Species Diversity and Distribution Patterns of Woody Plants in Adaba-Dodola Afromontane Forest, Oromia, Ethiopia

Abstract

p = 0.014). Jaccard coefficient of similarity showed that there was 73% of dissimilarity between standing vegetation and soil seed bank. The poor species correspondence might be emanated from many reasons. It could be due short-lived seed bank as a result of lack of dormancy mechanisms to persist in the soil, or due to endocarp dormancy (e.g. Juniperus procera and Olea europaea species) that prevented early germination, or high seed predation by animals (insects) due to fleshy nature of seeds that attract predators (e.g. Juniperus procera, Ekebergia carpesis and Podocarpus falcatus seeds) or poor viability due to soil borne fungal diseases attacks (e.g. Podocarpus falcatus

recent study of seed viability in Podocarpus falcatus showed 2.1% viability). Generally, based on soil seed bank evaluation, about 68 % of the woody species of the above ground vegetation was not represented in the seed bank. Based on the regeneration analysis, about 41.27 % of the total species was not regenerating at all, 25.40% species had fair regeneration and 9.52 % had poor regeneration status in Adaba-Dodola Afromontane forest. Moreover, the ecological dominance analysis also revealed that about 81% of the total woody species require immediate conservation priority. The important findings of the present study were that the joint effect of anthropogenic disturbances and environmental factors had been demonstrated to be a likely mechanism shaping the overall elevational pattern of species distribution, richness and diversity both in aboveground vegetation and soil seed bank. Effect of disturbance, especially cutting was found to be the main factor for the decline of species diversity at low elevations while that of the upper end elevation zone was mainly related to environmental factors rather than the disturbance activities. This implied that the observed patterns of vegetation were not attributed to a single universal explanation but rather to a combination of environmental and anthropogenic factors. Thus, it is highly important to take anthropogenic effects into account when describing ecological processes along elevational gradients. As a general conclusion, depending only on natural regeneration might not be an effective strategy for restoration of the degraded Adaba-Dodola Afromontane forest implying that enrichment planting with adaptable seedlings is required for successful vegetation restoration, especially for species having low importance value index and poor regeneration status. The forest restoration or conservation efforts need to focus both on maintaining the quantity as well as the quality (species composition) of species to diversify the species-poor sites. Mitigation of grazing and harvesting disturbances should be considered as one of the main measures that need to be taken into account in forest conservation planning in the study site.

Adaba-Dodola Afromontane forest is one of the biodiversity priority areas of Ethiopia. Nevertheless, this forest is among the most degraded forests in Ethiopian Afromontane forests with remnant fragmented forest patches. Mitigation and restoration of the disturbed forests are taken as priority task for biodiversity conservation and maintaining ecosystem functions. To generate relevant information that helps for maintaining the remaining forest, this study was aimed to explore how the forest species composition, distribution pattern, richness, diversity, structure and regeneration varied with the environment and anthropogenic influences and to evaluate the relative importance of these explanatory variables at different elevation zones. Three forest sites were considered for the study. 90 plots (20 m × 20 m) were sampled along elevational gradients from two transects of about 1 km apart from all sites. Soil seed bank germination test was performed in greenhouse for six months. Two multivariate analysis models, Canonical Correspondence Analysis (CCA) and Redundancy Analysis (RDA) models, were performed to examine the relationships between vegetation patterns and explanatory variables. Species diversity indices were used to analyze the pattern of species diversity, richness and evenness. The CCA/ RDA in CANOCO and stepwise multiple regression models were used for variation partitioning among explanatory variables. Analysis of variance (ANOVA) was used for means separation among the significant response variables. Species-environment ordination computed by CCA/RDA indicated that the variations in woody plant communities were explained by environmental and disturbance factors. The environmental and anthropogenic disturbance data sets in the entire vegetation, respectively explained 19 % (p = 0.005) and 26 % (p = 0.005) of the total variations (TVE) in species diversity and 37 % (p = 0.045) and 24 % (p < 0.045) of the total variation in species richness. The species richness (R2 = 0.36

p < 0.006) and diversity (R2 = 0.47

p < 0.001) showed a pronounced shift at mid elevation gradient exhibiting a hump shape pattern. The mid elevation maximum richness and diversity patterns were presumed to be emanated from many factors as indicated by multivariate models and analysis of variance. One of the main reasons might be the less anthropogenic activities such as cutting, grazing, settlement and canopy openness as compared to the low and high elevation sites. The less anthropogenic disturbances in the mid elevation in turn were attributed to the high topographic roughness (steep and undulating slope) that might have limited the ease accessibility of the forest for anthropogenic activities. Moreover, the middle elevation is the transitional zone between the lower and the upper zones that might have enabled the site to share some climatic conditions from both elevation zones. According to our climate data, this site was supposed to have moderate climatic conditions (1059 mm mean annual rainfall and 11 oC mean annual temperature) that might be favorable to sustain more number of species to coexist. At low elevation, the variance partitioning models indicated that pure anthropogenic factor exerted significant (p <0.005) effect on the species diversity while pure environmental factor did not have significant effect. Cutting disturbance had the most pronounced effect on species diversity at low elevations. However, at the upper end of elevation, the multivariate model analysis indicated that only pure environment had significant influence on the decline of species diversity implying that decline of diversity at the upper end elevation zone was mainly attributed to environmental factors rather than the anthropogenic disturbances. The cool temperature at high elevation (1.7 oC mean annual minimum temperature) might have accounted for the decline of species richness and diversity by slowing the growth and regeneration rate of plants. In the soil seed bank, the largest variation in species diversity was explained by cutting (R2adj =10.8 %

p = 0.038) followed by soil pH (R2adj = 5.1%

p= 0.029) and aspect (R2adj = 5.0%