Getting started

Versión en español

Getting started

All you need is to install the tool you are going to use: otherwise, MonitorEO computes in Google Earth Engine with its shared account, with no account of your own and no sign-in. With your Earth Engine account and your Google Cloud project, the computations count against your quota and you can also export to Google Drive. The Earth Engine app is not even installed: it opens in the browser (manual).

1. Earth Engine account and project

MonitorEO computes everything in Google Earth Engine, with one of these two accounts:

• The shared account of MonitorEO, the default when you do not give a project: you need no Earth Engine account and no sign-in. It computes with the Google Cloud project of MonitorEO maps, time series, trends, comparisons, fires and GeoTIFFs that are downloaded to your computer. It does not export to Google Drive, and each user has 25 EECU-h of computation per month, their research EECUs, for research only (Concepts): when you have spent them, or if the account does not respond for a while, use your own. It is a non-commercial project, so it is good for the same uses described below: for the operational management of a public administration, commissioned work or the work of a company, use your own commercial project.
• Your account and your project: the computations count against the quota of your project and you can export to Google Drive (maps of large regions at full resolution, long time series). It is enough to give your project in the tool.

To use your account you need:

1. A Google account. In a public administration, Google requires the institutional Google account of the body.
2. A Google Cloud project registered in Earth Engine. For non-commercial use (research, teaching, NGOs, personal non-profit use) it is free, with a monthly computing quota. It is created at https://code.earthengine.google.com/register (choose Unpaid usage and the type of use) or from the Google Cloud console; details at https://developers.google.com/earth-engine/guides/access. According to Google's terms, companies generally need a commercial project (Paid usage), and an administration can use it for free only for academic research (papers, theses, reports), except those of less developed countries and indigenous governments: operational management (ongoing monitoring, tools or products for management) and internal testing need a commercial project. Non-commercial use does not cover commissioned or paid work for administrations or companies either. Google reviews each case.
3. The project ID (e.g. my-ee-project, not its display name): you give it in each tool or, once for all of them, in the environment variable EE_PROJECT.

If you already use the Earth Engine code editor, you already have a project: it is the one shown at the top right.

2. Authentication

Only to use your account, once per computer. The credentials are stored in ~/.config/earthengine/credentials and are valid for all the tools:

• QGIS: the Authenticate… button of the MonitorEO panel, which opens the browser to sign in with your Google account. It is the recommended way in QGIS: no terminal needed.

• Python, command line, MCP and R: in a terminal with the MonitorEO environment activated,

  earthengine authenticate

  (or rgee::ee_Initialize() if you already use rgee). The browser opens to sign in with your Google account.

3. Installation

Tool	What to install
Earth Engine app	Nothing, and no Earth Engine account.
QGIS plugin	The plugin zip. It has everything inside: no Python or conda needed.
Python package and Command line	The monitoreo package (best with the conda environment)
MCP server	The monitoreo package and an entry in the configuration of Claude Desktop, Claude Code or another client. Without knowing Python: step-by-step installation
Assistant for managers	Its installer: double click on Windows, one command in the Terminal of macOS or Ubuntu (or the conda environment). The first time, the assistant downloads its language model and its engine (about 5 GB)
R package	The Python package monitoreo and the R package monitoreo

QGIS plugin

1. Download monitoreo_qgis-<version>.zip from the latest release.
2. In QGIS: Plugins > Manage and Install Plugins > Install from ZIP, choose the zip and Install Plugin.
3. Open the panel with the MonitorEO button. With the Project field empty it already computes with the shared account. To use your account, press Authenticate…, sign in with your Google account in the browser that opens, type your project and press Connect.

It works in QGIS 4.0.3 or later, on Windows, macOS and Linux. Details in QGIS plugin.

Python package (also for the MCP and R)

This section assumes you know how to use a terminal and conda. If not, follow the step-by-step installation, which explains each command.

With Miniforge (conda with conda-forge only; the defaults channel of Anaconda requires a license for large organizations):

curl -O https://raw.githubusercontent.com/monitoreo-obsnev/monitoreo/main/environment.yml
conda env create -f environment.yml
conda activate monitoreo
earthengine authenticate        # only to use your Earth Engine account

The monitoreo environment brings Python 3.12, Earth Engine, GDAL (to read vector layers and write GeoTIFF), matplotlib (charts), the MCP server and the monitoreo package from GitHub.

Or with pip, in any Python 3.9 or later (without GDAL, regions from files do not work):

pip install "monitoreo[mcp,plot] @ https://github.com/monitoreo-obsnev/monitoreo/archive/refs/heads/main.zip"

Check that everything works:

monitoreo parks "Sierra Nevada"

It should list the protected areas with that name and their WDPA ID, computed with the shared account; with your account, monitoreo --project my-ee-project parks "Sierra Nevada".

Updating

• Python and R: repeat the pip installation adding --force-reinstall (or mo_install() in R).
• MCP and assistant: they update themselves to the latest version from GitHub every time they start.
• QGIS: download the zip of the new release and install it over the old one.

monitoreo --version (or monitoreo.__version__) gives the installed version.

Next step

Read Concepts (variables, regions, periods and filters) and follow the manual of your tool.