earthengine-py-notebooks

A collection of 360+ Jupyter Python notebook examples for using Google Earth Engine with interactive mapping

Contact: Qiusheng Wu (https://wetlands.io)

1. Description

This repository is a collection of 360+ Jupyter Python notebook examples. I developed these examples by converting my other repo qgis-earthengine-examples from Python scripts to Jupyter notebooks. Now you can display Earth Engine data layers interactively in Jupyter notebooks without having to install QGIS. Several Python packages are being used in these examples, including the Earth Engine Python API, folium, ipyleaflet, and geemap. The geemap Python package is built upon the folium and ipyleaflet packages and implements several methods for interacting with Earth Engine data layers, such as Map.addLayer(), Map.setCenter(), Map.centerObject(), and Map.setOptions().

2. Usage

Step 1: Explore this repository and open any available Jupyter notebook in your browser (e.g., template.ipynb)
Step 2: When the selected Jupyter notebook is open, click the Notebook Viewer link to view the interactive map.
Step 3: If you would like to execute cells in the notebook interactively, you will need to Sign up for a Google Earth Engine account. Then you can click either Run in Google Colab to execute code interactively. This will allow you to add your own Earth Engine Python script.

3. How to automatically convert Earth Engine JavaScripts to Python scripts?

The following examples require the geemap package, which can be installed using pip install geemap. Check the Installation section for more information.

Launch an interactive notebook with Google Colab. Keep in mind that the conversion might not always work perfectly. Additional manual changes might still be needed. ui and chart are not supported. The source code for this automated conversion module can be found at conversion.py.

import os
from geemap.conversion import *

# Create a temporary working directory
work_dir = os.path.join(os.path.expanduser('~'), 'geemap')
# Get Earth Engine JavaScript examples. There are five examples in the geemap package folder. 
# Change js_dir to your own folder containing your Earth Engine JavaScripts, such as js_dir = '/path/to/your/js/folder'
js_dir = get_js_examples(out_dir=work_dir) 

# Convert all Earth Engine JavaScripts in a folder recursively to Python scripts.
js_to_python_dir(in_dir=js_dir, out_dir=js_dir, use_qgis=True)
print("Python scripts saved at: {}".format(js_dir))

# Convert all Earth Engine Python scripts in a folder recursively to Jupyter notebooks.
nb_template = get_nb_template()  # Get the notebook template from the package folder.
py_to_ipynb_dir(js_dir, nb_template)

# Execute all Jupyter notebooks in a folder recursively and save the output cells.
execute_notebook_dir(in_dir=js_dir)

4. Demo

5. Examples

The Table of Contents below mimics the structure of the Google Earth Engine API Documentation. I strongly encourage you to check out the API Documentation if you need an in-depth explanation of each Python example. Please note that the list below does not include all the Python examples contained in this repository. You are welcome to explore the repository and find more examples to suit your needs.

Get Started

Machine Learning

Supervised Classification Algorithms
- Classification and Regression Trees (CART) | Support Vector Machine (SVM) | Confusion Matrix
Unsupervised Classification Algorithms
- KMeans Clustering

Image

ImageCollection

Geometry, Feature, FeatureCollection

Reducer

Reducer Overview
ImageCollection Reductions
Image Reductions
Statistics of an Image Region
Statistics of Image Regions
Statistics of Image Neighborhoods
Statistics of FeatureCollection Columns
Raster to Vector Conversion
Vector to Raster Conversion
Grouped Reductions and Zonal Statistics
- Statistics by group | Zonal Statistics
Weighted Reductions
Linear Regression

Join

Array

Specialized Algorithms

Landsat Algorithms
- Radiance | Surface Reflectance | Simple cloud score | Simple composite
Sentinel-1 Algorithms
Resampling and Reducing Resolution
- Resampling | Reducing Resolution
Linear fit
Pattern recognition
- Center-pivot Irrigation Detector

Asset Management

How Earth Engine Works

Filter

Visualization

Datasets

Tutorials

Global Surface Water

Gena's Examples

Map.addLayer() | Map.centerObject() | Map.setCenter()

image = ee.Image('LANDSAT/LC08/C01/T1_SR/LC08_038029_20180810') landsat_vis = { 'bands': ['B4', 'B3', 'B2'], 'gamma': 1.4 } Map.addLayer(image, landsat_vis, "LE7_TOA_5YEAR/1999_2003", True, 0.7) # Draw any shapes on the map using the Drawing tools before executing this code block feature = Map.draw_last_feature if feature is None: geom = ee.Geometry.Polygon([[[-115.413031, 35.889467], [-115.413031, 36.543157], [-114.034328, 36.543157], [-114.034328, 35.889467], [-115.413031, 35.889467]]]) feature = ee.Feature(geom, {}) roi = feature.geometry() import os out_dir = os.path.join('/root/Downloads') filename = os.path.join(out_dir, 'landsat.tif') geemap.ee_export_image(image, filename=filename, scale=90, region=roi, file_per_band=True)

Downloading Large Images Mosaic from GEE

Hi Professor, I hope you are doing good . I was creating a mosaic for larger area from sentinel 2 collection . Isn't there a way to download those images by chipping into smaller ones ? There is a limit of 1gb but I was trying to download whole cloud free mosaic .

Thank you for your help !

opened by MrZombie69232 3

Export landset image...

Hello, professor,

I want to use your GEEMAP library to export pictures. There are 300 examples in your tutorial. I am a bit confused and I don’t know how to use Google Earth.

Can you give some suggestions based on the picture I gave and which examples should I use?

There are two key points:

Add label of localtion, like A in the pic.
Display the image acquisition time.

Thank you very much. 20211122113953

Another question: I use this script in jupyterlab, there are some error..

Error like this:

Generating URL ...
An error occurred while downloading.
Total request size (42441000 bytes) must be less than or equal to 33554432 bytes.

opened by lovechang1986 1

Scripts in "HowEarthEngineWorks" seem to be template

Noticed that the scripts in /HowEarthEngineWorks don't have what (I assume) is meant to be the meat of the content. Maybe I'm missing something obvious?

opened by TimNagle-McNaughton 1
Can i open a image from gee and change to matrix in jupyter

I performed below： ds = gdal.Open('COPERNICUS/S2_SR20170328T083601_20170328T084228_T35RNK.tif') myarray = np.array(ds.GetRasterBand(1).ReadAsArray()) But fails to open it.

Is there other way to achieved my purpose.

opened by xiaoxiaokuishu 1
Classifier notebooks use old notation

Machine learning notebooks use the old classifier names.

Here are the update ones --> https://groups.google.com/g/google-earthengine-announce/c/rCu4FP_Cn08/m/DqC192X9BAAJ

In any case, I was not able to display the results after using the new classifiers.

opened by RamiroMF 1