🛰️

Robbi Bishop-Taylor robbibt

🛰️

Earth Observation Scientist at @GeoscienceAustralia. Using satellites 🛰️, data science 📊, geospatial 🗺️ and dataviz 🎨 to map our changing planet

robbibt / dcload_inline_config.py

Last active September 24, 2019 07:03

Specify custom dc.load config details in-line inside a notebook

	# Code via David Gavin

	import datacube

	# Connect to standard database for Sentinel 2 ARD
	dc = datacube.Datacube()

	# Set up a custom config to access s2tsmask database
	config = {'db_hostname': 'test.test.org.au',
	'db_port': 1234,

robbibt / matplotlib_hillshading.py

Last active July 29, 2021 08:15

Applying a hillshade to an array using Matplotlib

	# Code via Alex Leith

	from matplotlib.colors import LightSource
	import matplotlib.pyplot as plt

	# Sample elevation array
	elevation = data_cube.elevation.isel(time=0).values

	# Create hillsahde based on the array
	ls = LightSource(azdeg=315, altdeg=45)

robbibt / facet_plot_seaborn.py

Created October 8, 2019 01:01

Creating a simple faceted plot using Seaborn

robbibt / rayshader_africa.R

Last active October 10, 2019 04:27

R script for making pseudo-topography 3D animations using Rayshader and Datacube data


	library(rayshader)
	library(magrittr)
	library(raster)
	library(lubridate)
	library(dplyr)

	smooth_sequence = function(from=0, to=1, cycles=2, n=180) {

	#' Produces a smooth sequence of any length between two values

robbibt / wofs_stats.py

Created October 21, 2019 05:41

Compute valid WOfS statistics from WOFLs

	# Source: https://bitbucket.org/kirill-ga/wofs-summary/src/0f1e4284085e2a65bd7a0b9fdb00e0f65c36f713/kk/wofs_summary/_wofs_stats.py?utm_term=file&utm_source=bb-slack&utm_medium=referral-external&atlOrigin=eyJpIjoiMWRiZjlmZjhkYmE3NDg0Mzk3NWI3ODZhZjczNGQyODQiLCJwIjoiYmItY2hhdHMtaW50ZWdyYXRpb24ifQ#lines-114:136

	def wofs_stats(xx):
	attrs = {'crs': xx.crs}

	xx_wet = make_mask(xx.water, wet=True).sum(dim='time', dtype='int16')
	xx_dry = make_mask(xx.water, dry=True).sum(dim='time', dtype='int16')
	xx_clear = xx_wet + xx_dry
	xx_freq = xx_wet.astype('float32')/xx_clear

robbibt / check_product_metadata.py

Created October 30, 2019 05:41

Get a list of available product metadata that can be filtered on load

	ds, = dc.find_datasets(product='ga_ls5t_ard_3', limit=1)
	dir(ds.metadata)

robbibt / import_multiple_geotiffs.py

Last active November 14, 2019 23:51

Import multiple geoTIFFs to xarray

	def time_index_from_filenames(filenames, string_slice=slice(0, 10)):
	'''
	Helper function to generate a Pandas datetimeindex object
	from dates contained in a file path string
	'''

	date_strings = [os.path.basename(i)[string_slice] for i in filenames]

	return pd.to_datetime(date_strings)

robbibt / from_xr_to_dc.py

Created December 4, 2019 06:15

Convert plain xarray to datacube xarray with geobox

	# Code from Kirill Kouzoubov, Digital Earth Australia

	import xarray as xr
	import datacube
	def from_xr_to_dc(xx):
	crs = xx.data_vars.get('crs')
	if crs is None:
	print('no crs var')
	return xx
	crs = crs.attrs.get('crs_wkt', None)

robbibt / DEA_hotspots_download.R

Last active February 13, 2020 04:25

Downloading DEA Hotspots data via WFS in R

	# Set up parameters to load Hotspots data
	from_date = '2020-01-27'
	to_date = '2020-02-12'
	y_max = -35.38
	x_min = 148.68
	y_min = -36.02
	x_max = 149.32
	min_confidence = 0 # minimum confidence value of hotspots to load from WFS
	max_features = 900000 # maximum features to load from WFS

robbibt / product_metadata.py

Last active August 18, 2025 08:08

Load product metadata as an xr.DataArray that matches dc.load data

	# Code adapted from original by Andrew Hicks

	import numpy as np
	import datacube
	from datacube.api.query import query_group_by
	from datacube.model.utils import xr_apply
	dc = datacube.Datacube(app='Metadata')

	# Spatiotemporal query
	query = dict(product='ga_ls8c_ard_3',

	import seaborn as sns
	import pandas as pd

	# Create test data
	pard_std_df = pd.DataFrame(data=[['Burdekin', 'Midshelf', 0.6, 0.5, 0.4],
	['Burdekin', 'Offshore', 0.3, 0.5, 0.8],
	['Burdekin', 'Coastal', 0.2, 0.4, 0.9],
	['Fitzroy', 'Midshelf', 0.6, 0.5, 0.4],
	['Fitzroy', 'Offshore', 0.3, 0.5, 0.8],
	['Fitzroy', 'Coastal', 0.1, 0.2, 0.3]],