Livestock exclusion and belowground ecosystem responses in riparian meadows of Eastern Oregon

Ecological restoration of riparian zones that have been degraded by decades of overgrazing by livestock is of paramount importance for the improvement of water quality and fish and wildlife habitats in the western United States. An increasingly common approach to the restoration of habitats of endangered salmon in the Columbia Basin of the Pacific Northwest (USA) is to exclude livestock from streamside communities. Yet, few studies have examined how ending livestock grazing changes ecosystem properties and belowground processes in herbaceous-dominated riparian plant communities (meadows). Along the Middle Fork John Day River, Oregon, we compared ecosystem properties of dry (grass and forb-dominated) and wet (sedge-dominated) meadow communities at three sites that had been managed for sustainable livestock production with three sites where livestock had been excluded for 9-18 years as a means of riparian and stream restoration. Profound differences in the belowground properties of grazed and exclosed communities were measured. In dry meadows, total belowground biomass (TBGB consisting of roots and rhizomes) was similar to50% greater in exclosures (1105 and 1652 g/m(2) in the grazed and exclosed sites, respectively). In exclosed wet meadows, the TBGB was 62% greater than in the grazed sites (1761 and 2857 g/m(2), respectively). Soil bulk density was significantly lower, and soil pore space was higher in exclosed sites of both meadow types. The mean infiltration rate in exclosed dry meadows was similar to 13-fold greater than in grazed dry meadows (142 vs. 11 cm/h), and in wet meadows the mean infiltration rate in exclosures was 233% greater than in grazed sites (24 vs. 80 cm/h). In exclosed wet meadows, the rate of net potential nitrification was 149-fold greater (0.747 vs. 0.005 mug NO3-N-[g soil](-l).d(-1)), and the rate of net potential mineralization was 32-fold greater (0.886 vs. 0.027 mug N.[g soil](-l).d(-1), respectively) when compared to grazed sites, though changes observed in dry meadows were not significant. Livestock removal was found to be an effective approach to ecological restoration, resulting in significant changes in soil, hydrological, and vegetation properties that, at landscape scales, would likely have great effects on stream channel structure, water quality, and the aquatic biota.