openEO Overview

openEO Back-end

This section covers the usage of the openEO back-end for Gaussian Process Regression (GPR) using PyEOGPR.

class pyeogpr.Datacube[source]

Bases: object

sensorSENTINEL2_L1C, SENTINEL2_L2A, SENTINEL3_OLCI_L1B, SENTINEL3_SYN_L2_SYN

Satellite sensor to process the data with.

biovarBiophysical variable to process. The selected variable’s map will be retrieved.

Currently “built-in” variables available for each sensor:

  • for own model, simply put the directory of your model on your machine.

bounding_boxlist

Your region of interest. Insert bbox as list. Can be selected from https://geojson.io/ (e.g.: [-4.55, 42.73,-4.48, 42.77])

temporal_extentlist

Your temporal extent to be processed. (e.g.: [“2021-01-01”, “2021-12-31”])

cloudmaskBoolean

If “True” the Sentinel 2 cloud mask will be applied (only to S2 data), with Gaussian convolution to have smoother edges when masking.

Methods

construct_datacube([composite])

Build the datacube with optional temporal compositing and cloud masking.

process_map([gapfill, fileformat])

Process the datacube into maps, by applying GPR algorithm on the spectral image stack.
construct_datacube(composite=None)[source]

Build the datacube with optional temporal compositing and cloud masking.

Parameters:

composite (str, optional) – Temporal compositing interval (e.g., ‘month’, ‘dekad’).

process_map(gapfill=False, fileformat='tiff')[source]

Process the datacube into maps, by applying GPR algorithm on the spectral image stack.

Parameters:

  • gapfill (bool, default=False) – Apply Savitzky-Golay interpolator for gap filling.

  • fileformat (str, default='tiff') – Output file format (‘nc’ or ‘tiff’).

Example Usage

import pyeogpr

bounding_box = [
    17.897539591074604,
    46.59810244496674,
    17.96594608650338,
    46.639078751019014
  ]

time_window = ["2020-07-01", "2020-07-10"]

dc = pyeogpr.Datacube(
    "SENTINEL2_L1C",
    "Cm",
    bounding_box,
    time_window,
    cloudmask=False)

dc.construct_datacube("dekad")

dc.process_map(gapfill=False, fileformat="tiff")

To download the GPR processed map go to the openEO portal:

download

You can use QGIS or Panoply to visualize. IMPORTANT: The data range is off, due to few pixels being outliers. Set the data range manually for the corresponding variable e.g. FVC –> 0 to 1.

map