Interception of precipitation by mesquite dominated rangelands in the rolling plains of Texas

Abstract

Interception of precipitation by mesquite (Prosopis glandulosa Torr.) canopy was analyzed using the rainfall, throughfall and stemflow data collected at the Wagon Creek Spade Ranch from August 1986 to March 1990. This ranch is located 22 km north of Throckmorton, Texas. The objective was to develop an in-depth understanding and predictive capability of the rainfall interception process. Canopy and foliar area of the sample trees were obtained using 35 mm slides and video images. An image of each sample tree was captured and analyzed by the computer software. The effect of storm and plant variables on the process of interception was analyzed by applying statistical procedures. Regression models were developed and the Gash model validated. The results of the study indicate that 80.9% of the rainfall was in the form of throughfall while 2.8% was in the form of stemflow. Application of stepwise regression determined that rainfall was the major factor contributing to the variation in throughfall (r^2 = 0.98) and stemflow (r^2 =0.71), followed by season, duration and intensity. Among the climatic factors affecting the interception process no significant difference for throughfall or stemflow was attributable to season, duration, intensity, nor to vegetation factors like canopy density, total tree height or stem circumference. Storm size was the principal determinant of throughfall and stemflow. Finally, the amount of rainfall not available to mesquite due to the interception process was calculated to be 16.3%.