Identifying critical vegetation types for biodiversity conservation in the Americas

The Americas contain highly biodiverse yet vulnerable ecosystems, with many threatened species inadequately protected. Finer-scale, localized habitat assessments are crucial for effective conservation planning, but continental-scale high-resolution vegetation maps remain limited. This study addresses this gap by identifying critical vegetation types across the Americas using the standardized framework of the International Vegetation Classification (IVC) system at the macrogroup level, representing the finest vegetation classification available across the region, as well as the highest-resolution Area of Habitat (AOH) maps currently available for birds and mammals. By combining these high-resolution IVC macrogroup maps with detailed AOH maps, we highlight at-risk vegetation types based on 1) threatened and macrogroup-associated species (species that have at least 50% of their AOH in one macrogroup), 2) current protection levels, and 3) projected threats from land use changes, and 4) develop a conservation value index (CVI) that accounts for all these factors. The results highlighted the remarkable diversity of high conservation value macrogroups across the Americas, emphasizing their significance in regions such as the Andes, montane Mesoamerica, the Caribbean, Brazil's Cerrado, and the Atlantic Forest. Among the highest-scoring macrogroups, the Northern Andean Montane & Upper Montane Humid Forest emerged as critically important, harboring a high number of threatened and macrogroup-associated species. Other macrogroups of immediate conservation concern include the Brazilian Atlantic Montane Humid Forest, Pacific Mesoamerican Seasonal Dry Forest, Caribbean Lowland Humid Forest, and Central Midwest Oak Forest, Woodland and Savanna. However, the study revealed that nearly three-quarters of the over 300 macrogroups in the Americas fall below the global target of 30% protection. Notably, a fifth of all species were macrogroup-associated species, including over 40% of threatened species. Our findings emphasize the need for targeted conservation strategies that consider finer-scale habitat classifications and paired with high-quality species distribution data to guide conservation strategies for biodiversity across the Americas.

Methods
Data Collection
Vegetation Types
The study utilized the International Vegetation Classification (IVC) system, focusing on the macrogroup level consisting of 321 natural vegetation types. This systematic approach enables a detailed vegetation analysis based on ecological criteria. The size of these macrogroups ranged significantly, with the vegetation map having a spatial resolution of 90 x 90 meters, reflecting vegetation as of 2010. All maps were resampled for consistent resolution.

Species Distribution
Area of Habitat (AOH) maps from Lumbierres et al. (2022) were used to assess terrestrial birds and mammals' habitat patterns in the Americas. The dataset included 6,204 species, with 1,868 mammals and 4,336 birds, offering a detailed 100 x 100-meter resolution. Migratory birds had separate AOH maps for different life stages, which were combined for analysis.

Protected Areas
Polygon data from the World Database of Protected Areas (WDPA) was employed to evaluate the protection status of IVC macrogroups. The database includes 273,263 polygons across six protection classes, and all terrestrial areas and OECMs were considered.

Projected Land-use
Future anthropogenic land-use was projected using high-resolution models from Chen et al. (2020), covering various SSP and RCP scenarios. The SSP3-RCP4.5 scenario was specifically chosen for its moderate outlook. Land types were grouped, and the projected percentage of anthropogenic land use by 2030 was calculated to align with the Aichi Biodiversity Targets.

Data Processing
An overlay analysis of IVC macrogroup and AOH data determined species composition within each macrogroup. Migratory birds were handled separately, combining three AOH maps. Species were then filtered by threatened status using IUCN Red List categories. Macrogroup-associated species were defined as having at least half of their AOH within one macrogroup. Protected area coverage was assessed by overlaying macrogroup and protected area data. Future land use data was integrated to estimate land conversion probabilities by 2030. These steps were used to develop a Conservation Value Index (CVI).

Conservation Value Index
The CVI was calculated using a formula that equally weighs the number of threatened species, macrogroup-associated species, the percentage of macrogroup protection, and average land use within the macrogroup. The index offers a composite metric for assessing vegetation types for conservation, with adaptable parameters suitable for various contexts.

美洲拥有兼具高度生物多样性与高度脆弱性的生态系统，诸多受威胁物种未获得充分保护。精细化的本地化栖息地评估对于高效的保护规划至关重要，但跨大陆尺度的高分辨率植被地图依然相对匮乏。本研究针对这一研究空白，基于国际植被分类系统（International Vegetation Classification, IVC）的植被大类层级标准框架，识别美洲范围内的关键植被类型——该层级是当前该区域可获取的最精细植被分类等级，同时结合了目前针对鸟类和哺乳动物的最高分辨率栖息地面积（Area of Habitat, AOH）地图。本研究将高分辨率IVC植被大类地图与精细化AOH地图相结合，从四个维度识别存在保护风险的植被类型：1）受威胁且与该植被大类绑定的物种（即至少50%的栖息地面积分布于某一植被大类内的物种）；2）当前保护水平；3）土地利用变化带来的潜在威胁；同时构建了综合上述所有因素的保护价值指数（Conservation Value Index, CVI）。

研究结果凸显了美洲范围内高保护价值植被大类的显著多样性，强调了安第斯山脉、中美洲山地、加勒比地区、巴西塞拉多以及大西洋森林等区域的重要性。在得分最高的植被大类中，北安第斯山地与上山地湿润森林展现出极高的保护优先级，其栖息的受威胁物种与植被绑定物种数量众多。其他亟需关注的植被大类包括巴西大西洋山地湿润森林、太平洋中美洲季节干性森林、加勒比低地湿润森林以及中西部橡树林、林地与稀树草原。

然而本研究显示，美洲境内的300余个植被大类中，近四分之三未达到全球30%保护率的目标值。值得注意的是，五分之一的物种为植被绑定物种，其中包括超过40%的受威胁物种。本研究结果强调，需制定针对性的保护策略，考虑精细化的栖息地分类，并结合高质量的物种分布数据，以指导美洲全域的生物多样性保护工作。

研究方法
数据采集
植被类型
本研究采用国际植被分类系统（IVC），聚焦包含321种自然植被类型的植被大类层级。该系统性方法可基于生态标准开展精细化植被分析。各植被大类的面积差异显著，所用植被地图的空间分辨率为90×90米，反映的是2010年的植被覆盖情况。所有地图均经过重采样以保证分辨率统一。

物种分布
本研究采用Lumbierres等人（2022）发布的栖息地面积（AOH）地图，用以评估美洲境内陆生鸟类与哺乳动物的栖息地格局。该数据集涵盖6204个物种，其中哺乳动物1868种、鸟类4336种，空间分辨率达100×100米。候鸟拥有针对不同生活史阶段的独立AOH地图，分析时将其合并处理。

保护区
本研究采用世界保护区数据库（World Database of Protected Areas, WDPA）的多边形数据，用以评估IVC植被大类的保护现状。该数据库包含273263个多边形，涵盖6个保护等级，所有陆地区域与其他有效的区域保护措施（Other Effective Conservation Measures, OECMs）均被纳入考量。

潜在土地利用变化
本研究采用Chen等人（2020）发布的高分辨率模型，预测未来人为土地利用变化情况，覆盖各类共享社会经济路径（Shared Socioeconomic Pathways, SSP）与典型浓度路径（Representative Concentration Pathways, RCP）情景。本研究选取SSP3-RCP4.5情景，因其预测结果较为适中。研究对土地类型进行了归类，并计算了至2030年的人为土地利用占比，以契合《爱知生物多样性目标》的要求。

数据处理
通过对IVC植被大类与AOH数据进行叠加分析，确定每个植被大类内的物种组成。针对候鸟单独处理，合并其三张AOH地图。随后根据世界自然保护联盟红色名录（IUCN Red List）类别对物种进行受威胁状态筛选。植被绑定物种的定义为：至少一半的栖息地面积分布于某一个植被大类内的物种。通过叠加植被大类与保护区数据，评估其保护覆盖情况。整合未来土地利用数据，估算至2030年的土地转化概率。上述步骤被用于构建保护价值指数（CVI）。

保护价值指数
保护价值指数（CVI）的计算公式为：受威胁物种数量、植被绑定物种数量、植被大类保护占比以及植被大类内平均土地利用占比的等权综合得分。该指数为评估植被类型的保护优先级提供了综合量化指标，其参数可根据不同研究场景进行调整。