Find hot spots

A hot spot analysis resulting in a new layer with statistically high population values by USA county.

What is a hot spot analysis?

A hot spot analysis is the process of determining if there are statistically significant clusters in spatial data. To execute the analysis, use the spatial analysis service and the FindHotspot operation.

The analysis uses the (Gi* de Getis-Ord) equation to find groups of clusters of either point or polygon data based on high (hot) or low (cold) values, which are defined as the resultant z-scores and p-values. For a feature to be in a hot spot cluster, it must have a high value and be surrounded by other clusters with high values. For a feature to be in a cold spot cluster, the feature must have a low value and be surrounded by other clusters with low values.

A hot spot analysis helps to find spatial trends and patterns in the data that may not be visible at first glance.

Real-world examples of this analysis include the following:

Finding high and low clusters of crime or traffic crash incidents.
Discovering the distribution of high and low property values in an area.
Determining where people are spending the most money in an area compared to another.

How to perform a hot spot analysis

Review the parameters for the FindHotSpots operation.
Send a request to get the spatial analysis service URL.
Execute a job request with the following URL and parameters:
- URL: https://<YOUR_ANALYSIS_SERVICE>/arcgis/rest/services/tasks/GPServer/FindHotSpots/submitJob
- Parameters:
  - analysisLayer: Your dataset as a hosted feature layer or feature collection.
  - analysisField: A numeric field.
  - shapeType: Fishnet or Hexagon.
  - outputName: A string representing the name of the hosted feature layer to return with the results.
Check the status.
Get the output layer results.

To see examples using ArcGIS API for Python, ArcGIS REST JS, and the ArcGIS REST API, go to Examples below.

URL request

Use dark colors for code blocksCopy
http://<YOUR_ANALYSIS_SERVICE>/arcgis/rest/services/tasks/GPServer/FindHotSpots/submitJob?<parameters>

Required parameters

Name	Description	Examples
`f`	The format of the data returned.	`f=json` `f=pjson`
`token`	An access token (API key or OAuth 2.0) with the required Spatial Analysis > Feature analysis privilege.	`token=<ACCESS_TOKEN>`
`analysisLayer`	The point or polygon feature layer.	`{url:"https://services3.arcgis.com/GVgbJbqm8hXASVYi/arcgis/rest/services/Average_PDX_home_values/FeatureServer/0"}`
`analysisField`	Use if the analysis layer contains polygons.	`counts`, `rates`, `averages`

Key parameters

Name	Description	Examples
`shapeType`	Use if the analysis layer contains points.	`hexagon`, `fishnet`
`boundingPolygonLayer`	When the analysis layer contains points and no `analysisField` is specified, you can provide a boundary for the analysis.	`{url:"https://services3.arcgis.com/GVgbJbqm8hXASVYi/arcgis/rest/services/Delaware_County_Municipal_Boundaries/FeatureServer/0"}`
`outputName`	A string representing the name of the hosted feature layer to return with the results. NOTE: If you do not include this parameter, the results are returned as a feature collection (JSON).	`{"serviceProperties": {"name": "<SERVICE_NAME>"}}`
`context`	A bounding box or output spatial reference for the analysis.	`"extent":{"xmin:", "ymin:", "xmax:", "ymax:"}`

Code examples

Find traffic crash hot spots

This example uses the FindHotSpots operation to determine where there are statistically significant clusters of Traffic crashes counted within a fishnet grid. The red clusters represent a statistically high number of crashes. The blue spots represent a statistically low number of crashes.

In the analysis, the analysisLayer value is the Traffic crashes hosted feature layer. The points in the layer are counted within a fishnet, which was set in the shapeType parameter.

Hot spot analysis showing statistically signifcant areas of traffic crashes.

APIs

ArcGIS API for Python

ArcGIS REST JS

Expand

Use dark colors for code blocks

Copy

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
traffic_crashes = "https://services3.arcgis.com/GVgbJbqm8hXASVYi/arcgis/rest/services/Traffic_Crashes/FeatureServer/0"
fishnet = "fishnet"

results = find_hot_spots(
    analysis_layer=traffic_crashes,
    shape_type=fishnet,
    # Outputs results as a hosted feature layer.
    output_name="Hot spot analysis results",
)

result_features = results.layers[0].query()

print(f"The hot spot layer has {len(result_features.features)} new records")

Expand

Go to tutorial

Service requests

Request

HTTP

cURL

Use dark colors for code blocksCopy
POST arcgis.com/sharing/rest/portals/self HTTP/1.1
Content-Type: application/x-www-form-urlencoded

&f=json
&token=<ACCESS_TOKEN>
Run in Postman

Response (JSON)

Use dark colors for code blocksCopy
{
  "helperServices": {
    // Other parameters
    "analysis": {
      "url": "https://<YOUR_ANALYSIS_SERVICE>/arcgis/rest/services/tasks/GPServer"
    },
    "geoenrichment": {
      "url": "https://geoenrich.arcgis.com/arcgis/rest/services/World/GeoenrichmentServer"
    }
  }
}

Request

HTTP

cURL

Use dark colors for code blocks

Copy

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
POST <YOUR_ANALYSIS_SERVICE>/arcgis/rest/services/tasks/GPServer/FindHotSpots/submitJob HTTP/1.1
Content-Type: application/x-www-form-urlencoded

&f=json
&token=<ACCESS_TOKEN>
&analysisLayer={"url":"https://services3.arcgis.com/GVgbJbqm8hXASVYi/arcgis/rest/services/Traffic_Crashes/FeatureServer/0"}
&shapeType=fishnet
&returnProcessInfo=true
&outputName={"serviceProperties":{"name":"Find hot spot results"}}
&context={}

Run in Postman

Response (JSON)

Use dark colors for code blocksCopy

{
  "jobId": "<JOB_ID>",
  "jobStatus": "esriJobSubmitted",
  "results": {},
  "inputs": {},
  "messages": []
}

Request

HTTP

cURL

Use dark colors for code blocks

Copy

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
POST <YOUR_ANALYSIS_SERVICE>/arcgis/rest/services/tasks/GPServer/FindHotSpots/jobs/<JOB_ID> HTTP/1.1
Content-Type: application/x-www-form-urlencoded

&f=json
&token=<ACCESS_TOKEN>

Run in Postman

Response (JSON)

Use dark colors for code blocksCopy
// Executing job
{
  "jobId": "<JOB_ID>",
  "jobStatus": "esriJobExecuting",
  "results": {},
  "inputs": {},
  "messages": []
}
// Job succeeded
{
  "jobId": "<JOB_ID>",
  "jobStatus": "esriJobSucceeded",
  "results": {
      "hotSpotsResultLayer": {
          "paramUrl": "results/hotSpotsResultLayer"
      },
      "processInfo": {
          "paramUrl": "results/processInfo"
      }
  },
  "inputs": {},
  "messages": []
}

Request

HTTP

cURL

Use dark colors for code blocks

Copy

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
POST <YOUR_ANALYSIS_SERVICE>/arcgis/rest/services/tasks/GPServer/FindHotSpots/jobs/<JOB_ID>/results/hotSpotsResultLayer HTTP/1.1
Content-Type: application/x-www-form-urlencoded

&f=json
&returnType=data
&token=<ACCESS_TOKEN>

Run in Postman

Response (JSON)

Use dark colors for code blocksCopy
{
  "paramName": "hotSpotsResultLayer",
  "dataType": "GPString",
  "value": {
      "url": "<SERVICE_URL>",
      "itemId": "<ITEM_ID>"
  }
}

Identify high and low home values

This example uses the FindHotSpots operation to determine where there are statistically significant hot and cold spots for home values in Portland. The hot and cold spots, shown as red and blue respectively, indicate whether a home's value is in a statistically high or low cluster.

In the analysis, the analysisLayer value is the Enriched Portland hexagon bins hosted feature layer. The feature layer was created using generated hexagon bins that were enriched using data from the GeoEnrichment service. To analyze home values, you set the analysisField with the AVG_CY attribute.

To learn how to generate hexagon bins, go to generate tessellations.

Hot spot analysis showing statistically significant areas of high home values.

APIs

ArcGIS API for Python

ArcGIS REST JS

Expand

Use dark colors for code blocks

Copy

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
pdx_homes = {
    "url": "https://services3.arcgis.com/GVgbJbqm8hXASVYi/arcgis/rest/services/Enriched_PDX_hex_bin_tessellations/FeatureServer/0",
    "filter": "AVGVAL_CY <> 0",
}
value = "AVGVAL_CY"

results = find_hot_spots(
    analysis_layer=pdx_homes,
    analysis_field=value,
    # Outputs results as a hosted feature layer.
    output_name="Hot spot analysis results",
)

result_features = results.layers[0].query()

print(f"The hot spot layer has {len(result_features.features)} new records")

Expand

Go to tutorial

Service requests

Request

HTTP

cURL

Use dark colors for code blocksCopy
POST arcgis.com/sharing/rest/portals/self HTTP/1.1
Content-Type: application/x-www-form-urlencoded

&f=json
&token=<ACCESS_TOKEN>
Run in Postman

Response (JSON)

Use dark colors for code blocksCopy
{
  "helperServices": {
    // Other parameters
    "analysis": {
      "url": "https://<YOUR_ANALYSIS_SERVICE>/arcgis/rest/services/tasks/GPServer"
    },
    "geoenrichment": {
      "url": "https://geoenrich.arcgis.com/arcgis/rest/services/World/GeoenrichmentServer"
    }
  }
}

Request

HTTP

cURL

Use dark colors for code blocks

Copy

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35

POST <YOUR_ANALYSIS_SERVICE>/arcgis/rest/services/tasks/GPServer/FindHotSpots/submitJob HTTP/1.1
Content-Type: application/x-www-form-urlencoded

&f=json
&token=<ACCESS_TOKEN>
&analysisLayer={"url":"https://services3.arcgis.com/GVgbJbqm8hXASVYi/arcgis/rest/services/Enriched_PDX_hex_bin_tessellations/FeatureServer/0","filter":"AVGVAL_CY <> 0"}
&analysisField=AVGVAL_CY
&returnProcessInfo=true
&outputName={"serviceProperties":{"name":"Find hot spot results"}}
&context={"extent":{"xmin":-13701588.357587751,"ymin":5636873.660541155,"xmax":-13554599.952195555,"ymax":5730585.457218655,"spatialReference":{"wkid":102100,"latestWkid":3857}}}

Run in Postman

Response (JSON)

Use dark colors for code blocksCopy

{
  "jobId": "<JOB_ID>",
  "jobStatus": "esriJobSubmitted",
  "results": {},
  "inputs": {},
  "messages": []
}

Request

HTTP

cURL

Use dark colors for code blocks

Copy

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
POST <YOUR_ANALYSIS_SERVICE>/arcgis/rest/services/tasks/GPServer/FindHotSpots/jobs/<JOB_ID> HTTP/1.1
Content-Type: application/x-www-form-urlencoded

&f=json
&token=<ACCESS_TOKEN>

Run in Postman

Response (JSON)

Use dark colors for code blocksCopy
// Executing job
{
  "jobId": "<JOB_ID>",
  "jobStatus": "esriJobExecuting",
  "results": {},
  "inputs": {},
  "messages": []
}
// Job succeeded
{
  "jobId": "<JOB_ID>",
  "jobStatus": "esriJobSucceeded",
  "results": {
      "hotSpotsResultLayer": {
          "paramUrl": "results/hotSpotsResultLayer"
      },
      "processInfo": {
          "paramUrl": "results/processInfo"
      }
  },
  "inputs": {},
  "messages": []
}

Request

HTTP

cURL

Use dark colors for code blocks

Copy

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
POST <YOUR_ANALYSIS_SERVICE>/arcgis/rest/services/tasks/GPServer/FindHotSpots/jobs/<JOB_ID>/results/hotSpotsResultLayer HTTP/1.1
Content-Type: application/x-www-form-urlencoded

&f=json
&returnType=data
&token=<ACCESS_TOKEN>

Run in Postman

Response (JSON)

Use dark colors for code blocksCopy
{
  "paramName": "hotSpotsResultLayer",
  "dataType": "GPString",
  "value": {
      "url": "<SERVICE_URL>",
      "itemId": "<ITEM_ID>"
  }
}