First let's check how a deforestation event looks with Optical data. We will explore a deforestation event in Papau New Guinea as detected by RADD Forest Disturbance Alert.
There is availability of ALOS Palsar ScanSAR data in GEE over Oceania and South East Asia that we can use in combination with Sentinel-1 data for this excercise.
The location of interest is Lat: -4.011° , Lon: 141.303°. First we will locate the area of interest through the point location and create a rectangle around it.
//Create variable with the point location of interest
var defPaG = ee.Geometry.Point([141.303, -4.011])
// Define a rectangular area of interest, by listing coordinates
var defaultStudyArea = ee.Geometry.Polygon(
[[[141.2500740578722,-4.04103621945203],
[141.3599373391222,-4.04103621945203],
[141.3599373391222,-3.9691145182502945],
[141.2500740578722,-3.9691145182502945]]], null, false);
Map.addLayer(defaultStudyArea, {}, 'aoi');
Map.addLayer(defPaG, {}, 'point of deforestation')
// Zoom to area of of interest
Map.centerObject(defaultStudyArea, 13);Then, we will define the dates of interest:
var firstdate = '2022-01-01'
var seconddate = '2022-12-31'First we will visualize optical data over the area of interest and the time period selected. We will use Sentinel-2 Optical data and will visualize on the Map viewer the first image available:
var sentinel2 = ee.ImageCollection('COPERNICUS/S2_SR_HARMONIZED')
.filterDate(firstdate, seconddate)
.filterBounds(defaultStudyArea)
.limit(60) //limit for animation purposes
print(sentinel2, 'sentinel2 collection') //Explore properties of Sentinel-2 data
// Define visualization arguments for Optical S2 data.
var visArgsSen2 = {bands: ['B4', 'B3', 'B2'], min: 300, max: 2500};
Map.addLayer(sentinel2.first(), visArgsSen2, 'First Sentinel2 image') What is the date of the first image available for Sentinel-2?
Prepare and visualize animation for the Sentinel-2 Image Collection
// Define a function to convert an image to an RGB visualization image and copy
// properties from the original image to the RGB image.
var visFun = function(img) {
return img.visualize(visArgsSen2).copyProperties(img, img.propertyNames());
};
// Map over the image collection to convert each image to an RGB visualization
// using the previously defined visualization function.
var s2colVis = sentinel2.map(visFun);
//Adding date to animation
var text = require('users/gena/packages:text'); // Import gena's package which allows text overlay on image
var annotations = [
{position: 'left', offset: '1%', margin: '1%', property: 'label', scale: 40} //large scale because image if of the whole world. Use smaller scale otherwise
]
//Function to add text (date) to animation
function addText(image){
var timeStamp = ee.Date(image.get('system:time_start')).format().slice(0,10) // get the time stamp of each frame. This can be any string. Date, Years, Hours, etc.
var timeStamp = ee.String('Date: ').cat(ee.String(timeStamp)); //conveee.String('Date: ').cat(ee.String(timeStamp));rt time stamp to string
var image = image.visualize(visArgsSen2).set({'label': timeStamp}) // set a property called label for each image
var annotated = text.annotateImage(image, {}, defaultStudyArea, annotations); // create a new image with the label overlayed using gena's package
return annotated
}
var s2colVis2 = sentinel2.map(addText) //add time stamp to all Sentinel 2 images
// Define GIF visualization parameters.
var gifParams = {
'region': defaultStudyArea,
'dimensions': 700,
'crs': 'EPSG:3857',
'framesPerSecond': 5,
'maxPixels': 46046000
};
//Visualize animation in Console and get URL for .gif animation
print(ui.Thumbnail(s2colVis2, gifParams)); // Sentinel-2 Optical
print(s2colVis2.getVideoThumbURL(gifParams)); //URL The following animation should appear in the Console. In addition, a URL will be printed and you can download the .gif animation
Figure 2. Sentinel-2 Animation .
Can you identify the date on which the major deforestation took place?
Now, we will create similar animations using Sentinel-1 and Alos Palsar data. Get image collections for both SAR datasets
//Get Functions that will be used to process SAR datasets
var functions = require('users/africa_uah/SAREdX:00.Functions')
var todBall = functions.todBall
//Get ALOS Palsar data
var collectionALOS = ee.ImageCollection('JAXA/ALOS/PALSAR-2/Level2_2/ScanSAR')
.filterBounds(defPaG)
.filterDate(firstdate, seconddate)
.filter(ee.Filter.eq('Polarizations', ["HH", "HV"]))
var ALOSdB = collectionALOS.map(todBall) //add dB bands for HH and HV
print(ALOSdB, 'ALOSdB Image Collection')
var hhdB = ee.Image( ALOSdB.select('dBHH').mean()); //calculate average value at pixel level for HH
var hvdB = ee.Image( ALOSdB.select('dBHV').mean()); //calculate average value at pixel level for HV
//Visualize average image for HH and HV
Map.addLayer( hhdB, {min:-30, max:0}, "1.Mean HH ")
Map.addLayer(hvdB, {min: -30, max:0}, "2.Mean HV ")How many Alos Palsar images are available for the time period of interest?
Get Sentinel-1 data
var sentinel1 = ee.ImageCollection('COPERNICUS/S1_GRD');
//Get descending mode
var ddata = sentinel1
// Filter to get images with VV and VH dual polarization.
.filter(ee.Filter.listContains('transmitterReceiverPolarisation', 'VV'))
.filter(ee.Filter.listContains('transmitterReceiverPolarisation', 'VH'))
// Filter to get images collected in interferometric wide swath mode.
.filter(ee.Filter.eq('instrumentMode', 'IW'))
.filter(ee.Filter.eq('orbitProperties_pass', 'DESCENDING'))
.filterDate(firstdate, seconddate)
.filterBounds(defaultStudyArea);
print(ddata, 'Sentinel-1 Descending')
//Calculate average image and visualize it for VV and VH polarizations
var vv=ee.Image(ddata.select('VV').mean());
Map.addLayer(vv, {min: -30, max:0}, "3. Mean VV ")
print('vv',vv);
var vh = ee.Image(ddata.select('VH').mean());
print('vh',vh)
Map.addLayer(vh, {min: -30, max:0}, "4. Mean VH ")The following code covers:
- Pre-process Sentinel-1 GRD data to get simulated Radiometric Terrain corrected data
- Apply functions to corrected Sentinel-1 data to derive RGBs
- Create and visualize animations of RGBs
Note: try to keep all the code snippets in the same script, since we're using variables and functions defined above.
//Get Functions that will be used to process SAR datasets
var functions = require('users/africa_uah/SAREdX:00.Functions')
var afn_raiseToPowerForSAR = functions.afn_raiseToPowerForSAR
var powerToDb = functions.powerToDb
var dbToPower = functions.dbToPower
var terrainCorrection = functions.terrainCorrection
var gammaMap = functions.gammaMap
var afn_SARtoRGB = functions.afn_SARtoRGB //to create RGB following Freeman image decomposition
//Correct Sentinel-1 GRD data
var ICofRGBs0 = ddata.map(terrainCorrection)
var ICofRGBs1 = ICofRGBs0.map(gammaMap)
//Function to apply afn_SARtoRGB to individual S1 images
function afn_oneSARtoRGB (oneImage){
var vv = ee.Image(oneImage.select('VV'));
var vh = ee.Image(oneImage.select('VH'));
var OnecrossPolThenCoPol = vh.addBands(vv)
return(afn_SARtoRGB(OnecrossPolThenCoPol))
}
//Create Image collection of Sentinel-1 RGBs
var ICofRGBs = ICofRGBs1.map(afn_oneSARtoRGB)
//Visualization of RGBs for S1 and Alos Palsar
var visArgsSen1 = {bands: ['AR', 'AG', 'AB'], min: 30, max: 210};
//Adding date as text to S1 RGBs animations
function addTextS1(image){
var timeStamp = ee.Date(image.get('system:time_start')).format().slice(0,10) // get the time stamp of each frame. This can be any string. Date, Years, Hours, etc.
var timeStamp = ee.String('Date: ').cat(ee.String(timeStamp));
var image = image.visualize(visArgsSen1).set({'label': timeStamp}) // set a property called label for each image
var annotated = text.annotateImage(image, {}, defaultStudyArea, annotations); // create a new image with the label overlayed using gena's package
return annotated
}
var tempCol = ICofRGBs.map(addTextS1)//adding date annotation for S1
// Define arguments for animation function parameters.
var videoArgs = {
dimensions: 768,
region: defaultStudyArea,
framesPerSecond: 7,
min: 20,
max: 100.0,
};
//See animation and obtain URL of Sentinel-1 RGBs .gif
print(ui.Thumbnail(tempCol, videoArgs)); //Sentinel 1 RGBs
print(tempCol.getVideoThumbURL(videoArgs));
Figure 3. Sentinel-1 Animation - RGBs.
It is difficult to visualize a change in this animation. Hence, we will visualize directly one of the polarization to assess if a change is visible.
//Let's create an animation with the VV polarization from Sentinel-1
var visArgsSen1VV = {
bands: ['VV'],
min: -15,
max: -5,
palette: [
'000000', '222222', '333333', '666666',
'999999', 'CCCCCC', 'EEEEEE', 'FFFFFF']
};
//Add date as text
function addTextS1VV(image){
var timeStamp = ee.Date(image.get('system:time_start')).format().slice(0,10) // get the time stamp of each frame. This can be any string. Date, Years, Hours, etc.
var timeStamp = ee.String('Date: ').cat(ee.String(timeStamp)); //conveee.String('Date: ').cat(ee.String(timeStamp));rt time stamp to string
var image = image.visualize(visArgsSen1VV).set({'label': timeStamp}) // set a property called label for each image
var annotated = text.annotateImage(image, {}, defaultStudyArea, annotations); // create a new image with the label overlayed using gena's package
return annotated
}
//Add date to IC of Sentinel-1 VV polarizations
var S1VV = ICofRGBs1.select('VV').map(addTextS1VV)
var videoArgsdB = {
dimensions: 768,
region: defaultStudyArea,
framesPerSecond: 7,
};
print(ui.Thumbnail(S1VV, videoArgsdB));
print(S1VV.getVideoThumbURL(videoArgsdB));
Figure 3. Sentinel-1 Animation - VV.
Modify the code to visualize animation of the VH polarization Tip: You may need to modify min and max values of the visualization arguments
*Are there any differences between the VV and VH polarizations? What could be the explanations for any differences? *
Finally, let's check Alos Palsar data
//Visualizing RGBs
function afn_oneSARtoRGBALOS (oneImage){
var hh = ee.Image(oneImage.select('dBHH'));
var hv = ee.Image(oneImage.select('dBHV'));
var OnecrossPolThenCoPol = hv.addBands(hh)
return(afn_SARtoRGB(OnecrossPolThenCoPol))
}
var ICofRGBsALOS = ALOSdB.map(afn_oneSARtoRGBALOS)
var tempColAlos = ICofRGBsALOS.map(addTextS1)//adding date annotation for Alos
print(ui.Thumbnail(tempColAlos, videoArgs)); //Alos Palsar RGBs
print(tempColAlos.getVideoThumbURL(videoArgs));
//Visualizing HH polarization
var visArgsALOSHH = {
bands: ['dBHH'],
min: -15,
max: -5,
palette: [
'000000', '222222', '333333', '666666',
'999999', 'CCCCCC', 'EEEEEE', 'FFFFFF']
};
function addTextAHH(image){
var timeStamp = ee.Date(image.get('system:time_start')).format().slice(0,10) // get the time stamp of each frame. This can be any string. Date, Years, Hours, etc.
var timeStamp = ee.String('Date: ').cat(ee.String(timeStamp)); //conveee.String('Date: ').cat(ee.String(timeStamp));rt time stamp to string
var image = image.visualize(visArgsALOSHH).set({'label': timeStamp}) // set a property called label for each image
var annotated = text.annotateImage(image, {}, defaultStudyArea, annotations); // create a new image with the label overlayed using gena's package
return annotated
}
var AlosHH = ALOSdB.select('dBHH').map(addTextAHH)
print(ui.Thumbnail(AlosHH, videoArgsdB)); // HH band Alos Palsar
print(AlosHH.getVideoThumbURL(videoArgsdB));
Figure 4. Alos Palsar Animation - HH.
Modify the code to visualize animation of the HV polarization for Alos Palsar
*Are there any differences between the HH and HV polarizations in visualizing this type of change? What could be the explanations for any differences? *