The vegetation composition, structure and regeneration status of Gole Natural Forest, West Arsi Zone, Oromia Regional State, Ethiopia

This study was conducted in Gole natural forest (Dodola) West Arsi Zone of Oromia Regional State, Ethiopia. The study was intended to investigate the vegetation composition, structure, community types and the regeneration status. To collect the vegetation data, systematically 62 plots 20 m × 20 m (400 m2) were established at 100 m interval, starting from the top of the mountain. Tree and shrub species were counted and their cover abundance value was estimated. The data for herbaceous species were collected from five 2 m × 2 m sub-plots laid at the four corners each and one at the centre of the main plot. Height and diameter at breast height (DBH) of all woody species taller than 1.5 m and thicker than 2 cm were measured. R package was applied for cluster analysis. Indicator species analysis was performed in R Interpolated species accumulation curves. Estimate S 8.2 Software and Microsoft Excel were used to analyze the data. Rarefaction was applied to compare the species richness of the plant communities in the study area. Sorensen’s similarity coefficient was used to detect similarities and dissimilarities among communities.

A total of 114 plant species belonging to 57 families and 94 genera were identified. The most dominant families were Asteraceae, followed by Acanthaceae and Lamiaceaae. Out of 114 species 17 were endemic to Ethiopia. The study showed that high density was seen at lower height and DBH classes. Five plant community types were identified. The rarefaction revealed that there is difference in species richness among communities. The Sorensen’s similarity index showed that, there was a difference in the distribution of plant species composition among the five plant communities.

Methods
Systematically, quadrats of 20 m x 20 m (400 m2) were established. The first plot was established randomly starting from the top of the mountain to the lower side, and then the remaining plots were established 100 m interval along transect lines. A total of 62 quadrats were laid for vegetation data collection. Trees and shrubs were collected from the main plots. Five subplots of 2 x 2 m one at the center and four at the corner of the main (20x20m) plot were laid to collect data of herbaceous plants (Hailemariam MB. and Temam TD., 2018). Voucher specimens were collected for all plant species and were identified at National Herbarium (ETH). Environmental factors (altitude) and geographical coordinates were measured using Garmin GPS in the middle of the main plots.  The types of disturbance were recorded for each plot. Disturbance could be grazing, number of trees and shrubs cut, number of foot trails, and number of seedlings trampled. The intensity of anthropogenic disturbance (grazing) in each plot was estimated as a sum (cumulative effect) of the following scale: 0 = no disturbance, 1 = slightly disturbed, 2 = moderately disturbed, 3 = highly disturbed and 4 = destructive. Cover-abundance values were estimated using the modified Braun Blanquet scales (Van der Maarel E., 2005).

Classification and ordination methods were used to describe vegetation types and to examine the relationship between vegetation types and environmental variables. R statistical package (R Development Core Team, 2009), was used for cluster and ordination analysis. Indicator species analysis was performed to find indicator species characterizing the communities. Indicator species analysis was performed in R using package labdsv (Roberts DW., 2012). Box plots and One-way analysis of variance (ANOVA) were used to assess the relationships between plant communities and elevation as well as plant communities with disturbance intensity. Tukey’s test was performed to detect significant differences among the different means of the environmental parameters of each community types. Sorenson's index of similarity (Ss) was computed to assess the floristic similarity between communities.

本研究于埃塞俄比亚奥罗米亚州西阿尔西区戈莱天然林（多多拉）开展。本研究旨在探究植被组成、结构、群落类型及其更新现状。为采集植被数据，研究人员以山地山顶为起点，按100米间隔系统布设了62个20 m × 20 m（400 m²）的样方。对乔木与灌木物种进行计数并估算其盖度-多度值。草本植物数据则采集自主样方四角及中心布设的5个2 m × 2 m的小样方。对所有株高超过1.5 m、胸径（DBH）大于2 cm的木本物种，测量其株高与胸径。本研究采用R软件包开展聚类分析，在R环境中进行指示种分析及插值物种累积曲线分析。数据统计分析采用Estimate S 8.2软件与Microsoft Excel完成。采用稀疏分析比较研究区域内植物群落的物种丰富度，利用索伦森相似性系数检测群落间的相似性与异质性。

本研究共鉴定出114种植物，隶属于57科94属。优势科为菊科（Asteraceae），其次为爵床科（Acanthaceae）与唇形科（Lamiaceae）。在114个物种中，有17种为埃塞俄比亚特有种。研究结果显示，较低株高与胸径级别的木本植物密度更高。本研究共识别出5种植物群落类型。稀疏分析结果表明，各群落间的物种丰富度存在差异。索伦森相似性指数结果显示，5个植物群落的植物物种组成分布存在显著差异。

### 研究方法
系统布设20 m × 20 m（400 m²）的样方：首个样方以山地山顶为起点随机向山下布设，后续样方沿样线以100米间隔依次布设。本次研究共布设62个样方用于植被数据采集。乔木与灌木数据均采集自主样方。在20 m × 20 m主样方的四角及中心各布设1个2 m × 2 m的小样方，共计5个，用于采集草本植物数据（Hailemariam MB. 与 Temam TD., 2018）。采集所有植物物种的凭证标本，并在国家标本馆（ETH）完成物种鉴定。利用Garmin GPS在主样方中心位置测定环境因子（海拔）与地理坐标。记录每个样方的干扰类型，包括放牧、砍伐乔灌木数量、人行步道数量以及被踩踏的幼苗数量。采用如下分级求和法估算每个样方内人为干扰（放牧）的强度：0=无干扰，1=轻度干扰，2=中度干扰，3=重度干扰，4=破坏性干扰。采用改良的布朗-布朗凯特盖度等级法估算盖度-多度值（Van der Maarel E., 2005）。

采用分类与排序方法描述植被类型，并探究植被类型与环境变量之间的关联。本研究采用R统计软件包（R Development Core Team, 2009）开展聚类与排序分析。开展指示种分析以筛选表征各植物群落的指示物种，该分析通过R语言的labdsv软件包完成（Roberts DW., 2012）。采用箱线图与单因素方差分析（ANOVA）评估植物群落与海拔、以及植物群落与干扰强度之间的关联。通过Tukey检验检测不同群落类型的环境参数均值间的显著差异。计算索伦森相似性指数（Ss）以评估各群落间的区系相似性。