Poster | CoMet Toolkit

VH-RODA workshop 2025

Wed, 01 Oct 2025 00:00:00 +0000

We’re pleased to invite you to a hands-on tutorial session on the CoMet (Community Metrology) Toolkit, taking place during ESA’s Very High-resolution Radar & Optical Data Assessment (VH-RODA) 2025 Workshop which will be held at ESA-ESRIN, Frascati (Italy), from 17-21 November 2025. The tutorial itself will take place:

🗓 Wednesday 19th November

🕒 12:50 – 13:40

📍 ESA-ESRIN (Frascatti, Italy), Big hall

The CoMet Toolkit (www.comet-toolkit.org) is an open-source suite of Python tools for handling and propagating uncertainties and error-correlation in measurement data. Originally designed for Earth Observation (EO) applications, CoMet’s flexible tools can be applied to any dataset involving uncertainties — making it highly relevant for anyone working with measurement functions and data analysis in Python.

🗸 In this tutorial, we will:

Introduce the key concepts behind metrological uncertainty propagation
Walk through the core components of the toolkit:
- punpy – for robust uncertainty propagation with support for error-correlation structures
- obsarray – for managing and storing uncertainty metadata in a self-describing, traceable format
Work through a sensor calibration example using Google Colab-based Jupyter notebooks

This session will be tailored to implement rigorous, traceable uncertainty handling in satellite Cal/Val processing chains.

🗸 No installation is required beforehand – all examples will run in Google Colab!

Living Planet Symposium 2025

Thu, 01 May 2025 00:00:00 +0000

We’re pleased to invite you to a hands-on tutorial session on the CoMet (Community Metrology) Toolkit, taking place during ESA’s Living Planet Symposium 2025:

🗓 Sunday 22nd June

🕒 15:30 – 16:50

📍 Vienna, Living Planet Symposium (Room TBC)

The CoMet Toolkit (www.comet-toolkit.org) is an open-source suite of Python tools for handling and propagating uncertainties and error-correlation in measurement data. Originally designed for Earth Observation (EO) applications, CoMet’s flexible tools can be applied to any dataset involving uncertainties — making it highly relevant for anyone working with measurement functions and data analysis in Python.

🗸 In this tutorial, we will:

Introduce the key concepts behind metrological uncertainty propagation
Walk through the core components of the toolkit:
- punpy – for robust uncertainty propagation with support for error-correlation structures
- obsarray – for managing and storing uncertainty metadata in a self-describing, traceable format
Work through a sensor calibration example using Google Colab-based Jupyter notebooks

We’ll also provide support to get you started using CoMet with your own example use-case if time permits — so feel free to bring a Python example from your work involving measurement uncertainty.

Whether you’re working in satellite Cal/Val, EO data processing, or any other measurement-driven field, this session will help you implement rigorous, traceable uncertainty handling in your processing chains.

🗸 No installation is required beforehand – all examples will run in Google Colab!

HYPERNETS Science Conference

Mon, 07 Nov 2022 00:00:00 +0000

The HYPERNETS project has developed a new hyperspectral radiometer integrated in automated networks of water and land bidirectional reflectance measurements for satellite validation. A detailed uncertainty budget was measured in the lab for these instruments, and these uncertainties are propagated from product to product using the CoMet toolkit within the hypernets_processor. All HYPERNETS products are provided as obsarray-compatible digital effects tables. For the HYPERNETS vicarious calibration study, the uncertainties in these files are also propagated to TOA using punpy.

❔ What is HYPERNETS?

The HYPERNETS project has developed a new hyperspectral radiometer integrated in automated networks of water and land bidirectional reflectance measurements for satellite validation.

❕ How is the CoMet Toolkit contributing to HYPERNETS?

A detailed uncertainty budget was measured in the lab for the HYPERNETS instruments. These uncertainties were then propagated from product to product using the CoMet Toolkit within the hypernets_processor.

All HYPERNETS products are provided as obsarray-compatible digital effects tables. For the HYPERNETS vicarious calibration study, the uncertainties in these files are also propagated to TOA using punpy.

Living Planet Symposium

Thu, 12 May 2022 00:00:00 +0000

ESA’s Living Planet Symposium (LPS) is amongst the biggest Earth observation (EO) conferences in the world. There has been a growing interest in the EO comunity towards using approached from metrology (the measurement of science) to better quantify the uncertainties and error-correlation information on satellite and in-situ EO data. Having accurate uncertainties is paramount for making these data reliable, interpretable and actionable, which is key in understanding and reducing climate change. The CoMet poster being presented at LPS will provide a first introduction to our new toolkit, which aims to enable easier handling of covariance information. The 2022 LPS will be held at the World Conference Center, in Bonn, Germany on 23-27 May 2022. It promises to be bigger and wider ranging than ever before. We look forward to discussing our tools with such an extensive audience.

VH-RODA workshop

Wed, 11 May 2022 00:00:00 +0000

The Very High-resolution Radar & Optical Data Assessment (VH-RODA) will provide an open forum (for the new space, commercial and institutional space sectors) for presenting and discussing about the current status and future developments related to Earth Observation (EO) data quality, calibration and validation of space-borne very high-resolution SAR and Optical sensors and data products, with a dedicated focus on commercial EO data providers and related Cal/Val activities, synergies between optical and SAR communities, presentation of standards and best practices for data quality. Uncertainties are of paramount importance to this field, and the CoMet toolkit could be a very useful tool for this community.