H2CM - daily simulations

Dataset overview This archive contains the daily, global output of our Hybrid Hydrological Carbon‑Cycle Model (H2CM) from a 10‑fold cross‑validation experiment. Each of the ten NetCDF files corresponds to one held‑out fold and provides model predictions on a regular 1°×1° latitude–longitude grid for the period 1 January 2001 through 31 December 2019. Spatial and temporal resolution Spatial grid: 180 latitudes (from 89.5° N to 89.5° S) × 360 longitudes (from 179.5° W to 179.5° E) Temporal resolution: daily time steps (6939 days total) Variables and units Variable Description Units gpp Gross primary productivity g C m⁻² day⁻¹ npp Net primary productivity g C m⁻² day⁻¹ nee Net ecosystem exchange (gpp – ter) g C m⁻² day⁻¹ Ra Autotrophic respiration g C m⁻² day⁻¹ Rh Heterotrophic respiration g C m⁻² day⁻¹ prec_actual Predicted actual precipitation mm day⁻¹ T Transpiration mm day⁻¹ ET Evapotranspiration mm day⁻¹ snow_acc Snow accumulation mm day⁻¹ snow_melt Snow melt mm day⁻¹ swe Snow water equivalent (snowpack storage) mm Ei Interception evaporation mm day⁻¹ Es Soil evaporation mm day⁻¹ r_soil Soil recharge mm day⁻¹ r_gw Groundwater recharge mm day⁻¹ runoff_surface Surface runoff mm day⁻¹ baseflow Baseflow mm day⁻¹ runoff_total Total runoff (surface runoff + baseflow) mm day⁻¹ SM Soil moisture storage mm sm_max Maximum soil moisture capacity mm GW Groundwater storage mm tws Total water storage (all pools) mm fapar Fraction of absorbed photosynthetically active radiation - Purpose and usage These files support evaluation of H2CM’s ability to predict carbon and water fluxes under a cross‑validation framework. By providing ten independent model realizations—each omitting one fold of training data—users can: Assess spatial and temporal uncertainties in both carbon and hydrological estimates Reproduce key results on model performance in tropical, arid, and temperate regions Integrate these outputs into further analyses of terrestrial ecosystem responses to climate variability Users may load any single fold for standalone analysis or combine multiple folds to derive ensemble means, confidence intervals, and sensitivity metrics. Additional information If you encounter any issues, have questions about how to load or interpret the data, or would like access to additional variables or alternative aggregations that aren’t included here, please don’t hesitate to get in touch. We’re always happy to help troubleshoot, share more detailed outputs, or discuss how best to integrate these model results into your research.