Lagranto.MPAS - Lagrangian analysis tool for MPAS

This is the version of LAGRANTO used to compute air-parcel trajectories based on the output of the Model for Prediction Across Scales (MPAS).

The Lagrangian analysis tool LAGRANTO is widely used in the atmospheric sciences, for instance, to identify flow structures in extratropical cyclones (e.g., warm conveyor belts), long-range transport pathways of moisture, or to study the physical processes underlying the formation of potential vorticity anomalies.

LAGRANTO is developed at the Institute for Atmospheric and Climate Science at ETH Zurich by Dr. Michael Sprenger and Prof. Heini Wernli.

For more information see http://www.lagranto.ethz.ch.

Installation

Download:

git clone https://github.com/steidani/lagranto.MPAS.git /path/to/local/lagranto.mpas

Using install.csh

Compiling the Lagranto code on a Linux platform:

>./code/install.csh [lib|core|all|clean|test]

Fortran compiler PGI is used.

Tutorial

1 Meteorological Data

The meteorological fields are on netCDF files named latlon.current.YYYY-MM-DD_00:00:00.nc containing 4 timesteps (every 6 hour: 00, 06, 12, 18).

Needed fields are:

• zonal wind (uzonal, in m/s)
• meridional wind (umeridional, in m/s)
• vertical wind (w, in m/s)
• geometric height (height, in m)
• additional fields (temperature, specific humidity, ..) to be traced along the trajectories.

2 Starting Position

Starting positions are specified in a file startf:

> more example/startf_z
0.00    -35    47    1000.00
0.00    -35    47    1500.00
0.00    -35    47    1000.00

The different columns are: time (relative to initation date), longitude, latitude, height (in m).

3 Tracing Meteorological Fields

Meteorological fields (in addition to time, lon, lat, z) to be traced are listed in a file tracevars:

> more example/tracevars
temperature    1.    P

The different columns are: name, scaling factor, location of field (P = Primary). In this case, P is latlon.current.YYYY-MM-DD_00:00:00.nc.

4 Trajectory Calculation

In this example, the trajectories are calculated 3 days (72) forward in time, stored every 6 h (-o 360), with starting date (-ref) 18 UTC, 28 December 1988. The command is:

> caltra startf_z 72 lsl_forward.1 -ref 19881228_1800 -o 360

The staring positions are taken from startf_z and the output is written to lsl_forward.1. All input files (netCDF, startf, tracevars) are expected in the directory where Lagranto was called (see example/run_lagranto.sh).

 > more example/lsl_forward.1
Reference date 19881228_1800 / Time range    4320 min

time       lon      lat     z temperatu
------------------------------------------

 0.00   325.000   46.000  1500   274.301
 6.00   325.637   49.153  2722   266.827
12.00   326.411   52.902  4382   253.898
18.00   327.010   56.194  5748   242.070
24.00   328.731   60.388  7045   228.991
30.00   334.821   65.331  7559   222.908
36.00   349.509   67.713  7301   225.142
42.00     8.087   63.862  6967   227.987
48.00    17.993   56.261  6932   228.178
54.00    22.249   49.719  6931   228.324
60.00    25.509   45.767  7007   227.827
66.00    29.933   44.689  6997   227.905
72.00    35.660   47.792  6794   228.377

4.1 Blocking Trajectory Calculation

The shell script script/run_lagranto.sh provides an example on calculating backward trajectories starting from blocking anticyclones. Starting position are taken from startfiles named startf_YYYYMMDD_HH_FLAG. These startfiles are created using the output "FLAG" from the atmospheric blocking tracking index ConTrack.

5 Analysis and Plotting of Trajectories

A Python library to work with trajectories can be found here: https://lagranto.readthedocs.io/en/latest/.