RFF-SP scenarios with FaIR v2.1

This dataset contains output from the 10,000 Resources for the Future Socioeconomic Projections, run with the FaIR reduced complexity climate model (v2.1.0) using an IPCC Sixth Assessment Report consistent calibration of 1,001 probabilistic ensemble members (v1.0). A total of 10,010,000 climate projections are produced. Climate projections are produced for 1750 to 2301. FaIR is run using stochastically generated internal variability. The RFF-SPs contain CO2, CH4 and N2O emissions. The scenarios have been infilled using the Silicone package (Lamboll et al. 2020) to decompose the total CO2 into fossil and land-use components, and to infill emissions of other greenhouse gases and short-lived climate forcers, following the same strategy used to infill scenarios in the IPCC Sixth Assessment Report Working Group 3 from a large database of integrated assessment model pathways (see Kikstra et al. 2022). To extend scenarios beyond 2100 - the time horizon of IAM pathways - the approach consistent with extending the SSPs for CMIP6 is used (Meinshausen et al. 2020, sec. 2.3). Code and instructions to reproduce the results is available at https://github.com/chrisroadmap/rff-fair2.1. Dataset contents: <strong>output[0-9].zip</strong>: RFF-SP projections, in batches of 1,000 individual netCDF files (data_output/stochastic/run?????.nc) where ????? is in the range 00001 to 10000). Each file contains: Global mean near-surface air temperature, rebased to 1850-1900 mean Ocean heat content change since 1750 Effective radiative forcing with respect to pre-industrial (IPCC Sixth Assessment Report Working Group 1 convention of anthropogenic components using a 1750 baseline and natural components using a long pre-1750 mean) CO2 concentrations (ppm) CH4 concentrations (ppb) N2O concentrations (ppb) <strong>infilled_extended.zip</strong>: These are the infilled emissions used to run FaIR, containing 53 emissions species (data_processed/infilled_extended/emissions?????.csv). Time period covered is 2015 to 2300. Pre-2015 emissions can be obtained from the Reduced Complexity Model Intercomparison Project, using the CMIP6 historical emissions from 1750 to 2014. Bridging from 2015 to 2020 was performed using SSP2-4.5, as in the original RFF-SPs for CO2, CH4 and N2O. <strong>ssp.zip</strong>: For completeness and comparison, the eight main SSP scenarios (ssp119, ssp126, ssp245, ssp370, ssp434, ssp460, ssp534-over, ssp585) are run with the same emissions scenarios and calibration dataset. Notes about the data All variables in each file have dimension (timebounds, ensemble member) except for N2O which is unaffected by climate in this calibration of FaIR, having dimensions of (timebounds). Scenario 00001 runs from 1750 to 2301. Other scenarios run from 2020 to 2301, as the historical period is the same for every scenario. This approximately halves the size of the data output. Note that all variables in the output files are on <em>timebounds</em>. 2020 corresponds to 2020-01-01, 2021 to 2021-01-01, and so on. To approximate midyear values, take the mean of consecutive years. Infilled emissions are on <em>timepoints</em>, i.e. are representative of midyear values. See here for an explanation.本数据集包含未来资源研究所（Resources for the Future, RFF）10000项社会经济预测项目的输出结果，该预测采用FaIR简化复杂度气候模型（reduced complexity climate model，v2.1.0）运行，使用符合政府间气候变化专门委员会（Intergovernmental Panel on Climate Change, IPCC）第六次评估报告的校准方案，包含1001个概率集合成员（v1.0）。总共产出10010000份气候预测结果，预测时间跨度为1750年至2301年。FaIR模型运行时纳入了随机生成的内部变率。 RFF-SPs包含二氧化碳（CO₂）、甲烷（CH₄）与氧化亚氮（N₂O）排放数据。本数据集采用Silicone工具包（Lamboll et al. 2020）对情景进行补全：将总CO₂排放拆解为化石燃料与土地利用来源组分，并补全其他温室气体及短期气候强迫剂的排放数据，其策略与IPCC第六次评估报告第三工作组从大型综合评估模型（Integrated Assessment Model, IAM）路径数据库中补全情景的方法一致（参见Kikstra et al. 2022）。为将情景延伸至2100年之后（即IAM路径的时间上限），本数据集采用了与耦合模式相互比较计划第六阶段（CMIP6）共享社会经济路径（Shared Socioeconomic Pathways, SSPs）延伸方案一致的方法（Meinshausen et al. 2020, sec. 2.3）。 复现该结果的代码与操作说明可访问：https://github.com/chrisroadmap/rff-fair2.1。 ### 数据集内容说明 1. **output[0-9].zip**：包含RFF-SP预测结果，以每1000个独立netCDF文件为一批（存储路径为data_output/stochastic/run?????.nc，其中?????的取值范围为00001至10000）。每个文件包含以下变量： - 以1850-1900年平均值为基准的全球近地表平均气温 - 自1750年以来的海洋热含量变化 - 相对于工业化前的有效辐射强迫（遵循IPCC第六次评估报告第一工作组的约定：人为组分以1750年为基线，自然组分采用1750年前长期平均值） - 二氧化碳浓度（ppm） - 甲烷浓度（ppb） - 氧化亚氮浓度（ppb） 2. **infilled_extended.zip**：包含用于运行FaIR模型的补全后排放数据，涵盖53种排放物种（存储路径为data_processed/infilled_extended/emissions?????.csv），时间跨度为2015年至2300年。2015年之前的排放数据可从简化复杂度模型比对项目（Reduced Complexity Model Intercomparison Project）获取，即采用CMIP6的1750-2014年历史排放数据。2015年至2020年的衔接工作采用SSP2-4.5情景完成，与原始RFF-SPs中CO₂、CH₄和N₂O的处理方式一致。 3. **ssp.zip**：为便于对比与完整性验证，本数据集采用相同的排放情景与校准数据集，运行了8种主流共享社会经济路径（SSP）情景：ssp119、ssp126、ssp245、ssp370、ssp434、ssp460、ssp534-over、ssp585。 ### 数据说明 每个文件中的所有变量均采用`timebounds`维度，维度为（时间边界，集合成员），仅氧化亚氮除外：在本次FaIR校准方案中，氧化亚氮不受气候影响，其维度为（时间边界）。 情景00001的时间跨度为1750年至2301年，其余情景的时间跨度为2020年至2301年，因所有情景共享历史时段，此举约将数据输出规模缩减一半。 需注意：输出文件中的所有变量均以`timebounds`为时间坐标：2020对应2020-01-01，2021对应2021-01-01，以此类推。若需近似年中值，可对连续两年的变量值取平均。 补全后的排放数据采用`timepoints`时间坐标，即代表年中值。相关解释可参见此处。