Current NOAA NDFD forecasts

The National Weather Service creates forecasts for a grid of cells 2.5km on each side that span the entire United States. That means that if you select any point in the continental United States, it'll be a part of one of these grid cells. You can see it on the web using their web portal.

All the data on that web portal is also exposed through their APIs. I use those APIs to serve regularly-updated weather forecasts through the website and mobile app.

The URLs to query these gridpoints, the centroids of each of these cells, have the local station identifier as well, i.e.:

https://api.weather.gov/gridpoints/PQR/138,104/forecast

Because of this, I first need to query the points/ API endpoint with a series of lat/lon coordinates to find the URLs to later ping for the actual forecasts.

find_gridpoints.py takes a LineString geometry and a distance and retrieves the URLs for the grid cells corresponding to a point along that LineString every dist meters.

lambda.py is a small function that is run on AWS lambda to very cheaply update my S3 bucket with the latest version of the forecasts. Even without the Lambda free tier, the code takes about 4 minutes to run, and if I ran it every 4 hours, it would cost 15 cents per month.

lambda.py uploads both individual GeoJSON files and zipped archives of GeoJSONs per trail section. Individual GeoJSON files would be easiest to transmit via the web API, and the zipped files would allow for bulk downloads for offline use in the mobile app.

get_grid_geojson.py takes the grid cells found in find_gridpoints.py and computes a single GeoJSON with all given cells in a single FeatureCollection.

Usage

> python find_gridpoints.py --help
Usage: find_gridpoints.py [OPTIONS] FILES...

  Find NOAA NDFD gridpoints along LineString

Options:
  -d, --dist FLOAT      Distance along LineString to query for grid position
                        (in meters)  [required]
  --projection INTEGER  EPSG code for projection used when creating buffer.
                        Coordinates must be in meters.  [default: 3488]
  --help                Show this message and exit.

So to get the gridpoints for 10km intervals of the five sections of the PCT:

export DATA_DIR=...
python code/find_gridpoints.py \
    --dist 10000 \
    $DATA_DIR/CA_Sec_A_tracks.geojson \
    $DATA_DIR/CA_Sec_B_tracks.geojson \
    $DATA_DIR/CA_Sec_C_tracks.geojson \
    $DATA_DIR/CA_Sec_D_tracks.geojson \
    $DATA_DIR/CA_Sec_E_tracks.geojson \
    > data/ca_south.txt
python code/find_gridpoints.py \
    --dist 10000 \
    $DATA_DIR/CA_Sec_F_tracks.geojson \
    $DATA_DIR/CA_Sec_G_tracks.geojson \
    $DATA_DIR/CA_Sec_H_tracks.geojson \
    $DATA_DIR/CA_Sec_I_tracks.geojson \
    $DATA_DIR/CA_Sec_J_tracks.geojson \
    $DATA_DIR/CA_Sec_K_tracks.geojson \
    > data/ca_central.txt
python code/find_gridpoints.py \
    --dist 10000 \
    $DATA_DIR/CA_Sec_L_tracks.geojson \
    $DATA_DIR/CA_Sec_M_tracks.geojson \
    $DATA_DIR/CA_Sec_N_tracks.geojson \
    $DATA_DIR/CA_Sec_O_tracks.geojson \
    $DATA_DIR/CA_Sec_P_tracks.geojson \
    $DATA_DIR/CA_Sec_Q_tracks.geojson \
    $DATA_DIR/CA_Sec_R_tracks.geojson \
    > data/ca_north.txt
python code/find_gridpoints.py \
    --dist 10000 \
    $DATA_DIR/OR_Sec_B_tracks.geojson \
    $DATA_DIR/OR_Sec_C_tracks.geojson \
    $DATA_DIR/OR_Sec_D_tracks.geojson \
    $DATA_DIR/OR_Sec_E_tracks.geojson \
    $DATA_DIR/OR_Sec_F_tracks.geojson \
    $DATA_DIR/OR_Sec_G_tracks.geojson \
    > data/or.txt
python code/find_gridpoints.py \
    --dist 10000 \
    $DATA_DIR/WA_Sec_H_tracks.geojson \
    $DATA_DIR/WA_Sec_I_tracks.geojson \
    $DATA_DIR/WA_Sec_J_tracks.geojson \
    $DATA_DIR/WA_Sec_K_tracks.geojson \
    $DATA_DIR/WA_Sec_L_tracks.geojson \
    > data/wa.txt

Then to get the selected grid as a GeoJSON:

python code/get_grid_geojson.py data/*.txt > data/grid.geojson

Note that the NWS sometimes returns 404 from the API, so you might not be able to get a complete grid on a single try.

Then compress the GeoJSON and upload it to S3:

brotli -c data/grid.geojson > data/grid_compressed.geojson
aws s3 cp \
    data/grid_compressed.geojson s3://tiles.nst.guide/pct/ndfd_current.geojson \
    --content-type application/geo+json \
    --content-encoding br \
    `# Set to public read access` \
    --acl public-read

nst-guide/ndfd_current

Current NOAA NDFD forecasts

Usage