download NDVI

download NDVI.py

import timeit
import sys
import json
import ee

# Authenticate to google earth engine
service_account = '[email protected]'
key_path = 'path_to_your_service_account_key.json'
credentials = ee.ServiceAccountCredentials(service_account, key_path)
ee.Initialize(credentials)

# function to retrieve NDVI data from a single multipolygon

def getNDVI(fire_polygon_json):
    
    featureCollection = ee.FeatureCollection(json.loads(fire_polygon_json)) #json.loads(fire_bounds_geom))

        # Any region of the world
    # polygon = ee.Geometry.Polygon([112.0, 1.0,
    #                               112.0, 1.5,
    #                               112.5, 1.5,
    #                               112.5, 1.0,
    #                               112.0, 1.0])
    startDate = '2021-01-01'
    endDate = '2023-04-01'

    # An ImageCollection is a stack or sequence of images
    modisNDVI = ee.ImageCollection('MODIS/MOD09GA_006_NDVI').select('NDVI').filterDate(startDate, endDate)
        
    def getEarthEngineData(n):
        date = ee.Date(startDate).advance(n,'month')
        m = date.get("month")
        y = date.get("year")
        dic = ee.Dictionary({ # Create the earth engine dictionary
            'Date':date.format('yyyy-MM')
        })
        
        tempNDVI = (modisNDVI.filter(ee.Filter.calendarRange(y, y, 'year'))
                    .filter(ee.Filter.calendarRange(m, m, 'month'))
                    .mean()
                    .reduceRegion(
                        reducer = ee.Reducer.mean(),
                        geometry = featureCollection.geometry(),
                        scale = 250))
        return dic.combine(tempNDVI)

    modis_YrMo = ee.List.sequence(0, 12*2-1).map(getEarthEngineData)

    dataframe = gpd.GeoDataFrame(modis_YrMo.getInfo())

    return dataframe
 
#################################################
# Function to pass a list of polygons to getNDVI() 
# and save the results to a file

def collectAllNDVI(gdf):

    for i in range(len(gdf)): 
        
        # counters
        record = i
        next = record + 1

        # extract GeoDataFrame geometry to json
        fire_polygon_json = gdf.iloc[record:next].geometry.to_json()

        # extract the objectid
        objectid = gdf.iloc[record:next,0].values[0]

        # pass the json record for the fire to our getNDVI function 
        # to query Google Earth Engine for NDVI
        try: 
            
            start_time = timeit.default_timer()
            print(f'Getting NDVI data for objectid: {objectid}')

            # call NDVI function to retrieve NDVI as a pandas dataframe for a given polygon
            firedf = getNDVI(fire_polygon_json)
            elapsed = timeit.default_timer() - start_time
            print(f'Successfully retrieved NDVI data for objectid: {objectid} in {elapsed} seconds')
            
            # Add the OBJECTID to each record
            firedf['OBJECTID'] = objectid

            # save to file
            firedf.to_csv(f'ndvi data/{objectid}.csv', index=False)
            print(f'Saved NDVI data for objectid {objectid} at ndvi data/{objectid}.csv')
            # write confirmation to log file
            with open("output.out", 'a') as f:
                f.write(f'Saved NDVI data for objectid {objectid} at ndvi data/{objectid}.csv in {elapsed} seconds.\n')
                f.close()
        
        except Exception as error:
            print(f'Error: could not retrieve NDVI data for objectid: {objectid}')
            print(error)
            with open("output.out", 'a') as f:
                f.write(f'Error: could not save NDVI data for objectid {objectid}\n')
                f.close()
 

#################################################
# Execute functions

# Define batch download based on row number of GeoPandas dataframe
# I found that batches of 250-500 would complete without issues
# Some images take longer to download than others based on the size of the polygon
ndvi_gdf_run = gdf.iloc[0:250]

# Collect all 
collectAllNDVI(ndvi_gdf_run)
print('Complete.')