2004 Denver Annual Meeting (November 7–10, 2004)

Paper No. 56
Presentation Time: 6:30 PM-8:30 PM


TOWNSEND, M.A., Kansas Geological Survey, The Univ of Kansas, 1930 Constant Ave, Lawrence, KS 66047 and DAVIS, John C., Department of Mathematics, Baker University, Baldwin City, KS 66006, townsend@kgs.ku.edu

The use of real field data can be valuable for reinforcing concepts in hydrogeology and other disciplines. Many hydrologic studies, ranging from field to regional scales, have been published but it is unusual for supporting data to be reported in an article. Water quality publications typically include tabulations of water quality data, but ancillary information on water levels, precipitation, fertilizer application records, crop rotations, land use, and geologic logs may only be referenced or summarized in graphs. Educators wishing to use published reports for a complete picture of a field study find that crucial components are missing.

We used a published three-year study of nitrate in the vadose zone and ground water at a test site in south-central Kansas for an exercise in field hydrogeology and for a class project in environmental modeling. Water quality data from 18 soil lysimeters and five wells was available in the form of Excel spreadsheets of nitrate analyses, accompanied by sampling dates and depths of each lysimeter and well. Spreadsheets and other records provided precipitation, water level, crop use, fertilizer application times, and geographic coordinates of sampling sites.

Students in the field hydrology class were instructed to write a report evaluating the movement of nitrate through the vadose zone, based on the available data. Reports were judged on the student's understanding of the geology of the test site, their explanations of the effects of rainfall and irrigation on the movement of nitrate, and their observations concerning the downward movement of nitrate through the vadose zone to ground water.

Students in the environmental modeling class used multivariate statistical methods to test the significance of differences between depths and differences between years. Attempts by the students to apply time-series techniques were frustrated by the uneven sampling intervals of the original study, providing valuable lessons on the need for rigorous design of experiments prior to collecting data, and the difficulties of working in real-world situations. Because no formal statistical testing had been done in the original study, students in the environmental modeling class are writing a paper for publication on the results of their analyses.