CEDAR-GPP: A Spatiotemporally Upscaled Dataset of Gross Primary Productivity Incorporating CO2 Fertilization

Overview:
----------
CEDAR-GPP is a global Gross Primary Productivity (GPP) data product, including monthly GPP estimates at 0.05º spatial resolution. These datasets were generated via upscaling eddy covariance measurements with machine learning and satellite data. CEDAR-GPP uniquely incorporated the direct CO2 fertilization effect (CFE) using both data-driven and theoretical approaches. GPP estimates were produced from ten different model setups that vary by temporal span, direct CFE incorporation method, and GPP partitioning approaches. CEDAR stands for upsCaling Ecosystem Dynamics with ARtificial intelligence.

CEDAR-GPP consists of GPP estimates from ten model setups, differing by temporal range, methods for quantifying CO2 fertilization effects, and the partitioning methods used to derive GPP from eddy covariance measurements. Users are encouraged to refer to the user manual for a structured approach to selecting the most appropriate dataset.

Authors:
----------
Yanghui Kang, Maoya Bassiouni, Max Gaber, Xinchen Lu, Trevor Keenan

File Structure:
----------
Each zip file contains GPP data from a CEDAR model setup.

File Naming Convention:
----------
All netCDF files follow this naming convention:
CEDAR-GPP_<version>_<model-setup>_<YYYYMM>.nc

Where:
<model-setup> comprises of <temporal_span>_<CFE_option>_<GPP_partitioning>
<temporal_span>: ST denotes short-term (2001 to 2020); LT denotes long-term (1982 to 2020)
<CFE_option>: 'Baseline' indicates no direct CO2 fertilization effect, 'CFE-ML' represents direct CO2 fertilization incorporated by ML, 'CFE-Hybrid' implies direct CO2 fertilization incorporated by theory
<GPP_partitioning>: 'NT' for night-time GPP partitioning method, 'DT' for day-time GPP partitioning method

NetCDF characteristics:
----------
- Spatial Resolution: 0.05 degree
- Temporal Resolution: Monthly
- Temporal Coverage: Short-term (ST): 2001-2020; Long-term (LT): 1982 - 2020
- Image Dimension: Rows: 3600, Columns: 7200
- Units: gCm^-2day^-1
- Fill Value: -9999
- Multiply By Scale Factor: 0.01
- Data Type: uint16
- File Size: Approximately 99 MB per file

Data variables:
----------
- GPP_mean: monthly gross primary productivity (gCm^-2day^-1), mean from 30 model ensemble
- GPP_std: standard deviation of 30 model ensemble

Support Contact:
----------
For any queries related to this dataset, please contact:

Name: Yanghui Kang
Email: kangyanghui@gmail.com