Community Land Model synthetic meteorology simulation model output

Terrestrial processes such as photosynthesis and the movement of water through soils can influence climate from local to global scales by controlling land-to-atmosphere fluxes of water, energy, and carbon. Terrestrial processes are also influenced by climate, as demonstrated by the large body of research exploring how the terrestrial water and carbon cycles respond to climate change. Biogeophysical land-atmosphere feedbacks can therefore potentially modulate changes in land surface water and carbon fluxes. However, the influence of land-atmosphere feedbacks on terrestrial processes has been underexplored. Most previous studies evaluate the biogeophysical impact of land surface changes either in a land only context (i.e., not accounting for land-atmosphere feedbacks at all) or in a fully coupled context (i.e., quantifying the net change in land fluxes without the ability to attribute how much of the response is from feedbacks). While some coupled studies invoke land-atmosphere feedbacks as important drivers of the net coupled land surface changes, it is rare for coupled modeling studies to unambiguously disentangle the extent to which (or mechanisms through which) land-atmosphere feedbacks contribute to the net coupled land response. In isolation, neither coupled nor land-only simulations alone are able to directly disentangle the influence of land-atmosphere feedbacks on the overall coupled change in land water and carbon fluxes. We ran idealized model experiments in the Community Land Model version 5 (CLM5) that can be used to disentangle the atmosphere-to-land branch of the overall land-atmosphere feedback. Methods We ran idealized simulations in the Community Land Model version 5 (CLM5) to isolate how different atmospheric drivers influence terrestrial processes. In each idealized simulation (as described in table below), we modified a different atmospheric driver in the atmospheric forcing data. This method allows us to quantify the terrestrial response to mean state climate changes, but does not assess the terrestrial response to shorter timescale atmospheric variability. In this data repository we include simulations from the reference case and the "Increase temp", "Decrease temp", and "Increase humidity" idealized simulations. Our synthetic meteorology simulations were based on the land-only preindustrial control from a perturbed parameter ensemble in CLM5 (https://doi.org/10.5061/dryad.0k6djhb73). In this control simulation, CLM5 was run with default parameters under preindustrial conditions where the atmospheric state was prescribed at 3-hourly resolution to match the atmosphere simulated by the Community Atmosphere Model version 6 in a preindustrial coupled equilibrium simulation in the Community Earth System Model version 2 (CESM2). The idealized simulations in this repository branched from year 119 of that preindustrial control simulation and were run for 20 years. Simulation Variable Name Variable description Perturbation Increase temp a2x3h_Sa_tbot Temperature at the lowest model level +1 K Decrease temp a2x3h_Sa_tbot Temperature at the lowest model level -1 K Increase precip a2x3h_Faxa_rainl Large-scale precipitation rate 10% a2x3h_Faxa_rainc Convective precipitation rate 10% a2x3h_Faxa_snowl Large scale snow rate 10% a2x3h_Faxa_snowc Convective snow rate 10% Decrease precip a2x3h_Faxa_rainl Large-scale precipitation rate -10% a2x3h_Faxa_rainc Convective precipitation rate -10% a2x3h_Faxa_snowl Large scale snow rate -10% a2x3h_Faxa_snowc Convective snow rate -10% Increase SWdown a2x1hi_Faxa_swndr Direct near-infrared incident solar radiation 10% a2x1hi_Faxa_swvdr Direct visible incident solar radiation 10% a2x1hi_Faxa_swndf Diffuse near-infrared incident solar radiation 10% a2x1hi_Faxa_swvdf Diffuse visible incident solar radiation 10% Increase LWdown a2x3h_Faxa_lwdn Downward longwave heat flux 10% Increase humidity a2x3h_Sa_shum Specific humidity at the lowest model level 10%