Feature: CO2 Emissions Tracking for MaRDI Activities

[M-dev-stack]

Add functionality to measure and track CO2 emissions from MaRDI activities, including research travel, computational work, and infrastructure operations.
This will support MaRDI's commitment to research sustainability and help quantify the environmental impact of our work.

Motivation & Context

• Growing need for research institutions to measure environmental impact
• Transparency and accountability in research operations
• Support for sustainability goals and carbon reduction strategies
• Enable data-driven decision making on resource allocation

Goals (Phase 1)

• Design data collection workflow for travel emissions
• Establish CO2 conversion factors and methodology
• Create simple user input form or interface
• Build basic dashboard/visualization of aggregate metrics
• Document process and publish initial findings

Scope - What to Track (Initial)

Priority 1:

• Travel emissions
  • Conference attendance
  • Research visits between institutions
  • Transportation modes: plane, train, car, public transport

Priority 2 (Future):

• Computational/processing emissions
• Data center operations
• Cloud service usage

Technical Considerations

• Integration with existing MaRDI portal (MediaWiki + Knowledge Graph)
• Data storage: use existing database?
• Frontend: Wiki extension vs. separate dashboard

Acceptance Criteria

• Form/interface for users to log travel activities
• CO2 calculations working correctly
• At least one visualization (chart/dashboard) of results
• Documentation of methodology and conversion factors
• Initial pilot testing with willing team members

Estimated Effort

• Phase 1: 8 weeks

Proposed Implementation Steps

1. Define scope and finalize CO2 measurement methodology
2. Determine technical architecture (wiki extension)
3. Design data collection interface
4. Implement CO2 calculation engine
5. Create visualization dashboard
6. Pilot test and gather feedback

[M-dev-stack]

Infrastructure Details (Updated)

Current Setup:

• On-premise servers with Kubernetes clusters
• No existing power consumption monitoring
• Main workloads:
  • Data import/export
  • Knowledge graph processing
  • Web portal traffic

Implication: We need to establish baseline energy measurement
before we can calculate CO2 emissions.

[timconrad]

ZIB uses only green energy sources. We do not have any CO2 emissions.

[M-dev-stack]

how about embodied carbon ? See https://docs.green-coding.io/docs/measuring/carbon/embodied-carbon/

[timconrad]

That looks like it does plays a role.