| Summary: | This dataset provides the hourly output of the firn model developed for Vandecrux et al. (2020, https://doi.org/10.1017/jog.2020.30). In that study, we filtered and gap-filled weather data from nine Greenland Climate Network (GC-Net) stations: Crawford Point 1 (CP1), Dye-2, NASA-SE, NASA-E, Saddle, South Dome, NASA-U, Summit and Tunu-N. We used these data to calculate the surface energy and mass balance which is used to force the firn model. The firn model is decribed in Vandecrux et al. (2018, https://doi.org/10.1029/2017JF004597, 2020, https://doi.org/10.1017/jog.2020.30) and its code is available at https://github.com/BaptisteVandecrux/SEB_Firn_model. The firn model has 200 layers. each composed of snow, ice and water compartments. Layers are managed in a Lagrangian framework with a splitting/merging strategy that keeps the resolution higher close to the surface. During snowfall, fresh snow is added as a new 4 centimeter (cm) w.e. thick layer with a density of 315 kilogram per cubic meter (kg m-3) to the top of the model. During melt, mass is taken from snow and ice compartments of the top layer and transferred to the water compartment. Downward flow follows Darcy's law as described by Langen et al. (2017, doi: 10.3389/feart.2016.00110) and Vandecrux et al. (2018, https://doi.org/10.1029/2017JF004597, 2020, https://doi.org/10.1017/jog.2020.30) . When the underlying layer is below freezing point, the water is refrozen, moved to the ice compartment of that layer until either the layer reaches melting point or the layer's bulk density reaches ice density. In the rare cases when it occurred, ponding of water over a layer of reduced permeability was allowed. No lateral runoff was allowed. The firn density is updated every time step for compaction as calculated from the overburden pressure (Vionnet et al., 2012, https://doi.org/10.5194/gmd-5-773-2012).
|
|---|