SIGNAL No.17 — Forest carbon credits are being sized for a climate that no longer exists

A new Nature study finds the buffer pools that backstop forest carbon credits have been sized for a climate the world no longer lives in — and the gap is now mappable in space and time.

Today's signal sits across three movements: fashion and culture, where the Yorkshire Fashion Archive treats digitisation as a knowledge problem rather than a photography problem and Formula D reports that experiential, multi-sensory museum design has crossed into the standard kit; technology and bodies, where Frontiers in VR shows minimal wearable haptics already produce real rehabilitation gains via multisensory integration, MIT Technology Review names mechanistic interpretability a 2026 breakthrough as Anthropic and OpenAI move it into operational safety tooling, and the US Congress's bipartisan Strategic Subsea Cables Act puts the cables that carry 95% of international data into the named-infrastructure column; and climate and the carbon books, where Nature exposes the forest-credit buffer-pool gap, PNAS shows extreme-weather attribution messaging measurably raises public climate-policy support, and Johns Hopkins' BioDIGS consortium begins a national reference catalogue of the US soil microbiome.

Today's stories:

1. Forest carbon protocols underestimate climate risks — Nature
2. Memories of style: Digitizing the Yorkshire Fashion Archive — Cambridge Core
3. 6 Experiential Design Trends Transforming Museums in 2026 — Formula D
4. Wearable haptics for VR rehabilitation — Frontiers in Virtual Reality
5. Mechanistic Interpretability: 10 Breakthrough Technologies 2026 — MIT Technology Review
6. US introduces Strategic Subsea Cables Act of 2026 — Submarine Networks
7. Extreme weather attribution predicts climate policy support across the world — PNAS
8. Science army mobilizes to map US soil microbiome — Phys.org / Johns Hopkins

Hosted by Oli. Original sources are at sorah.io.

SIGNAL by SORAH — daily morning briefing on the signal between art and data.