The forecast has become the fact. NOAA's Climate Prediction Centre has moved its ENSO alert to an El Niño Advisory, with El Niño already observed in the tropical Pacific and strengthening — not a probability anymore, but a present condition. Forecasters now put the odds of a very strong (“super”) event during the Northern Hemisphere fall and winter above 90%, with some outlooks flagging a chance of the strongest event since records began in 1950.
For the drought-prone peatlands of Southeast Asia, the Amazon, and Southern Africa, the anticipatory window described in “When the forecast is the warning” is no longer a window to plan through — it is the season we are now standing in.
That earlier piece made the case for acting on the forecast: mapping risk, keeping peat wet, training brigades, before the dry season bites. That case still holds, but it is no longer the whole story. Once El Niño-driven drought is established, prevention only gets you so far — some ignitions are inevitable. The question shifts from how we stop this fire from starting to how we stop it from becoming a peat fire. That is a different phase of the same framework, and it runs on a different clock.
2. The IFM wheel has more than one gear
Integrated Fire Management is usually drawn as a cycle: reduction (prevention), readiness (preparedness), response (suppression), and recovery, sometimes with a fifth “review” segment feeding lessons back into the loop. The Working on Fire programme's version of this wheel, referenced in the earlier UN-REDD post, is designed to be read as a whole system rather than a single stage — each segment supports the others, and neglecting one weakens the rest.
Figure 2. The IFM Wheel as an operating model..
How the wheel is applied
- Reduction (50%): peat rewetting, risk reduction, community prevention and alternatives to burning.
- Readiness (30%): forecasting, detection, dispatch, training, equipment, interoperable SOPs and ICS.
- Response (15%): Golden Hour initial attack, coordinated ground operations and needs-based aviation support.
- Recovery (5%): burn-severity assessment, peat hydrology repair, restoration and finance-ready recovery plans.
That allocation is deliberate, not accidental. Reduction and readiness together account for the bulk of the effort — peat rewetting, risk reduction, community-level prevention and alternatives to burning on one side; forecasting, detection, dispatch, training, equipment and interoperable standard operating procedures on the other — because prevention, hydrological resilience, trained institutions and rapid detection are what determine whether the response segment can succeed at all.
Response and recovery account for smaller shares of ongoing effort, but that does not make them optional: they are where the pre-season investment either pays off or doesn't, in the span of a single incident. Response, in particular, is most effective when it is tightly connected to the proactive side of the wheel rather than treated as a stand-alone emergency function — and on peat landscapes, that connection is tested inside the Golden Hour of initial attack.
The first blog sat mostly in reduction and readiness: the pre-season work that a forecast makes possible. With the prevention window closed for this cycle, response is where that preparation gets tested — and on peat landscapes, it is tested within a very narrow margin. It is a matter of one specific hour.
2.1 The Golden Hour and the re-burn phenomenon
Research by Wilson and colleagues (2016), reinforced by operational material from the Working on Fire programme, describes how a peatland fire escalates if it is not brought under control immediately. A surface fire starting with a modest fuel load — on the order of 3 to 5 tonnes per hectare — can begin moving down into the sub-surface peat layer within roughly the first hour. That first hour is the “Golden Hour”: the point at which a response can still keep the fire on the surface, where it is comparatively fast to control and releases comparatively little carbon.
Miss that window, and the trajectory changes sharply. As the peat structure beneath a stand weakens, trees begin to fall, and the fallen biomass can push surface fuel loads past 150 tonnes per hectare — thirty to fifty times the load the fire started with. That timber does not stay wet for long in drought conditions; within roughly 3 to 12 hours it dries enough to re-ignite, and the resulting burn is hot and sustained enough to drive combustion deeper into the peat itself, where fuel stocks are measured in the thousands of tonnes per hectare. From there, peat combustion becomes largely self-sustaining, smouldering on with limited oxygen until the fuel runs out or heavy rain intervenes — which, in an El Niño year, may be a long wait.
Put simply: the same fire, left unattended, moves from a fuel pool of single-digit tonnes to one of thousands of tonnes per hectare over the course of a single day. Emissions, suppression costs, and risk to responders do not rise gradually along that curve — they climb steeply once the fire transitions from surface to subsurface fuel.
Figure 1. Conceptual relationship between response time and cumulative GHG emission potential. The curve is illustrative; it does not report measured tCO2e. Phase timing and fuel-load anchors are adapted from Wilson et al. (2016) and the Working on Fire presentation.
Read against that curve, the interpretation is straightforward: the first hour is the Golden Hour. Delayed response lets a small surface fire access progressively larger pools of biomass and peat carbon, and once it does, emissions, safety risk, suppression effort and cost all rise sharply together — which is exactly why response earns outsized attention in a season like this one, even within a wheel that spends most of its energy on prevention and readiness.
3. Why this is a carbon story, not just a fire story
The earlier post noted that wildfire emissions can undermine forest carbon gains, reduce credit supply, and increase buffer pool requirements. The Golden Hour is where that risk is actually decided. A surface fire suppressed within the first hour draws on a few tonnes of fuel per hectare. The same fire left to move into peat draws on fuel stocks orders of magnitude larger, and — because smouldering peat can burn for weeks — extends the emissions event over a much longer period. For jurisdictions relying on forest-carbon finance, the difference between a same-hour response and a same-day response is not a matter of degree; it is close to the difference between an incident and a reversal event.
This is also why response, in an IFM sense, has to mean more than firefighters arriving. It means detection systems and fuel-moisture monitoring sensitive enough to flag ignition within minutes rather than hours; pre-positioned crews and equipment close enough to reach a hotspot inside sixty minutes; and, underneath all of it, the groundwater management — canal blocking, hydrological restoration, the 40 cm threshold referenced for Riau — that determines whether a given hectare of peat is even capable of sustaining that second, larger burn.
4. The window that matters now
The forecasting window described in the earlier post has done its job: institutions that used it had months of notice. The window that matters for the rest of this El Niño event is much shorter and repeats with every new ignition — the first hour after a fire is detected. Preparedness built during the calm months either buys a fast enough response to hold a fire at the surface, or it doesn't. Integrated Fire Management was never meant to stop at planning; the same coordination, resourcing, and political priority that went into pre-season readiness now has to show up as speed on the day. For peatland regions moving deeper into this event, that measure will determine how much carbon this season actually costs.
This post follows on from “When the forecast is the warning: wildfires, El Niño, and the case for acting early”, UN-REDD Programme, June 2026. Phase timing and fuel-load figures are adapted from Wilson et al. (2016) and Working on Fire programme materials; the response-time/emissions relationship is illustrative and does not report measured tCO2e.
