The GMT Advisory Group meeting was held September 10, 2026. The following
is a mostly AI generated summary of the meeting.
Meeting Summary
Quick recap
The meeting began with a discussion about hosting a GMT in-person summit with a plan to develop a list of conventions or conferences that potential attendees might consider participating in if some travel funding were available. Remko Sharroo noted he built a MacPorts build for the GMT v6.7.0. The meeting focused on how AI is changing the use and support of GMT, with David describing how he used an AI interface to help debug code and generate GMT plots interactively. Rob and Solar discussed the growing importance of high-quality, accurate documentation for GMT, since LLMs rely on it, and noted that AI has already sped up debugging and identifying issues in the code. The group considered the idea of agent-based interfaces for GMT to make interactions more consistent and robust, and Solar mentioned the use of Jupyter environments as a model for more interactive workflows. They reviewed the recent Asian-friendly meeting, which covered GMT 6.7, access to Claude for core developers, and Joaquim’s new graphical interface for GMT built on Julia and Windows, as well as his efforts to integrate MB System tools. Remko and Walter highlighted the unique strengths of GMT for high-quality, publication-ready graphics, shared examples of unexpected uses such as aviation mapping and medical imaging, and discussed the value of showcasing such stories in AGU posters. The conversation ended with participants noting the informative nature of the discussion and the need to keep promoting GMT’s capabilities.
Next steps
David
- Continue working on code modifications to the MB System to enable it to build in Joaquim’s Windows environment, as discussed.
Solar
-
Follow up offline with Remko to gather more information about the strengths of GMT and its quality compared to other tools, to inform the content of the AGU poster.
-
Monitor participation in the next “Asian friendly” meeting (scheduled for November) and discuss in the forum whether it is worth continuing these meetings based on the level of participation.
Summary
GMT Summit Discussion
Proposed action is to come up with a list of conventions and ask our advisory group and even the community to vote on which ones they feel they would visit if they could receive monetary travel support. Venues such as Fall AGU, SSA, EGU, and NSF NGF all came up as possibilities.
AI Interfaces for GMT Software
The group discussed how AI interfaces are changing the way people interact with GMT software. David shared his experience using Claude to help with code debugging and generate GMT plots, which helped him isolate issues in his code. Rob emphasized the importance of good documentation to ensure AI models provide accurate instructions for using GMT. The discussion also touched on potential future developments, including the creation of agent-mediated interfaces for GMT and the integration of AI tools with Jupyter notebooks for data analysis. Walter raised concerns about the potential impact of AI on GMT’s popularity, noting that if AI systems find GMT difficult to use, they might steer users toward alternative tools.
Asian Team Meeting Updates
Walter inquired about a previous meeting with Asian colleagues that he missed, specifically asking about attendance and outcomes. Solar confirmed the meeting was held in August with about five or six attendees, including representatives from China, and discussed setting up license access for Dongdong, Joaquim, and Federico to use Claude. The meeting also covered the recent GMT 6.7 release and Joaquim’s development of a graphical tool interface.
GMT Graphical Interface Development Updates
The team discussed Joaquim’s development of a new GMT graphical interface using Julia, which allows interactive manipulation of data views and access to data sources. David mentioned ongoing work to integrate MB System tools as GMT modules, though progress has been slower than expected due to environment differences. The group also briefly discussed earlier attempts at creating graphical interfaces for GMT workflows, with Remko describing Joaquim’s previous interface that used menu-based interactions rather than command-line inputs.
Meeting Scheduling and GMT Discussion
The team discussed scheduling a meeting for November, with Solar expressing concerns about low participation from Asian participants and questioning whether it would be worth continuing. Walter raised concerns about time zone challenges, particularly for European participants who would need to attend at 3 AM. Remko shared details about a GMT wallpaper featuring satellite altimeter tracks and praised GMT’s superior quality compared to alternatives like MATLAB for publications and graphics.
AGU Poster Planning Discussion
The group discussed elements for a poster at AGU, with Remko suggesting they include data on the high number of scientific publications using GMT. They shared an example of a Lufthansa pilot named Christoph who created custom maps for emergency flight routes using GMT, which led to a discussion about specialized aviation symbols that needed to be encoded in GMT. Walter mentioned that Christoph had attended their summit in Norway before Paul’s death and had worked with them to add specific aviation chart symbols to the GMT system.
Unexpected GMT Applications Discussion
The group discussed unexpected uses of GMT , including its application in medical imaging and historical research on Earth’s rotation rate using ancient Chinese documentation of battles. Walter expressed interest in having the guest speaker return for another meeting to share updates on their work. The conversation ended with Rob planning to discuss the conversation with Solar to prepare for future presentations or posters.
Attendees
-
Rob Casey
-
Solar Smith
. Unfortunately saw too late that the protocol from the previous meeting states the potential next meeting date.