Compound Extremes Data Benchmarking (ComExDBM)
Access by contacting [email protected]
Project Name: Compound Extremes Data Benchmarking (ComExDBM)
Date of last update: 2023-06-12 Revision history: 2023-05-03 : Initial creation data 2023-06-12 : Cleaned up formatting
- Data description
Extreme weather events such as fires, heatwaves(HWs), and droughts result in significant socioeconomic and environmental damage around the world. Mechanistic and predictive understanding of extreme weather events are crucial for the detection, planning and response to these extremes and mitigating their impacts. Records of historical extreme weather events provide an important data source for understanding present and future climate risks, but the data are sparse, unevenly distributed, and of multi-fidelity from multiple sources; in addition, there are many nonstandard metrics defining the levels of severity or impacts of extremes. In this study, we develop a benchmark data inventory of US extreme weather events (i.e., fires, HWs, and droughts) with daily temporal resolution and spatial resolution of 0.5°×0.5° (~55km×55km) using data from multiple sources.
The data inventory of US extreme weather events were created using heatwaves(HWs), fires, droughts, and meteorological variables data collected from multiple sources. The resulting datasets include daily temperature anomaly, heatwave labels, fire related features(i.e., fire hot spots, fire burned area),drought indices and co-located meteorological variables. All collected data are compiled and summarized to match the daily temporal resolution and 0.5°×0.5°(~55km×55km) spatial resolution. The output datasets are saved in netcdf files, which can be easily accessed and applied to machine learning (ML)-based research and would encourage ML/AI research in extreme weather and facilitate further work in understanding and mitigating the negative effects of these extremes.
- Folder structure
01_Input contains the input data information and description for all the input data from external sources
02_Code contains all code files
03_Analysis contains the exploratory data analysis results, which include figures and docx files
04_OutputData contains seperate sub-folders for all the final output data
04_OutputData/01_HeatWave contains netcdf files for daily temperature as well as the heatwave labels
with spatial resolution of 0.5°×0.5°(~55km×55km)
04_OutputData/02_FireData contains netcdf files for fire related features with daily temporal resolution and
spatial resolution of 0.5°×0.5°(~55km×55km)
04_OutputData/03_DroughtData contains netcdf files for drought indices with daily temporal resolution and
spatial resolution of 0.5°×0.5°(~55km×55km)
04_OutputData/04_Meteorological_Variables contains netcdf files for meteorological variables with daily
temporal resolution and spatial resolution of 0.5°×0.5°(~55km×55km)
- File naming schema
File type: netcdf data files Filename schema: ComExDBM_YYYY_DT_status.nc Schemakey: ComExDBM = project name YYYY = year of data DT = data type (i.e., HeatWave, Fires, Drought, Metvars) status = data version Example filename : ComExDBM_2001_HeatWave_V01.nc
- File abbreviation
Project abbraviation: ComExDBM = Compound Extremes Data Benchmarking
Filename abbraviation: Metvars = Meteorological Variables
- Data sharing/access information a) Licenses/restrictions placed on the data: b) Links to the publications that cite or use the data
- Information about the founding source
This research has been funded by the Laboratory Directed Research and Development(LDRD) Program at Pacific Northwest National Laboratory (PNNL). Data processing was performed using resources available through Research Computing at PNNL. PNNL is operated by Battelle for the U.S. Department of Energy under Contract DE‐AC05‐76RL01830.