The Python notebooks included in this repository were developed for a training workshop that aimed to introduce a heterogeneous community of weather scientists, meteorologists, remote sensing experts, agronomists, etc. to current developments in using Earth Observation (EO) data for crop monitoring. Tho this end, several Python codes and self-directed Python notebooks were developed.

The notebooks review some concepts that are important in crop monitoring using Earth Observation. These include:

• Accessing meteorological data from reanalysis services.
• Showcasing contemporary EO data from e.g. Sentinel 2.
• Illustrating high level, agronomically-relevant, EO-derived parameters over large areas.
• Intuitively build-up towards dynamic crop models, considering how to model the effect of e.g. weather.
• A simple familiarisation with a well-known and widely used crop model (WOFOST).
• An example of assimilating observational streams (leaf area index, soil moisture) into the WOFOST model using an Ensemble Kalman Filter.

Accessing the materials

The materials are available directly on GitHub, but can also be run directly on the browser. For any problems related to these materials, please open a new issue on Github.

See below for links to directly run individual notebooks.

To run the notebooks on the browser click on the links below:

• A brief exploration of meteorological data from an agroclimatology perspective.
• Exploring MODIS LAI data products over Ghana.
• A brief illustration of Sentinel-2 data over northern Ghana.
• This notebook develops the intuition of a very simple production efficiency model (PEM).
• A notebook demonstrating the use of the WOFOST crop model, applied to maize in northern Ghana.
• Using data assimilation (DA) with crop growth models (WOFOST), an example using the Ensemble Kalman Filter (EnKF)