FAO Uganda drought: May 2026 activation + FAO's 2025-26 flood-side work - #601
Conversation
…rnment declaration; add FAO's 2025-26 flood-side work and the national AA roadmap context The framework page carried activations: [] and a schema_strain note saying no dated activation had been found. It activated in May 2026 (110,000 people, Kotido and Kaabong worst affected) before the government declared a six-month drought emergency in July. Also captured, because anyone searching FAO + Uganda + flood lands here: the Japan-funded OPM/FAO flood early-warning project (Mar 2025 - Mar 2026, Rwenzori and Mount Elgon, 10 hydro-climatic stations, no trigger), and the national AA Roadmap 2026-2031 / U-MHIEWS. Found while reviewing external frameworks for the Uganda flood AA trigger design (ocha-dap/ds-aa-uga-flooding). Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
…three triggers Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
| - https://allafrica.com/stories/202608250108.html | ||
| - https://allafrica.com/stories/202606050689.html | ||
| - https://reliefweb.int/report/uganda/uganda-roadmap-anticipatory-action-2026-2031 | ||
| - "FAO_Uganda_Flood AAP for Elgon (draft docx, 2026-09-07, FAO Uganda + FAO SWALIM; not public)" |
There was a problem hiding this comment.
Privacy: content from a source this line marks "not public" is inlined into the public repo.
extra.fao_uganda_flood_side (lines 63-76) reproduces the draft AAP's substance: the three triggers with thresholds, the districts (Sironko, Manafwa, Mbale, Butaleja, Namisindwa), 100,000 households, and the budget (USD 150k readiness + USD 1,530,000 activation).
docs/PRIVACY.md: "internal-by-default for anything drawn from internal sources. When unsure, classify internal", and its classification explicitly covers "non-public framework docs"; docs/INGESTION.md fixes external-framework pages as web-sourced public. This detail belongs in ds-knowledge-base-internal with an access-gated pointer here (or should wait until the AAP is published) — the public page should carry only what the cited public sources (e.g. the allAfrica story) independently establish.
| 🤖 I can help with this PR if wanted — mention `@kb-steward` with a question and I'll answer, or ask me explicitly to make a change (resolve conflicts, apply review feedback) and I'll push it. Otherwise I'll stay out of the way. |
…blic page
Per review: the page is visibility: public, and extra.fao_uganda_flood_side reproduced an
unpublished draft's scope, triggers, districts, household target and budget — which
docs/PRIVACY.md classifies internal ('non-public framework docs', 'when unsure, classify
internal'). The public page now carries only what the cited public sources establish (the
closed OSRO/UGA/070/BEL project and the Japan-funded OPM/FAO EWS project), plus an
access-gated pointer noting a draft exists and where the detail is held. It gets a framework
page when the AAP is published.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
|
Docs check — this PR touches machinery whose describing doc didn't change:
If the docs genuinely don't need an update (internal refactor, comment fix), ignore this — it's a reminder, not a gate. Capture-as-you-go: the author has the context now; the drift bots only catch it weeks later. |
|
@zackarno you're right, and this was a real breach rather than a borderline call. The page is Fixed in the latest commit. The public page now carries only what the cited public sources independently establish (the closed OSRO/UGA/070/BEL project, and the Japan-funded OPM/FAO early-warning project from the allAfrica story), plus a pointer recording that a draft exists, that it is not published, and where the detail is held. It gets a framework page once the AAP itself is published. I also removed the draft from the Two follow-ups, because the same mistake went further than this page: I'm stripping the same detail from the public |
…blic repo and site The Mt Elgon plan a partner circulated in September is not published, and this repo and its GitHub Pages site are public. Its thresholds, districts, household target and budget were being served on the results and coverage pages, carried in src/frameworks.py, the research notes and a committed figure. Flagged in review on the knowledge-base side (OCHA-DAP/ds-knowledge-base#601) and the same defect applies here, more visibly. Removed from everything published. The framework registry entry is gone, so the plan no longer appears on the coverage map, the harmonisation table or the per-organisation sections. The results page now states only that a partner draft exists, that we backtested it, and that the detail sits with the country team — plus the finding that is ours to make, that a rainfall threshold on these slopes has a low ceiling at any level. The backtest still runs: analysis/fao_elgon_triggers.py reads the parameters from config/partner_triggers.local.json, which is gitignored, and exits with an explanation when it is absent. Its figure is untracked. This also restores the convention already stated on the coverage page, that documents shared with the team are quoted at district level and not republished. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2 participants
The FAO Uganda drought page carried
activations: []and aschema_strainnote saying no dated activation had been found. It has since activated.What changed
schema_strainpoint (3) marked resolved.extra.fao_uganda_flood_side, because anyone searching FAO + Uganda + flood lands on this page: the closed OND-2023 one-off (OSRO/UGA/070/BEL), and its de facto successor, the Japan-funded OPM/FAO project "Enhancing Flood Management in Uganda with Integrated Early Warning Systems" (USD 1.13M, Mar 2025 - Mar 2026, Rwenzori and Mount Elgon, 10 hydro-climatic stations, 2 flood early-warning centres). Neither carries a trigger, so neither is a framework page of its own; the only documented flood trigger in Uganda remains the URCS/IFRC EAP.extra.national_context_2026: the AA Roadmap 2026-2031 and U-MHIEWS launched July 2026 with sub-national early-warning centres in Karamoja, Teso, Mount Elgon and Rwenzori, and the FAO/WFP joint El Nino appeal (June 2026).last_checkedbumped to 2026-09-04; five sources added.check_docs.pyandcheck_links.pyare clean for this page.Found while reviewing external frameworks for the Uganda flood AA trigger design in ocha-dap/ds-aa-uga-flooding, where the FAO/OPM station network is a candidate backstop for the Mount Elgon slopes.
🤖 Generated with Claude Code