. The majority of the data was collected throughout July and August of 2001, but some has been collected since.

Data files (.ssf) were transferred weekly to Trimble Pathfinder Office. In Pathfinder, the .ssf files underwent differential correction. For this project, the researcher downloaded base station files from the Oklahoma State University Department of Geography website. Once corrected, the .ssf files became .cor files. The .cor files were combined and then exported to ESRI ArcView 3.3 as shapefiles (.shp). Once in ArcView, the shapefiles could be manipulated again. The generic lines file was split into two separate shapefiles, boundaries and roads. The generic areas file was divided up into shapefiles for individual types of buildings, plots, and brush piles.

An image file was added to the ArcView project for reference. A satellite photograph of the OSUBG area called a .sid file was downloaded from www.geoou.edu. No geo-referencing was necessary since because .sid files are pre-referenced.

In May of 2002, the initial version of the on-line map was launched. Shapefiles from ArcView were loaded into ArcIMS, ESRI’s internet map-hosting program.

There are a few different limitations that have emerged during this project. The first limitation appeared during data collection. In order to accurately record a position for a plant specimen you must be able to get as close to its center as possible. With some specimens this became very difficult. Either the canopy blocked the satellite signal or the density of the foliage did not allow us to get as close to the center as we needed to. In some cases we used a GPS offset, but their accuracy is questionable. Erroneous offsets were manipulated in Pathfinder Office. Some specimens were accessioned individually, but had grown together so closely that individual positions could not be taken. When that occurred, the individuals were entered as a tree, shrub, or herbaceous area and given a temporary identification number.

Specimens that have not been added to the PIM database can cause problems. Accessioned specimens, plants within the database, can easily be recognized as being a part of the collection. Plants that are not tagged are entered by the researcher’s discretion. This means that volunteer plants may have accidentally been entered when they shouldn’t have and some plants may have been left out that needed to be added. Unaccessioned trees along the boundary of the arboretum caused problems because it was difficult to determine which plants were in and which were out.

Updates for the on-line map could potentially become a problem. The OSUBG is constantly changing. New specimens are constantly coming in and specimens can die or be removed. Some unaccessioned plants with temporary identification numbers have since been accessioned. These changes need to be added or adjusted in the GPS database. If they aren’t, the map becomes useless.

Back to Home