Vegetation Diversity and Soil Physico-chemical Properties Under Traditional Management of Rangeland in Eastern Ethiopia

The changes of herbaceous vegetation (composition, diversity, richness, evenness, and biomass production) and soil physico-chemical attributes of Harshin rangelands, eastern Ethiopia were examined under continuously open grazed areas and three ages of enclosures group arranged along chronological sequence into: < 5 years (young), 5-10 years (mid) and > 10 years (old). The herbaceous species composition showed a clear pattern of difference between the enclosure and open access grazed areas with desirable species being more prevalent in the enclosures and the undesirable ones being dominant in the open access grazed areas. The principal component analyses (PCA) of herbaceous species composition were accounted distinct across the open grazing young, mid and old enclosure management types. The diversity, richness and biomass production of herbaceous species were significantly different in the study area. The principal component analysis (PCA) of soil physico-chemical attributes showed distinct separation in relation to open grazing and three age enclosure areas. Overall, we found that the increased biomass production in the old enclosures may threat the herbaceous species diversity by dominating by only few species and hence affected the plants which are susceptible for species inter-competition.


Introduction
Quantifying the impact of livestock grazing on natural communities (forages) has become a major issue in the management of rangelands especially where grazing is very widespread and its impacts may be in conflict with conserving biodiversity (Friedel, Stuart-Hill, & Walsh, 2004).Grazing animals may exert beneficial or mutual influences on the vegetation for their own good (Kamau, 2004).Livestock can exert a considerable change on the diversity, composition, structure, and development of native plant communities (Hailu, 2017).However, the degree of change is highly dependent upon the ecosystem and plant community, the current environmental conditions, and the intensity and timing of grazing (Milchunas & Lauenroth, 1993).Large concentration of grazing animals may often have harmful effects on the plants because of selectivity and overgrazing (Kamau, 2004).Much of the literature indicates that the change has been more drastic and evident in ecosystems where native grazing ungulates were historically scarce or absent (Mcginty & Banner, 2009).Grazing increased, reduced or lacked consistent effect on plant diversity (Kamau, 2004).These contrasting patterns of response have frequently been attributed to differences in grazing intensity, with greatest diversity expected at intermediate level of grazing (Grime, 1973).Usually diversity is low in environments with very low availability of resources (i.e.where few species can survive) and increases with increasing resource availability (Hailu, 2017).The impact of grazing on diversity differs along gradients of primary productivity (Milchunas & Lauenroth, 1993), but there is no general consensus about the process involved in this interaction.Grazing intensity affects the quantity and quality of organic and mineralized materials that are delivered to the litter and soil as vegetative residues, urine, and manure (Abule, Snyman, & Smit, 2007).Grazing also affects the quality of herbage in the sward and this ultimately impacts degradation rate of litter, soil organic matter, and soil nutrient mineralization rates (Allsopp, Laurent, Debeaudoin, & Igshaan Samuels, 2007).Kamau (2004) predicted that grazing could change diversity in opposite ways in resource-poor versus resource-rich ecosystems.
Vegetation changes in response to livestock removal are occurred on prolonged temporal scales in semi-arid rangelands (Fuhlendorf, Engle, Elmore, Limb, & Bidwell, 2012).The rangelands in eastern Ethiopia have been exposed to extensive livestock grazing pressures which have led to a reduction in vegetation cover and preferred grass species, depletion of soil nutrients and accelerated erosion (Kassahun, Snyman, & Smit, 2008).In response to these declining rangeland resources, establishment of enclosures in communal rangelands used by pastoralists is a widespread practice in eastern Ethiopia.Napier and Desta (2011) outlined the types of rangeland enclosures in Somali region into three forms of enclosures: Private, Cooperative and community (established by community, NGOs and government facilitation).The enclosures used in the present study were established privately by individual pastoralist.According to Napier and Desta (2011), the driving factors for the emergence of private enclosures in the region are to secure dry season reserve for livestock in times of increasing scarcity of grazing species.Management of the enclosures are similar to some East African pastoralists and agro-pastoralists , which is protected from livestock grazing during the wet season and served as dry-season reserve pasture for an individual's own livestock (Angassa & Oba, 2010;Verdoodt, Mureithi, & Van Ranst, 2010).
Differences in plant diversity inside and outside of enclosures were observed for several community types in the Serengeti (Briske, Fuhlendorf, & Smeins, 2003).Diversity declined in all enclosure while equitability dropped sharply, although species richness was not significantly different (Kamau, 2004).Milchunas, Sala, and Lauenroth (1988) attributed this to replacement by tall species in ungrazed areas formerly dominated by short species when the area was being grazed.Pattern diversity, rather than point diversity was found to be the major contributor of stand diversity.The exclusion of domestic livestock from grazed rangeland often initiates secondary succession.If left undisturbed this succession may culminate in a plant community with a relatively stable composition (Brand & Goetz, 1986).Range managers have often used enclosures, cemeteries, churchyards and other areas inaccessible to livestock to determine the potential for improvement on grazed rangeland.According to Woldu and Mohammed Saleem (2000), floristic richness and the efficiency of the vegetation as an energy trapping system tends to decrease as aboveground biomass becomes concentrated in a few species.The opposite apparently occurs in the grazed sites.Hence, a greater aboveground herbaceous production and a greater grass height characterize ungrazed vegetation (Kamau, 2004).
The change of vegetation and soil physico-chemical characteristics following the establishment of enclosures from extensively managed rangelands in eastern Ethiopia has been a controversial issue.Some studies documented improvements in vegetation, soil and water infiltration inside enclosures (Yayneshet, Eik, & Moe, 2009).The previous research in the present study area by Haftay, Yayneshet, Animut, and Treydte (2013) also noted the enclosures aged from five to eight years have positive vegetation changes in herbaceous species diversity and biomass production.In contrast, other studies reported site-specific and minor differences between protected and adjacent grazed areas (Brand & Goetz, 1986).This indicates the need to interpret results of studies involving short and long-term livestock removal from rangelands in relation to specific influential factors (Simons & Allsopp, 2007).However, researches on the herbaceous vegetation and soil physico-chemical changes following the establishments of rangeland enclosures management practices and the changes among the different enclosure ages in the semiarid rangelands of eastern Ethiopia are limited.Therefore, this study was designed to elucidate the changes of herbaceous vegetation (composition, diversity, richness, evenness, and biomass production) and soil physico-chemical characteristics among continuously open grazed areas and three chronological sequence age enclosures (< 5 years, 5-10 years and > 10 years).

Description of the Study Area
The study was conducted in the Harshin rangelands of the Somali Regional State in Eastern Ethiopia located in 9°12′ N and 43°31′ E at about 950 km east of Addis Ababa (Figure 1).The rainfall in the district is bimodal with a short rainy season from April to May, and the main rainy season from June to August.The average minimum temperature is about 20 o C and the average maximum temperature is 35 o C. The average annual rainfall is 560 mm (Haftay et al., 2013).jas.ccsenet. .The mula

Selecti
(2) ) was (3) where, S is the total number of species.Species richness was also determined by counting all species with in the sample plots.Herbaceous aboveground biomass was estimated by harvesting at ground level using sickle from each 1 m 2 quadrates.The cut samples were weighed immediately and 20-30% of the total weight retained for determination of dry matter content (oven drying at 105 °C for 24 hours. Soil samples (to depth of 10-15 cm) were taken using an auger from each study plot at five nested quadrates per plot.Each set of five soil samples was mixed properly, air dried and passed through 2 mm sieve to make ready for analysis.Texture was determined using hydrometer method.Soil pH was measured using a 1:2.5 soil water relation extract method.Soil organic carbon was determined using walkey-black method (Baxter, Hastings, Law, & Glass, 2008).The percent soil organic matter (SOM) was calculated by multiplying the percent organic carbon by a factor of 1.724 (Brady, Smith, Beam, & Cravotta III, 1990).Total nitrogen was determined by the Kjeldahl method as described by (Campbell & Racz, 1975).The exchangeable cations, viz sodium (Na), potassium (K), and calcium (Ca) were extracted with ammonium acetate and analyzed using atomic absorption spectroscopy (Brady et al., 1990).Available phosphor (P) was analyzed using Olsen method (Campbell & Racz, 1975), at Haramaya University Soil Chemistry Laboratory.
Data collected within each plot were averaged for statistical analysis.Herbaceous species composition was subjected to principal component analysis (PCA) ordination to explore the main components variation across the management types.Frequency and desirability of species composition among the open grazing area, young, mid and old ages of enclosures was analyzed using descriptive statistics.The distribution of species between open grazed and enclosures area were computed using the Sørensen's Similarity Index: where, ISS is Sørensen's Similarity Index; a representing the number of species unique to sample a; b representing the number of species unique to sample b; and c representing the number of species in common.Data on vegetation diversity, richness, evenness, biomass production were subjected to one-way ANOVA to determine the changes by management type (open grazing area, young, mid and old ages of enclosures) on vegetation after Levene's equality of variance test using SPSS (version 20) software (Norusis, 2010).Differences were considered significant at P < 0.05.Soil physico-chemical attributes were subjected to principal component analysis (PCA), to explore the main components of variation in the soil attributes among the management types.Principal component analysis (PCA) ordination of vegetation and soil physico-chemical attributes was conducted using PAST (version 3) software (Ryan, Hammer, & Harper, 2001).

Herbaceous Species Composition
The principal component analyses (PCA) of herbaceous species composition were accounted distinct separation (PC 1 = 36.05%;PC = 32.90%)across the open grazing young, mid and old enclosure management types (Table 1).A total of 37 grass and forbs were identified in the study area (Table 1), of which 19%, 36%, 21% and 24% were highly desirable, desirable, less desirable and undesirable herbaceous species, respectively.From the identified herbaceous species about 38% and 62% were found in the open grazing and enclosure areas respectively (Table 2).
Sørensen's Similarity Index between the enclosures and the open access communal grazing areas were 0.48.From the identified perennial herbaceous species about 62% were found in the enclosure areas.However, the numbers of annual herbaceous species in the open grazing and enclosure areas were the same.A total of 29 herbaceous species were found in the three ages category of enclosures.Of these 65.5%, 55.2% and 48.3% were found in the young, mid and old enclosures respectively.Note.Young = < 5 years; Mid 5-10 years; Old = > 10 years. jas.ccsenet.
Highly desirable grass species such as Cenchrus ciliaris, Cynodon dactylonn and Panicum atrosanguineum were found in the enclosures and open access grazed areas.The presence of highly desirable species like C. dactylonin the open access grazed areas is because of their creeping nature of growth (Abate, Ebro, & Nigatu, 2012;Gebremeskel, 2018).However, most of the vegetation species dominated the enclosures areas where undesirable forbs and less desirable grasses species like Setaria verticillata, Sporobolus marginatus, Hibiscus macranthus and Indigo feraarrecta dominated the open access communal grazing area.Therefore, overgrazing in the study area might have led to a loss of grass species that have important feeding value.Gemedo-Dalle, Maass, and Isselstein ( 2006) noted that the limited spatial distribution of the highly desirable forage grasses with grazing pressure might be an indicator of the deteriorating condition of rangelands.
The species composition based on frequency of occurrence of individual herbaceous species was generally low among the age of the enclosures.Grass species of Bothriochloa insculptawas the most dominant in the young age enclosure areas.While grass species of Aristida adoensis, Eragrostis superb, Cenchrus ciliaris and Bothriochloa insculpta were dominated in the mid age enclosure.In the old enclosures grass species of Aristida adoensis, Cynodon dactylon and Digitaria abyssinica were dominated.
Herbaceous species diversity, richness and biomass in the enclosures were greater than the openly grazed areas.Aboveground herbaceous biomass in the enclosures was more than twice of the adjacent grazed areas.This is attributed to the presence of high and unregulated grazing pressure on the freely accessed areas.Tefera, Snyman, and Smit (2007) reported that rangeland vegetation dynamics in the arid and semi-arid African rangelands have been subjected to controversy on equilibrium and non-equilibrium theories.Note.Young = < 5 years; mid 5-10 years; Old= > 10 years; A = annual, P = perennial; HD = highly desirable, D = desirable, LD = less desirable.UI = unidentified, UD = Undesirable.
In equilibrium models vegetation dynamics are regulated by livestock grazing, and the vegetation moves in a predictable manner along a series of livestock densities showing a strong relationship between stocking rate and rangeland productivity (Briske, Bestelmeyer, Stringham, & Shaver, 2008;Briske et al., 2003).The vegetation dynamics by the effect of management in the present study may exhibit typical of equilibrium model of vegetation change.With moderate levels of stress both species resistant to stress and the species susceptible to stress are able to survive and reproduce resulting in maximum species diversity and biomass productivities.Holechek, Gomes, Molinar, Galt, and Valdez (2000) noted that, in equilibrium model, grazing is a root cause of rangeland degradation where heavy stocking occurs.Inline to our study, Angassa and Oba (2010), and Haftay et al. (2013) had reported herbaceous species diversity and richness were higher in the enclosure than adjacent open access communal grazing areas.The high diversity, richness and biomass measured in the enclosures might be explained by the increased litter accumulation, improved soil organic matter and other nutrients inside the enclosures (Angassa, Oba, Treydte, & Weladji, 2010).In the present study area high soil organic matter was found in the three ages of enclosures in relative to open access communal grazing areas.
In the present study, the vegetation change among the age of enclosures may be due to species inter/intra-competition.The herbaceous species diversity and richness were declined with increase age of the enclosures (i.e., young enclosures had more diversity than the old enclosures), in inverse, the herbage biomass increases with increases the age of enclosures.The moderate level of species competition at the early regeneration maybe attributes to survive and reproduce maximum species diversity and richness in the young enclosures, while the increased biomass production in the old enclosures may able to threat the herbaceous species diversity and richness by dominated only few species and affected the plants which are susceptible for species inter-competition.Zhang et al. (2005) suggested that short-term exclusion promoted herbaceous species richness, while long-term resting was not beneficial.The decline in species diversity and richness at a high biomass level is a major concern for rangeland conservation and management (Holmes, 2002).Grime (1973) hypothesized that species diversity could be plotted as a bell-shaped curve along gradients of environmental stress, as environmental stress increases, the species adapted to low-environmental stress lose their competitive advantage and species more resistant to environmental stress are allowed to increase in abundance.

Conclusions
Herbaceous species composition was accounted distinct separation across the open grazing, young, mid and old enclosure management types.High number of herbaceous species was identified in the open grazing area with 36.05% components variation.The highest diversities were recorded in the young and mid enclosures respectively.Aboveground herbaceous biomass in the old enclosures was more than three times from the open grazed areas.Overall, we found that increases the age of enclosures allowed maximum herbs biomass production that support the local pastoral economy in the Harshin rangelands of eastern Ethiopia.However, the increased biomass production in the old enclosures may able to threat the herbaceous species diversity by dominated only few species and affected the plants which are susceptible for species inter-competition.The soil physico-chemical attributes were accounted distinct separation among the open grazing and three age enclosure areas.From the selected soil attributes available phosphor (P), cations electrical conductivity (CEC) and calcium (Ca) were explored high variation components, contributed as sample plots with higher soil P content to sample plots high soil CEC and sample plots with greater Ca to sample plots with soil P and CEC. Figure

Table 1 .
Principal component analysis (PCA) ordination of herbaceous species composition among open grazing area and young mid and old enclosure ages in Harshin rangelands of Eastern Ethiopia

Table 2
the management systems.The species of Cenchrus ciliaris, Setaria verticillata and Indigo feraarrecta were having high percentage composition in the open grazed communal areas while species Aristida adoensis, Cenchrus ciliaris and Cynodon dactylon were having a high percentage composition in the enclosure areas.The desirability of herbaceous species composition showed a clear pattern of difference between the enclosure and open access grazed areas with desirable species being more prevalent in the enclosures and the undesirable ones being dominant in the open access grazed areas (Table

Table 3 .
Percent composition of individual herbaceous species (1/m 2 ) among open grazing area and young mid and old enclosure ages in Harshinrangelands of Eastern Ethiopia