Processing and Downloading NDVI Data Using Google Earth Engine

Loading the Country’s Boundary Data

First, we will load the boundary data for Pakistan. This data has already been added to GEE as an asset.

// Define the region of interest: Pakistan's boundary
var pakistan = ee.FeatureCollection("projects/ee-mohsinramay/assets/pak_adm0");

Loading and Filtering MODIS Data

Next, we will load the MOD13A1.061 Terra Vegetation Indices 16-Day Global 500m dataset from MODIS, focusing on data for the year 2023, and filter it to include only the area within Pakistan’s boundaries.

The MOD13A1 V6.1 product provides Vegetation Index (VI) values on a per-pixel basis. It includes two primary vegetation layers: the Normalized Difference Vegetation Index (NDVI), which serves as a continuity index to the NOAA-AVHRR derived NDVI, and the Enhanced Vegetation Index (EVI), which minimizes canopy background variations and maintains sensitivity over dense vegetation conditions. These indices are computed from atmospherically corrected bi-directional surface reflectances, which have been masked for water, clouds, heavy aerosols, and cloud shadows.

For more detailed information, refer to the official MOD13A1 dataset documentation.

// Load the MODIS Vegetation Indices data for 2023
var dataset = ee.ImageCollection('MODIS/061/MOD13A1')
                  .filterDate('2023-01-01', '2023-12-31')
                  .filterBounds(pakistan);

Selecting the NDVI Band

We will then select only the NDVI band from the dataset.

// Select the NDVI band (Normalized Difference Vegetation Index)
var ndvi = dataset.select('NDVI');

Clipping the NDVI Imagery to Pakistan

To focus on our Area of Interest, we will define a function to clip the NDVI imagery to Pakistan’s boundaries. This function will be used in the subsequent step when calculating the mean NDVI for each month.

// Clip the data to the Pakistan boundary
var clippedNDVI = ndvi.map(function(image) {
  return image.clip(pakistan);
});

// Group NDVI by month and calculate the mean for each month
var monthlyNDVI = ee.ImageCollection.fromImages(
  ee.List.sequence(1, 12).map(function(month) {
    return clippedNDVI
      .filter(ee.Filter.calendarRange(month, month, 'month'))
      .mean()
      .set('month', month);
  })
);

Combining Monthly NDVI Bands

We will now combine the monthly NDVI bands into a single image, where each band represents the mean NDVI for a specific month.

// Combine all monthly mean images into a single image with multiple bands
var combinedMonthlyNDVI = monthlyNDVI.toBands();

Renaming Bands for Later Processing

Next, we will rename the bands according to the month abbreviations, which will facilitate later processing in R. We will also print the combinedMonthlyNDVI data to ensure it aligns with our expectations.

// Corrected band renaming logic using map
var monthNames = ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'];
var newBandNames = ee.List(monthNames).map(function(name) {
  return ee.String('NDVI_').cat(name);
});
combinedMonthlyNDVI = combinedMonthlyNDVI.rename(newBandNames);

print(combinedMonthlyNDVI)

Visualizing the NDVI Data

Now, let’s visualize the NDVI data to inspect its quality. Below is a plot of the mean NDVI for January 2023.

// Set visualization parameters for NDVI
var ndviVis = {
  min: 0.0,
  max: 9000,
  palette: [
    'ffffff', 'ce7e45', 'df923d', 'f1b555', 'fcd163', '99b718', '74a901',
    '66a000', '529400', '3e8601', '207401', '056201', '004c00', '023b01',
    '012e01', '011d01', '011301'
  ],
};

// Add the layer to the map
Map.centerObject(pakistan, 6);
Map.addLayer(pakistan, {}, 'Pakistan');
Map.addLayer(combinedMonthlyNDVI.select('NDVI_Jan'), ndviVis, 'Pakistan NDVI January 2023');

Exporting the Data for Local Processing

Once we are satisfied with the data, we can export the multiband image to Google Drive. From there, it can be downloaded locally and further processed in R or Python.

// Export the combined image to Drive
Export.image.toDrive({
  image: combinedMonthlyNDVI,
  description: 'Pakistan_Monthly_NDVI_2023',
  folder: 'Remote Sensing',
  scale: 500,
  crs: 'EPSG:4326', // Specifying WGS 84 as the CRS
  region: pakistan.geometry(),
  fileFormat: 'GeoTIFF',
  formatOptions: {
    cloudOptimized: true
  },
  maxPixels: 1e13  // Increase maxPixels if necessary
});

That’s it!

Feel free to reach me out if you got any questions.