Metadata-Version: 2.4
Name: RadGEEToolbox
Version: 1.7.5
Summary: Streamlined Multispectral & SAR Analysis for Google Earth Engine Python API
Home-page: https://github.com/radwinskis/RadGEEToolbox
Author: Mark Radwin
Author-email: Mark Radwin <markradwin@gmail.com>
License: MIT
Project-URL: Homepage, https://github.com/radwinskis/RadGEEToolbox
Project-URL: Documentation, https://radgeetoolbox.readthedocs.io/en/latest/
Project-URL: Issues, https://github.com/radwinskis/RadGEEToolbox/issues
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Science/Research
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Scientific/Engineering :: GIS
Classifier: Topic :: Scientific/Engineering :: Information Analysis
Classifier: Operating System :: OS Independent
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE.txt
Requires-Dist: earthengine-api
Requires-Dist: numpy
Requires-Dist: pandas
Dynamic: author
Dynamic: home-page
Dynamic: license-file
Dynamic: requires-python

[![PyPI version](https://badge.fury.io/py/RadGEEToolbox.svg)](https://pypi.org/project/RadGEEToolbox/)

# RadGEEToolbox 🛠

![](ReadMeCover.png)

### 🌎 Streamlined Multispectral & SAR Analysis for Google Earth Engine Python API

### [See documentation here](https://radgeetoolbox.readthedocs.io/en/latest/)

**RadGEEToolbox** is an open-source Python package that **simplifies the processing and analysis of satellite imagery using the Google Earth Engine Python API**. It provides **ready-to-use tools for filtering, masking, mosaicking, image/band management, spectral index calculations, and extracting statistics** from multispectral (Landsat, Sentinel-2) and SAR (Sentinel-1) datasets.

Designed for both new and advanced users of Google Earth Engine, RadGEEToolbox minimizes repetitive scripting, accelerates common remote sensing workflows, and aims to maximize efficiency within the constraints of the Google Earth Engine API. Whether you’re building a time series of vegetation indices or extracting surface properties along transects, this package helps get results faster.

Although similar packages exist (eemont, geetools, etc.), `RadGEEToolbox` extends functionality and provides cohesive, chainable methods for research oriented projects working with Landsat TM & OLI, Sentinel-1 SAR, and/or Sentinel-2 MSI datasets (Table 1). The ultimate goal of `RadGEEToolbox` is to make satellite image processing easier and faster for real world applications relying on the most commonly utilized remote sensing platforms.

As of version `1.7.5`, `RadGEEToolbox` supports any generic image collection via the `GenericCollection` module which allows for utilization of the same data management, temporal reduction, zonal statistics, and data export tools available for the `LandsatCollection`, `Sentinel1Collection`, and `Sentinel2Collection` modules. This allows users to provide their own image collection of choice, such as PRISM or MODIS data, to benefit from the tools available with `RadGEEToolbox`.

***Table 1.*** *Comparison of functionality between RadGEEToolbox, eemont, and geetools.*

| Capability | **RadGEEToolbox** | **eemont** | **geetools** |
|:--------------:|:---:|:---:|:---:|
| **Dataset & Workflow Specific API's** | **YES** | NO | NO |
| **Synthetic Aperture Radar (S1) Support** | **YES** | NO | NO |
| **Zonal Time-series Extraction** | **YES** | **YES** | **YES** |
| **Area Time-series Extraction** | **YES** | NO | NO |
| **Transect Time-series Extraction** | **YES** | NO | NO |
| **Mann-Kendall & Sen's Slope Trend Analysis** | **YES** | NO | NO |
| **Comprehensive Preprocessing Operations** | **YES** | **YES** | **YES** |
| **Reflectance Scaling** | **YES** | **YES** | **YES** |
| **Land Surface Temperature Calculation (Landsat)** | **YES** | NO | NO |
| **Narrowband to Broadband Albedo Calculation** | **YES** | NO | NO |
| **Built-in Spectral Index Calculations** | **YES** | **YES** | **YES** |
| **Anomaly (Deviation from Mean) Calculations** | **YES** | NO | NO |
| **Image Masking for Classified Images** | **YES** | NO | NO |
| **Merging of Multiple Collections** | **YES** | NO | NO |
| **Image Selection by Date or Index** | **YES** | **YES** | NO |
| **Visualization Presets/Tools** | **YES** | NO | NO |
| **Batch Export to GEE Asset** | **YES** | **YES** | **YES** |
| **Batch Export to Google Drive** | **YES** | **YES** | **YES** |

_________
## Getting Started with Google Earth Engine

**RadGEEToolbox requires [access](https://developers.google.com/earth-engine/guides/access) to [Google Earth Engine (GEE)](https://earthengine.google.com/) and proper [authentication](https://developers.google.com/earth-engine/guides/auth) with the Python API.**

For more details, see the official [Google Earth Engine Python API Getting Started Guide](https://developers.google.com/earth-engine/guides/quickstart_python) or [Python Installation Guide](https://developers.google.com/earth-engine/guides/python_install).


### 1. Sign Up for Earth Engine
Apply for access here: [https://earthengine.google.com/signup/](https://earthengine.google.com/signup/) using a Google account. Approval typically takes a few days.

### 2. Install the Earth Engine Python API
**The Earth Engine API is installed automatically when you install `RadGEEToolbox`.** However, if you would like to install manually:

```bash
pip install earthengine-api
```

### 3. Authenticate & Initialize
Prior to using the Earth Engine Python client library, you need to authenticate and use the resultant credentials to initialize the Python client. The authentication flows use Cloud Projects to authenticate, and they're used for unpaid (free, noncommercial) use as well as paid use.


**Run the following during your first use:**
```python
import ee
ee.Authenticate()
```
***NOTE: Your GEE credentials will not permanantly be stored on your PC and you will periodically need to re-run `ee.Authenticate()` when `ee.Initialize()` returns an authentication error.***

`ee.Autheticate()` will select the best authentication mode for your environment, and prompt you to confirm access for your scripts. To initialize, you will need to provide a project that you own, or have permissions to use. This project will be used for running all Earth Engine operations:

```python
ee.Initialize(project='my-project')
```
Replacing 'my-project' with the name of the Google Cloud Project you created on sign-up or any Google Cloud Project that has the GEE API enabled.

### 4. Authentication Best Practices
It is reccomended to use the following authentication procedure once you have completed your initial authentication:
```python
import ee

try:
    ee.Initialize()
except Exception as e:
    ee.Authenticate()
    ee.Initialize()
```

### 5. Troubleshooting
`AttributeError: module 'ee' has no attribute 'Initialize'`

➤ Ensure that earthengine-api is installed properly.

`403 Access Denied`

➤ Your GEE account might not be approved, or the API is not enabled for your project.

`Initialization fails`

➤ Try using `ee.Initialize(project='your-project-id')` explicitly in case you are just calling `ee.Initialize()`

See the official [GEE documentation for authentication »](https://developers.google.com/earth-engine/guides/auth)

_________

## Key Features

- Modular tools for processing **Landsat, Sentinel-1 SAR, and Sentinel-2** imagery
- Efficient filtering, **cloud/shadow/water masking, threshold masking, spectral classification, and mosaicking** of Earth Engine image collections
- Collection management such as **collection merging and band renaming**
- Built-in support for computing **spectral indices** (see below for complete list)
- **Temporal reductions** (mean, median, min, etc.) and **monthly median/mean/min/max** functionality
- Image anomaly calculation
- SAR utilities for **multilooking**, **speckle filtering**, and **backscatter conversion**
- Automated and flexible extraction of **transect and zonal statistics (supporting multiple coordinates or geometries)** across image collections
- Calculation of Mann-Kendall and Sen's Slope timeseries trend analysis
- Easy conversion between RadGEEToolbox and standard Earth Engine objects
- Server-side–friendly workflows and caching for **faster, scalable** processing
- Automatically **batch export image collections to GEE assets or to Google Drive**

**List of all spectral index calculations available for 
Landsat (TM & OLI) and Sentinel-2 (MSI) imagery:**

- Normalized Difference Water Index (NDWI)
- Modified Normalized Difference Water Index (MNDWI)
- Normalized Difference Vegetation Index (NDVI)
- Enhanced Vegetation Index (EVI)
- Soil Adjusted Vegetation Index (SAVI)
- Modified Soil Adjusted Vegetation Index (MSAVI)
- Normalized Difference Moisture Index (NDMI)
- Normalized Difference Turbidity Index (NDTI)
- Chlorophyll-a Index (different for Landsat vs Sentinel-2)
- Normalized Burn Ratio (NBR)
- Normalized Difference Snow Index (NDSI)
- Land Surface Temperature (LST) in Celsius (Landsat only)
- Halite Index (Radwin & Bowen, 2021)
- Gypsum Index (modified from Radwin & Bowen, 2021)
- Broadband Albedo 



Features of `RadGEEToolbox` will be expanded in the future - if there is something you would like to see implemented in `RadGEEToolbox`, please open an issue or discussion on GitHub.

🔍 For a full breakdown of available features and tools, see the [RadGEEToolbox documentation »](https://radgeetoolbox.readthedocs.io/en/latest/)

_____________

## Installation Instructions

### Prerequisites

- **Python**: Ensure you have version 3.8 or higher installed.
- **pip**: This is Python's package installer. 
- **conda-forge**: Community led Conda package installer channel

### Installing via pip

To install `RadGEEToolbox` version 1.7.5 using pip (NOTE: it is recommended to create a new virtual environment):

```bash
pip install RadGEEToolbox==1.7.5
```

### Installing via Conda

To install `RadGEEToolbox` version 1.7.5 using conda-forge (NOTE: it is recommended to create a new virtual environment):

```bash
conda install conda-forge::radgeetoolbox
```

### Manual Installation from Source

1. **Clone the Repository**: 
   ```bash
   git clone https://github.com/radwinskis/RadGEEToolbox.git
   ```

2. **Navigate to Directory**: 
   ```bash
   cd RadGEEToolbox
   ```

3. **Install the Package**:
   ```bash
   pip install .
   ```

### ✅ Verifying the Installation

To verify that `RadGEEToolbox` was installed correctly:

```python
python -c "import RadGEEToolbox; print(RadGEEToolbox.__version__)"
```

You should see `1.7.5` printed as the version number.

### Want to Visualize Data? Install These Too

The core functionality of this package is focused on the processing of geospatial and remote sensing data using Google Earth Engine. It does **not** include interactive mapping functionality by default.

**If you wish to view GEE data on interactive maps (e.g., following along with the mapping portions of the [example notebooks](https://github.com/radwinskis/RadGEEToolbox/tree/main/Example%20Notebooks)), you will need to install the [optional `geemap` package](https://geemap.org) in the environment that `RadGEEToolbox` is installed:**

```python
pip install geemap
```
or 

```bash
conda install conda-forge::geemap
```
**If you plan to use `geemap` in a Jupyter Notebook, you will also need to install `ipykernel`:**

```python
pip install ipykernel
```
or 

```bash
conda install anaconda::ipykernel
```
________
# Usage Example

Below is an example use case using the `LandsatCollection` module to create an NDWI image collection, create water classification maps, and create a time series of water area.

### 1. Create a Filtered Landsat Image Collection

```python
# 1. Import necessary packages and modules
import ee
from RadGEEToolbox import LandsatCollection

# 2. Authenticate & Initialize GEE API
ee.Authenticate()
ee.Initialize(project='my-cloud-project-ID') #replace with your Google Cloud project ID

# 3. Define study area boundary - in this case Lake Powell, Utah
study_area = ee.Geometry.Polygon(
        [[[-111.35875055487008, 37.19999663127137],
          [-111.35875055487008, 37.00119876939416],
          [-111.12048456365915, 37.00119876939416],
          [-111.12048456365915, 37.19999663127137]]])

# 3. Create a Landsat image collection for a given time range and study area
# includes filtering imagery based on areal percentage of cloud cover
# WRS-2 tile(s) can also be used for filtering instead of an ROI geometry
collection = LandsatCollection(
    start_date='2020-01-01', #date formats of 'YYYY-MM-DD'
    end_date='2025-01-01',
    cloud_percentage_threshold=10, #filtering to <10% cloud coverage
    boundary=study_area #ee.Geometry() of your study area
    )

# 4. Check collection by printing the dates of all images in the collection
dates = collection.dates 
print(dates)
```
```
['2020-01-13', '2020-02-14', '2020-05-04', '2020-05-20', '2020-07-07', '2020-08-08', '2020-08-24', '2020-09-25', '2020-10-11', '2020-10-27', '2020-11-28', '2020-12-30', '2021-01-15', '2021-01-31', '2021-04-05', '2021-05-07', '2021-06-08', '2021-07-10', '2021-08-27', '2021-10-30', '2021-11-10', '2021-11-10', '2021-12-01', '2021-12-17', '2022-01-02', '2022-01-10', '2022-01-26', '2022-02-11', '2022-02-19', '2022-02-27', '2022-03-23', '2022-04-08', '2022-05-02', '2022-05-18', '2022-05-26', '2022-06-11', '2022-06-27', '2022-07-13', '2022-08-06', '2022-08-22', '2022-08-30', '2022-09-07', '2022-09-23', '2022-10-01', '2022-10-09', '2022-10-17', '2022-10-25', '2022-11-10', '2023-01-21', '2023-03-18', '2023-04-11', '2023-06-22', '2023-06-30', '2023-07-08', '2023-07-16', '2023-09-18', '2023-09-26', '2023-10-04', '2023-10-20', '2023-11-13', '2023-11-21', '2023-12-07', '2023-12-15', '2023-12-31', '2024-01-16', '2024-03-20', '2024-04-13', '2024-04-21', '2024-05-07', '2024-05-23', '2024-06-08', '2024-06-16', '2024-07-02', '2024-07-10', '2024-07-26', '2024-08-19', '2024-08-27', '2024-09-04', '2024-09-12', '2024-09-20', '2024-09-28', '2024-10-06', '2024-10-22', '2024-10-30', '2024-11-07', '2024-11-15', '2024-12-01']
```
### 2. Apply a Cloud Mask and Compute NDWI
```python
# 1. Mask clouds 
cloud_masked_collection = collection.masked_clouds_collection
# 2. Create a collection of singleband NDWI images with band names of 'ndwi'
cloud_masked_NDWI_collection = cloud_masked_collection.ndwi
# BONUS - 3. Create water classification maps using a user-set binary NDWI threshold
water_classification_maps = cloud_masked_collection.ndwi_collection(
      threshold=0
      )
```
Then, explore images from `water_classification_maps` as shown below either by exporting to a GEE asset for download, or by using `geemap` 👇

![Visualization of true color and classified water (in blue) from one of the dates in the collection](LakePowellNDWI.png)

***Visualization of true color and classified water (in blue) from one of the dates in the collection. To see the code used to display this image, please view [this Example Notebook](https://github.com/radwinskis/RadGEEToolbox/blob/main/Example%20Notebooks/Complete_ReadMe_Example.ipynb)***

### 3. Calculate Water Area Time Series
```python
calculate_water_area = cloud_masked_NDWI_collection.PixelAreaSumCollection(
      band_name='ndwi', #specify band to use from collection
      geometry=study_area, #ee.Geometry() of your study area
      threshold=0, #binary classification threshold for unclassified rasters,
      scale=90 #pixel size for zonal statistics
      )
water_area_time_series = calculate_water_area.ExtractProperties('ndwi')
```
Then, directly plot `water_area_time_series` using Matplotlib as shown below 👇


![Plotted Results from Above Example - All Processed in Less Than 5 Seconds!](LakePowellPlot.png)

***To see the code used to display this plot, please view [this Example Notebook](https://github.com/radwinskis/RadGEEToolbox/blob/main/Example%20Notebooks/Complete_ReadMe_Example.ipynb)***

For details about Sentinel-1 SAR and Sentinel-2 MSI modules, and all other available Landsat or cross-module functions, please refer to the [RadGEEToolbox documentation](https://radgeetoolbox.readthedocs.io/en/latest/). You can also explore [`/Example Notebooks`](https://github.com/radwinskis/RadGEEToolbox/tree/main/Example%20Notebooks) for more usage examples.

> NOTE for those newer to Python: when you see ***`property`*** next to the function name in the [RadGEEToolbox documentation](https://radgeetoolbox.readthedocs.io/en/latest/), you do not need to add parentheses at the end of the function when calling the function. For example: ```my_image_collection.add_date_property``` calls the `add_date_property` property function upon the `my_image_collection` image collection - no parentheses needed. All other non-property functions require one or more arguments to be specified by the user, and will require the use of parentheses to define the function arguments.

________


## Contributing

Contributions are welcome! If you’d like to suggest a feature, report a bug, or contribute code or documentation, please visit the [GitHub Issues](https://github.com/radwinskis/RadGEEToolbox/issues) page to get started or view the `CONTRIBUTING.md` file in the main directory.

____ 

## License

RadGEEToolbox is released under the MIT License.


