H3 Geohashing

Enrich a Point-Based Table with H3 Indexes

Create an H3 index for pickup locations in the NYC taxi data set.

• Lat/Lon-Based Table
• WKT-Based Table

1
2
3
4
5
6
7
8

CREATE OR REPLACE TABLE example_geospatial.nyctaxi_h3 AS
(
    SELECT
        pickup_latitude,
        pickup_longitude,
        H3_XYTOCELL(pickup_longitude, pickup_latitude, 8) AS h3_index
    FROM example_geospatial.nyctaxi_xy
)

1
2
3
4
5
6
7

CREATE OR REPLACE TABLE example_geospatial.nyctaxi_h3 AS
(
    SELECT
        pickup_location,
        H3_GEOMTOCELL(pickup_location, 8) AS h3_index
    FROM example_geospatial.nyctaxi_wkt
)

Create an Aggregated View with WKT Geometries

Use the table created above to generate an H3 grid view with the counts of pickups within each cell.

1
2
3
4
5
6

SELECT
    h3_index,
    H3_CELLTOBOUNDARY(h3_index) AS h3_cell,
    COUNT(*) AS total_pickups
FROM example_geospatial.nyctaxi_h3
GROUP BY h3_index

The H3 grid output as a class break render: