Visualising vegetation phenology

This notebook describes how to generate a phenology plot of NDVI, the example shown here is for the Gilbert River estuary in Queensland. The climate is dry tropics with a distinct wet season from November - April. The phenology plots are most applicable to areas that have a strong annual growth cycle. Thanks to Chris Holden from Boston University for providing the phenology plotting script.

Date: May 2018

Author: Leo Lymburner

Tags: phenology, NDVI, ipywidgets

%pylab notebook
import datacube
import xarray as xr
from import masking
from import mask_to_dict
from matplotlib import pyplot as plt
from IPython.display import display
import ipywidgets as widgets
import datetime as dt
import matplotlib as mpl
Populating the interactive namespace from numpy and matplotlib
dc = datacube.Datacube(app='Phenology Plot')
#Use this to manually define an upper left/lower right coords

#Define temporal range
start_of_epoch = '1987-01-01'
end_of_epoch =  '2017-12-31'

#Define wavelengths/bands of interest, remove this kwarg to retrieve all bands
bands_of_interest = [#'blue',

#Define sensors of interest
sensors = ['ls8', 'ls7', 'ls5']

query = {'time': (start_of_epoch, end_of_epoch)}
lat_max = -17.3
lat_min = -17.35
lon_max = 140.95
lon_min = 140.85

query['x'] = (lon_min, lon_max)
query['y'] = (lat_max, lat_min)
query['crs'] = 'EPSG:4326'

Retrieve the NBAR and PQ for the spatiotemporal range of interest

#Define which pixel quality artefacts you want removed from the results
mask_components = {'cloud_acca':'no_cloud',
'cloud_shadow_acca' :'no_cloud_shadow',
'cloud_shadow_fmask' : 'no_cloud_shadow',
'cloud_fmask' :'no_cloud',
'blue_saturated' : False,
'green_saturated' : False,
'red_saturated' : False,
'nir_saturated' : False,
'swir1_saturated' : False,
'swir2_saturated' : False,
#Retrieve the NBAR and PQ data
sensor_clean = {}
for sensor in sensors:
    #Load the NBAR and corresponding PQ
    sensor_nbar = dc.load(product= sensor+'_nbar_albers', group_by='solar_day', measurements = bands_of_interest,  **query)
    sensor_pq = dc.load(product= sensor+'_pq_albers', group_by='solar_day',  **query)
    #grab the projection info before masking/sorting
    crs =
    crswkt =
    affine = sensor_nbar.affine
    cloud_free = masking.make_mask(sensor_pq, **mask_components)
    good_data = cloud_free.pixelquality.loc[start_of_epoch:end_of_epoch]
    sensor_nbar = sensor_nbar.where(good_data)
    sensor_clean[sensor] = sensor_nbar
#Conctanate measurements from the different sensors together
nbar_clean = xr.concat(sensor_clean.values(), dim='time')
time_sorted = nbar_clean.time.argsort()
nbar_clean = nbar_clean.isel(time=time_sorted)
nbar_clean.attrs['crs'] = crs
nbar_clean.attrs['affine'] = affine
#calculate the normalised difference vegetation index  (NDVI)
all_ndvi_sorted = ((nbar_clean.nir - +
/g/data/v10/public/modules/agdc-py3-env/20171016/envs/agdc/lib/python3.6/site-packages/xarray/core/ RuntimeWarning: divide by zero encountered in true_divide
  if not reflexive
print ('The number of time slices at this location is '+ str([0]))
The number of time slices at this location is 1644

Plotting an image, select a location for extracting the phenology plot

The interactive widget allows you to select a location (x, y coordinates), the plot will then show all of the time series that fall into the same x coordinate.

#select time slice of interest - this is trial and error until you get a decent image
time_slice_i = 1602
rgb = nbar_clean.isel(time =time_slice_i).to_array(dim='color').sel(color=['swir1', 'nir', 'green']).transpose('y', 'x', 'color')
#rgb = nbar_clean.isel(time =time_slice).to_array(dim='color').sel(color=['swir1', 'nir', 'green']).transpose('y', 'x', 'color')
fake_saturation = 4500
clipped_visible = rgb.where(rgb<fake_saturation).fillna(fake_saturation)
max_val = clipped_visible.max(['y', 'x'])
scaled = (clipped_visible / max_val)
/g/data/v10/public/modules/agdc-py3-env/20171016/envs/agdc/lib/python3.6/site-packages/xarray/core/ RuntimeWarning: invalid value encountered in less
  if not reflexive
#Click on this image to chose the location for time series extraction
w = widgets.HTML("Event information appears here when you click on the figure")
def callback(event):
    global x, y
    x, y = int(event.xdata + 0.5), int(event.ydata + 0.5)
    w.value = 'X: {}, Y: {}'.format(x,y)

fig = plt.figure(figsize =(12,6))
plt.imshow(scaled, interpolation = 'nearest',
           extent=[scaled.coords['x'].min(), scaled.coords['x'].max(),
                   scaled.coords['y'].min(), scaled.coords['y'].max()])

fig.canvas.mpl_connect('button_press_event', callback)

#I'd like to get the date tidy
#this converts the map x coordinate into image x coordinates
image_coords = ~affine * (x, y)
imagex = int(image_coords[0])
imagey = int(image_coords[1])
## Now plot the phenology ##
ts = all_ndvi_sorted.isel(x=[imagex],y=[imagey]).dropna('time', how = 'any')'seaborn-dark')

# Given ``data`` is an xarray.Dataset with dimensions ('time', 'y', 'x')
pix = ts

fig, ax = plt.subplots()

# Grabbing the day of year (for xaxis) and year (for color) made easier with xarray
sp = ax.scatter(pix['time.dayofyear'], pix, c=pix['time.year'],
      , marker='o', edgecolor='none', s=35)

# If you want the months to show up as minor ticks...
month_doys = [, m, 15).timetuple().tm_yday for m in range(1, 13)]
months = mpl.ticker.FixedLocator(month_doys)
months_fmrt = mpl.dates.DateFormatter('%b')

ax.xaxis.set_ticks(month_doys, minor=True)
ax.tick_params(axis='x', which='minor', direction='in', pad=-10)

# Labels and axis limits are good
ax.set_xlim((0, 366))
ax.set_ylim((0, 1))
ax.set_xlabel('Day of Year')
ax.set_title('Phenology "DOY Plot"')
Text(0.5,1,'Phenology "DOY Plot"')
[ ]: