Clustering with visual variables

Explore in the sandbox

View live

This sample demonstrates how clustering can be used to group features into clusters based on their spatial proximity to one another. Clustering is enabled on the layer's featureReduction property.

Since featureReduction is independent of the renderer, the symbol and popupTemplate of each cluster graphic can be used to summarize features within the cluster. See Clustering styles and configurations for more information about the various ways clusters can summarize the features they represent.

The layer's renderer in this sample visualizes weather data with three visual variables, each representing a different attribute:

Color: TEMP
Size: WIND_SPEED
Rotation: WIND_DIRECT

The renderer object looks like the the following:

Calculate cluster statistics
Use dark colors for code blocksCopy
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
95
      const renderer = {
        type: "simple", // autocasts as new SimpleRenderer()
        symbol: {
          type: "simple-marker", // autocasts as new SimpleMarkerSymbol()
          // Arrow marker
          path: "M14.5,29 23.5,0 14.5,9 5.5,0z",
          color: [50, 50, 50],
          outline: {
            color: [0, 0, 0, 0.7],
            width: 0.5,
          },
          angle: 180,
          size: 15,
        },
        visualVariables: [
          {
            type: "rotation",
            // Use {cluster_avg_WIND_DIRECT} in the
            // featureReudction.popupTemplate to
            // display the average temperature of all
            // features within the cluster
            field: "WIND_DIRECT",
            rotationType: "geographic",
          },
          {
            type: "size",
            // Use {cluster_avg_WIND_SPEED} in the
            // featureReudction.popupTemplate to
            // display the average temperature of all
            // features within the cluster
            field: "WIND_SPEED",
            minDataValue: 0,
            maxDataValue: 60,
            minSize: 8,
            maxSize: 40,
          },
          {
            type: "color",
            // Use {cluster_avg_TEMP} in the
            // featureReudction.popupTemplate to
            // display the average temperature of all
            // features within the cluster
            field: "TEMP",
            stops: [
              { value: 20, color: "#2b83ba" },
              { value: 35, color: "#abdda4" },
              { value: 50, color: "#ffffbf" },
              { value: 65, color: "#fdae61" },
              { value: 80, color: "#d7191c" },
            ],
          },
        ],
      };

To enable clustering on the layer, you simply have to set the featureReduction property to type cluster,

Use dark colors for code blocksCopy
layer.featureReduction = {
  type: "cluster",
};

This will cluster the features using the default clusterRadius (80px), and assign a symbol to each cluster conforming to the visualVariables in the renderer in the following manner:

Color: average TEMP of all features in the cluster
Size: average WIND_SPEED of all features in the cluster
Rotation: average WIND_DIRECT of all features in the cluster

No feature reduction	Clustered features
	)

If a UniqueValueRenderer were used, the symbol of the predominant uniqueValueInfo among the features in the cluster would be used to represent the cluster.

To provide more meaning to the app, you can configure a popupTemplate to display the average values of all the fields used by the renderer when the user clicks or taps the cluster.

Configure clustering based on renderer
Use dark colors for code blocksCopy
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
140
      const clusterConfig = {
        type: "cluster",
        popupTemplate: {
          content: [
            {
              type: "text",
              text: "This cluster represents <b>{cluster_count}</b> weather stations.",
            },
            {
              type: "fields",
              fieldInfos: [
                {
                  fieldName: "cluster_avg_WIND_SPEED",
                  label: "Average wind speed (km/h)",
                  format: {
                    places: 0,
                  },
                },
                {
                  fieldName: "cluster_avg_WIND_DIRECT",
                  label: "Average wind direction (degrees)",
                  format: {
                    places: 0,
                  },
                },
                {
                  fieldName: "cluster_avg_TEMP",
                  label: "Average temperature (°F)",
                  format: {
                    places: 0,
                  },
                },
              ],
            },
          ],
        },
      };