Scientist Report - 10/29/96 update

Hydrology Update, October 31, 1996
Analysis of Phase 1 Data

Dear Students,

Our Hydrology team is now busy analyzing the GLOBE Phase I water temperature and pH data that were collected by over 509 schools beginning in April 19th, 1995. First, I would like to introduce graduate student Chris Gutmann, who is is working with me on the analysis.

Chris graduated from Middletown High School in Middletown, CT. He then went to attended Carleton College in Northfield, MN, getting a B.A. degree in Geology. Chris has been studying hydrology here at the University of Arizona since August, 1995. When not working on GLOBE, Chris is taking classes here at the university, studying the water quality in the Tokapah basin of Sequoia National Park, or occasionally playing in Ultimate frisbee tournaments.

Chris has started by dividing the whole data set into regions, defined by surface geology. We do this because the geology of an area is a major factor affecting the chemical composition of water. So far the groupings include the following areas:

He has plotted all of the data as time series graphs, like the examples below. Before we can discern trends and compare data from different stes, we are going through the data carefully and looking for outliers. For example, notice in the Sutter, CA graph below, that one temperature reading lies outside the range of the others. This is probably an error, and we will remove this point before continuing the analysis.

In addition, pH readings that deviate significantly from the average are suspect. For the Middletown, CA graph, note the single pH 4 reading, with the rest of the pH's being in the 6-9.5 range.

Some additional items of interest can be seen looking at these graphs. The graph from Middletown, CA shows what appears to be a pH trend gradually climbing over the course of the record. The pH's seem to be more scattered than would be expected. Why do you suppose this is the case? At the site of the school in Sutter, CA, we see a more typical variation in pH values, with a gradually increasing trend. This might be a problem associated with a buffer solution that was losing its accuracy, or it might actually represent a real pattern in nature! We encourage schools to look at their own data, and to try to minimize the possibility that calibration error or measurement inconsistancies may be influencing data.


Chris is making great progress, and hopes to have another report ready in early November. If you have any questions to ask about the data set and how the analysis is going, feel free to send email to Chris (cgutmann@hwr.arizona.edu) or me (roger@hwr.arizona.edu).

Sincerely,

Roger Bales
Professor of Hydrology & Water Resources
University of Arizona