Merge pull request #9 from edwardoughton/Week-10-11-Exercises

Initial-Week-10-Exercise

notebooks/10.01-filters-reducers-clip-operator-loops-GEE.js

@@ -0,0 +1,215 @@
+/*
+----------------------------------------------------------------------------
+
+---------- Working with Vector and Raster data collections in GEE ----------
+
+Topics covered:
+1) Filters
+2) Reducers
+3) Clipping
+4) Operators
+5) For-Loops (Map Function) in GEE
+
+----------------------------------------------------------------------------
+*/
+
+
+
+//---------------------------------------------------------------------------
+// 1) Filtering the data collection
+// 1.1) Filtering a feature from feature collection
+
+// Load US Census Counties 2018 boundary layer
+// available at: https://developers.google.com/earth-engine/datasets/catalog/TIGER_2018_Counties
+var us_shp = ee.FeatureCollection('TIGER/2018/Counties');
+Map.addLayer(us_shp, {}, 'US States Boundaries', false);
+
+// But first prepare a dictionary containing visualization parameters for Fairfax city boundary
+var fairfax_shp_vis = {
+  fillColor: 'b5ffb4',
+  color: 'FF6833',
+  width: 1.0,
+};
+// Now filtering the Fairfax city boundary based on metadata, to visualize only FairFax
+var fairfax_shp = us_shp.filterMetadata('NAMELSAD', 'equals', 'Fairfax city');
+Map.addLayer(fairfax_shp, fairfax_shp_vis, 'fairfax city boundary', false);
+Map.centerObject(fairfax_shp, 13);
+// More Filtering Feature Collection examples available at: 
+// https://developers.google.com/earth-engine/guides/feature_collection_filtering
+//--------------------------------------------------------------------------
+
+
+
+
+// --------------------------------------------------------------------------
+// 1.2) Filtering an Image from Image collection
+
+// In this example we will be using sentinel 2 data
+var s2_data = ee.ImageCollection('COPERNICUS/S2_SR');
+
+// first type of filter (By Date)
+var s2_data = s2_data.filter(ee.Filter.date('2021-06-01', '2021-08-31'));
+
+// Second type of filter (Filter by area of interest)
+var s2_data = s2_data.filterBounds(fairfax_shp);
+
+
+// Applying preprocessing
+function maskS2clouds(image) {
+  var qa = image.select('QA60');
+
+  // Bits 10 and 11 are clouds and cirrus, respectively.
+  var cloudBitMask = 1 << 10;
+  var cirrusBitMask = 1 << 11;
+
+  // Both flags should be set to zero, indicating clear conditions.
+  var mask = qa.bitwiseAnd(cloudBitMask).eq(0)
+      .and(qa.bitwiseAnd(cirrusBitMask).eq(0));
+
+  return image.updateMask(mask).divide(10000);
+}
+
+
+var s2_data = s2_data.map(maskS2clouds); // dont worry about .map() function, well cover it later in For-Loops section
+
+// Now by printing the filtered data, we'll see how many tiles of Sentinel-2 are present for fairfax in three months
+print('Total Sentinel-2 Images Available:', s2_data.size()); //(see in console, nearly 72 images are available)
+// You can also print just s2_data to see all metadata properties of 72 images
+// --------------------------------------------------------------------------
+
+
+
+// ----------------------------------------------------------------------
+
+
+/*
+
+2) Reducers:
+
+Background:
+When writing parallel computing code, a Reduce operation allows you to compute statistics on a large amount of inputs.
+In Earth Engine, you need to run reduction operation when creating composites, calculating statistics, doing 
+regression analysis etc. The Earth Engine API comes with a large number of built-in reducer functions 
+(such as ee.Reducer.sum(), ee.Reducer.histogram(), ee.Reducer.linearFit() etc.) that can perform a variety of 
+statistical operations on input data. You can run reducers using the reduce() function. Earth Engine supports 
+running reducers on all data structures that can hold multiple values, such as Images (reducers run on different bands), 
+ImageCollection, FeatureCollection, List, Dictionary etc.
+*/
+
+// Next we'll find middle pixel values for month of June-July-August using mean reducer
+// by using either mean or median we will be able to get composite single image form our image collection of 72 images
+// this step is used to ensure best pixel converage and also to make composite/mosaic of images
+var s2_m = s2_data.median(); // this will take middle value for each pixel in 72 images
+
+// Now if we print the mean_s2 data, it will show only one image because now we have applied mean reducer
+print('Average Sentinel-2 Image: ', s2_m);
+
+// Now we'll visualize the median image over fairfax
+//Defining sentinel-2 visualization parameters
+var s2_vis = {
+
+  "bands":["B4","B3","B2"],
+  "min": 0.026,
+  "max": 0.318
+
+};
+// Adding Sentinel-2 (Median Image) Image over map
+Map.addLayer(s2_m, s2_vis, 'S2 Full Images (Median)', false);
+
+// ----------------------------------------------------------------------
+
+/*
+3) Clipping
+It is often desirable to clip the images to your area of interest. 
+You can use the clip() function to crop an image using a geometry.
+
+NOTE:
+While in a Desktop software, clipping is desirable to remove unnecessary portion 
+of a large image and save computation time, in Earth Engine clipping can actually 
+increase the computation time. As described in the Earth Engine Coding Best Practices 
+guide, avoid clipping the images or do it at the end of your script.
+*/
+
+// Clipping the Sentinel-2 over Fairfox shp (earlier filtered)
+var s2_clip = s2_m.clip(fairfax_shp);
+
+// Visualize the clipped image
+Map.addLayer(s2_clip, s2_vis, 'Clipped S2 Image', false);
+
+// ----------------------------------------------------------------------
+
+/*
+4) Operators
+
+Just like in Python, mathematical operations such as divide, multiply, add, 
+substract, square, etc can be done in GEE using various methods
+Below we'll evaluate NDVI using operators
+
+*/
+
+// For applying any mathematical operation over an image we,ll use image.expression('expression', {})
+
+var s2_ndvi = s2_clip.expression(
+  '((NIR - RED) / (NIR + RED))', {
+    'NIR': s2_clip.select('B8'),
+    'RED': s2_clip.select('B4')
+  });
+// Note: This (evaluating NDVI of an imnage) can also be performed using simple
+// formula provided by GEE which is image.normalizedDifference(['NIR-Band', 'RED-Band'])
+// Defining NDVI visualization
+var s2_ndvi_vis = {
+  min: 0.0,
+  max: 1.0,
+  palette: [
+    'FFFFFF', 'CE7E45', 'DF923D', 'F1B555', 'FCD163', '99B718', '74A901',
+    '66A000', '529400', '3E8601', '207401', '056201', '004C00', '023B01',
+    '012E01', '011D01', '011301'
+  ],
+};
+// Visualizing NDVI over map
+Map.addLayer(s2_ndvi, s2_ndvi_vis, "S2 NDVI", false);
+
+// ----------------------------------------------------------------------
+
+/*
+5) Loops in GEE
+
+Just like Python (For-loops), GEE  support ilterating over data using a 
+function called .map() over feature/image collection.
+
+In case of feature/image collection, we first define a dictionary ourself and
+then map it over whole image collection
+Whereas in case of single feature or image, we directly apply the loop
+
+Examples of both scenearios are given below:
+
+*/
+// ---- First Method ----
+// Here we'll apply NDVI function over all 72 sentinel images 
+// individually between june-august-2021 (s2_data)
+// Thus in the end, all 72 images will contain an extra band named "ndvi"
+
+
+// First defining the function to map later
+// This function will add an extra bad named "ndvi" in each image of our image collection
+// Additionally we can forward the clipping command to each image (this will use much larger processing power of GEE)
+//
+function S2NDVI(image) {
+  var ndvi = image.normalizedDifference(['B8', 'B4']).rename('ndvi').clip(fairfax_shp);
+  return image.addBands(ndvi);
+}
+
+// Now once we define a dictionary based function, we can easily map it over large image collection
+var s2_ndvi_all = s2_data.map(S2NDVI);
+
+// Now if we print and see the properties of each image, there will be an extra added
+// band named "ndvi"
+
+print('NDVI added band in all images: ', s2_ndvi_all); // See 23rd band in console of each image
+
+// We can also visualize any image from previous 72 images using system index ID
+// Selecting random 32th image with its band "ndvi"
+var s2_ndvi_test = s2_ndvi_all.filterMetadata("system:index", "equals", "20210712T155819_20210712T160546_T18STJ")
+                              .select('ndvi');
+// Adding the random selected test NDVI image from Image Collection
+Map.addLayer(s2_ndvi_test, s2_ndvi_vis, "Mapped S2 image containing NDVI", false);