Backward trajectories within a circle around the ship using the LAGRANTO model, for the expedition ARTofMELT, Arctic Ocean, 2023

Sonja Murto, Michael Tjernström

7-day backward trajectories from within a circle around the ship track of the Swedish icebreaker Oden during the ARTofMELT (Atmospheric Rivers and the Onset of Arctic Melt) expedition in the Fram Strait pack ice during the spring transition, 8 May to 13 June of 2023.

The trajectories were calculated with the Lagrangian Analysis Tool LAGRANTO on ERA5 data — fifth generation of ECMWF atmospheric reanalysis data — initialised at locations 30 km equidistant from each-other within a circle (with origo at the ship and a radius of 100 km) every 6-hour at 14 different vertical starting levels.

The data tracked and stored in 1-hour intervals along the trajectories include 4D location (longitude, latitude and pressure), air temperature, humidity variables and distance from start, whereas additional surface variables, such as sea ice fraction, surface pressure and surface temperature, are interpolated along the tracks.

The data enable analysis of air mass origin and air mass transformation of an ensamble of air parcels ending up at Oden during the ARTofMELT expedition that occurred in the Arctic spring of 2023. A measure of the spread and uncertainty of the trajectories can also be obtained by the data.

Related objects

Datasets

• Trajectories — raw data
• Backward trajectories from the ship
• Forward trajectories from the ship
• Forward trajectories from within a circle around the ship

Code

• Scripts for creating and plotting netCDF files

Citation

Sonja Murto, Michael Tjernström (2024) Backward trajectories within a circle around the ship using the LAGRANTO model, for the expedition ARTofMELT, Arctic Ocean, 2023. Dataset version 1. Bolin Centre Database. https://doi.org/10.17043/oden-artofmelt-2023-trajectories-backward-circle-1

References

Hersbach H, Bell B, Berrisford P, Biavati G, Horányi A, Muñoz Sabater J, Nicolas J, Peubey C, Radu R, Rozum I, Schepers D, Simmons A, Soci C, Dee D, Thépaut J-N (2023) ERA5 hourly data on single levels from 1940 to present. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). https://doi.org/10.24381/cds.adbb2d47 (Accessed on 18-Jan-2024 and on 05-Feb-2024)

Murto S, Tjernström M (2024a) Air parcel trajectories for the expedition ARTofMELT, Arctic Ocean, 2023 — raw data. Dataset version 1. Bolin Centre Database. https://doi.org/10.17043/oden-artofmelt-2023-trajectories-raw-1

Murto S, Tjernström M (2024b) Backward trajectories along the ship track using the LAGRANTO model, for the expedition ARTofMELT, Arctic Ocean, 2023. Dataset version 1. Bolin Centre Database. https://doi.org/10.17043/oden-artofmelt-2023-trajectories-backward-ship-1

Murto S, Tjernström M (2024c) Forward trajectories along the ship track using the LAGRANTO model, for the expedition ARTofMELT, Arctic Ocean, 2023. Dataset version 1. Bolin Centre Database. https://doi.org/10.17043/oden-artofmelt-2023-trajectories-forward-ship-1

Murto S, Tjernström M (2024d) Forward trajectories within a circle around the ship using the LAGRANTO model, for the expedition ARTofMELT, Arctic Ocean, 2023. Dataset version 1. Bolin Centre Database. https://doi.org/10.17043/oden-artofmelt-2023-trajectories-forward-circle-1

Murto S (2024) Scripts for creating and plotting netCDF files from air parcel trajectories for the expedition ARTofMELT, Arctic Ocean, 2023. Software version 1.0.0. Bolin Centre Code Repository. https://doi.org/10.57669/oden-artofmelt-2023-trajectories-1.0.0

Sprenger M, Wernli H (2015) The LAGRANTO Lagrangian analysis tool – version 2.0. Geoscientific Model Development 8:2569–2586. https://doi.org/10.5194/gmd-8-2569-2015

Data description

This dataset includes one netCDF (.nc) with a total file size of about 830 MB (compressed 457 MB).

Data in this dataset are for 7-day backward trajectories started every 6-hour at 14 different vertical pressure levels within a circle that has the origo at Oden. The number of trajectories initialised within the circle varies between 31 and 37 (normally 35) depending on the location of the ship at each initialisation time.

A list of all variable names, with short explanations and units, can be found in an associated Readme_circleTrajs_bw.txt file. The readme file also provides further details of the trajectory run set-up.

Variables

Primary variables along the trajectories are 4D variables of location (longitude, latitude and pressure), specific humidity and air temperature. Additional parameters include model-calculated variables from the primary variabels (air potential temperature, relative humidity and spherical distance) and afterwards interpolated surface variables along the trajectory tracks (e.g., near surface temperature, sea ice concentration and boundary layer height).

• Primary variables
  • day_of_year Decimal day of year of the trajectory initialisation time (in UTC) [time]
  • start_longitude Longitude at the location of the ship (at initialisation time (degrees_east). Note! This is the origo of the circle where trajectories are started [time]
  • start_latitude Longitude at the location of the ship (at initialisation time) (degrees_north). Note! This is the origo of the circle where trajectories are started [time]
  • nr_circle_trajectories Number (between 31 and 37) of trajectories initialised within the circle, shown for each initialisation time [time]
  • time_along_trajectories Real time along the trajectories, in hourly time intervals [time/ref_time_along_trajectories]
• Primary variables traced by the model
  • longitude Longitude along trajectories (degrees_east) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
  • latitude Latitude along trajectories (degrees_north) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
  • air_pressure Air pressure along trajectories (hPa) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
  • air_temperature Air temperature along trajectories (K) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
  • specific_humidity Specific humidity along trajectories (kg / kg) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
• Calculated variables traced by the model
  • air_potential_temperature Air potential temperature along trajectories (K) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
  • relative_humidity Relative humidity along trajectories (%) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
  • distance_from_start Spherical distance from the trajectory initialisation location (m) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
• Interpolated surface variables
  • sea_ice_area_fraction Sea ice area fraction (0-1) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
  • near_surface_air_temperature Near-surface (2m) air temperature (K) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
  • surface_temperature Surface skin temperature (K) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
  • surface_air_pressure Surface pressure (hPa) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
  • air_pressure_at_mean_sea_level Mean sea level pressure (hPa) [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
  • atmosphere_boundary_layer_thickness Height of atmospheric boundary layer (m), calculated with the bulk Richardson number [time/vertical_level/circle_trajectories/ref_time_along_trajectories]
  • land_sea_mask Land/sea mask (0-1), portion of land (1) to ocean and inland waters. A value > 0.5 relates to land and inland water mixture but no ocean, whereas a value of 0.5 and below are water surfaces [time/vertical_level/circle_trajectories/ref_time_along_trajectories]

Dimensions

Main dimension for the fields are time, referring to the trajectory initialisation time (as a datetime object; every 6 h from 8 May 00 UTC to 13 June 18 UTC 2023).

4D-variables obtain the following dimensions:

• time Trajectory initialisation time (as above)
• vertical_level Number (1⁠ – ⁠14) for the vertical starting pressure level: (1⁠ – ⁠5) every 10 hPa between 10 and 50 hPa AGL, (6⁠ – ⁠10) every 50 hPa between 100 and 300 hPa AGL and (11⁠ – ⁠14) every 100 hPa between 600 and 300 hPa.
• circle_trajectories Number (1⁠ – ⁠37) of trajectories started at one vertical level within the circle for each initialisation time. Note that the non-existing trajectories are indicated with NaN (number of trajectories initialised within the circle varies by initialisation time, between 31 and 37)
• ref_time_along_trajectories Reference time since start along each trajectory (in negative hours for backward trajectories; from 0 to −168 for 7-day backward trajectories)

Trajectories are initialised every 6 h and the time step for each trajectory is 1 h.

Comments

Trajectory model

Trajectories are calculated by the LAGRANTO model (Sprenger & Wernli, 2015) on ERA5 data on model levels. Trajectory computations are performed on ECMWF’s high-performance computing facility (HPCF). This final dataset is created from the native output text files provided by LAGRANTO and from ERA5 data on surface levels (Hersbach et al. 2023) that are interpolated along the computed trajectory tracks. (See "Interpolated surface variables" as listed above). The trajectory native text files are available as a separate dataset.

Quality control

No quality control is applied except for within the trajectory model, a jumping flag is utilised to allow for trajectories to continue if they hit the ground (it is raised from the ground a little in order to continue).

Associated raw data, processed data and code

The raw data (trajectory text files), from which this dataset is derived, was created by Murto & Tjernström (2024a) and is available separately. The raw-data dataset is provided in a .zip file containing text files of four different sets of trajectory runs. User-friendly datasets, in NetCDF format, derived from the raw text files, have been created for all four trajectory runs:

1. Backward trajectories started every hour along the ship track, by Murto & Tjernström (2024b)
2. Forward trajectories started every hour along the ship track, by Murto & Tjernström (2024c)
3. Backward trajectories started every 6-hours within a circle around the ship (this dataset)
4. Forward trajectories started every 6-hours within a circle around the ship, by Murto & Tjernström (2024d)

The current dataset is a combination of the original raw data files, with additional surface variables interpolated along the provided trajectory tracks. The computer codes in Python, written by Murto (2024), used to derive this dataset in NetCDF format from the raw data files are also available. In the same code repository, we also provide python codes for reading in the netCDF files and plotting the trajectories, as well as example figures to illustrate the trajectories.

Data creator's role

Trajectories were computed post expedition based on the ship's GPS location along the track of the ARTofMELT 2023 expedition. The set-up for the trajectory runs as well as the creation of this dataset in NetCDF format were performed by one of the members of the Stockholm University Meteorology team onboard Oden:

• Sonja Murto (dataset creator)

Additionally:

• Michael Tjernström (chief scientist; project PI)

Acknowledgements

Anna Lewinschal (scientific programmer at Stockholm University) provided with example raw-code for runs on the ECMWF's supercomputing facility. The original setup of the LAGRANTO model to the supercomputer facility was done by Moritz Deinhard (researcher at Karlsruher Institut für Technologie, Germany).

The expedition

Data were produced for the ARTofMELT 2023 expedition on board the Swedish icebreaker Oden, which was organized by the Swedish Polar Research Secretariat from 8 May through 14 June, 2023.