El Niño 2026–27 Watch

Start here

Five things to know

1. It is big and it is now

   A very strong El Niño, likely the strongest on record. Peak October–December 2026, lasting into early 2027.

2. Same event, opposite problems

   Drought in Southern Africa, Central America, Southeast Asia and the Pacific. Floods on the Peru–Ecuador coast and in East Africa.

3. Plans exist, money does not

   Many appeals are 55–80% unfunded. Local groups get money 3–6 months late, through middlemen.

4. The real gap is the last step

   Forecasts are good. Warnings fail to become decisions by a named person in a district or village, in their language, in time.

5. Where we fit

   Facilitation, local-language translation, matchmaking needs with offers, and moving Nordic money to local groups. Norad's new civil society grants open for feedback in December.

What we do next

• Pick one region to focus on.
• Register for the Global Dialogue Platform, 20–22 Oct (free online).
• Send the "7-day introduction" offer to the first 50 local groups and partners.
• Prepare input for Norad's grant consultation (December 2026).

Find your way

What is happening now

A very strong El Niño is under way. It will probably peak October–December 2026 and last into early 2027. It may become the strongest on record.

Same event, opposite problems

A positive Indian Ocean Dipole (~+0.9 °C) is also forecast. It adds to East Africa rain and Australia/Indonesia drought.

The ocean and the air, in numbers RP

• Fastest rise on record. The weekly Niño-3.4 reading hit +3.0 to +3.05 °C by mid-September, climbing faster than in any earlier event. Off the South American coast (Niño 1+2) it stayed above +4 °C, causing marine heatwaves.
• Fuel in the tank. Water 50–150 m down is +8 to +10 °C above normal. That heat keeps the surface warm through the 2026–27 winter.
• The air has flipped too. The Southern Oscillation Index fell to −14.6 in August and −20.1 by mid-September. The Pacific trade winds have weakened or reversed. Scientists call this a collapse of the Walker Circulation.
• Australia confirmed it. The Bureau of Meteorology declared El Niño on 16 June 2026. On 27 September its index read +2.72 °C, the second-highest weekly value since 2008.
• Indian Ocean Dipole: +0.76 °C in late September, peaking October–November. Together with El Niño it makes southern Australia and Southern Africa hotter and drier, and the Horn of Africa wetter.
• New baseline. This event sits on top of record global ocean heat. Past events (1997, 2015) may not predict it well.

The money

Plans exist almost everywhere. Funding them is the bottleneck.

Response plans already short: Somalia 27% funded, South Sudan 34.5%, Sudan 39.6%. Global humanitarian funding fell by ~$5B in 2024.

Who controls the money RP: mostly UN pooled funds, CERF and the Red Cross DREF. Local NGOs usually get money as subcontractors, 3–6 months later, after it passes through UN agencies and international NGOs.

Latin America & Caribbean

Thin evidence for Mexico and Nicaragua. PAHO runs an "El Niño and Health" community of practice for the Americas.

Africa

Not confirmed for 2026: African Risk Capacity drought insurance policies (it paid $62M after the 2024 drought).

RP GHACOF74 (Kigali, Aug 2026) launched ICPAC's first climate-and-health early warning, for extreme heat and disease. SADC's Climate Services Centre warns of below-normal rain in the south.

Asia

Pacific & Timor-Leste

SPREP declared El Niño for the region on 12 June. 15 Pacific Island Countries have declared El Niño status RP. A fuel-price crunch makes water pumping harder.

Forgotten impacts

The second-order effects. These are where small, focused help is rare.

• HEALTH Cholera follows drought (188k cases in southern Africa, 2024). Dengue follows heat and rain (13M cases in the Americas, 2024). Haze from fires (~100k early deaths, Indonesia 2015).
• POWER Hydropower dams run dry. Then clinics, water pumps, cold chains and phone-based warnings fail too.
• PRICES Export bans (India rice 2023), cocoa and coffee spikes, Panama Canal limits.
• PROTECTION Child marriage, school dropout and gender-based violence rise in drought. UN Women issued a gender alert for Asia-Pacific (Sep 2026).
• MENTAL HEALTH Farmer suicide rises with drought months. Youth trauma after floods.
• CONFLICT & MIGRATION New civil conflict risk in the tropics doubles in El Niño years. Dry seasons push migration from Central America.
• FROST RP Dry soil loses heat fast at night. That brings hard ground frost to highland and temperate farms, killing flowering crops before drought is even declared (PNG highlands, Australian cereals).
• HEAT RP ICPAC issued its first extreme-heat early warnings in August 2026 to prevent heat deaths.
• GLOBAL SUPPLY RP Peru's anchoveta fishery collapses in marine heatwaves, cutting global fishmeal and fertiliser. Drought hits coffee, cocoa and sugar. Low water in Panama's Gatun Lake limits ship depth and daily transits, raising freight rates.
• HERDERS RP Drought pushes herders' routes into farmland in the Sahel, Lake Chad and Liptako-Gourma. Competition over water and grazing triggers violence and more displacement.
• INSURANCE Payout triggers can miss real losses. Malawi 2016: the model assumed a maize variety farmers no longer grew.

Lessons from past El Niños

• Warnings were not the problem. Action was. Horn of Africa 2011: months of warning, action only after the rains failed.
• Interpreters decided who acted. 2015–16: groups with someone to turn forecasts into decisions acted early.
• Early pays. FAO 2023–24: about $7 saved per $1 spent early. Cash before a flood beat the same cash after.
• Language matters. After Cyclone Idai, over 40% of affected people could not read the Portuguese warnings.
• 2026: the delay is a choice. RP The climate signal was stable from February 2026. The Red Cross Climate Centre (28 Sep): the forecast is as good as forecasting gets; every day of delay is an institutional choice, not a gap in knowledge.
• False starts kill seed. RP In 2015–16 and 2023–24 thousands of farmers planted after early, isolated rains and lost their seed when the dry spell came back. 2026 advisories in Southern Africa now focus on not planting too early.
• Money leaves too soon. RP Funding surges during relief and disappears in early recovery, leaving communities exposed to the next shock.
• Outsiders can harm. Local groups get under 10% of humanitarian money (3.8% directly). New middlemen, parallel warning channels and careless data use make it worse.

Unknown unknowns RP

Risks that standard El Niño preparedness tends to miss.

1. Record ocean heat underneath

   This El Niño sits on top of record global ocean temperatures from 2025–26. Past events may not predict how the atmosphere responds, so expect compound extremes no model shows.

2. Frost from dry soil

   Assessments watch daytime heat and rain totals. Dry soil loses heat at night, so hard frost can wipe out flowering crops before drought is declared.

3. Debt crisis removes the buffer

   Governments in debt distress cannot pay food, fuel and fertiliser subsidies. Price shocks then land directly on households.

4. AI data centres compete for water

   New energy- and water-hungry data centres near dry cities compete with people for groundwater during drought. Nobody has modelled this.

5. Green grids fail twice

   Drought cuts hydropower while heat and haze cut solar output and power-line capacity, at the same time.

6. Grain export bans

   Geopolitical tension makes sudden bans on rice, wheat or maize more likely at the first sign of a poor harvest, driving fast global price spikes.

7. Climate and conflict feed each other

   In the Sahel, Lake Chad and eastern DR Congo, drought pushes herders into farmland, triggering violence. Armed groups recruit displaced youth.

The biggest need after money: turning warnings into action

The forecasts are good. Most of the deadliest weather disasters this century were forecast in advance. The failure happens in the last step: a warning must become a decision by a named person in a district or village, in time.

Why warnings do not become action

• Fear of acting in vain. Blame for acting on a wrong forecast is bigger than credit for acting early. Horn of Africa 2011: famine warning in March, money in July. Same pattern again in 2022.
• Nobody has the written right to decide. Plans name agencies, not people. A forecast does not legally trigger an order. Reported example: Nepal floods, 26 Aug 2026. The weather office had no authority to order evacuation; community phone calls saved 900 students. (single source, verify)
• Plans stop at national level. Policies exist but local authorities lack capacity to run them (REAP 2023–24).
• Money is released late. Kenya's drought contingency funds are often released after the situation gets worse.
• The forecast is not farm advice. RP "75% chance of below-normal rain" does not tell a farmer when to plant, which seed to choose (short-season, drought-tolerant sorghum) or how to avoid a false start.
• Wrong language, wrong channel. RP Forecasts go out in English, Spanish or French on websites. Remote, indigenous and pastoralist people listen to community radio in local languages.
• Local reality is invisible. RP Big climate models miss frost hollows in valleys, falling groundwater and local grain price spikes. Local actors lack simple tools to link weather, market prices and malnutrition.
• Rigid triggers pay too late. RP Some funds need several weather thresholds confirmed before money moves. By then the dry spell has already damaged young crops.
• Local groups cannot reach the money. RP Trusted community and faith groups fail strict compliance and audit rules for pooled funds.
• The warning does not land. 41% of Cyclone Idai survivors could not read the Portuguese warnings. Standard forecast charts are read correctly only 57% of the time. In Nepal, 51% of women got warnings by word of mouth. 86% of people with disabilities feel left out of disaster planning.

What is unfinished right now (2026)

What works: the same four parts every time

Channels that reach people

Safe model for AI translation: AI drafts, a native speaker checks every message, the community pretests it, and a fixed glossary is agreed before the crisis. A mistranslated warning can kill. See CDAC Network "SAFE AI" (highest risk tier).

Five places to push, ranked

1. Write down who decides

   At district level: who orders action when threshold X is hit.

2. Turn plans into triggered plans

   Add the trigger and the stop rule to existing district plans (Mangochi, Zambia, Madagascar). Cheap and fast. Deadline: October planting.

3. Pre-sign the money route

   Use existing safety nets or county funds with pre-approved release, so no new approval cycle is needed.

4. Accept acting in vain, in advance

   Agree a tolerated false-alarm rate. List no-regret actions. Protect officials from blame.

5. Rehearse the decision, not only the evacuation

   Tabletop drills where officials get a simulated threshold breach and must issue orders.

A 90-minute workshop we can offer

"From El Niño warning to our decisions." A local co-facilitator leads. We handle forecast translation, timing and notes. Local language. No per diem beyond real costs.

Within 48 hours: one-page advisory and a voice note in the local language, checked by the local lead. Other ready-made methods: PICSA, FAO simulation toolkit, GNDR locally led guide, Climate Centre games.

Pitfalls to avoid

• Big per diems attract the wrong people and "workshops for the allowance".
• Workshop fatigue is often a signal of unequal partnership, not apathy.
• Token participation that ignores local and indigenous knowledge.
• Never run a session without a named decision owner or funder for the actions it produces. Give results back within 2 weeks.
• Using indigenous knowledge without free, prior and informed consent.

How our group can add value

1. Forecast-to-action interpreter

   Turn WMO, FEWS NET and regional outlooks into one-page briefs per sector or district. Include the forgotten risks. Coordinate with the national weather service.

2. Facilitated pre-mortems

   Run "It's March 2027 and we failed. Why?" sessions with local NGOs and consortia. Agree triggers and no-regret actions. Fits our facilitation skills directly.

3. Move money to local actors

   Steer Nordic philanthropy and diaspora giving into Start Network funds, local consortia or early cash. Zero overhead.

4. Language and AI, with humans in the loop

   Help local translators make voice notes and radio scripts in minority languages. Partner with CLEAR Global. Follow the Signal Code on data.

5. Forgotten sectors

   Mental health, gender-based violence, school dropout, clinic power and water backup. Connect specialists to Anticipation Hub working groups.

6. Open map data

   Organise mapping sprints for hotspots (Malawi, Mozambique, Madagascar) with the Humanitarian OpenStreetMap Team.

7. Real-time learning log

   Track "warning given vs action taken" for the 2027 lessons review. Monitoring and learning capacity is scarce.

8. Norway as a bridge

   Brief Norad, NORCAP and Norwegian NGOs on locally reported gaps. Push for flexible funding agreed before the shock.

When you reach out, lead with a specific offer ("we can run 3 pre-mortem workshops"), not "how can we help?".

Five pathways from the Research Plan RP

Rules for outsiders (Grand Bargain localisation, do no harm): no parallel coordination structures, no poaching experienced local staff, no proprietary software that locks people in, and no warnings without local validation.

The coordination map idea

The idea: one open map where anyone adds plans, live signals, resources and offers, finds each other, gets matched, and searches with AI.

What already exists

Why platforms like this usually stall

• Data goes stale. Even HDX's curated crisis data lost 6 points of freshness in one year. A volunteer-filled map decays faster.
• Pilots never become real use. Most Ushahidi deployments were tests or training.
• Parallel tools get ignored. Haiti 2010: volunteer data flooded in, but responders could not use it because it ran outside the official system.
• Community-owned lasts, outsider-built fades. Start Network's CLIP programme: 74% of community-led solutions reached a scaling path.
• Local groups value contacts most. A Start Network hub leader named "the ease of getting in touch with other organisations" as the key value.

The first thing to test: a matchmaking service done by people

A local group posts one need, for example "drought cash design help, Malawi, Chichewa". Within 7 days we introduce them to someone who can help. Success = a call actually happens.

1. Week 1Pick one region (Southern Africa drought or Central America Dry Corridor). Contact 30 local groups through Start Network hubs and NEAR, plus 20 people offering help. Collect needs and offers with a simple form. Register for the Berlin Dialogue Platform.
2. Week 2Put them in a spreadsheet as "need / offer cards". Make 10 introductions by hand (email, WhatsApp). Publish a simple map at province level only.
3. Week 3Follow up. Count calls held and next steps agreed. Test AI search over the cards plus Anticipation Hub documents, inside the test group only.
4. Week 4Interview 10 participants. Continue only if 5 of 10 introductions led to a call and 3 local groups want to renew their card. Take results to the Anticipation Hub and Norad as a partnership request.

Stop rule: if nobody renews their card after 60 days, stop. Cards expire after 60 days unless renewed; this keeps the data fresh.

Do-no-harm rules for any map

• No individuals. No locations more precise than district level.
• No protection, gender-violence, ethnic or displacement data.
• Organisations opt in. Contact goes through a relay, not public emails. Clear takedown route.
• AI searches public documents only, never personal data.
• Use existing standards (HXL, IATI, OpenStreetMap boundaries) and link to existing tools instead of copying them.

Guides: ICRC data protection handbook, Signal Code (Harvard Humanitarian Initiative), OCHA data responsibility guidelines.

Open map & agent layer

The idea: do not own one central map. Open-source a thin layer where anyone plugs in their own maps, data APIs and AI agents.

Map projects active now

Not verified: any 2026 El Niño map portals from Peru, Philippines or Kenya governments; FEWS NET's current status.

Standards that already connect maps

Existing AI connectors (MCP servers) for HDX, ReliefWeb and OpenStreetMap are all early 0.x versions by individual developers. None is official. No connector for STAC or OGC API was found, which is a small, real gap.

Lessons from earlier open projects

• OpenStreetMap works because contributors are spread out but the data model is one shared database.
• HDX works because OCHA curates it and pays a team. It is a central catalogue, not true federation.
• GeoNode federation ended as islands of agency portals with patchy upkeep.
• Sahana faded once volunteer maintenance stopped. Ushahidi survived with a hosted service and grants.
• Open Referral shows a standard plus a community forum can last. The real cost is keeping records fresh.
• A layer built only on open standards could register as a Digital Public Good, which helps UN and donor uptake.

Risks

• Nobody owns accuracy. Every item needs source, date and an expiry. Stale offers are worse than none.
• Prompt injection. Any contributed text can hide instructions to AI agents. Treat all incoming data as untrusted. Agents only read; they never act on it.
• Sensitive locations. Needs data can expose vulnerable people. Publish needs only at district level or coarser.
• Maintenance. Every connector breaks when the other side changes its API. A small team cannot maintain many.

A minimal first version: write little, reuse most

Order: run the 4-week matchmaking test by hand first (see Coordination map). Write the code only if people renew their cards.

Existing project: OSIRIS

Best match found for "Osiris, live location". Not confirmed as the project the group meant.

Useful: the hazard layers (fire, weather, earthquakes) and the live push-and-stream design. Its map code could be reused, since the licence is open.

Caution: it is built as a surveillance tool (CCTV, network scanning, data-breach lookups). Linking a humanitarian project to it is a reputation and do-no-harm risk. Its API is open with no login. Reuse ideas or map code, keep our data in our own open format, and never send people's data to it. Also check first: its GitHub description seems to end with a crypto token address (unverified). A token link would be a reputation risk.

Other projects with the same name: asuramaya/osiris, an MCP memory server for AI agents with an append-only log. It is a small one-person project, but a useful pattern for the shared ledger.

Talk to first

1. OCHA Centre for Humanitarian Data

   Its data science team already tracks anticipatory action and writes a knowledge base for AI. Ask: can a needs/offers profile sit on HDX? Data responsibility rules.

2. Open Referral

   Does a disaster-preparedness profile or "needs" extension already exist? Would they host one?

3. HOT regional hubs

   LATAM and East & Southern Africa teams: local partners, ChatMap community data.

4. Anticipation Hub + IFRC GO / Alert Hub

   Warning feeds and the Red Cross societies that hold real needs data.

5. Digital Public Goods Alliance

   Early check on eligibility. Also contact the developers of the existing HDX, ReliefWeb and OSM connectors to work together.

Idea: an AI plugin for a shared index

A plugin people add to the AI assistant they already use (Claude, ChatGPT, Gemini and others). Through it they add their needs and offers to one shared, public index, and find each other.

How it works for a user

1. Add"Add our offer: we run drought cash workshops in Chichewa, Malawi, until March." The AI drafts a card. The user checks it and confirms. The card gets a source, a district-level place and a 60-day expiry.
2. Find"Who helps with seed distribution in southern Zimbabwe?" The AI searches the index and the open map sources, and shows matches with dates and sources.
3. Connect"Introduce us." The index sends an intro request through a relay. Contact details are shared only if both sides say yes.
4. RenewAfter 60 days the AI asks: "Is this still true?" No answer means the card expires.

What the AI search finds

Every answer shows its source and date. Signals are linked from the official source, never rewritten.

Live location: four levels

Level 4 is off by default and never available in conflict areas. Live location of people in a crisis can get them hurt. Every level above 1 needs a named data owner.

The shared index ("ledger")

• Cards, not profiles. One need or one offer per card, using the Open Referral fields plus the small "need" format above.
• Public history. Every change is kept and visible: who added what, when. A plain database with an append-only log or a public git repository does this. No blockchain needed.
• Open data, open code. Anyone can copy the index or run their own plugin against it.

Risks and the fix for each

What we would build

• One MCP server with five tools: add_card, search, match, request_intro, renew.
• One small database with full change history.
• Listing in the Claude and ChatGPT plugin directories; a config snippet for other tools.

Test first without code: a shared sheet of cards plus a ChatGPT custom GPT or Claude Project that reads it. Give it to the 10 people in the 4-week test. Build the real plugin only if they use it.

Build ideas: one backend, many doors

We build one shared backend. Every plugin and bot is only a door into it. People use whatever they already have.

The backend (built once)

• Shared index: needs and offers cards with full change history, 60-day expiry, district-level places.
• MCP server: tools add_card, search, match, request_intro, renew, plus read-only access to the open map sources.
• Matching: plain rules (place + type + time). AI only explains.
• Intro relay: passes contact only when both sides say yes.

The doors

Order: by-hand test first, then MCP server, then ChatGPT and Claude doors (cheap), then WhatsApp (reaches the most local people, costs more).

Idea: a task adapter

Every group already has its own to-do tool. Nobody will move to ours. So the system syncs tasks into the tools people already use.

• One task card format: title, owner, due date, status, trigger, funding source, place (district level), link back to the need or offer card and the source meeting.
• One small adapter per tool, four functions: pull (read tasks), push (create or update), map_status (their "Done" = our "done"), on_change (webhook when something changes).
• Clear rule for conflicts: each field has one home. Status lives with the task owner's tool; trigger and funding live in our index. No silent overwrites.
• Same MCP server: add list_tasks, create_task, update_task, so AI assistants can read and update tasks too, always with the person confirming.

Risks: duplicate tasks and sync loops (every task carries one ID across tools), wrong permissions, personal data in task text. Start with two tools only: GitHub Issues plus one spreadsheet. Add more when a real group asks.

A shared GitHub repository: super-el-nino

Created (private for now) and connected to Vercel: every push updates super-el-nino.vercel.app. Next: invite Artem and build in the open. Suggested layout as it grows:

super-el-nino/
  README.md          what, why, how to join
  docs/              this overview, research notes
  schema/            need and offer card formats (Open Referral based)
  data/sources.json  catalogue of open map and data sources
  server/            shared index + MCP server
  doors/
    chatgpt/
    claude/
    whatsapp/
  adapters/          task sync: github, sheets, trello, ...
  LICENSE            Apache-2.0 for code, CC-BY for data
  SECURITY.md        how to report problems; data rules

Start with only README, docs and schema. Add folders when there is code to put in them.

Open-source code to build on

GitHub repositories found in this research. Reuse these before writing anything new.

Licences: check each repository before reuse. AGPL code requires sharing changes if we run it as a service.

Burners

The Burning Man community has disaster experience, off-grid skills and a global network. Its relief arm already works the way the research recommends: support local people, do not parachute in.

Burners Without Borders (BWB)

• Since 2005. Started with Hurricane Katrina. Then Pisco, Peru 2007 (earthquake, 154 volunteers from 23 nations), Fukushima, Sandy, Puerto Rico solar after Maria, Ukraine, Acapulco, Maui, Helene, LA fires.
• New model (Dec 2025): no more own teams. Support "people and organizations already active in their communities".
• Small: 2025: 47 active groups, 900+ volunteers, 13 grants totalling $14,000 in 10 countries. Do not expect money.
• No El Niño or anticipatory work found. This is a gap our group could fill.

Where burners already are

Burner skills that fit El Niño

Five ways to connect burners

1. Propose an "El Niño Ready" project to BWB

   Tools and training for chapters in Mexico, Colombia, South Africa, Australia, plus Peru and the Philippines. Use the create-a-project form; apply for a Civic Ignition grant.

2. Off-grid input for the map

   Meshtastic / Burning Mesh as a way to send needs and check-ins with no phone network. Hardware workshops at regional burns.

3. Reuse burner tools

   Ranger IMS and Clubhouse for incidents and rosters; cobudget for micro-funding local groups.

4. A "Dream" at The Borderland 2027

   Closest big burn to Norway. Recruit Nordic burners for remote mapping and needs/offers matching shifts.

5. An online session for regional contacts

   Teach anticipatory action through BWB's calls. Pair each willing regional with a local NGO.

Contacts

• BWB team (Jaymie "Sticky" Braun, Christopher Breedlove): bwb@burningman.org
• BWB Colombia: colombia@burnerswithoutborders.org · BWB Mexico: Roberto Cerda Zuniga, contact@restorecoral.org
• BWB South Africa / AfrikaBurn: Gerhard Gibbons, gerhard@afrikaburn.com
• The Borderland: talk.theborderland.se
• Burner Disaster Response (Andy Owens, founded after Helene 2024): burnerdisasterresponse.org

Risks

• "Disaster tourism" and the festival image. Burning Man's rich-elite reputation can hurt trust.
• Radical self-expression does not fit a crisis zone. Local hosts set the terms.
• Do not use BWB or Burning Man names without permission; BWB disclaims projects it does not run.
• Volunteers who fly in need food and housing too. Prefer remote work and local burners.

Not verified: Norwegian burner groups, Australian burner bushfire response 2019–20, status of European Disaster Volunteers.

Meetings

Records of the Super El Niño Response group's calls, newest first.

Who to talk to

First ten conversations

1. Anticipation Hub

   Front door to the whole field. Ask which working group needs help. info@anticipation-hub.org

2. NORCAP (NRC, Oslo)

   Climate-services experts at WMO Africa. Close to home.

3. Red Cross Red Crescent Climate Centre

   Forecast interpretation; 14 African early action protocols.

4. Start Network

   Fast pooled funds for local NGOs. National hubs in Indonesia (YEU) and Philippines (CDP).

5. CLEAR Global

   Language and early-warning partnerships.

6. Norad, humanitarian section

   Gave NOK 30M in July. Ask about remaining reserve and gaps.

7. Humanitarian OpenStreetMap Team

   Eastern & Southern Africa hub (OMHESA) for mapping priorities.

8. Norwegian Church Aid

   Local partners in Malawi, Zambia, Zimbabwe. Water and gender-violence work.

9. CICERO

   Climate-services research in Malawi and Tanzania. Possible evaluation partner.

10. REAP

    100+ partner network, open to new members. Convening at COP31.

Norwegian money and partners RP

• Norad gave NOK 30M to El Niño early action (3 Jul), on top of Norway's ~NOK 6bn humanitarian budget for 2026.
• Norwegian Church Aid stepped up El Niño preparedness in late September across East and Southern Africa: early warning, stockpiles, water and food security, with local faith actors. NCA news
• NORCAP places climate and protection experts in UN agencies. Norwegian Red Cross supports emergency funds. met.no and CICERO support climate modelling and forecasting.

More contacts from the Research Plan RP

On the ground, by region

Dates that matter

1. 8 Oct 2026Next NOAA El Niño update.
2. October 2026Seed and inputs must reach southern African farmers before planting.
3. 6–8 Oct 20261st National Anticipatory Action Dialogue, Madagascar.
4. 20–22 Oct 2026Global Dialogue Platform on Anticipatory Action, Berlin + online. Best single place to meet most contacts above. Register now.
5. 28–30 Oct 20262nd Eastern Africa Dialogue Platform, Mombasa.
6. Oct–Dec 2026Expected El Niño peak. East Africa short rains, Peru/Ecuador coastal rain begins.
7. 9–20 Nov 2026COP31, Antalya. Australia hosts key negotiation tracks. Loss and damage finance, adaptation money, Early Warnings for All review RP.
8. December 2026Norad open consultation on the new civil society grant scheme (Oslo + online) RP.
9. 17–19 Nov 2026Asia-Pacific Dialogue Platform, Bangkok (listed, not confirmed elsewhere).
10. Nov 2026 – May 2027Mekong salt-water intrusion season.
11. Jan–Feb 2027GHACOF 75 and SARCOF mid-season reviews: East and Southern Africa forecasts updated, early-action plans adjusted RP. Next moment to repeat a workshop.
12. Early 2027Norad call for proposals under the new civil society scheme RP.
13. Feb–Mar 2027Malawi's World Bank money pays out (after impact).
14. March 2027FAO–WFP appeal period ends. Lessons-learned window opens.

Deep research prompt

Paste this into a deep research tool to go further. Fill in the date and the optional line first.

You are a senior humanitarian and climate-risk researcher. Today is [DATE]. Research the 2026–27 El Niño. Use only sources from 2026 for "current" claims. Mark anything older as [OLDER].

## Who we are
We are a small group based in Norway. Our skills: tech and AI, facilitation, community building, regenerative business. We want to support communities and organisations that are preparing for El Niño. We are outsiders. We must not duplicate or bypass local actors.
[Optional: our focus region is ___. Our existing contacts are ___. Our budget/time is ___.]

## Questions
1. Forecast: What do NOAA CPC, WMO, BoM and regional climate centres say now? Strength, peak timing, end date, Indian Ocean Dipole. Give bulletin dates.
2. Impacts by region: Which countries face drought, floods, fire/haze, heat, disease? Which are hit hardest relative to their capacity to cope?
3. Who is preparing, per region (Latin America, Africa, Asia, Pacific):
   - Governments: agencies, task forces, emergency decrees, budgets.
   - UN and multilaterals: OCHA/CERF, FAO, WFP, UNICEF, WHO, IFRC, World Bank, regional bodies.
   - Civil society: local NGOs, community groups, indigenous groups, farmer and fisher cooperatives, faith networks, women's networks, community radio.
4. Money: All appeals and pre-arranged funds. Amount asked, amount funded, gap. Who decides how it is spent?
5. Gaps: What do LOCAL actors say is missing? Last-mile warnings, translation, funding access, data, coordination, drought protocols.
6. Second-order effects people forget: health (dengue, cholera, malaria), hydropower and electricity cuts, food prices and export bans, coffee/cocoa, fisheries, shipping (Panama Canal), school dropout, gender-based violence, migration, conflict, mental health.
7. Lessons: What went wrong in 1997–98, 2015–16 and 2023–24? Where did early warning fail to become early action? Cite evaluations.
8. Value for outsiders: Where have small outside groups added real value? Where did they cause harm? Cover localisation and do-no-harm principles.
9. Nordic links: Norad, Norwegian MFA, NORCAP/NRC, Norwegian Red Cross, Norwegian Church Aid, CARE Norge, CICERO, met.no, Nordic foundations. What are they funding or doing for El Niño 2026–27?
10. Upcoming events Oct 2026–Mar 2027: webinars, forums, working groups, conferences (e.g. COP31, regional climate outlook forums, Anticipation Hub sessions).

## Output format
- Executive summary: 10 bullets max.
- One table per region: Country | Hazard | Key government actor | Key local/civil actors | Funding status | Gaps | Source + date.
- Ranked list: 5–8 concrete ways our group can add value. For each: who benefits, what we do, what it costs us, risk of harm.
- Ranked list: first 15 organisations or people to contact. For each: why them, entry point (URL, network, event), how to approach.
- "Unknown unknowns": 5–10 things we did not ask but should know.
- Every claim has a source URL and a date. State clearly when you cannot verify something.

Videos

Explainers and updates worth watching. Summaries are ours; transcripts are from YouTube captions via Apify.

Sources and limits

Compiled 1 Oct 2026 from 2026 sources: NOAA CPC (10 Sep), WMO (3 Sep), ICPAC GHACOF-74, SARCOF-33, OCHA ROSEA alert, FAO–WFP appeal, UNICEF, IFRC / Climate Centre (28 Sep), Anticipation Hub, Norad, national agencies and press in Spanish, Portuguese, French and English. Some pages (OCHA, BoM) blocked direct access, so those figures come from press summaries. Ecuador's $649M plan and Peru's S/8.2bn bill are reported, not confirmed. Treat figures as leads to verify before quoting.

Research Plan sources RP

Items tagged RP come from the "Global Climate Risk Assessment and Operational Response Framework for the 2026–2027 El Niño Event" (the group's El Niño Research Plan, Oct 2026). Its 59 cited sources are below. A few figures were cut off in the file (a Guatemala commitment, one UN appeal amount) and are left out. Figures not cross-checked.

1. Climate Prediction Center: ENSO Diagnostic Discussion - NOAA
2. ENSO Forecast
3. Southern hemisphere monitoring - BoM
4. Season of extremes: Why India and the world must take the 2026 El Nino seriously
5. UNDP calls for urgent anticipatory investment as El Niño worsens an
6. Emerging 2026 El Niño: Potential Implications for Agriculture, Food
7. El Niño impacts Greater Horn of Africa
8. The 2026/27 rainfall season is predicted to be *below normal with
9. Zimbabwe - Key Message Update: Forecast El Niño-related below
10. Southern Africa prepares for Super El Niño as erratic rainfall looms
11. SADC Outlook Favours Drier Conditions Across Much of Southern
12. Latin America & The Caribbean Weekly Situation Update as of 25
13. Latin America & The Caribbean Weekly Situation Update as of 18
14. IOM Appeals for USD 110 Million to Protect 4.9 Million People
15. El Niño FAO–WFP Joint Anticipatory Action Appeal June 2026
16. UNICEF Latin America and Caribbean Region Humanitarian
17. El Niño 2026-2027: Climate Risk & Humanitarian Outlook for Latin America and the Caribbean - As of September 2026
18. The grant scheme for Civil Society Strengthening is changing
19. AU: As El Niño strengthens Australian farmers could face fierce frosts
20. Honduras: Drought - May 2026 Updates - ReliefWeb
21. Global public health situation analysis El Niño 2026 - PreventionWeb
22. As a Super El Niño Builds, ICPAC Launches Its First Climate and
23. Preventing the worst impacts of El Niño | Norad.no
24. Climate Prediction Center: ENSO Diagnostic Discussion - NOAA
25. 2026–2027 El Niño event - Wikipedia
26. El Niño/La Niña Update (August 2026)
27. BoM forecasts strongest El Niño on record as research points to
28. Global Seasonal Climate Update for August-September-October 2026
29. El Niño 2026: what the evidence says, and what it means for Australia
30. El Niño–Southern Oscillation (ENSO) | State Seasonal Update
31. El Niño: what it means for Australia's climate - BoM
32. Regional Climate - El Niño / La Niña - ASMC
33. UNICEF Latin America and Caribbean Region Humanitarian Situation Report No.1 (El Niño Multi-Country Preparedness and Response), 11 June - 18 September 2026
34. UN Office for the Coordination of Humanitarian Affairs (OCHA)
35. El Niño 2026–27: no excuse for inaction
36. High-Level Ministerial Forum on El Niño Preparedness in the Horn
37. With El Niño predicted for the 2026/2027 agricultural season, can
38. FAO HIGHLIGHTS ZAMBIA'S CLIMATE RESILIENCE EFFORTS - nbc
39. El Nino rises: India's post-monsoon outlook turns dry, below-normal rain ahead
40. Anticipation Hub: Home
41. How Norway Will Allocate Humanitarian Funding in 2026 - Norad.no
42. Norwegian Church Aid Strengthens Preparedness for El Niño
43. Climate, food, environment and energy | Norad.no
44. Potentially historic El Niño to come, analysis shows humanitarian toll
45. El Niño 2026–2027: Preparedness Resources for Asia Pacific
46. Forebygger skadeomfanget til El Niño - Norad.no
47. Call for speakers for the 14th Global Dialogue Platform
48. 14th Global Dialogue Platform on Anticipatory Humanitarian Action
49. Dialogue Platforms on anticipatory action - Anticipation Hub
50. COP31 – UN Climate Change Conference - SER-Europe Chapter
51. 2026 UN Climate Change Conference (UNFCCC COP 31)
52. 2026 United Nations Climate Change Conference - Wikipedia
53. COP31 - Antalya, Türkiye, 9-20 November 2026
54. December (OND) 2026 season as El Niño strengthens - ICPAC
55. 14th Global Dialogue Platform on Anticipatory Humanitarian Action
56. Grants Portal - Ministry of Foreign Affairs
57. El Niño 2026: Anticipatory Action in Latin America and the Caribbean
58. Global Dialogue Platform on Anticipatory Humanitarian Action
59. The record-breaking summer of 2026 also reached the ocean

More research

• SENRI wiki: Understand El Niño. The Super El Niño Response group's own wiki.
• Super El Niño site (ChatGPT). Companion site.
• Gemini deep research (shared conversation). Added by the group. Not reviewed in this overview; the link could not be opened automatically.
• RSD framework space (Sutra). Group discussion space where these links were shared.
• NewStories.org. Bob Stilger's project on community-led renewal. Relevant to facilitation and locally led recovery.
• A Paradise Built in Hell (Rebecca Solnit, 2009). How ordinary people self-organise and help each other after disasters. Background for the mutual-aid and "find the others" ideas.
• El Niño Ready Nations. Framework for countries to become El Niño-ready.
• Mickey Glantz. El Niño and society researcher; argues forecasts must fit what communities can act on.
• Weather-Ready Nation Ambassadors. NOAA programme for organisations.
• European Movement Strategy Conference report. Plans for a rapid response network to climate disasters.
• Guardian: El Niño bursts through temperature record (21 Sep 2026) · Guardian explainer (26 Sep 2026).
• Planetary Health Check.
• This El Niño Is About To Cause a Global Catastrophe (video, The World, The Universe And Us).
• ClimateWatch: El Niño strengthens, October update (video, WeatherWatchTV).
• Why El Niño 2026 has scientists worried (video, BBC News · The Global Story). Summaries of all three videos are in the Videos section.
• Indy Johar: How do we prepare the world for what's coming? · conversation with Sarah Wilson · The Collapse of Self.
• Collective intelligence example: crowd-built clue map and open data from a Norwegian mass search game.
• Mutual Aid LA Network. A connector and information hub for mutual aid groups across Los Angeles. A working example of the needs/offers matchmaking idea at city scale.