Second workshop on synthetic satellite data generation and application to hydrological modelling

Location: Online Date: 25 February 2025, 9h-12h GMT Organisers: Grégoire Mariéthoz (UNIL), Loïc Gerber (UNIL), Said Obakrim (UNIL), Moctar Dembélé (IMWI)

Background

To assess climate change impacts on hydrology, long-term data is essential. While modern satellites offer detailed current observations, historical records are often incomplete. This project proposes a framework to generate synthetic satellite time series from by learning from present-day satellite and reanalysis data. Using machine learning approaches, it aims to reconstruct past hydrological processes and support long-term climate impact studies, notably related to hydrology.

The organizers

The proposed workshop is part of a research project carried out at the Geostatistical Algorithms & Image Analysis research group (GAIA lab) of the University of Lausanne (Switzerland) and International Water Management Institute (IWMI) West Africa Office. Our work is aimed at studying the possibility of generating synthetic hydrological data consistent with climate reanalysis and assessing their potential for long-term hydrological modelling. We have experience in remote sensing, hydrological modelling, and geostatistics.

Goal of the workshop

The goal of this second workshop is to demonstrate some of image generation tools that we have developed, in an interactive online session. The method presented is an improved and more user-friendly version of the prototype codes used in the first iteration of the workshop in 2024. No particular software installation is required as the entirety of the developed codes are based exclusively on the online platforms Google Colab and Google Earth Engine. However participants will need to have a valid Google account as well as a Google Cloud project registered for noncommercial use (follow procedure here https://console.cloud.google.com/earth-engine/configuration, it takes a couple of minutes).

Target participants

This workshop is designed for professionals and researchers in various disciplines including, but not limited to: hydrologists, climatologists, meteorologists, geo-statisticians, geoscientists, data scientists, remote sensing specialists and environmentalists.

Expected outcomes

By the end of the workshop, participants are expected to: • Gain a thorough understanding of the methodology for generating synthetic satellite image time series. • Acquire practical skills necessary for implementing the framework in their respective areas of work. • Foster collaboration and knowledge exchange among professionals from diverse disciplines.

Here is the video where you can re-live the workshop:

https://unils-my.sharepoint.com/:v:/g/personal/gregoire_mariethoz_unil_ch/IQBqDniWqwbbQrPEVFAOMKIiAV36JXsohxyOpRzTw8c-xzs?e=gsDr5N

And are the slides of the workshop:

https://unils-my.sharepoint.com/:b:/g/personal/gregoire_mariethoz_unil_ch/IQB8-xN1CV6sSLZAGZy3d8GPAepIHqNK4fh2imNSElzoSLQ?e=43K9wQ

Contact

For more information, please contact:
Dr Moctar Dembélé (IWMI): moctar.dembele@cgiar.org
Prof. Grégoire Mariéthoz (UNIL) : gregoire.mariethoz@unil.ch
Mr. Loic Gerber (UNIL) : loic.gerber.2@unil.ch