Human disturbances lead to forest degradation and a drastic reduction in forest area. In Africa, the most affected continent by this phenomenon, selective cutting of trees remains the main forest management strategy. However, the effects of management on biodiversity are insufficiently known, particularly in Africa. We investigated how forest management affects tree species diversity, composition, size structure and carbon biomass of mature and juvenile trees in the sudanian domain of Senegal by comparing unmanaged forests and three types of managed forests, while considering the disturbance level of each stand. We collected floristic data on five and fifteen unmanaged and managed forest stands, respectively. We calculated species richness, the Shannon-Wiener diversity Index (alpha, beta and gamma), as well as carbon stocks of trees for each forest stand. Then we fitted linear models to estimate the differences between forest types for each index. We also analyzed tree size structure and species composition of highly valuable species. In total, 26,009 mature and juvenile trees in 183 species were recorded. Our findings showed that management status and disturbance level affect tree species in different ways and that disturbance level explains a greater proportion of the variation in species diversity than management status. Considering alpha, beta and gamma diversity, we found no significant association between any of these metrics and management status, for either mature or juvenile trees. Disturbance level was only significantly associated with the gamma diversity of mature trees. Species composition of juvenile trees of highly valuable species was significantly associated with both management status and disturbance level, unlike mature tree species composition where the associations were not significant. The distribution of mature tree diameter forms an inverted J-shape for each management category and disturbance level. However, neither the median tree diameter nor the median absolute deviation presented significant differences as a function of management status or disturbance level. For carbon stocks, none of the differences observed by management status and disturbance level are statistically significant. Our findings indicated that forest management in the sudanian zone affects species composition more than diversity and that mature trees respond differently than juvenile trees. Disturbances more than forest management were the underlying process for biodiversity changes both in managed and unmanaged forests. These findings suggest a better protection of unmanaged forests, and also a development of specific conservation action plans for highly valuable species, especially for species that are threatened at national or global levels in order to minimize their risk of local extinction.

中文翻译：

---

探索西非塞内加尔苏丹地区森林管理对树木多样性、群落组成、种群结构和碳储量的影响

人类干扰导致森林退化和森林面积急剧减少。在受这一现象影响最严重的非洲大陆，选择性砍伐树木仍然是主要的森林管理策略。然而，管理对生物多样性的影响尚不清楚，特别是在非洲。我们通过比较未管理的森林和三种类型的管理森林，同时考虑每个林分的干扰水平，研究了森林管理如何影响塞内加尔苏丹地区成熟树和幼树的树种多样性、组成、尺寸结构和碳生物量。我们分别收集了 5 个和 15 个非管理和管理林分的植物区系数据。我们计算了物种丰富度、香农-维纳多样性指数（α、β 和 γ）以及每个林分树木的碳储量。然后我们拟合线性模型来估计每个指数的森林类型之间的差异。我们还分析了高价值树种的树木大小结构和物种组成。总共记录了 183 种 26,009 棵成熟树和幼树。我们的研究结果表明，管理状态和干扰水平以不同的方式影响树种，并且干扰水平比管理状态更能解释物种多样性的变化。考虑到α、β和γ多样性，我们发现无论是成熟树还是幼树，这些指标和管理状态之间都没有显着关联。干扰水平仅与成熟树木的伽马多样性显着相关。高价值树种幼树的物种组成与管理状况和干扰水平显着相关，而成熟树种组成则与相关性不显着。对于每个管理类别和干扰水平，成熟树直径的分布形成倒J形。然而，树木直径中值和绝对偏差中值均未随管理状态或干扰水平呈现显着差异。对于碳储量，管理状态和干扰水平观察到的差异均不具有统计显着性。我们的研究结果表明，苏丹地区的森林管理对物种组成的影响大于多样性，而且成熟树木的反应与幼树不同。无论是管理森林还是非管理森林，生物多样性变化的根本过程不仅仅是森林管理的干扰。这些发现表明，应更好地保护未管理的森林，并为高价值物种制定具体的保护行动计划，特别是在国家或全球层面受到威胁的物种，以尽量减少其局部灭绝的风险。