Bewerken

Delen via


geo_point_to_h3cell()

Applies to: ✅ Microsoft FabricAzure Data ExplorerAzure MonitorMicrosoft Sentinel

Calculates the H3 Cell token string value of a geographic location.

Read more about H3 Cell.

Syntax

geo_point_to_h3cell(longitude, latitude, [ resolution ])

Learn more about syntax conventions.

Parameters

Name Type Required Description
longitude real ✔️ Geospatial coordinate, longitude value in degrees. Valid value is a real number and in the range [-180, +180].
latitude real ✔️ Geospatial coordinate, latitude value in degrees. Valid value is a real number and in the range [-90, +90].
resolution int Defines the requested cell resolution. Supported values are in the range [0, 15]. If unspecified, the default value 6 is used.

Returns

The H3 Cell token string value of a given geographic location. If the coordinates or levels are invalid, the query will produce an empty result.

Note

  • H3 Cell can be a useful geospatial clustering tool.
  • H3 Cell has 16 levels of hierarchy with area coverage ranging from 4,250,547km² at the highest level 0 to 0.9m² at the lowest level 15.
  • H3 Cell has a unique hexagon shape and this leads some unique properties:
  • Hexagons have 6 neighbors
  • Hexagons allow us to approximate radiuses easily and all neighbors are equidistant
  • Hexagons are visually pleasant
  • In some rare cases the shape is pentagon.
  • H3 Cell has a rectangular area on a plane surface.
  • Invoking the geo_h3cell_to_central_point() function on an H3 Cell token string that was calculated on longitude x and latitude y won't necessarily return x and y.
  • It's possible that two geographic locations are very close to each other but have different H3 Cell tokens.

H3 Cell approximate area coverage per resolution value

Level Average Hexagon Edge Length
0 1108 km
1 419 km
2 158 km
3 60 km
4 23 km
5 8 km
6 3 km
7 1 km
8 460 m
9 174 m
10 66 m
11 25 m
12 9 m
13 3 m
14 1 m
15 0.5 m

The table source can be found in this H3 Cell statistical resource.

See also geo_point_to_s2cell(), geo_point_to_geohash().

For comparison with other available grid systems. see geospatial clustering with Kusto Query Language.

Examples

print h3cell = geo_point_to_h3cell(-74.04450446039874, 40.689250859314974, 6)

Output

h3cell
862a1072fffffff

The following example finds groups of coordinates. Every pair of coordinates in the group resides in the H3 Cell with average hexagon area of 253 km².

datatable(location_id:string, longitude:real, latitude:real)
[
    "A", -73.956683, 40.807907,
    "B", -73.916869, 40.818314,
    "C", -73.989148, 40.743273,
]
| summarize count = count(),                                         // Items per group count
            locations = make_list(location_id)                       // Items in the group
            by h3cell = geo_point_to_h3cell(longitude, latitude, 5)  // H3 Cell of the group

Output

h3cell count locations
852a100bfffffff 2 [
"A",
"B"
]
852a1073fffffff 1 [
"C"
]

The following example produces an empty result because of the invalid coordinate input.

print h3cell = geo_point_to_h3cell(300,1,8)

Output

h3cell

The following example produces an empty result because of the invalid level input.

print h3cell = geo_point_to_h3cell(1,1,16)

Output

h3cell

The following example produces an empty result because of the invalid level input.

print h3cell = geo_point_to_h3cell(1,1,int(null))

Output

h3cell