Heating Our Planet: The Causes and Effects of Global Warming, Measured Honestly

Introduction

I dived a reef off Indonesia in 2021 and did not go back until 2024. Three years is nothing. Reefs are supposed to outlast the people who visit them. I remember dropping in expecting the same wall of staghorn and table coral I had photographed on the earlier trip, and finding instead something that read, at ten metres, like an overexposed photograph — structure intact, colour gone. Whole colonies stark white. Not dead yet. Just emptied.

That was not a bad week or a bad site. Surveys by the Zoological Survey of India recorded that between April and May 2016, 83.6 per cent of hard corals across the Bay Islands had bleached, with the worst figures — up to 91.5 per cent. When the follow-up surveys came in later that year, mean recovery among the bleached corals was measured at around 23.7 per cent. Roughly three-quarters did not come back.

I have spent the years since trying to understand what put that heat in the water. This is what the measurements actually say.

What is actually heating the planet

The mechanism is not in dispute and has not been for a long time. Carbon dioxide and other greenhouse gases trap outgoing heat, and we have added a great deal of both.

NOAA’s Global Monitoring Laboratory has measured atmospheric CO2 continuously at Mauna Loa since 1958. In July 2026 the monthly mean stood at 429.12 ppm, against 427.87 ppm in July 2025. In May 2025 the seasonal peak passed 430 ppm for the first time in the record, at 430.5 ppm — a jump of 3.6 ppm in a single year. Pre-industrial concentrations sat around 280 ppm. The IPCC’s Sixth Assessment Report concluded that atmospheric CO2 is now higher than at any point in at least two million years.

The temperature response is showing up on schedule. The WMO’s consolidated analysis put 2024 at 1.55°C above the 1850–1900 baseline — the first calendar year past 1.5°C — and 2025 at 1.44°C (±0.13°C). The eleven years from 2015 to 2025 are the eleven warmest on record. NASA’s independent GISTEMP analysis put 2025 at 1.19°C above its own 1951–1980 baseline, with 2023, 2024 and 2025 as the three warmest years in a 146-year record. Different institutions, different baselines, same direction. AR6 attributed a best estimate of 1.07°C of observed warming to human activity, out of roughly 1.09°C observed.

The ice is where the signal is cleanest

Ice does not have opinions. It responds to energy.

NASA’s GRACE and GRACE Follow-On satellites show Greenland shedding on the order of 264 to 270 billion tonnes of ice per year since 2002. In March 2026, NASA and the National Snow and Ice Data Center found Arctic winter sea ice had tied its lowest maximum extent in the satellite record for the second consecutive year — and that today’s Arctic ice is far younger than in the 1980s, with roughly 47 per cent less multi-year ice than in 2005.

Mountain glaciers tell the same story. The World Glacier Monitoring Service recorded 408 gigatonnes of global glacier mass loss in 2025 and 450 gigatonnes in 2024. Since 1975 the cumulative loss is 9,583 ± 1,211 Gt, equivalent to about 26.4 mm of sea level rise, and six of the highest-loss years on record have occurred in the past seven.

Then there is permafrost, which is the part that keeps me up. Northern permafrost soils hold an estimated 1,400 to 1,600 billion tonnes of organic carbon — roughly twice what is currently in the atmosphere. NOAA’s Arctic Report Card has now concluded that the Arctic tundra has flipped from a net carbon sink to a net carbon source, driven by thaw and wildfire. That is not an impact of warming. That is warming feeding itself.

Weather that no longer fits the old averages

For years the honest answer to “was that storm climate change?” was “you can’t attribute single events.” That is no longer true. Attribution science now runs the observed event against modelled worlds with and without human emissions.

World Weather Attribution found that human-driven warming made June 2025’s heat in southeast England roughly 100 times more likely to exceed 32°C on a given day. A two-week Fennoscandian heatwave in July 2025 was found to be about 2°C hotter and around 33 times more likely. These are specific, published, replicable numbers about specific events.

What warm water does to everything downstream

NOAA confirmed a fourth global coral bleaching event in April 2024, and determined it likely ended in mid-2025. Across it, 84 per cent of the world’s reef area experienced bleaching-level heat stress, with mass bleaching documented in at least 83 countries and territories. The comparison is the part that matters: 21 per cent of reefs in the first global event in 1998, 37 per cent in 2010, 68 per cent across 2014–2017, and now 84 per cent.

Species are moving in response. The IPCC’s AR6 Working Group II report finds marine species distributions shifting poleward at an average of around 59 km per decade. Fish do not respect fishing licences or national boundaries. When a stock moves, so does the livelihood attached to it.

The human dimension

The Internal Displacement Monitoring Centre recorded 45.8 million internal displacements triggered by disasters in 2024 — the highest annual figure since it began monitoring in 2008, and more than double the average of the preceding decade. Twenty-nine countries and territories set their own records. India recorded 5.4 million displacements, the highest in twelve years, roughly two-thirds of them from floods. By the end of 2024, 83.4 million people were living in internal displacement worldwide.

A word on terminology, because it matters legally. UNHCR does not endorse “climate refugee.” The 1951 Refugee Convention does not cover people displaced by disaster or environmental change, which means most of the people in the figures above have no refugee status and no attached protection. Calling them refugees is emotionally accurate and legally empty.

What is actually possible

The IPCC’s AR6 Working Group III report is specific: limiting warming to around 1.5°C requires global greenhouse gas emissions to peak before 2025 and fall by at least 43 per cent by 2030 relative to 2019, and about 84 per cent by 2050. We did not clear the first condition on the published schedule, which means the remaining pathways are steeper than the ones we were offered.

That said, the energy data has genuinely turned. Ember’s Global Electricity Review 2026 found that renewables reached 33.8 per cent of global electricity generation in 2025, overtaking coal at 33.0 per cent for the first time in more than a century. Solar alone met 75 per cent of the growth in global electricity demand; solar and wind together met 99 per cent of it. Global fossil generation fell 0.2 per cent — small, but it is the first fall of its kind, and it was driven by reversals in China and India.

Which is where I want to be careful about individual action. Your flights and your consumption are real, but electricity systems, steel, cement, shipping and agriculture are structural. The individual choice that moves the most carbon is not the one you make as a consumer. It is the one you make as a voter, a shareholder, a professional, and a person who decides what is normal in the rooms they are in.

Conclusion

The reef I dived in 2021 is not the reef I dived in 2024. In that decade, atmospheric CO2 passed 430 ppm, the world posted its first calendar year above 1.5°C, the most extensive bleaching event on record ran through 84 per cent of the planet’s reefs, and the Arctic tundra started emitting more carbon than it absorbs. In the same decade, renewables overtook coal and fossil generation fell for the first time. Both of those things are true, and neither cancels the other.

What I have stopped doing is arguing about whether it is happening. The measurements are public, the instruments are calibrated, the institutions are named. What is left is a question of pace — whether the decarbonisation curve that finally bent upward can bend fast enough to matter to the reefs that are still alive.

So the question I would put to anyone reading this is not whether they believe the data. It is narrower and harder: in the specific system you have influence over — your company’s energy contract, your city’s transport plan, your institution’s procurement, your own vote — what is the single decision that would move the most carbon, and what is actually stopping you from making it?