Russian Arctic COSMO-CLM hindcast over 1980-2016 period

COSMO-CLM Russian Arctic hindcast over 1980 - 2016 years. All data are presented in directories containing 1 year data. Almost each directory contains 12 surface meteorlogical variables (at the moment) with 3 hours time step. These variables include 2 m temperature (T_2M.nc, K), 10 m zonal and merid...

Full description

Bibliographic Details
Main Authors: Platonov, Vladimir, Varentsov, Mikhail
Format: Article in Journal/Newspaper
Language:unknown
Published: figshare 2021
Subjects:
Online Access:https://dx.doi.org/10.6084/m9.figshare.c.5186714.v18
https://figshare.com/collections/Arctic_COSMO-CLM_reanalysis_all_years/5186714/18
Description
Summary:COSMO-CLM Russian Arctic hindcast over 1980 - 2016 years. All data are presented in directories containing 1 year data. Almost each directory contains 12 surface meteorlogical variables (at the moment) with 3 hours time step. These variables include 2 m temperature (T_2M.nc, K), 10 m zonal and meridional components of wind speed (U_10M.nc, V_10M.nc, m/s), total precipitation (kg/m2), 10 m wind speed maxima (VABSMX_10M.nc, m/s), sea level pressure (PMSL.nc, Pa), surface sensible heat flux (SHFL_S.nc, W/m2), surface latent heat flux (LHFL_S.nc, W/m2), surface net downward shortwave radiation (SOBS_RAD.nc, W/m2), direct downward shortwave radiation at the surface (SWDIRS_RAD.nc, W/m2), diffuse downward shortwave radiation at the surface (SWDIFDS_RAD.nc, W/m2), downward longwave radiation at the surface (THDS_RAD, W/m2). The number of variables will be increased in the near future. Besides the meteorological variables, each '$YEAR' directory contains 2 additional utulity files. There is latlon.nc file containing latitude and longitude grid data according to all model fields. And there is time${YEAR}.nc file (where ${YEAR} is the corresponding year) containing the correct time variable in format seconds from 01.January.$YEAR.The latter issue came from the specific experiments configuration, which included the monthly reinitialization runs. Therefore, in the whole year the time variable in the meteorological data files was written incorrectly, setting with start of each month to zero. The time${YEAR}.nc contains the correct time variable values ready to evaluations. IMPORTANT NOTE 1! 10 m zonal and meridional wind speed components data (U_10M.nc and V_10M.nc) are given with respect to the model grid coordinate system. Therefore, to get wind speed components and wind direction with respect to geographical coordinates (i.e., latitude and longitude), an additional transformation is necessary. More detailed information about required calculations are given in the COSMO Model Dynamics and Numerics Description, available at: http://www.cosmo-model.org/content/model/documentation/core/cosmo_dynamics_5.05.pdf. It is important to note that this transformation concerns wind direction only, and does not affect wind speed values, therefore the U_10M and V_10M variables could be used for wind speed calculations and analysis without any transformations. IMPORTANT NOTE 2! Since the data are given with 3 hours timestep, there is an issue concerning the TOT_PREC variable. The TOT_PREC output variable contatins 1-hour sums of precipitation, therefore, the presented TOT_PREC.nc files contains 1-hour sums of precipitation for each 3 hours, instead of 3-hours precipitaion sums. This issue will be fixed in the next hindcast version by extending the timestep up to 1 hour instead the current 3 hours. Sorry for inconveniences. You can learn more detailed information about experiments configuration and other hindcast issues in: http://dx.doi.org/10.1088/1755-1315/611/1/012044http://dx.doi.org/10.3390/atmos12030350