More Advanced Data Cube Usage

Interactive Crop Health

In this notebook you will learn about an interactive way of doing time series analysis of field, to understand the status of crops.

This notebook will require a number of external Python modules to run, which will be imported at the top of the notebook as his convention.

import sys
import datacube
import warnings
sys.path.append('/eOsphere/bin')
from dcFunctions import scal_2_tru, crop_health

As we have been showed in other notebooks, we need to initialise a datacube instance in order to have access to the Data Cube, and this can be done as follows:

dc = datacube.Datacube()

We need to define our area of interest and this can be done, either specifying the minimum and maximum latitude and longitude, or defining the centre coordinates plus a buffer; which will be the number of square degrees to load around the central latitude and longitude.

lat, lon, buffer = 42.3, 75, 0.5

latitude = (lat - buffer, lat + buffer) 
longitude = (lon - buffer, lon + buffer) 
print(latitude,longitude)

Define a query to load the data from the datacube

query = {
        'x': longitude,
        'y': latitude,      
        'output_crs':'EPSG:32643',
        'resolution':[-250, 250],
}

Load data from the already made NDVI 10-day product. You can change this so that you can do time series analysis on other datasets.

ds = dc.load(product='mcv_indices', measurements=['ndvi'], time=['2020-01-01','2020-10-31'], **query)
ds = scal_2_tru(ds) # Applying scaling to get real values
print(ds)

Now we can run the crop_health function available in dcFunctions.

As arguments you have to pass:

• the data layer
• latitude
• longitude
• method (statical methods to apply: 'mean', 'median', 'min', 'max', 'std', 'minmax'
• buffer (optional, only if central latitude and longitude are used)
• ds_rgb (optional, RGB layer from the MDC, on top of the base layer)

crop_health(data=ds, lat=latitude, lon=longitude, method='std')

If you want to use an RGB layer from the KDC then you will first need to load the data and then pass it to the crop_health function.

query_high = {
        'x': longitude,
        'y': latitude,      
        'output_crs':'EPSG:32643',
        'resolution':[-20, 20],
}

ds_rgb = dc.load(product='rgb',measurements=['red','green','blue'], time='2020-08-21', **query_high)

Now that you load the data from the KDC, you can pass to crop_health as an additional argument.

crop_health(data=ds, lat=lat, lon=lon, buffer=buffer, ds_rgb=ds_rgb, method='mean')