Main Model Configuration

Attention

The most up-to-date DEPHY forcing file (V2.4) may be found here.

Note

While we hope that all participants can apply the below specifications to their model(s), we do not want to discourage anyone from participating in the MIP! Please reach out to us if you have any concerns regarding the model setup or plan to submit outputs with specifications different from those below.

Model component	Setting
Horizontal grid cell spacing	LES: Dx=Dy=100 m SCM: use operational spacing
Horizontal domain dimensions/size	Small-domain LES: Nx=Ny=256; Lx=Ly=25.6 km (required) Large-domain LES: Nx=Ny=1280; Lx=Ly=128 km (desired but not required) - Note: if you are unable to perform a Large-domain simulation, then we will accept smaller domain configurations - File naming conventions may be found on this page.
Vertical grid	LES: according to input forcing file specifications - Variable name to use: zw_grid SCM: use operational grid
Domain top	LES: 7 km SCM: use operational height
Lateral boundary conditions	Doubly periodic
Start/end times	22 UTC on 12 March 2020 18 UTC on 13 March 2020
Initial profiles	Thermodynamic and kinematic soundings provided - Variable names to use: ps, thetas, lev, pressure, u, v, temp, theta, qv, o3
Initial perturbations for LES	Theta only: 0.1 K below 250 m
Surface forcing	Specified time-varying skin temperature* - Assume the surface is always water - Variable name to use: ts Compute surface fluxes interactively using fixed roughness lengths for the entirety of the simulation (results are relatively insensitive to reducing roughness lengths over ice) - Momentum, z0m = 9.0e-4 m - Temperature and moisture, z0t = z0q = 5.5e-6 m - A brief highlight on sensitivity to roughness lengths may be found on this page.
Large-scale forcing	Time- and height-varying pressure gradient forcing via the geostrophic wind - Variable names to use: ug and vg - Assume trajectory-averaged latitude for Coriolis: 74.5 N
Nudging**	None (not required)
Subsidence	None (results are relatively insensitive)
Radiation	Only interactive longwave - Ozone sounding specified in input file - CO2 = 420 ppm - No solar radiation (results are relatively insensitive)
Aerosol & microphysics	Fixed droplet - Nd = 20/cm3 Diagnostic ice formation - Ni = 25/L (minimum total) where qc+qr>1e-6 kg/kg and T<268.15 K - Only other ice crystal formation mechanism active is homogeneous drop freezing Prognostic aerosol^ - Initialize with three lognormal modes (see aerosol specification page overview), assuming vertically uniform number concentration in each mode (see table for modal parameters and kappa values); if participant is limited to one aerosol mode, please initialize with the accumulation mode only (Mode 2) - Apply sea spray emissions (see specifying surface aerosol emissions) to accumulation and sea spray modes (Modes 2 and 3) using recommended split (98.5% to Mode 2 and 1.5% to Mode 3); if participant is limited to one aerosol mode, please place 100% of emission flux into that mode (with Mode 2 characteristics) Prognostic ice^^ - Specified to assume that the only heterogeneous ice nucleation is on sea spray aerosol, and to further assume that the accumulation mode (Mode 2) is half sea spray (at all times) and the sea spray mode is 100% sea spray (Mode 3) - Participants are requested to choose their parameterization of preference; see aerosol-ice nucleating particle closure for an evaluation of several schemes using in situ measurements of aerosol PSDs and INP in cloud-free air at upwind and downwind sites

Notes

*From 0-2h we use a constant ice skin temperature (247.0 K), and from 2–20 h we use time-varying SST from ERA5.
**While the official specification does not call for any nudging, we include nudging variables extracted from ERA5 along the trajectory in the forcing file to: (1) easily enable sensitivity tests and (2) readily plot against them as quasi-observational data.
^It is anticipated that prognostic aerosol schemes will account for number losses to collision-coalescence (e.g., rain formation and riming) and potentially others (e.g., dry deposition).
^^Since this is intended to represent heterogeneous ice nucleation in the immersion mode, it is anticipated that the freezing process will operate by moving droplets to an ice class.