The hydrologic importance of grazing is receiving increased attention on rangelands in the United States. The literature on this topic is fragmented. This paper explores the available literature for information useful in understanding the hydrologic impacts of grazing intensity as related primarily to infiltration and runoff. Generally, data relative to range condition are not adequate for evaluating hydrologic impacts. Data relating grazing intensity to infiltration rates are available, yet distinct limitations are evident. These limitations are discussed in terms of identifying future research needs. The greatest need appears to be a detailed definition of the long-term effects of grazing (by year and season) on infiltration rates as a function of site, range condition, and grazing intensity. Once obtained, infiltration rates must be coupled with an appropriate method for generating runoff volumes, storm hydrographs, and long-term water yields.In general, rangelands (especially in the western United States) receive low precipitation, maintain sparse cover, are sometimes saline, have variable soils, and undergo a regular if not diffuse harvest. Classical overland flow from rainstorms, found to be infrequent on most forested lands, seems to occur with more regularity on grazing lands. The management of most rangeland is directed toward on-site conservation and use of the usually meager water resource, as contrasted with the espoused goal of water yield improvement from generously watered higher-elevation forest lands. THE HYDROLOGIC CYCLEExamination of the hydrologic cycle suggests that grazing impacts 'might be felt' throughout the surface and near-surface portions, i.e., interception, infiltration, runoff, groundwater recharge, and evapotranspiration. However, only the infiltration process has been studied enough to begin a preliminary analysis of any grazing impact on runoff. There is only limited Gifford et al., 1976]. This information is probably best suited for interpreting analyses conducted by using available (though limited) infiltration data. Thus the availability of infiltration data is both important and fortunate, since this process determines, indirectly, the water available for runoff, for recharging soil moisture, for deep seepage and groundwater recharge, and for plant growth. GRAZING INTENSITY AND RANGE CONDITIONGrazing activity and its effect are often described by either the grazing 'intensity' or the range 'condition.' Of these two descriptors, intensity (ungrazed, light, moderate, or heavy) is GRAZING GR ZING BEANS ENDS v • 0 TIME duration of the activity, regardless of the interim status. The path taken from A to B is unimportant. Boundary locations between the intensity classes are chosen locally, as are the key species. Paper number 8W0065.

Cowpies molded to a standard configuration and size were subjected to simulated rainfall, and the fecal coliform counts were determined using the most probable number (MPN) method of enumeration. The standard cowpie deposits were exposed to simulated rainfall once at ages 2 through 100 days. The effects of rainfall intensity and recurrent rainfall were also tested. Naturally‐occurring fecal deposits were also tested to compare their results with those from the standard cowpies. A log‐log regression was found to describe the decline in peak fecal coliform release with fecal deposit age. The 100‐day‐old fecal deposits produced peak counts of 4,200 fecal coliform per 100 milliliters of water. This quantity of release is minimal compared to the release from fresher fecal material. Rainfall intensity had little effect on peak fecal coliform release from fecal deposits that were 2 or 10 days old. At age 20 days the effect of rainfall intensity was significant; the highest intensity gave the lowest peak counts, and the lowest intensity gave the highest peak counts. The effect of rainfall intensity appears to be related to the dryness of the fecal deposits. Peak fecal coliform counts were significantly lowered when the fecal deposits were rained on more than once. This decline was thought to be produced by the loss of bacteria from the fecal deposits during the previous wettings. Standard cowpies produced a peak release regression that was not significantly different from the regression for the natural fecal deposits. Apparently, grossly manipulating the fecal deposits did not significantly change the release patterns.

This study evaluated the impact of selected soil surface characteristics on infiltration rates and sediment production from interrill erosion from loam soil. Treatments were two different grass species (crested wheatgrass and intermediate wheatgrass), three levels of grass cover (30, 50, and 80 percent), four levels of rock cover (5, 10, 15, and 20 percent), and six levels of simulated trampling (10 to 60 percent of the respective plot area by 10 percent increments). Results indicated that plots with sod forming grass infiltrated only slightly more water than plots with bunchgrass, though the differences were significant. Trampling reduced infiltration rates significantly. On uncompacted soil, infiltration rates increased as percentage of rock cover increased. Trampling gradually destroyed this relationship however. Rock cover did not significantly affect sediment production. The tradeoff between vegetal cover and rock cover was affected by simulated trampling. Once trampling disturbance reached 20 percent, no relationship between vegetal cover and rock cover existed. Trampling was the most important factor influencing infiltration rates, explaining 35 to 48 percent of the variation in infiltration rates. The most important factor influencing sediment production was grass cover, which explained 40 to 62 percent of the variations associated with sediment yield at various trampling percentages. Results strongly suggest that, for slopes and soils as used here, adequate watershed protection may be obtained by maintaining 50 percent protective ground cover. Additional validation studies are recommended.

The study examined the magnitude of a release of indicator bacteria (fecal coliform) from bovine fecal deposits that were rained on by a rainfall simulator at a rate of 6.1 ± 0.3 cm/h for 15 min, as affected by duration of rainfall and age of fecal deposits. Standard fecal deposits were placed on a platform and rained on with the runoff water being sampled at 5, 10, and 15 min. Samples were then examined by the most probable number (MPN) method for the presence of fecal coliforms.Results indicate the potential for bacterial pollution from bovine fecal deposits. An equilibrium in the concentration of fecal coliforms being released from the fecal deposit was reached within 10 min. Fecal deposits < 5 d of age released fecal coliforms on the order of millions/100 mL of water. Concentrations declined to 40,000/100 mL at 30 d of not‐rained on age. The decline followed a typical bacterial death curve.

Highlight:During 19 73 and 1974 wildland water quality analyses were performed on a semiarid, chained and seeded, pinyon-juniper site in southeastern Utah. The area was treated in 1967 and protected from grazing until 1974. In 1974 livestock grazing was introduced and investigations continued to determine if any deleterious land use effects were present from fecal contamination by cattle. No significant changes were noted in fecal and total coliform production (fecal pollution bacterial indicators) from grazing use. There is an element of risk involved whenever data generated from a small area are projected to larger land areas. However, it appears that this level of livestock grazing (2 hafA UM) did not constitute a public health hazard in terms of fecal pollution indicators onthe semiarid watershed. JOURNAL OF RANGE MANAGEMENT 29(2), March 1976

The study examined the magnitude of a release of indicator bacteria (fecal coliform) from bovine fecal deposits that were rained on by a rainfall simulator at a rate of 6.1 ± 0.3 cm/h for 15 min, as affected by duration of rainfall and age of fecal deposits. Standard fecal deposits were placed on a platform and rained on with the runoff water being sampled at 5, 10, and 15 min. Samples were then examined by the most probable number (MPN) method for the presence of fecal coliforms. Results indicate the potential for bacterial pollution from bovine fecal deposits. An equilibrium in the concentration of fecal coliforms being released from the fecal deposit was reached within 10 min. Fecal deposits < 5 d of age released fecal coliforms on the order of millions/ 100 mL of water. Concentrations declined to 40,000/100 mL at 30 d of not-rained on age. The decline followed a typical bacterial death curve.

Interception patterns of big sagebrush (Artemisia tridentata Nutt.) and shadscale (Atriplex confertifolia (Torr.) Wats.) were measured under two simulated rainfall intensities during three different seasons, Mean rainfall interception rate of individual plants of both species was 0.1.5 cm when averaged over all sampling dates and rainfall intensities. Interception during individual storms of at least 0.15 cm size by entire plant communities, based on measured vegetal cover, was calculated at 0.028 cm or less. On the average, about 4% of the total annual rainfall (not snowfall) would be intercepted by these plant communities.

No abstract