Google Earth Engine – Best Parameters for Visualizing Sentinel-2 for Same Region Over Different Times

google-earth-enginegoogle-earth-engine-javascript-api

I would like to visualize the sentinel-2 imagery for the same region but at different times and only use the same visual parameters in the google earth engine. My problem is that one image is quite dark and the other is very bright in my script. Is there any way to choose the suitable parameters for both imageries clearly seen? My script is below:

var S2_collection_2021 = ee.ImageCollection("COPERNICUS/S2_SR")
                    .filterDate('2021-01-10','2021-05-30')
                    .filterMetadata('MGRS_TILE', 'equals', '60HVC') print (S2_collection_2021,'collection');
    
var S2_collection_2022 = ee.ImageCollection("COPERNICUS/S2_SR")
                    .filterDate('2022-01-10','2022-03-10')
                    .filterMetadata('MGRS_TILE', 'equals', '60HVC')
                  
    
print (S2_collection_2022,'collection'); 
// VISUALIZE IMAGERY 
var visualize={'bands': ['B4', 'B3', 'B2'], 'min': 0, 'max': 1500,gamma:0.6} 
Map.addLayer(S2_collection_2021,visualize, 'S2_2021'); 
Map.addLayer(S2_collection_2022, visualize, 'S2_2022');

The link of code is here: https://code.earthengine.google.com/370f4a22a4622587f1d4cc8327440f90

Best Answer

You should use the harmonized Sentinel-2 collection. This collection matches the range of the image's values in both old and new images. You can find more details in the description of the collection.

ee.ImageCollection("COPERNICUS/S2_SR_HARMONIZED")

Related Solutions

Google Earth Engine – Grouping Weather Data by Region

It's a bit fiddly to do this. Here's one approach:

var days = endDate.difference(startDate, "days") // Number of days in date range
var dailyTempByCounty = ee.FeatureCollection(ee.List.sequence(0, days.subtract(1))
  .map(function(dayOffset) { // Calculate mean per day
    var date = startDate.advance(ee.Number(dayOffset), 'days')
    var dailyMean = tmpCollection
      .filterDate(date, date.advance(1, 'days'))
      .mean()
    return counties.map(function (county) {
       // Calculate mean for every county separately, 
       // so the date can be included
      var mean = dailyMean.reduceRegion({
        reducer: ee.Reducer.mean(), 
        geometry: county.geometry(), 
        scale: 4000
      })
      return county
        .set(mean)
        .set('date', date.format('yyyy-MM-dd'))
    })
  })
).flatten() // Flattens the collection of collection of features into a collection of features

https://code.earthengine.google.com/64937228d07d5c2d05cec1d5400aa914

It takes quite a lot of processing to do this. The linked script uses only data for a couple of days, and just two counties, to give a fast result that can be printed in the console.

Note that the end date is exclusive. Just like the date filter in collections - many people seem to get that wrong, and miss the last day of the year, just like in your question.

Google Earth Engine – No Images Found for June in Sentinel-2

In the dataset catalog for that collection:

The Sentinel-2 L2 data are downloaded from scihub. They were computed by running sen2cor. WARNING: ESA did not produce L2 data for all L1 assets, and earlier L2 coverage is not global.

You will have to fallback to the "COPERNICUS/S2 or COPERNICUS/S2_HARMONIZED collections, which unfortunately are not surface reflectance.

Related Question