Department of Zoology, University of Cambridge, The David Attenborough Building, Cambridge, UK.
Forestry England, Bristol, UK.
Ecol Appl. 2022 Oct;32(7):e2678. doi: 10.1002/eap.2678. Epub 2022 Aug 3.
Widespread afforestation is a crucial component of climate mitigation strategies worldwide. This presents a significant opportunity for biodiversity conservation if forests are appropriately managed. Within forests, structural and habitat diversity are known to be critical for biodiversity but pragmatic management recommendations are lacking. We make a comprehensive assessment of the effects of habitat variables on bird populations using data from over 4000 ha of forested landscape. We combine high-resolution remote sensing data with comprehensive management databases to classify habitat attributes and measure the response of six taxonomic and functional diversity metrics: species richness, Shannon diversity, functional richness, functional evenness, functional divergence, and functional dispersion. We use a novel approach that combines hierarchical partitioning analysis with linear models to determine the relative importance of different habitat variables for each bird diversity metric. The age class of forest stands was consistently the most important variable across all bird diversity metrics, outperforming other structural measures such as horizontal and vertical heterogeneity and canopy density. Shrub density and gap fraction were each significantly associated with one bird diversity metric. In contrast, variables describing within-stand structural heterogeneity (vertical and horizontal) were generally less important while tree species identity (e.g., conifer or broadleaved) was not significant for any bird diversity metric. Each of the six bird diversity metrics had different patterns of independent variable importance and significance, emphasizing the need to consider multiple diversity metrics in biodiversity assessments. Similarly, the optimal resolution for remote sensing metrics varied between structural variables and bird diversity metrics, suggesting that the use of remote sensing data in biodiversity studies could be greatly improved by first exploring different resolutions and data aggregations. Based on the results from this comprehensive study, we recommend that managers focus on creating habitat diversity at the between-, rather than exclusively within-stand scale, such as by creating a matrix of different age classes, to maximize bird diversity. This recommendation for forest managers is powerful yet pragmatic in its simplicity.
广泛的造林是全球气候缓解策略的重要组成部分。如果森林得到适当管理，这为生物多样性保护提供了一个重大机会。在森林中，结构和生境多样性被认为对生物多样性至关重要，但缺乏实用的管理建议。我们使用超过 4000 公顷林地的数据集，全面评估了生境变量对鸟类种群的影响。我们将高分辨率遥感数据与全面的管理数据库相结合，对生境属性进行分类，并测量了六个分类和功能多样性指标的响应：物种丰富度、香农多样性、功能丰富度、功能均匀度、功能发散和功能离散。我们使用一种新颖的方法，将层次划分分析与线性模型相结合，确定不同生境变量对每种鸟类多样性指标的相对重要性。林分的年龄类别始终是所有鸟类多样性指标中最重要的变量，优于其他结构措施，如水平和垂直异质性以及冠层密度。灌木密度和空隙分数与一个鸟类多样性指标显著相关。相比之下，描述林分内结构异质性（垂直和水平）的变量通常不太重要，而树木种类身份（例如针叶树或阔叶树）对任何鸟类多样性指标都不重要。六个鸟类多样性指标中的每一个都有不同的自变量重要性和显著性模式，这强调了在生物多样性评估中需要考虑多个多样性指标。同样，结构变量和鸟类多样性指标之间的遥感指标最佳分辨率也有所不同，这表明通过首先探索不同的分辨率和数据聚合，可以大大改进遥感数据在生物多样性研究中的应用。基于这项全面研究的结果，我们建议管理者专注于在林分之间而不是仅在林分内创造生境多样性，例如通过创建不同年龄类别的矩阵，以最大化鸟类多样性。这一建议对森林管理者来说既强大又简单实用。
