+
+ +
+

GOES CIRA Product Writer

+

Notebook +Python-AWIPS Tutorial Notebook

+
+
+

Objectives

+
    +

  • Use python-awips to connect to an EDEX server

    • +

  • Define and filter the data request specifically for new CIRA GOES16 +data +products

    • +

  • Resize the products to their native resolution

    • +

  • Write the individual bands (channels) locally

    • +

  • Combine and write the RGB product locally

    • +
+
+ +
+
+

1 Imports

+

The imports below are used throughout the notebook. Note the first +import is coming directly from python-awips and allows us to connect to +an EDEX server. The subsequent imports are for data manipulation and +visualization.

+
from awips.dataaccess import DataAccessLayer
+import cartopy.crs as ccrs
+import cartopy.feature as cfeat
+import matplotlib.pyplot as plt
+from datetime import datetime
+import numpy as np
+import os
+
+
+

Top

+
+
+
+

2 Initial Setup

+
+

2.1 EDEX Connection

+

First we establish a connection to Unidata’s public EDEX server. With +that connection made, we can create a new data request +object +and set the data type to satellite.

+
# Create an EDEX data request
+DataAccessLayer.changeEDEXHost("edex-cloud.unidata.ucar.edu")
+request = DataAccessLayer.newDataRequest()
+request.setDatatype("satellite")
+
+
+
+
+

2.2 Parameter Definition

+

After establishing the python-awips specific objects, we create a few +other parameters that will be used for the data query based off of known +values: projection, and extent.

+
# Create a projection for ECONUS and WCONUS
+# Set up the projection using known parameters (from the netcdf of GOES products)
+globe = ccrs.Globe(semimajor_axis=6378137.0, semiminor_axis=6356752.5, ellipse=None)
+sat_h = 35785830.0
+proj = ccrs.Geostationary(globe=globe, central_longitude=-75.0, satellite_height=sat_h, sweep_axis='x')
+
+# Define the extents for ECONUS and WCONUS in goes native coords
+# (originally taken from netcdf GOES data)
+extent = (-3626751., 1382263.5, 1583666.1, 4588674.)
+
+
+

Top

+
+
+
+
+

3 Function: set_size()

+

Here we’re defining a function that will allow us to pass in the +dimensions of the output file we desire in pixels. Default Python +methods require the size to be set in inches, which is confusing in our +case, since we know what the size of GOES images are in pixels. Also, +default Python functions add a padding when creating figures, and we +don’t want that.

+

This function allows the exact final image to be specified based in +pixels, with no padding or buffers.

+
def set_size(w,h, plt):
+    """ w, h: width, height in pixels """
+
+    # Convert from pixels to inches
+    DPI = plt.figure().get_dpi()
+    w = w/float(DPI)
+    h = h/float(DPI)
+
+    # Get the axes
+    ax=plt.gca()
+
+    # Remove the padding
+    l = ax.figure.subplotpars.left
+    r = ax.figure.subplotpars.right
+    t = ax.figure.subplotpars.top
+    b = ax.figure.subplotpars.bottom
+    figw = float(w)/(r-l)
+    figh = float(h)/(t-b)
+
+    # Set the final size
+    ax.figure.set_size_inches(figw, figh)
+
+    # Return the DPI, this is used when in the
+    # write_image() function
+    return DPI
+
+
+

Top

+
+
+
+

4 Function: write_img()

+

Next, we’re defining another function which takes the image data, file +name, projection, extent, reference time, and whether or not to print +out a footnote.

+

This method specifies the size of the output image and creates a plot +object to draw all our data into. Then it draws the GOES data, +coastlines, state boundaries, and lat/lon lines onto the image. +Additionally, if we want, it writes out a short footnote describing what +product we’re looking at. Finally, it writes out the figure to disk.

+

By default we’re specifying the output dimensions to be 5000x4000 +pixels, because that is the native GOES image size, but feel free to +modify these values if you wish to print out an image of another size +(you may want to keep the w:h ratio the same though).

+
def write_img(data, name, proj, extent, reftime, footnote):
+
+    # Specify the desired size, in pixels
+    px_width = 5000.0
+    px_height = 3000.0
+
+    # Create the plot with proper projection, and set the figure size
+    fig = plt.figure()
+    DPI = set_size(px_width, px_height, plt)
+    ax = plt.axes(projection=proj)
+
+    # Draw GOES data
+    ax.imshow(data, cmap='gray', transform=proj, extent=extent)
+
+    # Add Coastlines and States
+    ax.coastlines(resolution='50m', color='magenta', linewidth=1.0)
+    ax.add_feature(cfeat.STATES, edgecolor='magenta', linewidth=1.0)
+    ax.gridlines(color='cyan', linewidth=2.0, xlocs=np.arange(-180, 180, 10), linestyle=(0,(5,10)))
+
+    # Create and draw the footnote if needed
+    if footnote:
+        footnoteStr = '   CIRA-'+name[7:-4]+'-'+str(reftime)
+        plt.annotate(str(footnoteStr), (0,0), (0, 0), xycoords='axes fraction', textcoords='offset points', va='top')
+
+    # Write out the figure
+    plt.savefig(name, dpi=DPI, bbox_inches='tight', pad_inches=0)
+
+
+

Top

+
+
+
+

5 Get the Data and Write it Out!

+
+

5.1 Filter the Data

+

Define exactly what data we want to be printing out. This notebook is +designed to loop through and print out multiple images, so here we can +pick which images we’re wanting to print out. We’re specifying +ECONUS (for East CONUS), CLDSNOW, DBRDUST, and GEOCOLR +(for the new CIRA products) and the three channels for the RBG +composites.

+
+
+
Tip:
+
More information could be gathered by looking at all the available +location names (sectors), identifiers (entities), and parameters +(channels). To see those run the following lines of code after the +dataType has been set to satellite on the request object:
+
+
+
## Print Available Location Names
+print((DataAccessLayer.getAvailableLocationNames(request))
+
+## Print Available Identifiers and Values
+ids = DataAccessLayer.getOptionalIdentifiers(request)
+print(ids)
+for id in ids:
+    print(id, DataAccessLayer.getIdentifierValues(request, id))
+
+
+
# Define Location names
+sectors = ["ECONUS"]
+
+# Define creatingEntity Identifiers
+entities = ["CLDSNOW", "DBRDUST", "GEOCOLR"]
+
+# Define parameters
+ch1 = "CH-01-0.47um"
+ch2 = "CH-02-0.64um"
+ch3 = "CH-03-0.87um"
+channels = [ch1, ch2, ch3]
+
+
+
+
+

5.2 Define Output Location

+

Here we define a folder for where the satellite images will be written +to. The default directory is a new folder called ‘output’ that lives +whereever this notebook lives.

+
+
+
Tip:
+
If you specify the fully qualified path, it will no longer depend +on where this notebook is located. For example (for a Mac):
+
+
+
outputDir = '/Users/scarter/test_dir/output/'
+
+
+
# Define name of the desired end directory
+outputDir = 'output/'
+
+# Check to see if this folder exists
+if not os.path.exists(outputDir):
+    # If not, create the directory
+    print('Creating new output directory: ',outputDir)
+    os.makedirs(outputDir)
+else:
+    # If so, let the user know
+    print('Output directory exists!')
+
+
+
Output directory exists!
+
+
+
+
+

5.3 Write Out GOES Images

+
# First loop through the sectors (location names)
+for sector in sectors:
+
+    # Set the location on the request
+    request.setLocationNames(sector)
+
+    # Next loop through the Products (entities)
+    for entity in entities:
+        # Reset the time and channel variables since we're on a new product
+        time = None
+        R = None
+        G = None
+        B = None
+
+        # Set the product
+        request.addIdentifier("creatingEntity", entity)
+
+        # Cycle through the channels (parameters)
+        for channel in channels:
+            request.setParameters(channel)
+
+            # Set the time for this product if it hasn't been set
+            # If it has been set, then we proceed with that value
+            # so that all bands in for the one product are pulled
+            # from the same time
+            if(time is None):
+                times = DataAccessLayer.getAvailableTimes(request)
+                time = [times[-1]]
+                print("selected time:", time)
+
+            # Request the data from EDEX
+            response = DataAccessLayer.getGridData(request, time)
+            # Grab the actual data from the response
+            grid = response[0]
+            # Get the raw data from the response
+            data = grid.getRawData()
+            reftime = grid.getDataTime().getRefTime()
+
+            # Set the R,G,B channel
+            if(channel == ch1):
+                B = data
+            elif (channel == ch2):
+                R = data
+            elif (channel == ch3):
+                G = data
+
+            # Create the single channel name
+            name = outputDir+entity+'-'+sector+'-'+channel+'.png'
+
+            # Write out the single channel
+            print('writing',name)
+            write_img(data, name, proj, extent, reftime, False)
+
+        # --- End of channel loop
+
+        # Create the RGB product
+        # Apply range limits for each channel. RGB values must be between 0 and 1
+        R = np.clip(R, 0, 1)
+        G = np.clip(G, 0, 1)
+        B = np.clip(B, 0, 1)
+        RGB = np.dstack([R, G, B])
+
+        # Create RGB name
+        rgbName = outputDir+entity+'-'+sector+'-RGB.png'
+        # Write out the RGB image
+        print('writing', rgbName)
+        write_img(RGB, rgbName, proj, extent, time, False)
+
+    # --- End of entity loop
+#--- End of sector loop
+
+print('Done!')
+
+
+
selected time: [<DataTime instance: 2021-05-28 06:51:14 >]
+writing output/CLDSNOW-ECONUS-CH-01-0.47um.png
+writing output/CLDSNOW-ECONUS-CH-02-0.64um.png
+writing output/CLDSNOW-ECONUS-CH-03-0.87um.png
+writing output/CLDSNOW-ECONUS-RGB.png
+selected time: [<DataTime instance: 2021-05-28 06:51:14 >]
+writing output/DBRDUST-ECONUS-CH-01-0.47um.png
+writing output/DBRDUST-ECONUS-CH-02-0.64um.png
+writing output/DBRDUST-ECONUS-CH-03-0.87um.png
+writing output/DBRDUST-ECONUS-RGB.png
+selected time: [<DataTime instance: 2021-05-28 06:56:14 >]
+writing output/GEOCOLR-ECONUS-CH-01-0.47um.png
+writing output/GEOCOLR-ECONUS-CH-02-0.64um.png
+writing output/GEOCOLR-ECONUS-CH-03-0.87um.png
+writing output/GEOCOLR-ECONUS-RGB.png
+Done!
+
+
+
<Figure size 432x288 with 0 Axes>
+
+
+../../_images/GOES_CIRA_Product_Writer_25_2.png +
<Figure size 432x288 with 0 Axes>
+
+
+../../_images/GOES_CIRA_Product_Writer_25_4.png +
<Figure size 432x288 with 0 Axes>
+
+
+../../_images/GOES_CIRA_Product_Writer_25_6.png +

Top

+
+
+
+ +
+
+ + +
+ +