A month ago, Buffalo Creek behind my house stood at 5 feet, 8 inches deep. Last week it was below 2 feet, according to the U.S. Geological Survey, which operates the stream-flow and water-level gauge for Buffalo Creek.

A month ago, the flow in the creek was 588 cubic feet per second (4,400 gallons per second). Now it is down to 17 cubic feet per second -- only 127 gallons per second, 3 percent as much as a month ago, according to the USGS.

Since the creek is a sensitive indicator of rainfall over the 90-square mile drainage area, it is scary to think what might happen if this drought is extended -- or if changes in our climate are producing long-term changes in the rainfall pattern.

I just read an article by Pavel Ya. Groisman, and Richard W. Knightin the May 1 issue of the Journal of Climate in which they talk about "dry spells."

They say that if more rainfall comes in intense rainfalls during a period of climate warming, when there are longer growing seasons, then there will be more frequent dry spells -- long runs of days with little or no rain.

They ask, is this already happening in the United States? Their data show that in the eastern U.S., the number of such dry spells has more than doubled in the past four decades. It means that instead of a 30-day dry spell on average every 15 years, we will see such dry spells every six to seven years.

I decided to ask a similar science question about Buffalo Creek.

You can see the USGS measurements at http://waterdata.usgs.gov/nwis/uv?site_no=02039000. This from a data-rich Web site about water, whose home page is http://water.usgs.gov/. It reminds us that plenty of science data are available free on the Internet, ready for curious scientists to ask it questions.

From the Web site, I downloaded the daily stream data for the past 60 years into a spreadsheet. From there, I calculated runs of dry days, by looking at whether stream flow increased, stayed the same, or decreased. For the latter two, I assumed "no rain."

Looking at the resulting graph with plenty of annual variation, it is difficult to see much distinctive change at this one station. A trend line shows a slight increase in the length of the longest run of dry days, but the increase over 60 years is just two days, from 15 to 17.

But, such is science. Not all hypotheses are upheld; not all expectations are fulfilled. Not all questions have clean answers.

However, what is key is that many scientists, such as Groisman and Knight, are asking more and more sophisticated questions about climate change. In particular, they are asking about the implications of change -- in this case, the effect on food plants and agriculture from lengthening periods of drought, during warming, for plants in an extended growing season.

Such questions are the hallmark of scientists, whose task it is, in the words of physicist Richard Feynman, to find out, when something looks funny, just how funny is it.

Rivers, streams, and climate are covered in Virginia's science Standards of Learning in sections 4.6, 6.7, LS.11, ES.9, ES.11, ES.13. The simple experimental analysis of data from Buffalo Creek illustrates major goals in the SOLs: to enable students to make informed decisions about public policy issues, and to develop scientific thinking habits, including curiosity.

On the Web

The Buffalo Creek water monitoring station has live data every 15 minutes.

See a current map of drought areas in Virginia