— Project 1 (2022) —

— AI and Atmospheric Modelling —

Atmospheric modelling of planets within our solar system and the impact of space weather, are based on data sets that are sparse and disparate (different sensors, missions, altitudes, seasons, time, and latitudes). This disparity is making such modelling particularly difficult. If we were able to fully understand the impact of space weather on Mars and correlate such data with what’s happening on Venus and Earth, we might come a long way in understanding the impact of space weather on our own planet’s atmosphere, at a time when climate change is of the utmost importance.

Cross-mission and cross-planet correlation of data would be a major step towards this direction if the data could be normalised. Such comparative data analysis is particularly complex for humans.

But with recent successes of Artificial Intelligence (AI) in astronomical discoveries (discovery of exoplanets that were undetectable by humans), the use of AI could be a solution in achieving the impossible.

Utilising AI supercomputers, and computing experts, it might be possible to formulate such an AI task.

The purpose of this event is to bring together plasma-physicists and AI computing experts in a first attempt to create innovative new solutions that could tackle this space weather problem.

Utilising AI supercomputers, and computing experts, it might be possible to formulate an AI task / deep-learning algorithm that could analyse and make sense of sparse space weather data, and try to correlate in order to create a model that could prove useful in understanding space weather on Mars, but furthermore understand Venus and its runaway greenhouse effect, thus creating an overall understanding of space weather in multiple planetary bodies. This would assist in understanding our own atmosphere and the impact of solar weather to it, and subsequently identify methods to mitigate or reverse climate change.

We are requesting to supercomputing experts and industry to be involved with infrastructure and experts in this challenge that will kickoff with introductory meetings in the next weeks/months (as preparatory activities) leading to a culmination at the Cospar 2022 General Assembly in Athens, where we aim to hold an experts summit from both sides (super-computing/deep-learning/AI and plasma-physics/space-weather) in order to agree on possible methods to tackle this problem. We aim to identify datasets and algorithms that would then be utilised in a first such pilot task.

— Project 2 (2023) —

— Cubesats: Space-borne augmentation to SWx measurements —

To ensure continuous monitoring of Space Weather (SWx), CubeSats offer a promising solution for a medium-term "near" space monitoring program and a long-term possibility of exploring "deep" space. The altitude range of 50 to 400 km is crucial for studying magnetosphere-ionosphere-thermosphere coupling processes, but it is challenging to access through in-situ measurements. Traditional satellite missions face obstacles such as air drag, making them impractical for long-term operations. Meanwhile, rocket or balloon campaigns only provide sporadic datasets. By leveraging CubeSats, we can address these limitations and establish a sustained monitoring program for SWx, capturing valuable data from the challenging altitude range and facilitating a comprehensive understanding of space phenomena.

CubeSats provide a suitable solution for measuring the density, temperature, and local magnetic field of ionospheric plasma, as well as the currents Joule-heating the upper atmosphere. An effective approach involves placing the CubeSat in an orbit with a decay time exceeding 1 year. This allows for continuous data collection without the requirement for station keeping or orbital maneuvers, simplifying the satellite design significantly. By adopting this approach, we can efficiently gather the necessary information while maintaining a stable and streamlined CubeSat operation.

Expanding our capacity to forecast space weather for longer durations, extending into days or even weeks, is of utmost importance. To achieve this, we must address the sources and underlying causes of space weather phenomena, rather than solely monitoring the current conditions. To accomplish this goal, we require a multifaceted approach, which includes:

• Conduct a comprehensive review of the extensive and continuously expanding database, utilizing an innovative approach that involves the application of AI techniques. Our aim is to identify patterns and correlations within the data, leading to a more profound comprehension of the behavior of the solar wind and its interaction with our magnetosphere.
• Improve ground-based observations of solar wind interactions with the Earth's magnetosphere, utilizing an "ad hoc" space segment. Large satellites are unsuitable for low orbital insertions, necessitating the implementation of a CubeSat program with sampling capabilities. CubeSats, designed for low orbits, enable us to collect data from these regions. By combining ground-based and CubeSat observations, we can enhance our understanding of solar wind interactions with the Earth's magnetosphere.