grib2pf

This is a python script and C code which takes GRIB data (mainly MRMS) and outputs it to a placefile for usage with Supercell-Wx. This adds the ability for nation wide weather radar in Supercell-Wx.

Update Guide Windows

There are 2 ways to update this on Windows. The second way is will allways work but takes a bit more effort. The first ways is simpler

Simple Update

Unzip the new grib2pf folder from the zip overtop your preexisting folder, and select "Replace the files in the destination". This will save your settings.

Complete Update

1. Go into the grib2pf folder, then the _internal folder.
2. Copy settings.jsonc or settings to your Desktop, or somewhere outside of the grib2pf folder.
3. Delete the grib2pf folder.
4. Unzip the new grib2pf folder where you want it.
5. Move the settings.jsonc or settings file back into the _internal folder inside the grib2pf folder.

Setup Guide Windows

There are 2 main ways to run this in Windows: WSL or via an executable. For WSL follow the Linux instructions. These instructions are for the executable.

Windows Dependencies

You will need the latest Microsoft Visual C++ Runtime to run this program. If you have Supercell-Wx v4.5.0 running on your system, then you should already have them. Otherwise download them from https://aka.ms/vs/17/release/vc_redist.x64.exe .

Download The Executable

Download the executable form the latest release and extract it. You will want to extract it to a permanent location. This program does not install itself, so where you place it is where it will be installed.

Running the GUI

grib2pf-ui.exe or grib2pf-ui can be run by double clicking it. This will launch a settings UI. You can select a preset to start with by clicking Load Preset. From there you can modify the settings to your liking. Clicking Run will save your settings and run grib2pf. Once your settings are saved, you can simply run grib2pf directly, without the GUI.

Run It

For a basic run, you can simply double click on grib2pf.exe or grib2pf. This will put a basic settings file (settings.json) in the _internal folder. It should look like below, where {_internal folder} is replaced by the path to the _internal folder.

{
    "url": "https://mrms.ncep.noaa.gov/data/2D/MergedBaseReflectivity/MRMS_MergedBaseReflectivity.latest.grib2.gz",
    "imageFile": "{_internal folder}\\baseReflectivity.png",
    "placeFile": "{_internal folder}\\baseReflectivity.txt",
    "verbose": true,
    "refresh": 15,
    "regenerateTime": 60
}

This will generate a base reflectivity placefile. The "URL" for this placefile with simple be the path following placeFile (You may need to change \\ to \). This setup will update the placefile every minute. Updating the placefile is somewhat slow, and the MRMS data is only updated every two minutes, so this should be enough. In order to stop it use press CTRL-C in the window it launches.

Using Multiple Placefiles

One way to have multiple placefiles is to simply copy the grib2pf folder. Each folder will have its own settings files. Just make sure that the paths in settings.json do not overlap.

Another way to generate multiple placefiles by having multiple settings files. By default, grib2pf only uses the _internal/settings.json file, but this can be changed. Probably the easiest way is to create a shortcut to launch grib2pf with an argument. In the Create Shortcut dialog, simply enter the path to grib2pf.exe followed by a space, and then the path to your settings file. You may want to use quotes around the paths. Below is an example.

"C:\Users\user\Documents\grib2pf\grib2pf.exe" "C:\Users\user\Documents\grib2pf\settings2.json"

Double clicking on this shortcut will launch grib2pf using the given settings file.

Installing on Linux

To install on Linux you can simply download or clone the repository, then run the following command to install all dependencies.

git submodule update --init --recursive
mkdir build
cd build
cmake -DCMAKE_BUILD_TYPE=Release ..
make
cd ..
pip install -r requirements.txt

You can then run grib2pf.py and grib2pf-ui.py as normal scripts.

This will only work if your distro does not manage your Python packages. Otherwise, you can use a Python virtual environment. To do so, do the following.

python3 -m venv venv
./venv/bin/pip -r requirements.txt

In order to run grib2pf.py, use

./venv/bin/python grib2pf.py

If you are using WSL, you will want to modify the path to output the placefile and image in settings.json to a folder accessible by Windows.

Adding To Supercell-Wx

Copy the path to the placefile, and use it as the URL for a placefile in Supercell-Wx. It is recommended to place your grib placefile layer(s) below the radar layer. The local radar will have a higher resolution, and have more plots. You could also only enable the grib placefile layer(s) on specific panes.

Arguments

When run without an argument, grib2pf will run using the settings file in the install directory (grib2pf on Linux, and grib2pf\_internal on Windows). When run with one argument, grib2pf will use that argument as the path to a settings file. Otherwise, grib2pf can be run without a settings file, using arguments instead. The arguments documentation can be seen by running grib2pf with --help. The arguments align with the settings described below.

Settings

Below are all the settings supported by this program.

mainType: The overall type of plot you want generated. Right now only basic and MRMSTypedReflectivity are valid. basic just applies the color table to the data. MRMSTypedReflectivity applies different color tables to the data depending on the type precipitation. This is designed to be used with reflectivity, although nothing is stopping you from using something else.

aws: Boolean describing if AWS data should be used

product: AWS product to use

typeProduct: The AWS product to use for precipitation type for MRMSTypedReflectivity. It should always be end in /PrecipFlag_00.00/.

reflProduct: The AWS product to use for reflectivity for MRMSTypedReflectivity. Although it is called reflectivity, other products will work fine.

pullTime: How often to pull AWS for new data

url: The URL to pull from. Should probably come from "https://mrms.ncep.noaa.gov/data" "https://mrms.ncep.noaa.gov/data/2D/MergedBaseReflectivity/MRMS\_MergedBaseReflectivity.latest.grib2.gz" is a useful reflectivity plot. Descriptions of the plots can be found at (https://www.nssl.noaa.gov/projects/mrms/operational/tables.php)

imageFile: The file name were the image should be written. Should be an absolute path to a png

placeFile: The file name were the placefile should be written

palette: The path to a GRS color table to use for this plot. Useful for non reflectivity plots. Defaults to NOAA's Weather and Climate Toolkit reflectivity palette.

rainPalette: The path to a GRS color table to use for this plot in areas with rain. (MRMSTypedReflectivity only)

snowPalette: The path to a GRS color table to use for this plot in areas with snow. (MRMSTypedReflectivity only)

hailPalette: The path to a GRS color table to use for this plot in areas with hail. (MRMSTypedReflectivity only)

title: The title to display in Supercell-Wx

refresh: How often Supercell-Wx should refresh the placefile, in seconds

imageURL: The URL at which the image will be hosted. Unnecessary for local usage, but useful for web hosting

imageWidth: The width of the image to be generated. Only effects the resolution off the plot

imageHeight: The height of the image to be generated. Only effects the resolution off the plot

regenerateTime: How often to regenerate the image and placefile in seconds. Without this, it will only generate once

verbose: Print status messages if true

timeout: How long to wait for a responce from the URL in seconds. Defaults to 30s. No way to disable, because that will lock up the program

gzipped: If the GRIB file is gzip compressed. Defaults to true. Is true for MRMS data.

renderMode: How the renderer should select the data for each pixel. Average_Data averages the data under the pixel. Nearest_Data finds the nearest data point the the center of the pixel. Nearest_Data is good for integer or flag data. Otherwise Average_Data is probrably better.

Other Radar Viewers

If another radar viewer uses a Mercator projection and has placefile support, this project should work, although I give no guaranties.

Overview of how grib2pf Works

The MRMS grib files provide a grid (in latitude/longitude space) of data. Each grid point has a value, latitude, and longitude. Every point in a column has the same longitude, and every point in a row has the same latitude. grib2pf converts the latitudes and longitudes to x,y coordinates on a Mercator projection. In order to maximize the pixel density of the output image, these x,y coordinates are normalized such that 0,0 is the top left, and imageWidth - 1,imageHeight - 1 is the bottom left of the grib data (image coordinates). This is simply a linear transformation of normal Mercator projections, which is undone when the placefile is rendered because of the latitude and longitude coordinates saved in the placefile. The pixels are the average of all data points inside of them.