Scenarios simulations of Bergen (TURBAN - D06)

Basic information

This dataset contains simulation results for the so-called Danmarksplass domain, an area in the city of Bergen, Norway. Danmarksplass is the major trafic conjuction point in Bergen. Danmarksplass is located in a densily built-up and populated urban district subjected to many environmental challenges among them air pollution by NOx and aerosols (particulate matter PM2.5 and PM10) are considered as significant health threats. Detailed studies of the Danmarksplass meteorological conditions, air quality, and structure of the air polution could be found in Wolf et al. (2014a,b; 2017; 2020; 2021). 

This dataset contains two collections of new simulations of air quality at Danmarksplass. The simulations were performed with the PALM modeling system v23.04 with additional modules developed in the TURBAN project (Radovic et al., 2024; Resler et al. to be submitted). This two collections of the PALM runs spans two air pollution episodes :

Summer episode (2019-07-20 to 2019-07-27) of the record breaking summer heat wave in Bergen (on July 26, 2019; https://www.bt.no/nyheter/lokalt/i/QoE3rA/naa-er-334-den-nye-varmerekorden-i-bergen) when considerable problems (haze, smell) from the aerosol air pollution (PM2.5 and PM10) were noted (Esau et al., 2022). The main sources of pollution are ships in the port of Bergen and road traffic.

Winter episode (2021-02-04 to 2021-02-12) of the prolonged cold wave in Bergen (https://www.bt.no/nyheter/lokalt/i/lEOVm9/kulde-med-historisk-sus-i-bergen) when considerable problems (haze, smell) from the aerosol air pollution (PM2.5 and PM10) were noted (Esau et al., 2022). the main source of pollution is household wood combustion for heating.

Thus, the dataset presents multiple daily (24 h) runs driven by the results of model downscaling of ERA5 reanalysis with WRF model (produced by K. Eben and M. Bures). The aerosol emission sources are described in Wolf et al. (2020; 2021). The runs where combined into a single dataset in post-processing. 

The observational data for these two episodes collected in the TURBAN project are avaialble in Esau et al. (2023).

For more detailed description of the experiments see the TURBAN project website at https://www.project-turban.eu/.

General organisation

The dataset organized as follows. There are four folders named as "scenario_Bergen_{episode}_PALM_set0_{domain}_domain" where

{episode} is either the selected summer scenario {episode} = summer_2019-07 or the winter scenario {episode} = winter_2021-02

{domain} is either the larger coarse resolution domain smaller fine resolution domain {domain} = child 

Each scenario folder contains daily dynamic, static, chemistry drivers and the PALM configuration files in subfolders INPUT in folders designated by the day, e.g., dpc_set0_D0_D1_20210215/INPUT, to rerun all simulations if necessery. For convinience, the common static driver (set0_2021-02_static.nc) and the combined model output averaged over 3 h intervals (combined_set0_2021-02_av_xy.nc) are provided.

The simulation results for each scenario contain:

The combined PALM runs output average over 3 hours and presented on certain model levels (the files "*_av_xy*.nc"). The files are in the NetCDF4 format.

The maps in PNG format visualizing the most relevant results for stakeholds (files in folder NMAP)

In addition, we included the template of a Python script that is used to read the data and create Nmaps.

Modelled variables

Each subfolder includes 4 subfolders with variables. Variable kc_PM10 is the concentration of PM2.5 at the 3rd model level, theta_2m is the potential temperature at 2m above ground, tsurf is the surface temperature and wspeed_10m is the wind speed at 10m above ground.

File nomenclature

Each file (PNG) has the same nomenclature. An example (set0_kc_PM10_2021-02-04T0300.png) could be parsed as: domain name (set0), variable name (kc_PM10), date and time of the output (2021-02-04T0300) and averaged period (from (03:00 - 3h) to 03:00). So, the result is a map with 3 hourly averaged PM2.5 concentrations for 4 February 2021 between 00:00 and 03:00 UTC.

Important note

During the processing phase a few potentially important problems were identified and need to be analysed in detail. One of them are extremely overestimated concentrations due to stable conditions from boundary condition inputs. In certain situations it can happen that the best regional meteorological model can provide inappropriate input conditions for some episode. This needs to be checked in detail before any following interpretation.

References

Esau, I.: TURBAN – Observational datasets for studies of urban air quality hazard scenarios in Bergen, Norway, DataverseNO, V1, https://doi.org/10.18710/QHUAZ2, 2023.

Radović, J., Belda, M., Resler, J., Eben, K., Bureš, M., Geletič, J., Krč, P., Řezníček, H., Fuka, V., 2024. Challenges of constructing and selecting the “perfect” boundary conditions for the large-eddy simulation model PALM. Geosci. Model Dev. 17, 2901–2927. https://doi.org/10.5194/gmd-17-2901-2024

Wolf, T., Esau, I., Reuder, J., 2014a. Analysis of the vertical temperature structure in the Bergen valley, Norway, and its connection to pollution episodes. J. Geophys. Res. 119. https://doi.org/10.1002/2014JD022085

Wolf, T., Esau, I., 2014b. A proxy for air quality hazards under present and future climate conditions in Bergen, Norway. Urban Clim. 10, 801–814. https://doi.org/10.1016/j.uclim.2014.10.006

Wolf-Grosse, T., Esau, I., Reuder, J., 2017. The large-scale circulation during air quality hazards in Bergen, Norway. Tellus A Dyn. Meteorol. Oceanogr. 69, 1406265. https://doi.org/10.1080/16000870.2017.1406265

Wolf, T., Pettersson, L.H., Esau, I., 2020. A very high-resolution assessment and modelling of urban air quality. Atmos. Chem. Phys. 20, 625–647. https://doi.org/10.5194/acp-20-625-2020

Wolf, T., Pettersson, L.H., Esau, I., 2021. Dispersion of particulate matter (PM2.5) from wood combustion for residential heating: optimization of mitigation actions based on large-eddy simulations. Atmos. Chem. Phys. 21, 12463–12477. https://doi.org/10.5194/acp-21-12463-2021

Acknowledgements

The PALM simulations, and pre- and postprocessing were performed partially on the HPC infrastructure of the Norwegian SIGMA2 facilities. The work was performed within the project TURBAN (TO01000219; TURBAN – Turbulent-resolving urban modelling of air quality and thermal comfort) supported by Norway Grants and Technology Agency of the Czech Republic.