Time Series in Image Regions

The previous examples illustrated charting a series for a single region and charting spectra at a single time for multiple regions. To get series for multiple regions, use ui.Chart.image.seriesByRegion(). For example, the following code plots time series of land surface temperature in three regions representing three land cover types:

// Define a FeatureCollection: regions of the American West.
var regions = ee.FeatureCollection([
  ee.Feature(    // San Francisco.
    ee.Geometry.Rectangle(-122.45, 37.74, -122.4, 37.8), {label: 'City'}),
  ee.Feature(  // Tahoe National Forest.
    ee.Geometry.Rectangle(-121, 39.4, -120.8, 39.8), {label: 'Forest'}),
  ee.Feature(  // Black Rock Desert.
    ee.Geometry.Rectangle(-119.15, 40.8, -119, 41), {label: 'Desert'})
]);

// Load Landsat 8 brightness temperature data for 1 year.
var temps2013 = ee.ImageCollection('LANDSAT/LC8_L1T_32DAY_TOA')
    .filterDate('2012-12-25', '2013-12-25')
    .select('B11');

// Create a time series chart.
var tempTimeSeries = ui.Chart.image.seriesByRegion(
    temps2013, regions, ee.Reducer.mean(), 'B11', 200, 'system:time_start', 'label')
        .setChartType('ScatterChart')
        .setOptions({
          title: 'Temperature over time in regions of the American West',
          vAxis: {title: 'Temperature (Kelvin)'},
          lineWidth: 1,
          pointSize: 4,
          series: {
            0: {color: 'FF0000'}, // urban
            1: {color: '00FF00'}, // forest
            2: {color: '0000FF'}  // desert
}});

// Display.
print(tempTimeSeries);
    

The result should look something like Figure 1.

Figure 1. Time series of mean land surface temperature in three regions.