Data from: Using fuzzy logic to determine the vulnerability of marine species to climate change

Marine species are being impacted by climate change and ocean acidification, although their level of vulnerability varies due to differences in species' sensitivity, adaptive capacity and exposure to climate hazards. Due to limited data on the biological and ecological attributes of many marine species, as well as inherent uncertainties in the assessment process, climate change vulnerability assessments in the marine environment frequently focus on a limited number of taxa or geographic ranges. As climate change is already impacting marine biodiversity and fisheries, there is an urgent need to expand vulnerability assessment to cover a large number of species and areas. Here, we develop a modelling approach to synthesize data on species-specific estimates of exposure, and ecological and biological traits to undertake an assessment of vulnerability (sensitivity and adaptive capacity) and risk of impacts (combining exposure to hazards and vulnerability) of climate change (including ocean acidification) for global marine fishes and invertebrates. We use a fuzzy logic approach to accommodate the variability in data availability and uncertainties associated with inferring vulnerability levels from climate projections and species' traits. Applying the approach to estimate the relative vulnerability and risk of impacts of climate change in 1074 exploited marine species globally, we estimated their index of vulnerability and risk of impacts to be on average 52 ± 19 SD and 66 ± 11 SD, scaling from 1 to 100, with 100 being the most vulnerable and highest risk, respectively, under the ‘business-as-usual' greenhouse gas emission scenario (Representative Concentration Pathway 8.5). We identified 157 species to be highly vulnerable while 294 species are identified as being at high risk of impacts. Species that are most vulnerable tend to be large-bodied endemic species. This study suggests that the fuzzy logic framework can help estimate climate vulnerabilities and risks of exploited marine species using publicly and readily available information.海洋物种正受到气候变化与海洋酸化的影响，不过由于不同物种的敏感性、适应能力以及面临气候灾害的暴露程度存在差异，其脆弱性水平各不相同。由于多数海洋物种的生物学与生态学属性数据匮乏，且评估过程本身存在固有不确定性，海洋环境中的气候变化脆弱性评估往往仅聚焦于有限的类群（taxa）或地理范围。鉴于气候变化已对海洋生物多样性与渔业造成影响，亟需将脆弱性评估的覆盖范围扩展至更多物种与区域。本研究构建了一种建模方法，可整合物种特异性暴露评估数据以及生态学与生物学性状数据，以此对全球海洋鱼类与无脊椎动物的气候变化（含海洋酸化）脆弱性（包括敏感性与适应能力）及影响风险（结合灾害暴露与脆弱性）开展评估。我们采用模糊逻辑（fuzzy logic）方法，以适配数据可用性的差异，以及由气候预测与物种性状推断脆弱性等级时所伴随的不确定性。我们将该方法应用于全球1074种被开发利用的海洋物种，以估算其相对脆弱性与气候变化影响风险。结果显示，二者的指数得分介于1至100之间（分值越高代表脆弱性与风险越高），其平均值分别为52±19标准差（SD）与66±11标准差（SD）。在‘照常排放’温室气体排放情景（典型浓度路径8.5（Representative Concentration Pathway 8.5））下，分值为100即代表最脆弱且影响风险最高。本研究共鉴定出157个高脆弱性物种，以及294个面临高影响风险的物种。脆弱性最高的物种多为大型特有物种。本研究表明，该模糊逻辑框架可借助公开且易于获取的信息，辅助估算被开发海洋物种的气候脆弱性与影响风险。