RAMI4PILPS: An intercomparison of formulations for the partitioning of solar radiation in land surface models

Remotely sensed, multiannual data sets of shortwave radiative surface fluxes are now available for assimilation into land surface schemes (LSSs) of climate and/or numerical weather prediction models. The RAMI4PILPS suite of virtual experiments assesses the accuracy and consistency of the radiative transfer formulations that provide the magnitudes of absorbed, reflected, and transmitted shortwave radiative fluxes in LSSs. RAMI4PILPS evaluates models under perfectly controlled experimental conditions in order to eliminate uncertainties arising from an incomplete or erroneous knowledge of the structural, spectral and illumination related canopy characteristics typical for model comparison with in situ observations. More specifically, the shortwave radiation is separated into a visible and near‐infrared spectral region, and the quality of the simulated radiative fluxes is evaluated by direct comparison with a 3‐D Monte Carlo reference model identified during the third phase of the Radiation transfer Model Intercomparison (RAMI) exercise. The RAMI4PILPS setup thus allows to focus in particular on the numerical accuracy of shortwave radiative transfer formulations and to pinpoint to areas where future model improvements should concentrate. The impact of increasing degrees of structural and spectral subgrid variability on the simulated fluxes is documented and the relevance of any thus emerging biases with respect to gross primary production estimates and shortwave radiative forcings due to snow and fire events are investigated. Key Points Models‐to‐reference biases vary with illumination and background conditions Mean bias estimated at 4.1 W/m2 for 1‐D cases and 13.9 W/m2 for 3‐D cases Model divergences affect estimation of GPP and shortwave radiative forcing