By Petras Anaab Ali (MPhil)

The sun has gone down, but the room still feels hot. The fan is running, yet sleep does not come easily. Outside, people seek relief on verandas, in compound yards and along the roadside. For anyone who has spent such an evening in Accra, the question is simple: why does sunset bring so little comfort?

That question has become more urgent. On 30 October 2026, the Ghana Meteorological Agency (GMet) warned that a very strong and still intensifying El Niño is developing, and that Ghana should prepare for a hotter, drier and dustier dry season. Heat stress is expected to peak between February and April 2027, especially in the north. The ocean data behind that warning are striking. In mid-October, the weekly Niño 3.4 index, the standard measure of El Niño, reached +3.3°C. Scientists classify anything above +2.0°C as “very strong.”

For Accra, the warning raises a practical question: how prepared is a densely built city for additional heat? Answering it means looking at three things together: the global climate, the city we have built, and, less obviously, the whole column of atmosphere above us. My MPhil research in Physics at the University of Ghana followed the last strong El Niño upwards through that column. What it found is that El Niño does not only change the weather at ground level. It disturbs the entire sky.

A distant ocean, a connected atmosphere

El Niño begins with unusually warm surface water in the central and eastern equatorial Pacific. Normally, trade winds push warm water westward towards Asia. During El Niño those winds weaken, and the warm pool spreads east. Warm water heats the air above it, the air rises, and storm clouds form where they normally would not. That shift in where the atmosphere draws its heat rearranges winds and rainfall across the globe. Scientists call these long-distance links teleconnections.

El Niño strengthNiño 3.4 sea surface temperature anomalyWeak+0.5°C to +0.9°CModerate+1.0°C to +1.4°CStrong+1.5°C to +1.9°CVery strong+2.0°C or higher

The present event is far beyond the top of that scale. In September, the US Climate Prediction Center put the chance at 75% that it would exceed every El Niño recorded since 1950. The great events of 1982/83, 1997/98 and 2015/16 are remembered in Ghana for drought and bushfires, and the first two were followed by power rationing.

Two cautions belong here, because they keep the argument honest. First, El Niño changes the odds; it does not dictate the weather of every community. Ghana’s seasons also depend on the tropical Atlantic, and GMet notes that a cool Gulf of Guinea could shape how the 2027 rains behave. Second, the heat people feel today cannot automatically be blamed on a forecast for the coming months. Establishing how unusual Accra’s present heat is requires local observations compared with the same season in earlier years.

The planet is already running hot

According to the US National Oceanic and Atmospheric Administration (NOAA), September 2026 was the second-warmest September on record, with the global surface temperature 1.33°C above average. NOAA’s map for that month shows a bright red tongue of warm water stretching along the equator from South America across the Pacific: El Niño itself. Northern Canada and Siberia are deep red. Cooler-than-normal blue is now confined mostly to Antarctica and a few small patches.

Ghana and the Gulf of Guinea appear in light pink, roughly half a degree to one degree above normal. That should not reassure us. Global temperatures usually peak a few months after El Niño itself peaks, because the heat released by the Pacific takes time to spread through the atmosphere. That is what happened after the 2023/24 El Niño, when 2024 went on to break global temperature records. On the evidence so far, West Africa’s turn is still coming.

Accra already has a heat problem

El Niño will not arrive in a cool city. Local evidence over the past year gives the discussion a firm foundation.

  • Surfaces have heated sharply: A peer-reviewed study of the Greater Accra Metropolitan Area found that urban land surface temperature rose by an average of 4.07°C between 1991 and 2020, as built-up land grew from about 31% to 73% of the area and forest cover fell from about 21% to 11%.
  • Heat is now on the city’s agenda: In October 2025, the Accra Metropolitan Assembly (AMA) trained 30 data collectors to survey households, hawkers and outdoor workers in heat-vulnerable zones including Chorkor, Mamprobi, Korle Gonno, Kaneshie, Makola, Agbogbloshie, Old Fadama and Jamestown, to inform the city’s first operational Heat Action Plan.
  • Vulnerability is rising: In December 2025, AMA reported preliminary findings from its Urban Heat Vulnerability Assessment showing a sharp rise in land surface temperatures over three decades. The lead researcher said Accra is “heating faster than many imagine.”
  • No neighborhood is spared: A study presented in June 2026 found rising heat in low-, middle- and high-income communities alike, from East Legon and Cantonments to Chorkor and Abossey Okai. It found women more affected than men, and, perhaps surprisingly, indoor workers more vulnerable than outdoor workers.

One clarification matters. Land surface temperature is the temperature of roofs, roads and soil as seen by satellites. A 4°C rise in surface temperature does not mean the air everywhere in Accra has warmed by 4°C. But hot surfaces are exactly what keep a city warm after dark, which brings us back to that sleepless evening.

Why sunset brings little relief

The physics begins with the energy received during the day. Sunlight heats roofs, roads and walls. Dark surfaces absorb more of it, and dense materials such as concrete and block store it. After sunset, that stored energy keeps flowing into the surrounding air and into rooms, partly as infrared radiation we cannot see but can certainly feel. A room can stay uncomfortable long after the sun has gone because it is still releasing the day’s heat.

How fast that heat escapes depend on ventilation, shade, roof design and how closely buildings are packed. Iron roofing sheets heat quickly under the sun; poorly ventilated rooms hold heat in. Trees and wetlands work the other way: they shade surfaces and cool the air through evaporation. As Accra has replaced vegetation with concrete, it has removed some of its own natural cooling.

Humidity adds another layer. Our bodies cool mainly by evaporating sweat. In Accra’s humid coastal air, sweat evaporates slowly, so a night at 27°C can feel far more oppressive than the number suggests. A thermometer reading alone cannot describe the experience of heat.

Who can afford relief?

Heat is shared, but the ability to escape it is not. One household can run an air conditioner all night. Another depends on a single fan in a poorly ventilated room. A trader may have to stay at a roadside stall through the hottest hours because leaving means losing income. That helps explain why indoor workers in crowded, unventilated spaces can be more exposed than people working outside.

Accra’s authorities have begun to document this. AMA’s citywide Extreme Heat Perceptions Survey asks residents how heat affects their health, livelihoods, productivity, sleep, transport and daily routines. It complements a heat-mapping exercise in which temperature sensors were installed across the city to identify hotspots. Publishing those results in an accessible form would let residents understand their own exposure and let the public judge whether interventions reach the places of greatest need.

Electricity sits at the centre of this. Every fan and air conditioner draws power, and a hotter dry season will push demand up. The supply side carries its own El Niño risk. GMet says Lake Volta is in good shape for now, but warns that the real risk falls on the 2027 rainy season, which determines how much water will be available for hydropower in 2028. When power fails on a hot night, the fan stops, and the people with the fewest alternatives suffer most.

Following the last El Niño upwards

The heat we feel is only the bottom of the story. For my MPhil thesis, Investigation of the Potential Changes in Ionospheric Patterns Due to Climate Variability (University of Ghana, 2024), I compared the strong La Niña of 2020/21 with the strong El Niño of 2023/24. I used temperature profiles from the COSMIC-2 satellites, covering the atmosphere from the ground to 60 km, and navigation-satellite signals recorded in Bogotá (Colombia), Antananarivo (Madagascar) and Baie-Comeau (Canada). Those signals reveal the Total Electron Content (TEC) of the ionosphere, the electrically charged layer roughly 100 to 1,000 km up that every GPS signal must cross.

To be clear, that work did not measure temperatures in Accra’s streets or homes. But it showed that El Niño leaves a fingerprint at every level of the atmosphere.

LayerWhat was measuredLa Niña (2020/21)El Niño (2023/24)Lower atmosphere (2–5 km)Strength of early-morning temperature inversion5–8 K3–5 KLower atmosphereTime the morning inversion breaks upBaseline30–45 minutes earlierUpper troposphereSize of wave-driven temperature disturbances1–2 K3–4 KStratosphere (20–30 km)Tropical stratospheric temperatureBaselineAbout 2–3 K warmerIonosphereIrregularity in electron content1.7–2.1 TECU2.4–2.8 TECU (25–40% higher)IonosphereNight-time electron contentBaseline30–40% higher

In plain terms: during El Niño, the stable morning layer of air that acts as a lid on the lower atmosphere was weaker and broke up earlier, so convection started sooner and reached higher. That churning launches atmospheric gravity waves, ripples in the air that grow as they climb into thinner air, much as a whip’s crack builds towards its tip. A warmer stratosphere let more of those waves through, and when they reached the ionosphere, they stirred up its charged gas.

The Sun is the ionosphere’s main driver, so the two had to be separated. My La Niña year fell near solar minimum and my El Niño year near solar maximum. After statistically removing solar activity, El Niño still explained an additional 17–27% of the variation in the ionosphere’s electron content. The study was among the first systematic analyses of El Niño’s effect on the African equatorial ionosphere, and the African station responded as strongly as stations elsewhere in the tropics.

The Sun has been active this season too. The provisional Dst index, which tracks geomagnetic storms, recorded six disturbances between September and October 2026. The strongest, on 11 October, reached about −93 nanotesla, a moderate storm just short of the intense threshold. My thesis concluded that El Niño-linked ionospheric variability can rival a moderate geomagnetic storm of that kind. The difference is duration: a solar storm passes in a day or two, while El Niño’s influence persists for months.

Why should anyone in Accra care? Because a disturbed ionosphere bends and delays satellite signals. My analysis suggests El Niño-linked changes can produce position errors of a meter or more, and vertical errors of 2 to 3 meters for single-frequency receivers, the kind found in most everyday devices. That will not stop anyone finding a trotro station, but it matters for satellite-guided aircraft approaches, land surveying, precision farming, drones and radio communications. West Africa sits beneath the equatorial ionization anomaly, one of the most turbulent parts of the ionosphere, where irregularities after sunset are already common. My data showed night-time electron content rising 30 to 40% during El Niño, precisely when those irregularities are worst.

I must add one honest note. None of my stations was in Ghana, because long, continuous, publicly available records of the kind this study needed were not available here. A Ghanaian physicist studying the African sky had to use data from Madagascar.