Projected climate risk of aquatic food system benefits

AbstractAquatic foods from marine and freshwater systems are critical to the nutrition, health, livelihoods, economies and culture of billions of people worldwide – but climate-related hazards may compromise their ability to provide these benefits. This analysis estimates national-level aquatic food system climate risk using a fuzzy logic modeling approach that connects climate hazards impacting marine and freshwater capture fisheries and aquaculture to their contributions to sustainable food system outcomes, and vulnerability to losing those contributions. Estimates are presented for a high and a low emissions scenario in three different time windows (2030, 2050, 2090).  , MethodsThis analysis computes quantitative indices of climate risk for four aquatic food system outcomes – nutrition & health, economic, social, and environmental – adopting a fuzzy logic modeling approach to implement the risk assessment framework used by the Intergovernmental Panel on Climate Change. In this framework, climate risk results from the interaction between climate-change induced hazards, exposure to those climate hazards, and vulnerabilities of components of the aquatic food systems. For our purposes, we conceptualized climate hazards as the dominant climate variables that impact aquatic food production and supply chains, exposure as the degree to which aquatic foods contribute to the various food system outcomes at a national-level, and vulnerability as a combination of sensitivity to and adaptive capacity of the nationally-aggregated food systems in the face of the loss of aquatic food contributions. Through two rounds of virtual workshops, the team of co-authors – who were selected for their expertise spanning marine and freshwater ecosystems, fisheries and aquaculture production systems, and multiple food system outcomes – selected hazard, exposure, and vulnerability indicators based on their expert knowledge, published literature, and data availability for most of the countries included in this study. Climate hazards Climate hazard scores were calculated for six different components of aquatic food systems: marine fisheries, freshwater fisheries, marine aquaculture, freshwater aquaculture, brackish aquaculture, post-production processes. The following variables were selected for each of these components: Marine fisheries: Maximum catch potential (from an ecology model based on ocean temperature, circulation, dissolved oxygen, net primary production in the top 100m, salinity and sea ice) ; surface and bottom pH; marine heatwave frequency Freshwater fisheries: Near-surface air temperature; freshwater balance; percent extraction of renewable freshwater Marine aquaculture: Maximum mariculture potential (from an ecology model based on ocean conditions, suitable marine area for farming, fishmeal and fish oil production) ; marine heatwave frequency; percent of population inundated by sea level; cyclone strength in Low Elevation Coastal Zone; global cropland temperature; feed Crude Protein index Freshwater aquaculture: Near-surface air temperature; freshwater balance; percent extraction of renewable freshwater; global cropland temperature; fishmeal/fish oil availability; feed Crude Protein index Brackish aquaculture: Near-surface air temperature; percent of population inundated by sea level; cyclone strength in Low Elevation Coastal Zone; global cropland temperature; fishmeal/fish oil availability; feed Crude Protein index Post-production: Near-surface air temperature; percent of population inundated by sea level; cyclone strength in Low Elevation Coastal Zone; change in sea ice extent; % of landings from small-scale operations Where possible, projections from three different Earth system models (ESM) were used to represent uncertainties in projections of environmental changes, all available from the Coupled Models Intercomparison Project Phase 6 (CMIP6): Geophysical Fluid Dynamics Laboratory (GFDL)-ESM4, The Institut Pierre-Simon Laplace (IPSL)-CM6A-LR, and Max Planck Institute (MPI)-ESM1-2-HR. We calculated climate hazards using two contrasting scenarios – Shared Socio-economic Pathway (SSP) 1 - Representative Concentration Pathway (RCP) 2.6 (SSP1-2.6) and SSP5-8.5. The SSP1-2.6 and SSP5-8.5 represent a ‘strong mitigation’ low-emissions pathway and a ‘no mitigation’ high-emissions pathway, respectively. For the marine heatwave variable, CMIP6 results were not yet available so CMIP5 equivalents were used. Results were calculated for the near future (2021-2040), middle (2041-2060) and end (2081-2100) of the 21st century.  Exposure The following exposure indicators were selected for each of the four food system outcomes: Nutrition & health: Per capita supply of marine and freshwater aquatic foods; percentage of a nation’s consumption of vitamin B-12 and DHA+EPA fatty acids derived from aquatic foods Economic: Contribution of aquatic food production to Gross Domestic Product (GDP); economic multipliers of marine supply chains; net aquatic food trade balance relative to GDP Social: Contribution of marine fisheries, aquaculture, and inland fisheries to employment; ratio of indigenous to national-average consumption of seafood Environmental: Average greenhouse gas emissions, nitrogen and phosphorus emissions, land use and freshwater use of different types of wild-capture and farmed aquatic food production Vulnerability The following vulnerability indicators were selected for each of the four food system outcomes: Nutrition & Health: Percent of population below national poverty line; percent secondary educational attainment; percent stunted children under 5; Summary Exposure Values for Vitamin B-12 and omega-3 fatty acids;  Economic: GDP per capita; GINI coefficient; percent of population with access to bank account; R&D expenditures relative to GDP; average of Worldwide Governance Indicators; percent of landings from small-scale operations Social: Percent of population below national poverty line; GINI coefficient; percent of population with access to bank account; average of Worldwide Governance Indicators; percent secondary educational attainment; percent of landings from small-scale operations Environmental: GDP per capita; GINI coefficient; R&D expenditures relative to GDP; average of Worldwide Governance Indicators; Environmental Performance Index – Biodiversity & Habitat, Fisheries, and Climate Change; percent landings from small-scale operations Fuzzy logic system The fuzzy logic modeling system consists of three steps: Categorizing each indicator variable into one or more levels of ‘low’, ‘medium’, ‘high’ and ‘very high’ simultaneously, with the degree of membership defined by fuzzy membership functions (“fuzzification”) Accumulating the degree of membership associated with each level using the MYCIN algorithm for each of the subcomponents of climate risk (hazard, exposure and vulnerability) and applying a set of heuristic rules to combine the components into an aggregate risk score (“fuzzy reasoning”) Calculating a final score from the accumulated memberships in order to express climate risk on a scale from 1 to 100 (“defuzzification”). For more information on the fuzzy logic methodology, see:  Cheung, W. W. L., Pitcher, T. J. & Pauly, D. A fuzzy logic expert system to estimate intrinsic extinction vulnerabilities of marine fishes to fishing. Biol. Conserv. 124, 97–111 (2005). Cheung, W. W. L., Jones, M. C., Reygondeau, G. & Fr.licher, T. L. Opportunities for climate-risk reduction through effective fisheries management. Glob. Chang. Biol. 24, 5149–5163 (2018). Jones, M. C. & Cheung, W. W. L. Using fuzzy logic to determine the vulnerability of marine species to climate change. Glob. Chang. Biol. 24, e719–e731 (2018)., Usage notesThe datafile contains 6 tabs, for combinations of 2 emissions scenarios (SSP1-2.6, low-emissions; SSP5-8.5, high-emissions) and 3 time frames (2030, 2050, 2090). Each tab contains 23 variables for 240 countries and territories. All variables are of a unit-less score ranging from 1-100, where <25 indicates 'Low', 25-50 indicates 'Medium', 50-75 indicates 'High', and >75 indicates 'Very High'. Missing data are marked by empty cells, shaded grey. Countries and territories marked with an asterisk (*) are ones for which data availability was low, indicating reduced confidence in resulting risk scores. The following variables are included: Hazard - Aggregate: Climate hazard score aggregated across all production systems based on present-day production weights; in all countries the 'post-production' component is assigned a weight of 10% Hazard – Marine fisheries: Climate hazard score for marine fisheries Hazard – Freshwater fisheries: Climate hazard score for freshwater fisheries Hazard – Marine aquaculture: Climate hazard score for marine aquaculture Hazard – Freshwater aquaculture: Climate hazard score for freshwater aquaculture Hazard – Brackish aquaculture: Climate hazard score for brackish aquaculture Hazard – Post-production: Climate hazard score for post-production processes Exposure – Nutrition_Health: Exposure score for the Nutrition & Health food systems outcome Exposure – Economic: Exposure score for the Economic food systems outcome Exposure – Social: Exposure score for the Social food systems outcome Exposure – Environmental: Exposure score for the Environmental food systems outcome Exposure to Hazard – Nutrition_Health: Exposure to Hazard score for the Nutrition & Health food systems outcome; scores were aggregated across all production systems based on present-day production weights; in all countries the 'post-production' component is assigned a weight of 10% Exposure to Hazard – Economic: Exposure to Hazard score for the Economic food systems outcome; scores were aggregated across all production systems based on present-day production weights; in all countries the 'post-production' component is assigned a weight of 10% Exposure to Hazard – Social: Exposure to Hazard score for the Social food systems outcome; scores were aggregated across all production systems based on present-day production weights; in all countries the 'post-production' component is assigned a weight of 10% Exposure to Hazard – Environmental: Exposure to Hazard score for the Environmental food systems outcome; scores were aggregated across all production systems based on present-day production weights; in all countries the 'post-production' component is assigned a weight of 10% Vulnerability – Nutrition_Health: Vulnerability score for the Nutrition & Health food systems outcome Vulnerability – Economic: Vulnerability score for the Nutrition & Health food systems outcome Vulnerability – Social: Vulnerability score for the Nutrition & Health food systems outcome Vulnerability – Environmental: Vulnerability score for the Nutrition & Health food systems outcome Risk – Nutrition_Health: Climate risk score for the Nutrition & Health food systems outcome Risk – Economic: Climate risk score for the Nutrition & Health food systems outcome Risk – Social: Climate risk score for the Nutrition & Health food systems outcome Risk – Environmental: Climate risk score for the Nutrition & Health food systems outcome More information can be found in the associated README file.

摘要：源自海洋与淡水生态系统的水生食品，对全球数十亿人口的营养健康、生计发展、经济运转与文化传承至关重要——但气候相关灾害或会削弱其提供这些福祉的能力。本分析采用模糊逻辑建模方法，评估国家级水生食品系统的气候风险，该方法将影响海洋、淡水捕捞渔业与水产养殖的气候灾害，与其对可持续食品系统成果的贡献，以及丧失这些贡献的脆弱性联系起来。评估针对两种排放情景（高排放与低排放），并覆盖三个不同时间窗口（2030年、2050年、2090年）。

方法：本分析针对四大水生食品系统成果——营养与健康、经济、社会与环境——计算气候风险定量指数，采用模糊逻辑建模方法落实政府间气候变化专门委员会(Intergovernmental Panel on Climate Change, IPCC)所用的风险评估框架。在该框架中，气候风险源于气候变化引发的灾害、对这些气候灾害的暴露，以及水生食品系统各组成部分的脆弱性三者的交互作用。就本研究而言，我们将气候灾害定义为影响水生食品生产与供应链的主导气候变量；将暴露定义为国家级尺度下水生食品对各类食品系统成果的贡献程度；将脆弱性定义为国家层面汇总的食品系统在丧失水生食品贡献时的敏感性与适应能力的综合。

通过两轮线上研讨会，研究合著者团队（其遴选依据为涵盖海洋与淡水生态系统、渔业与水产养殖生产系统，以及多类食品系统成果的专业专长）基于专家知识、已发表文献，以及本研究涵盖的多数国家的数据可得性，筛选出灾害、暴露与脆弱性相关指标。

### 气候危害
针对水生食品系统的六大组成部分计算气候危害得分，分别为海洋捕捞渔业、淡水捕捞渔业、海水养殖、淡水养殖、咸淡水养殖，以及产后加工流程。各组成部分选取的变量如下：
- 海洋捕捞渔业：最大捕捞潜力（基于海洋温度、环流、溶解氧、表层100米净初级生产力、盐度与海冰的生态模型推导得出）；表层与底层pH值；海洋热浪发生频率
- 淡水捕捞渔业：近地面气温；淡水收支；可再生淡水抽取占比
- 海水养殖：最大海水养殖潜力（基于海洋环境、适宜养殖海域面积、鱼粉与鱼油产量的生态模型推导得出）；海洋热浪发生频率；受海平面上升淹没的人口占比；低海拔沿海区域的气旋强度；全球农田气温；饲料粗蛋白指数
- 淡水养殖：近地面气温；淡水收支；可再生淡水抽取占比；全球农田气温；鱼粉/鱼油供应量；饲料粗蛋白指数
- 咸淡水养殖：近地面气温；受海平面上升淹没的人口占比；低海拔沿海区域的气旋强度；全球农田气温；鱼粉/鱼油供应量；饲料粗蛋白指数
- 产后加工：近地面气温；受海平面上升淹没的人口占比；低海拔沿海区域的气旋强度；海冰范围变化；小型作业捕捞占总上岸量的比例

尽可能采用三种不同的地球系统模型(Earth System Models, ESM)的预测结果来表征环境变化预测的不确定性，这三种模型均来自耦合模式比较计划第六阶段(Coupled Models Intercomparison Project Phase 6, CMIP6)：地球物理流体动力学实验室(Geophysical Fluid Dynamics Laboratory, GFDL)-ESM4、皮埃尔-西蒙·拉普拉斯研究所(Institut Pierre-Simon Laplace, IPSL)-CM6A-LR，以及马克斯·普朗克研究所(Max Planck Institute, MPI)-ESM1-2-HR。我们采用两种对比情景计算气候危害：共享社会经济路径(Shared Socio-economic Pathway, SSP)1-典型浓度路径(Representative Concentration Pathway, RCP)2.6（SSP1-2.6）与SSP5-8.5。其中SSP1-2.6代表‘强减排’低排放路径，SSP5-8.5代表‘无减排’高排放路径。针对海洋热浪变量，暂未获取CMIP6的结果，因此采用CMIP5的等效数据。计算结果覆盖21世纪的近期（2021-2040年）、中期（2041-2060年）与末期（2081-2100年）。

### 暴露
针对四大食品系统成果选取以下暴露指标：
- 营养与健康：海洋与淡水水生食品的人均供应量；一国从水生食品中获取的维生素B12与DHA+EPA脂肪酸消费占比
- 经济：水生食品生产对国内生产总值(Gross Domestic Product, GDP)的贡献；海洋供应链的经济乘数；净水生食品贸易差额相对GDP的比值
- 社会：海洋捕捞渔业、水产养殖与内陆捕捞渔业对就业的贡献；原住民海鲜消费量与全国平均消费量的比值
- 环境：不同类型野生捕捞与养殖水生食品生产的平均温室气体排放量、氮与磷排放量、土地使用与淡水使用量

### 脆弱性
针对四大食品系统成果选取以下脆弱性指标：
- 营养与健康：国家贫困线以下人口占比；中等教育完成率；5岁以下儿童生长迟缓率；维生素B12与ω-3脂肪酸的总暴露值
- 经济：人均GDP；基尼系数；拥有银行账户的人口占比；研发支出相对GDP的比值；全球治理指标平均值；小型作业捕捞占总上岸量的比例
- 社会：国家贫困线以下人口占比；基尼系数；拥有银行账户的人口占比；全球治理指标平均值；中等教育完成率；小型作业捕捞占总上岸量的比例
- 环境：人均GDP；基尼系数；研发支出相对GDP的比值；全球治理指标平均值；环境绩效指数——生物多样性与栖息地、渔业、气候变化；小型作业捕捞占总上岸量的比例

### 模糊逻辑系统
模糊逻辑建模系统包含三个步骤：
1. 将每个指标变量同时划分为‘低’‘中’‘高’‘极高’四个等级，隶属度由模糊隶属函数定义（即‘模糊化’过程）
2. 针对气候风险的各子组成部分（灾害、暴露与脆弱性），采用MYCIN算法累加各等级的隶属度，并应用一系列启发式规则将各子组成部分整合为综合风险得分（即‘模糊推理’过程）
3. 从累加的隶属度中计算最终得分，将气候风险以1-100的量表形式表达（即‘去模糊化’过程）。
如需了解模糊逻辑方法论的更多细节，请参见：
Cheung, W. W. L., Pitcher, T. J. & Pauly, D. A fuzzy logic expert system to estimate intrinsic extinction vulnerabilities of marine fishes to fishing. Biol. Conserv. 124, 97–111 (2005).
Cheung, W. W. L., Jones, M. C., Reygondeau, G. & Frölicher, T. L. Opportunities for climate-risk reduction through effective fisheries management. Glob. Chang. Biol. 24, 5149–5163 (2018).
Jones, M. C. & Cheung, W. W. L. Using fuzzy logic to determine the vulnerability of marine species to climate change. Glob. Chang. Biol. 24, e719–e731 (2018).

### 使用说明
本数据集包含6个工作表，对应2种排放情景（SSP1-2.6，低排放；SSP5-8.5，高排放）与3个时间范围（2030年、2050年、2090年）的组合。每个工作表包含240个国家与地区的23个变量。所有变量均为无量纲得分，取值范围为1-100，其中得分<25代表‘低’，25-50代表‘中’，50-75代表‘高’，>75代表‘极高’。缺失数据以空白灰色单元格标记。带有星号（*）的国家与地区表示其数据可得性较低，对应风险得分的置信度有所降低。

包含的变量如下：
- 灾害-综合：基于当前生产权重聚合所有生产系统的气候灾害得分；所有国家的‘产后加工’组成部分均被赋予10%的权重
- 灾害-海洋捕捞渔业：海洋捕捞渔业的气候灾害得分
- 灾害-淡水捕捞渔业：淡水捕捞渔业的气候灾害得分
- 灾害-海水养殖：海水养殖的气候灾害得分
- 灾害-淡水养殖：淡水养殖的气候灾害得分
- 灾害-咸淡水养殖：咸淡水养殖的气候灾害得分
- 灾害-产后加工：产后加工流程的气候灾害得分
- 暴露-营养与健康：营养与健康食品系统成果的暴露得分
- 暴露-经济：经济食品系统成果的暴露得分
- 暴露-社会：社会食品系统成果的暴露得分
- 暴露-环境：环境食品系统成果的暴露得分
- 灾害暴露-营养与健康：营养与健康食品系统成果的灾害暴露得分；得分基于当前生产权重聚合所有生产系统；所有国家的‘产后加工’组成部分均被赋予10%的权重
- 灾害暴露-经济：经济食品系统成果的灾害暴露得分；得分基于当前生产权重聚合所有生产系统；所有国家的‘产后加工’组成部分均被赋予10%的权重
- 灾害暴露-社会：社会食品系统成果的灾害暴露得分；得分基于当前生产权重聚合所有生产系统；所有国家的‘产后加工’组成部分均被赋予10%的权重
- 灾害暴露-环境：环境食品系统成果的灾害暴露得分；得分基于当前生产权重聚合所有生产系统；所有国家的‘产后加工’组成部分均被赋予10%的权重
- 脆弱性-营养与健康：营养与健康食品系统成果的脆弱性得分
- 脆弱性-经济：经济食品系统成果的脆弱性得分
- 脆弱性-社会：社会食品系统成果的脆弱性得分
- 脆弱性-环境：环境食品系统成果的脆弱性得分
- 风险-营养与健康：营养与健康食品系统成果的气候风险得分
- 风险-经济：经济食品系统成果的气候风险得分
- 风险-社会：社会食品系统成果的气候风险得分
- 风险-环境：环境食品系统成果的气候风险得分

更多信息可参阅附带的README文件。