A Study of Diurnal Circumference Changes in Trees by Students in Clark Atlanta University's Research Experiences for Undergraduates Program

Transpiration from trees has been a project of the National Science Foundation's (NSF) Research Experiences for Undergraduates (REU) Program at Clark Atlanta University (CAU) ever since CAU's Earth System Science Program (ESSP) made first contact with the local United States Geological Survey (USGS) office in the Fall semester 2000. Dr. Norman (Jake) Peters of the Norcross, Georgia, USGS office introduced ESSP scientists and students to the effects of transpirative pumping of stream water by trees on a diurnal timescale. Trees near the stream would draw water from the stream during the heat of day, thereby reducing the stream height, and return water to the stream late at night, which increased the stream height. The signal produced by this cyclical forcing took the form of a nearly perfect sinusoidal function.

ESSP CIRE (NSF Collaboratives to Integrate Research and Education) students during the Spring semester 2001 made first measurements of tree circumference in order to correlate the diurnal circumference changes with diurnal changes in stream height for nearby trees. Subsequently four generations of REU students during the summers of 2001-2004 attempted to measure by hand as accurately as possible the expected expansion and contraction of local trees due to the diurnal solar forcing. In 2004 a team of eight REU students made a suite of measurements for six sweetgum trees at the CAU Aiken field property which included tree circumference, cloud cover, air temperature, relative humidity, soil moisture, insolation, rainfall rate, soil moisture and temperature profiles to 0.4 m, and occasional radiative ground and tree temperatures. High soil moisture conditions existed for the three trees near the stream during much of the five-week measurement period due to the rainfall surplus experienced in the first half of this period.

The methodology used by the students to record five weeks of data was to divide the day into three shifts, with two students working per shift and the supervisors working their shift plus extra hours in the daytime. The students spent one week prior to data acquisition preparing the field site by hacking trails through the thick underbrush, laying hundreds of feet of power cable, designing lighting arrangements at the two data acquisition sites (upper and lower), setting up the atmospheric sensors, digging the hole and inserting the soil temperature sensors at precisely the same depths as the soil moisture sensors, programming and checking the automated dataloggers, designing a protocol for recording the hand measurements, designing quality control procedures, and making the mistakes that should not be repeated during the five weeks of concentrated observation and data acquisition. Once the observation period began hand data was measured once per hour on the half hour. Automated measurements were made for the first two weeks at an hourly interval, and afterwards at a one-minute interval.

Limited results have been obtained to date. About a dozen preliminary spectral analyses completed and evaluated between July 2004 and May 2005 do not show strong evidence of a diurnal signal for 128-point FFT (fast Fourier transform) decompositions. However, when a suite of 64-point, 128-point, and 256-point FFTs were run block-by-block for each tree, a convincing number of these did show a noteworthy diurnal component. An example of a 256-point spectrum is shown below. Bin 11 approximately represents a diurnal (24 hour) periodicity.

Other important discoveries were made on further analysis. When the data are analyzed with regard to the frequent rainfall events that took place during the first half of the five-week observation period, a spectral component is found that may best be explained by interference with the diurnal period from these rainfall events. In addition to this, mean values of tree circumference are seen to decay in a manner similar to the decrease in subsurface soil moisture for periods up to about 48 hours after each of the stronger rainfall events. The most significant of these results will be presented in the poster. Detailed results will be presented by co-author Bryan Woods in a research poster at the 5th Annual AMS Student Conference.