El Niño forecasts: turn climate probabilities into a useful plan

Start with the dated forecast

The WMO's September 2026 El Niño update reports firmly established El Niño conditions and a very high likelihood of persistence into early 2027. This is the outlook available when this article was written. Readers planning a later event should consult the newest update rather than treating this article as a live forecast.

El Niño refers to a large-scale ocean–atmosphere pattern in the tropical Pacific. Its influence on temperature and rainfall varies across regions and seasons. A strong signal about the pattern does not specify the exact temperature, rainfall, or impact for one town on one day.

Keep three probabilities separate

First, how likely is the large-scale climate pattern? Second, given that pattern and other conditions, how likely is a relevant local weather outcome? Third, if that outcome occurs, how likely is your activity to be disrupted? Collapsing these into one number removes the part of the analysis most connected to your own circumstances.

For example, above-normal seasonal temperature is not a daily heat warning. A rainy season does not tell you which afternoon will be wet. Exposure also matters: two organisations in the same city can face different disruption because one works indoors and the other depends on outdoor equipment.

Ask what 'above normal' means

A forecast category depends on a reference period and a definition. 'Above normal' is a comparison with a climatological baseline, not automatically a statement about an absolute dangerous threshold. Read the legend, the region, the lead time, and the baseline before interpreting the colour on a map.

The base rate experiment develops the habit of keeping the starting frequency in view. It does not model climate. Use its lesson to ask what normally happens in the relevant place and season, then how the forecast changes that distribution. For local warnings and safety instructions, consult the national or local meteorological service.

A worked example: planning an outdoor workshop

Imagine a community organiser deciding whether to reserve an indoor backup for a workshop several months away. All costs below are invented planning units, not a weather forecast or market price. Suppose holding the backup costs 50 units, and a weather-related cancellation would cost 300 units if no alternative exists.

1. Define the disruption: weather that prevents the planned outdoor activity, not simply the presence of El Niño.
2. In a simplified comparison where the backup completely avoids a 300-unit loss, 50 / 300 gives a break-even disruption probability of about 17%. Treat that as a sensitivity calculation, not an estimated local probability.
3. Seek a regional outlook and test several plausible probabilities. Check the backup's limitations, cancellation terms, and the people affected before choosing a plan.

Prefer preparation that helps across scenarios

An action can be useful even if the specific forecast changes. Clarifying decision authority, checking equipment, agreeing communication channels, or choosing a flexible booking may help under several kinds of disruption. The optionality experiment offers a way to compare reversible choices with commitments that leave little room to respond.

Low-regret does not mean free or automatically correct. A backup can consume money, time, or scarce space. Write down both its cost and the scenarios in which it helps. If an option only protects against one narrow event, consider whether a simpler arrangement covers several risks.

Decide when new information should change the plan

Set review dates that match the decision. A seasonal outlook may inform early reservations; a short-range local forecast may inform the final schedule. Waiting for certainty can be costly if alternatives disappear, but committing too early can also waste flexibility. Identify the last useful date for each action.

Try Bayesian updating to practise changing a belief when genuinely new evidence arrives. Then use value of information to ask whether another update could change your choice. Repeatedly checking the same forecast is not necessarily a better preparation process.

• Name the local disruption you are planning for.
• Use the newest regional information and official warnings.
• Compare several plausible scenarios and their consequences.
• Set a review date and a clear trigger for changing the plan.

Use the right source for the right decision

WMO's El Niño and La Niña background explains the broader phenomenon. Local forecasting services supply the location-specific information a real operational decision needs. Neither this article nor Modic's experiments are weather, medical, or emergency-warning tools.

The transferable skill is to move from a headline probability to a defined exposure, a set of options, and an update schedule. You do not need to forecast every detail correctly to make a plan that handles more than one plausible future. You do need to know which assumptions your plan depends on and when those assumptions should be reviewed.