Seefeld Cold-Air Pool Experiment (SEECAP): WRF Simulation Output with snow cover January 16 2020 0000 UTC to January 17 2020 1200 UTC

The Seefeld Cold-Air Pool Experiment (SEECAP) focused on the cross-country skiing area Olympiaregion Seefeld and in particular the topographic setting in the Nordic ski arena which favors the formation of cold-air pools and took place between December 2019 and March 2020. The measurement data are described in Rudolph (2022) and Rauchöcker et al. (2024d) and meteorological measurement data associated with SEECAP are published in Rauchöcker et al. (2024c). This upload contains WRF simulation output data for the night between January 16 and January 17 2020 with snow cover and the plotting routines to reproduce the figures in Rauchöcker et al. (2024d). The night between January 16 and January 17 2020 initially featured ideal condition for cold-air pool formation followed by a disturbance around midnight. There is also an upload with simulation output for the same night, but without snow cover (Rauchöcker et al., 2024a). Also available in a different dataset are data from a simulation with snow cover for the night between January 12 and January 13 2020 (Rauchöcker et al., 2024b), which featured an undisturbed cold-air pool for almost the entire night. This case was considered to feature in Rauchöcker et al. (2024d), but a different case was chosen because some measurement data was not available during this period.

WRF Simulation Output

This Dataset includes data generated with WRFlux v1.4.1 (Göbel et al.,  2022), a fork of the Weather Research and Forecasting model WRF (Skamarock et al. 2021).  WRFlux allows to calculate the contribution of different processes to the potential temperature tendency at each grid point. The data published here is from the innermost simulation domain with 40m horizontal resolution and 10m vertical resolution close to the surface. The simulations were initialized at 00:00 UTC January 16 2020 and run until 12:00 UTC January 17 2020.

Three different simulations were performed: two simulations with modified snow cover as described in Rauchöcker (2022), one each with the MYNN 2.5-order and the SMS-3DTKE PBL parameterizations (a scheme that blends a PBL scheme and a LES subgrid parameteriztion in the greyzone of turbulence), and one without snow cover with the MYNN 2.5-order PBL parameterization. Otherwise the simulations were identical. This dataset includes the two simulation with snow cover, where  jan126_sms.zip contains the files relating to the simulations with the SMS-3DTKE scheme and jan16.zip those for the simulation with the MYNN 2.5-order PBL parameterization. A detailed description of the model setup can be found in Rauchöcker et al (2024d) and in the files namelist.input and namelist_sms.input that were used for the simulations. 

Standard WRF output can be found in wrfout_40m_jan16 and wrfout_40m_jan16_sms. The mean wind speed components, which were necessary to rotate the tendencies in a coordinate system that is aligned with the valley orientation, are contained in windout_40m_jan16 and windout_40m_jan16_sms. These variables were contained in the  unprocessed output files produced by WRFlux; the full files were unfortunately too large to be included here. The postprocessed tendencies are stored in tend_40m_jan16.nc and tend_40m_jan16_sms.nc.

Plotting routines

Python scripts and environment files to reproduce most figures in Rauchoecker et al. (2024d) are included in code.zip. To reproduce plots involving measurement data, which is available in Rauchöcker et al. (2024c), is also needed.

Due to conflicts between some packages, two different environment were needed. To reproduce Figure 2, install the orthoplot environment by running "conda env create orthoplot.yml" in a terminal window, activate it  ("conda activate orthoplot") and then run ortho_plot.py. All other plots require the wrfstuff environent (installed by running "conda env create wrfstuff.yml" and activated by "conda activate wrfstuff") and are produced by paper_plots.py. Functions used to load data and plot the figures are included in dataload.py and plotting_routines.py, respectively.

Two variables decide which figures are plotted for which dataset: dataname and doplot. The variable dataname defines the path to the dataset that should be used to produce the figures, while the value doplot defines which figure to reproduce. By setting doplot="fig1", Figure 1 is reproduced, while doplot="fig7" and doplot="fig9" reproduce Figures 7 and 9, respectively. For all other values for doplot, Figures 4, 5, 6, 8 and 11 are reproduced. We decided not to include a script to plot Figure 3 because the data the climatology is based on is owned by GeoSphere Austria - the agency operating the permanent weather station. Further, no script for reproducing Figure 10 is included because it was not created within the framework of Python.

Geofiles

The files included in geofiles.zip are needed to plot Figure 1 and Figure 2, although not of particular interest on their own. Included are output files from geogrid.exe, whicih are necessary to plot the domains overview (Figure 1), as well as an orthophoto (orthophoto.tif) and high-resolution digital elevation model (topo_hr.tif) which are both based on data from Land Tirol.