Infiltration, sediment, and erosion under grass and shrub cover in the southern high plains

On most rangeland watersheds, water is the limiting factor to plant production. Also, water that does not infiltrate and becomes surface runoff is the primary force for erosion on those watersheds. Arid and semiarid environments in the southwestern United States and northern Mexico have significant spatial and temporal variations in hydrologic and erosion processes. Different water and soil losses can occur at zones under plants, bare ground areas, and transition zones between plants and bare ground. Type and amount of vegetation are important factors in maintaining high infiltration rates and lower interrill erosion on rangeland watersheds. Two objectives were stated for this study: (1) to estimate spatial differences of infiltration rates, sediment concentration in runoff, and interrill erosion under grass and shrub cover in the Southern High Plains, and (2) to find soil and plant characteristics that influence infiltration, sediment concentration in runoff, and interrill erosion.

The research was conducted on the rangeland portion of the Texas Tech University Campus and on the Muleshoe National Wildlife Refuge, during the summers of 1992, 1993, and 1994. Treatments included clumps of blue grama {Bouteloua gracilis [H.B.K.] Lag. ex Griffiths), clumps of sideoats grama (Boutloua curtipendula [Michx.] Torr.), coppice dunes of broom, snakeweed {Gutierrezia sarothrae [Pursh.] Shirmers), transition areas of blue grama-bare gound, sideoats grama-bare ground, and broom snakeweed-bare ground, and bare ground areas.

A portable simulator was used to apply a rainfall with an intensity of 16.5 cm/hr during 40 minutes over 0.15 m^ plots. Runoff was collected at each 5-min interval, measured by volume, and transformed to runoff rates. Infiltration was estimated as the difference between applied rainfall and the collected runoff. Sediment concentration in runoff was determined in each interval by taking a sample of the collected runoff. These samples were filtered to collect sediment, oven-dried, and weighed. Interrill erosion was estimated by using the sediment concentration in runoff and plot areas, and from the amount of runoff. Finally, selected plant and soil variables were estimated before and after rainfall simulation for every plot.

A split-plot design with two factors (amount of canopy cover and time) and five replications were used to find difference between treatments on infiltration rates, sediment concentration in runoff, and interrill erosion. Tukey's comparison test was used to separate treatments means. Partial correlation and multiple regression analyses were used to find soil and plant variables affecting infiltration rates, sediment concentration in runoff, and interrill erosion.

In general, results showed that infiltration rates were greater in coppice dunes of broom snakeweed, foUowed by clumps of blue and sideoats grama, transition areas of blue and sideoats grama, transition areas of broom snakeweed-bare ground, and bare ground. Sediment concentration in runoff had a high variability between and within treatments, sediment concentration usually reached a greatest amount within the first 10 minutes and then decreased continuously in all treatments. Interrill erosion was lower under coppice dunes of broom snakeweed, followed by clumps of blue and sideoats grama, transition areas of blue and sideoats grama-bare ground, transition areas of broom snakeweed-bare ground, and bare ground of interspaces between plants. Results of the partial correlation and regression analyses indicated that bare ground, basal cover of grasses, cryptogams cover, sand content, clay content, and organic carbon were the most influential factors affecting infiltration rates, sediment concentration in runoff, and interrill erosion in the two shortgrasses.