vis4climate: Workshop on Building a Transdisciplinary Climate Vis Communitybookmark_border

Soil organic carbon (SOC) is a key indicator of soil health and one of the largest terrestrial carbon pools. Recent European initiatives, including the EU Soil Monitoring Law, highlight the need for scalable and cost-effective approaches for monitoring agricultural soils. This study investigates how visual analytics can support farmers in the long-term monitoring and management of soil carbon stocks. User requirements were elicited through exploratory interviews with farmers and incorporated into AgroScan, an open-source web platform that combines remote-sensing-based SOC prediction with visual analytics. The platform provides field-level SOC estimates and derived soil organic matter (SOM) values together with contextual information such as crop rotation history to facilitate the interpretation of long-term soil health trends. By increasing the transparency and contextualizing the model outputs, AgroScan enables farmers to explore relationships between management practices and soil carbon dynamics and supports adaptive field management decisions.

Construction accounts for 50% of European material consumption, 36% of EU waste, and 39% of global energy-related greenhouse gas emissions, making material reuse a pressing climate priority. Reclaimed components exist in abundance, but reuse remains impractical within typical project schedules: available stock is scattered across dozens of fragmented Digital Material Banks (DMBs) - web portals that list components salvaged from demolition or refurbishment - and the data on those platforms is frequently incomplete; critical attributes are buried in PDFs rather than exposed as searchable fields. Developed in close collaboration with practising architects, ReUseBIM addresses both obstacles. A dashboard aggregates listings from DMBs into a single metadata-searchable view and provides Building Information Modelling (BIM)-ready Industry Foundation Classes (IFC) files for drag-and-drop into any BIM-capable software, helping architects to efficiently search, compare, and select reclaimed components without leaving their existing workflow. Our enrichment pipeline uses an Large Language Model (LLM) to extract missing attributes from attached documents, presenting each detected value to the architect for confirmation before writing it into the IFC file. Together, these make specifying a reclaimed component as straightforward as specifying a new one.

Weather and climate foundation models produce high-dimensional forecasts whose learned relationships are difficult to inspect with static plots alone. GeoXplain is an interactive Python-based visualization toolkit for exploring geospatial attribution maps across climate variables, atmospheric pressure levels, and forecast time. The toolkit accepts attribution bundles containing attribution grids together with corresponding metadata and renders them in a notebook widget or browser with map and globe modes, linked timelines, pressure-level controls, target annotations, and optional physical-field overlays. We frame GeoXplain as a model-agnostic earth-system visualization toolkit and present the GeoXplain Aurora Adapter as its first computation backend. The adapter computes explanations for the Aurora foundation model, either in a local GPU process, through a GPU listener, or through a SLURM-backed listener, while preserving the same Python call site for analysts. It currently supports gradient saliency, Integrated Gradients, RISE, ViT-CX, multi-frame saliency and Integrated Gradients rollouts, and retrieval of ERA5 overlays. GeoXplain can be installed as a PyPI package with pip install geoxplain. The code is open-source and available at https://github.com/clemenskoprolin/geoxplain.

While the environmental impact of GenAI becomes increasingly significant, and a growing number of tools that aim to raise awareness of this issue, limited attention has been paid to how such information should be visualized and communicated to end users, especially in ways that are easy to understand and interpret. To this end, we designed EcoChat, a Chrome browser extension for the web-based ChatGPT that visualizes carbon emissions of its usage in real time. EcoChat uses various metaphorical representations to contextualize the carbon emissions data and enables users to set up ``carbon budgets'', which are carbon emission allowances representing their own personal goals. Users can compare their carbon budgets with their actual emissions to foster environmental awareness. We reflect on EcoChat's design choices in the context of the current landscape of AI eco-feedback tools, discussing how it addresses several identified shortcomings while also highlighting remaining challenges and future research directions.