GeoDAWG was formed in early 2021, in response to an observed need to bring together disparate expertise on data assimilative modeling in the geospace sciences. Our charter below describes the group in more detail

Current Co-Chairs are Tomoko Matsuo (Aerospace Engineering Sciences Department, University of Colorado Boulder, USA) and David Themens (University of Birmingham, UK)

Motivation and Background

Geospace, extending from Earth’s upper atmosphere to the magnetopause and bow shock to encompass Earth’s entire magnetosphere, is a large interconnected, nonlinear, dynamic system. Dynamics in this region are controlled by external forcing from the Sun and solar wind, as well as forcing from below from lower regions in Earth’s atmosphere. Physical regimes range from magnetohydrodynamics (MHD) to ion-neutral coupling to electron-scale kinetic physics. The disparate regimes are connected via waves, flows, and currents, and couple and feedback upon each other to produce a complex array of interactions occurring at multiple scales, often simultaneously.

Because Geospace is so large, observations are often limited. The scientific community has advanced, physics-based numerical models that can fill in the gaps, but in many cases our ability to simulate Geospace regimes vastly exceeds our ability to constrain or even validate them with observations.

Such large, interconnected systems require simultaneous, multipoint measurements. For the ionosphere these include remote sensing instrumentation such as GNSS, ionosondes, magnetometers, radars, and all sky imagers, while in situ measurements include particles, fields, drifts, and remote sensing (EUV and ENA, e.g.). In the coming years, constellations of spacecrafts are envisioned to complement and expand the multipoint observations of Geospace that are so badly needed. 

As our numerical models continue to advance, and constellations become a reality, there is a need to develop models that can incorporate multipoint measurements across all of Geospace. Data assimilative modeling, in which in situ and/or remote observations are used to actively constrain or guide the numerical solutions, are used with great success in disciplines with large numbers of measurements, such as terrestrial weather forecasting. However, even with large constellations, coverage throughout Geospace will remain sparse and/or unevenly sampled. This, together with non-local, collective interactions of plasmas and neutrals in Geospace, make data assimilative modeling particularly challenging. Yet, if we are to reap the benefits of our advanced numerical models and the breadth of data our community is producing, it is important to bring Data Assimilation (DA) capability to wide swaths of our community, and more tightly link data and modeling than has been traditionally done in the past.

Mission

Recognizing that data assimilative modeling using sparse measurements is an issue from the mesosphere throughout the magnetosphere, the working group is joint between Divisions II and III.

The broad goal is to exchange ideas and techniques on how to incorporate sparsely and/or unevenly sampled measurements into models, a problem we face throughout Geospace. Specific goals include:

• Provide a forum for groups to highlight modeling capabilities. Oftentimes disciplines may not be aware of modeling efforts in other disciplines.

• Discuss complexities of data assimilation unique to Geospace modeling, what can and cannot be adapted from other disciplines such as atmospheric modeling.

• Exchange methodologies across Geospace disciplines. What works and what doesn’t?

• Assist communication within disciplines, to overcome geographic barriers to collaboration

• Connect with groups external to Divisions II and III to learn about different numerical techniques, particularly new state of the art techniques from statistics, mathematics, computer science, and other physical disciplines.

• Share information on upcoming missions or measurements that could be of value to data assimilation modeling groups.

• Provide guidance on data sources, data reliability, & appropriateness for data assimilation. Even the best data assimilative model is only as good as the data that go into it.

• Discuss data sources that may be unique to Geospace such as remote sensing using energetic neutral atoms or auroral imaging.

Membership

Although this is a Division II+III working group, membership is open to any scientist interested in data assimilative modeling in Geospace.

Leadership structure

The working group is chaired by 1 chair and at least 1 co-chair, with at least 1 representative each from Division II and III. Leadership may be expanded as the group grows.

Approach

The approach of this working group is likely to evolve over time, with input from its members. To start, the WG’s approach is to:

Plan and coordinate topical joint sessions at IAGA meetings

Plan and hold regular in-person workshops to exchange data assimilation techniques and maintain awareness of the current state-of-the-art. This includes inviting modelers from communities external to Divisions II and III, including those from statistics, mathematics, and other physical disciplines.

Distribute a regular email newsletter to share research highlights and information relevant to the community, as well as to keep members of the group engaged