Global Daily FFDI Projections from HadCM3C Perturbed Physics Ensemble - RCP8.5
收藏资源简介:
This dataset contains daily McArthur Forest Fire Danger Index (FFDI) outputs derived from the HadCM3C Perturbed Physics Ensemble (PPE) under two future emissions scenarios. Data is split between two report, the RCP2.6 Data Repository here, while this one is the RCP8.5 Data Repository here. Each repository includes NetCDF files (.nc.gz) representing daily FFDI values for 57 ensemble members. The data spans from 1 January 1970 to 22 December 2097 on a global 3.75° x 2.5° grid using the following projection: +proj=longlat +R=6371229 File Naming Convention NetCDF files that contain daily FFDI values follow the structure: ffdi_joined_rcpX.x_XXXXX.nc.gz Where: rcpX.x refers to the emissions scenario (2.6 or 8.5). XXXXX represents the unique ensemble member ID. We also include three files that, for each ensemble member (first column), contain the start (2nd column) and end date (3rd column) of the 20-year running mean of the exceedance window of 1.5, 2 and 4°C global warming level. These filenames have the structure rcpX_xgwl_Y_y.txt Where: rcpXyx refers to the emissions scenario (2.6 or 8.5). Y_y represents the global warming level above pre-industrial Dataset Overview Variables: Daily FFDI values calculated from model outputs of temperature, relative humidity, wind speed, and precipitation. Temporal Coverage: 1970-01-01 to 2097-12-22. Spatial Coverage: Global grid at 3.75° longitude x 2.5° latitude resolution. Number of Ensemble Members: 57. Methodology The FFDI was computed using input variables from the HadCM3C PPE, which incorporates dynamic vegetation and carbon cycle feedbacks. This ensemble samples uncertainties in atmospheric feedbacks, land carbon processes, and ocean physics. The FFDI is calculated as: FFDI = 1.25 * DF * exp[((T - H) / 30.0) + 0.0234 * V] Where: T = daily maximum temperature at 1.5m (°C), H = daily relative humidity (%), V = daily wind speed at 10m (km/h), DF = drought factor, derived from soil moisture deficit and recent precipitation. For a detailed explanation of the methodology and the specific parameters used, refer to our associated publication once published:DOI: xxxxx Usage Notes Data Formats: Files are compressed NetCDF (.nc.gz) and can be accessed using tools like Python's xarray, R's ncdf4, or command-line tools like ncdump. Applications: This data is suitable for assessing future changes in fire weather conditions, global fire risk analysis, and climate impact assessments. Citation If you use this dataset, please cite: Taylor et al. (2024). Limiting global warming to 1.5°C minimises projected global increases in fire weather days, but adaptation to new fire regimes is still needed. . DOI: xxxxx
本数据集包含基于HadCM3C扰动物理集合(HadCM3C Perturbed Physics Ensemble, PPE)、在两种未来排放情景下生成的逐日麦克阿瑟森林火险指数(McArthur Forest Fire Danger Index, FFDI)结果。数据集分为两个子部分:其一为典型浓度路径(Representative Concentration Pathway, RCP)2.6数据集仓库,当前仓库则为RCP8.5数据集仓库。 每个数据集仓库均包含对应57个集合成员的逐日FFDI值网络通用数据格式(Network Common Data Form, NetCDF)压缩文件(.nc.gz)。数据集的时间范围为1970年1月1日至2097年12月22日,空间分辨率为全球3.75°经度×2.5°纬度,采用以下投影参数:+proj=longlat +R=6371229。 ### 文件命名规范 包含逐日FFDI值的NetCDF压缩文件遵循以下命名结构:`ffdi_joined_rcpX.x_XXXXX.nc.gz` 其中,`rcpX.x`代表排放情景(取值为2.6或8.5),`XXXXX`为唯一的集合成员编号。 本数据集还附带三个文件,分别对应1.5℃、2℃和4℃全球增暖阈值下的超标窗口20年滑动平均数据:文件每一行的第一列为集合成员编号,第二列为滑动平均窗口的起始日期,第三列为结束日期。此类文件的命名结构为`rcpX_xgwl_Y_y.txt`,其中`rcpX.x`代表排放情景(取值为2.6或8.5),`Y_y`代表相对于工业化前的全球增暖水平。 ### 数据集概览 - 变量:基于气温、相对湿度、风速和降水的模式输出计算得到的逐日FFDI值。 - 时间覆盖范围:1970年1月1日至2097年12月22日。 - 空间覆盖范围:全球3.75°经度×2.5°纬度分辨率格网。 - 集合成员数量:57个。 ### 研究方法 FFDI的计算输入变量来自HadCM3C扰动物理集合(PPE),该集合纳入了动态植被与碳循环反馈过程,采样了大气反馈、陆地碳循环过程以及海洋物理过程中的不确定性。FFDI的计算公式如下: FFDI = 1.25 * DF * exp[((T - H) / 30.0) + 0.0234 * V] 其中: T = 1.5米高度处的日最高气温(单位:℃), H = 日相对湿度(单位:%), V = 10米高度处的日风速(单位:km/h), DF = 干旱因子,由土壤水分亏缺和近期降水计算得到。 如需了解研究方法与所用具体参数的详细说明,请参阅待正式发表的关联研究论文:DOI: xxxxx ### 使用说明 - 数据格式:文件为压缩NetCDF格式(.nc.gz),可通过Python的xarray、R的ncdf4等工具,或命令行工具ncdump进行读取与处理。 - 应用场景:本数据集适用于评估未来火天气条件变化、全球火风险分析以及气候影响评估。 ### 引用方式 若您使用本数据集,请引用以下文献: Taylor等人(2024)。将全球变暖限制在1.5℃可最大程度降低预估的全球火天气天数增长,但仍需适应新的火险格局。DOI: xxxxx



