Casco Bay Database and GIS

One common research problem facing colleges is the disorganization associated with rapidly increasing amounts of data that are routinely created and used. In particular, collegiate departments that use GIS programs such as ArcGIS and GRASS need ways to organize their data. In the Bowdoin College geology department, we sought a way to keep data from several course and independent study projects organized and easily accessible. The data are regularly used by relatively inexperienced undergraduate students that often use the data in only one course or for a few semesters. The data needed to have continuity to allow students to learn how to access all of the data quickly and without extensive training. A MS Access database linked with ArcGis was chosen as the most versatile solution that was the easiest to access.

The MS Access database is split into 4 major tables: metadata, geochemical data, microscope photos and field photos. All tables have a sampleID field that links data from two or more tables. The metadata table includes latitude and longitude coordinates for each site along with the date collected, location description, field description, formation, mineralogy, strike, dip and the project associated with the specific sample. In addition to these fields the metadata table lists what other kinds of data are present for each sample. The geochemical data table consists of 64 oxide and trace element fields along with the name of the external lab that analyzed the data. Each of the photo tables have sampleID, name, description and a link to the picture that is posted on a website.

One of the strengths of this database is the simple interface that is used to add data from the field. With a little training the students can enter data collected in the field or during a course project. Once data is in the database it is relatively simple to move it into ArcGIS. This allows for a simple uniform method of access for those wishing to contribute to or access the database. Since all of the tables in the database have a common field, sampleID, linking multiple tables is extremely easy. First the tables of interest for the given project are exported and added to ArcGIS where they are then combined based upon sampleID. The tables are then converted into spatial data through use of the “Add XY data” feature. Keeping the different types of data in different tables allows for only the use of necessary data, while still allowing for more data to be added at a later date. When the database is updated the process can be easily redone to reflect the changes.

Once data is in ArcGIS there is a wide range of applications of the data. Concentrations of oxide and trace elements can be mapped for visualization of trends. In addition, the Spatial Analyst Tool can be used to determine relationships between different sample sites. Maps with interactive sites linked to field and microscope pictures can be used in presentations. These are just a few of the many possible uses of the geodatabase.

A webpage www.bowdoin.edu/~rbeane/gis/homepage.htm walks students through basic ArcGIS use, provides links to many data sources, and provides instructions on how to administer the database. The database has been growing steadily, with data being added from a variety of projects. There are currently 254 metadata entries, 90 geochemical data entries, 109 microscope photos and 34 field photos. Each table currently contains over 5300 cells, 5760 cells, 545 cells and 170 cells respectively, while containing large amounts of data the geodatabase can easily fit on a single CD. This makes the data easily transportable and accessible to everyone on campus since it is also accessible from the network.

The geodatabase currently designed for the Bowdoin College Geology Department contains 7 years worth of petrologic data. The geodatabase is easily accessible and readily linked to ArcGIS for spatial analysis and presentation. This geodatabase is anticipated to be the long term solution due to its breadth, small space requirements, scalability and simple integration with GIS applications.

Written and designed by Jeremy Huckins '06 as part of a summer research project and independent study.