Temperature Climate Projections from STAR Downscaling

The Climate Resilience Information System (CRIS) provides data and tools for developers of climate services. This image service provides access to downscaled climate projections for 27 threshold values of temperature for the contiguous United States for 2 SSP climate scenarios from 1950-2100. These services are intended to support analysis of climate exposure for custom geographies and time horizons. Sixteen downscaled global circulation models (GCMs) were chosen to be included in a weighted ensemble, optimized for the contiguous United States. More details on the models included in the ensemble and the weighting methodologies can be found in Understanding CRIS Data. Time Ranges Pixel values for each variable were calculated for each year from 2005 to 2100. Additionally, a modeled history runs from 1950 - 2005. The modeled history and future projections have been merged into a single time series. These annual increments support deriving a temporal average, such as a decadal or thirty-year period centered on a specific year. These time steps should not be used to make predictions about conditions for a specific year, especially at a pixel-level. Climate Scenarios Climate models use estimates of future greenhouse gas concentrations and human activities to predict overall change. These different scenarios are called the Shared Socioeconomic Pathways (SSPs). Two different SSPs are presented here: 2-4.5 and 5-8.5. The 2- or 5- represents the socioeconomic growth model. The 4.5 or 8.5 number indicates the amount of radiative forcing (watts per meter square) associated with the greenhouse gas concentration scenario in the year 2100 (higher forcing = greater warming). It is unclear which scenario will be the most likely, but SSP2-4.5 aligns closest with the international targets of the COP-26 agreement for no greater than 2oC average global warming. SSP3-7.0 may be the most likely scenario based on current emission trends. SSP5-8.5 acts as a cautionary tale, depicting a worst-case scenario if reductions in greenhouse gasses are not undertaken. Variable Definitions See the variable list and definitions here. Additional Services Three versions of the gridded climate projections are available from CRIS: LOCA2 Ensemble: a statistically downscaled 6-km resolution model. LOCA2 has SSP2-4.5, 3-7.0 and 5-8.5 STAR-ESDM Ensemble: a statistically downscaled 4-km resolution model. STAR-ESDM has SSP2-4.5 and 5-8.5 NCA5 Blended Ensemble: a merging of LOCA2 and STAR-ESDM ensembles at a 6-km resolution, as was done for the 5th National Climate Assessment (2023). NCA Blended Ensemble has SSP2-4.5 and 5-8.5 Using the Imagery Layer The ArcGIS Tiled Imagery Service has a multidimensional structure -- a data cube with variable, SSP, and time dimensions. Methods for accessing the different dimensions will depend on the software/client being used. For more details, please see the CRIS Developer’s Hub along with this instructional StoryMap. To run analysis, first use the multidimensional tools Aggregate or Subset in ArcGIS Pro to copy the necessary data locally. Data Export Data export is enabled on the services if using an ArcGIS client. NetCDF or Zarr files are also available from the NOAA Open Data Distribution system on Amazon Web Services.气候韧性信息系统（Climate Resilience Information System，CRIS）为气候服务开发者提供数据与工具。本影像服务可获取美国本土27个温度阈值的降尺度气候预测数据，涵盖2种共享社会经济路径（Shared Socioeconomic Pathways，SSP）情景，时间跨度为1950年至2100年。该服务旨在支持针对自定义地理区域与时间跨度的气候暴露度分析。 16个降尺度全球环流模式（Global Circulation Models，GCMs）被选中，组成针对美国本土优化的加权集合模式。关于集合模式所用模型与加权方法的更多细节，可参阅《Understanding CRIS Data》文档。 ### 时间范围 各变量的像素值按年份计算，时间跨度为2005年至2100年。此外，系统还提供1950年至2005年的模拟历史数据。模拟历史数据与未来预测数据已合并为单一时间序列。按年递增的时间粒度支持生成时间平均值，例如以特定年份为中心的十年或三十年滑动平均序列。但需注意，该时间粒度不可用于针对特定年份（尤其是像素级）的气候状况预测。 ### 气候情景 气候模式通过预估未来温室气体浓度与人类活动情况，预测整体气候变化。这类不同情景被称为共享社会经济路径（SSPs）。本次服务提供两种SSP情景：SSP2-4.5与SSP5-8.5。其中前缀2或5代表社会经济增长模式，后缀4.5或8.5代表2100年时该温室气体浓度情景对应的辐射强迫（radiative forcing）强度，单位为瓦/平方米，强迫强度越高，升温幅度越大。 目前尚无法确定哪一情景最符合未来实际发展，但SSP2-4.5最接近第26届联合国气候变化大会（COP-26）设定的全球平均升温不超过2摄氏度的国际目标。根据当前排放趋势，SSP3-7.0可能是最贴合实际的情景。SSP5-8.5则作为警示情景，刻画了未采取温室气体减排措施时的最坏情形。 ### 变量定义 变量清单与定义请参阅此处。 ### 附加服务 CRIS提供三种网格化气候预测数据版本： - LOCA2集合模式：经过统计降尺度的6千米分辨率模式，支持SSP2-4.5、SSP3-7.0与SSP5-8.5三种情景。 - STAR-ESDM集合模式：经过统计降尺度的4千米分辨率模式，支持SSP2-4.5与SSP5-8.5两种情景。 - NCA5混合集合模式：融合LOCA2与STAR-ESDM集合模式的数据，分辨率为6千米，其制作方案沿用自2023年发布的《第5次全国气候评估》（5th National Climate Assessment），支持SSP2-4.5与SSP5-8.5两种情景。 ### 影像图层使用方法 本ArcGIS切片影像服务采用多维数据结构——即包含变量、SSP与时间维度的数据立方体。访问不同维度的具体方法取决于所使用的软件或客户端。如需了解更多细节，请参阅CRIS开发者中心与本教程StoryMap。 如需开展分析，请首先在ArcGIS Pro中使用多维工具中的聚合（Aggregate）或子集（Subset）工具，将所需数据下载至本地。 ### 数据导出 若使用ArcGIS客户端，可在服务中启用数据导出功能。此外，用户还可在亚马逊云服务（Amazon Web Services，AWS）平台的美国国家海洋和大气管理局（National Oceanic and Atmospheric Administration，NOAA）开放数据分发系统中获取NetCDF或Zarr格式的数据文件。