Date of Award:


Document Type:


Degree Name:

Doctor of Philosophy (PhD)


Wildland Resources

Department name when degree awarded

Rangeland Resources

Committee Chair(s)

D. Layne Coppock


D. Layne Coppock


Allan Falconer


Brien E. Norton


Neil E. West


Helga Van Miegroet


Phil Urness


The dominant habitat-type of Omo National Park (ONP), Ethiopia, is grassland. This grassland supports a variety of wild herbivores and indigenous people; the latter hunt large herbivores for subsistence or graze their domestic livestock in the Park. Therefore, an understanding of grassland dynamics is a high priority for ONP management. Grazing and fire are major factors influencing species composition and vegetation change in East African grasslands. Rainfall regime, in turn, can influence both grassland response to grazing and fire. The ONP grasslands occur along a rainfall gradient ranging from subhumid to semiarid. Research objectives were to: (1) test the Milchunas, Sala, and Lauenroth grazing response model at three sites along the rainfall gradient; (2) determine how basal cover for dominant perennial grasses along the rainfall gradient was influenced by an extended period of intensive defoliation; and (3) determine if current distribution of grassland species associations was related to fire frequency.

Species composition changes associated with defoliation were consistent with predictions of the Milchunas, Sala, and Lauenroth model; large changes occurred in subhumid grasslands while minimal changes occurred in semiarid sites. Furthermore, an intermediate response was found at the site that received an intermediate level of rainfall. Basal cover of dominant grasses was not adversely affected by intensive defoliation, and this tolerance was expressed under all three moisture regimes. Significant reductions in basal cover were found, however, in >50% of dominant grasses as a result of protection from grazing and fire. Estimated fire frequency pattern was unrelated to the distribution of species associations within grasslands.

Conceptual models of vegetation change were developed for the three grasslands using research results. At the current level of management these models provide information that can be used to improve relations between Park staff and local pastoralists. Given the tolerance of most perennial grasses to intensive defoliation, ONP managers could modify the current policy of no grazing within the Park to one of limited grazing. If management capabilities were to improve, conceptual models could guide an active approach to grassland management.