Jet streams are narrow bands of strong winds in the upper troposphere and have dominant influences over short-term weather patterns, and long-term climate and global temperatures. The Jet Stream Visualization Tool (jsviz) is a graphical tool to help see the variable path of the strongest wind speeds in the upper atmosphere and how these maximum winds evolve with time and space. Combined with data layers of geopotential height and surface pressure, the tool helps provide understanding of the basic atmospheric dynamics associated with the jet stream. The tool is also used for evaluating the algorithm for determining the position of the jet stream and how it varies with time, latitude, longitude and altitude.
The tool provides a map and vertical section (along a given longitude) to view atmospheric pressure highs and lows and high wind features. ECMWF Reanalysis (ERA5) [1] data of wind speed (wspd), height of the geopotential surface (hgt), and mean sea-level pressure (msl) fields are displayed. Currently, the extents of the map and data focus on North America and hence the atmospheric jets in the northern hemisphere. These extents can be adjusted in code and data downloaded from ERA5 for other regions of the world.
The main map shows the average wspd for the 100-400 hPa levels in filled contours corresponding to the scale below the map, hgt (300 hPa level) in sold, black lines, and msl in grey lines (>=1013 hPa solid, <1013 hPa dashed) for North America. The section plot on the right displays the longitudinal slice through the upper atmosphere showing the detailed vertical structure of wspd and equal surfaces of hgt (solid black line for 300 hPa and solid blue for other) with altitude (km). By using the graphical interface, longitudes and times can be selected to show (or "play") the 4-dimensional (4D), dynamic nature of the upper atmosphere and the jet stream paths in northern hemisphere. Furthermore, the positions of the jet stream based on the algorithm (red dots) are also displayed, as they vary across the region with time.
Here is a screen image of the map and interface.
There are many different ways that jet stream locations are determined in the literature using reanalysis data. The algorithm employed in this version of the jsviz tool is as follows:
- Find maximum wind speeds in upper troposhere, using 4D ERA5 data, for each longitude at a given time.
- Apply further limitations when mulitple peaks are found
More details of the method are provided in the Jet Stream Characterization page.
You can clone the jsviz Github Repository or download the zipped code from Github:
git clone https://github.com/neaptide/jsviz.gitor download and unzip the code:
wget https://github.com/neaptide/jsviz/archive/master.zipOnce the code is downloaded locally, you will be using the environment.yml file in the code folder to get all the dependencies to build an environment with default name test_jsviz. You can use a different enviroment name by editing the environment.yml.
For this example the code is unzipped or cloned into C:\your\home\jsviz
Open a terminal that activates Anaconda base environment.
Change the working directory to where you unpacked the code and create the test_jsviz environment
(base) C:\your\home> cd jsviz
(base) C:\your\home\jsviz>conda env create -f environment.yml
Activate the test_jsviz environment
(base) C:\your\home\jsviz>conda activate test_jsviz
(test_jsviz) C:\your\home\jsviz>
You can now run jsviz.py in either IPython or in a Jupyter Notebook.
Open an Ipython terminal.
(test_jsviz) C:\your\home\jsviz>ipython
In Ipython, use the magic command %run to run the jsviz.py code for the specified year and month (YYYY_MM).
[1] %run jsviz.py 2018_01
This will initialize the interactive graph, previously described, to the first day and hour of that month on the map and you can begin using the interface. The veritcal section will iniatilize to the the first (left-most) longitude of the area. For example, if 2018_01 is used, the map and vertical section will show the data for 2018-01-01 at 00:00 (UTC) and the longitude of 140 W.
Start the Jupyter Notebook for your web browser.
(test_jsviz) C:\your\home\jsviz>jupyter-notebook
Find and open the run_jsviz.ipynb notebook.
You should see a code cell with the following code. Once the cell is run, it will load data for a specified year and month (YYYY_MM).
%matplotlib notebook
%run jsviz.py 2018_01
To run the cell block:
- Place the cursor within cell.
- Change
2018_01(YYYY_MM) to run a different year and month (1979-2019). - Press "Run" (
) to run the cell.
Once the "Figure 1" bar with the interaction button (
) is displayed, you can now use the graphical interface. The graph will be initialized to the first day and hour of that month on the map. The veritcal section will iniatilize to the the first (left-most) longitude of the area. For example, if 2018_01 is used, the map and vertical section will show the data for 2018-01-01 at 00:00 (UTC) and the longitude of 140 W.
- Select another longitude by moving the "Long" slider or pressing left- (<) and right-arrow (>) associated with it.
- Select a different time and date by moving the "Date" slider or pressing left- (<) and right-arrow (>) associated with it.
- See the Jet Stream Characterization description for "Local Maxima Detection" parameters and further peak limitation.
- Press "Jet Stream ON/OFF" button to toggle display of jet stream markers.
- Press "Limitation ON/OFF" button to enable/disable further limitation of jet stream algorithm.
jsviz has dependency on the following Python modules:
- numpy
- matplotlib
- metpy
- netCDF4
- scikit-image
- opencv-python
The jsviz tool is based upon work supported by National Science Foundation under grant OCE-1558920.
[1] Copernicus Climate Change Service (C3S) (2017): ERA5: Fifth generation of ECMWF atmospheric reanalyses of the global climate . Copernicus Climate Change Service Climate Data Store (CDS), date of access. https://cds.climate.copernicus.eu/cdsapp#!/home