A Number That Defies Imagination
In 1750, humanity released a negligible quantity of carbon dioxide from burning fossil fuels. By 2023, that figure had climbed to over 37 billion tonnes in a single year. That is not a gradual drift upward. It is one of the most dramatic transformations any single human-driven variable has ever undergone, and it is hiding in plain sight inside a freely available dataset on DataHub.io.
The dataset tracks global CO₂ emissions from fossil fuels going back to 1750, drawing on figures compiled by the Global Carbon Project and related research groups. It breaks emissions down by fuel type, including coal, oil, gas, cement production, and gas flaring. When you plot even a fraction of that history, the shape of the curve tells a story that no summary statistic can fully capture.
Coal Built the Baseline, Oil Changed the Game
For the first century and a half of the industrial era, coal was essentially the only fossil fuel that mattered at scale. Steam engines, iron foundries, and urban heating all ran on it. Emissions rose slowly through the 1800s, then accelerated sharply around the turn of the twentieth century as electrification spread and heavy industry expanded across Europe and North America.
Oil entered the picture meaningfully after the First World War and exploded after the Second. The postwar economic boom, cheap petroleum, and the rise of the automobile sent emissions on a steeper trajectory. Natural gas followed, growing steadily through the latter decades of the century as pipeline infrastructure matured and gas-fired power generation spread globally. By the time you look at the breakdown by fuel type in the twentieth and twenty-first centuries, you are looking at a three-layered system, each layer compounding the one beneath it.
The Surprises Hidden in the Recent Decades
Most people expect the emissions story to be one of relentless, unbroken growth. The data adds important nuance. There are visible dips: the global financial crisis of 2008 to 2009 caused a measurable drop, and the COVID-19 pandemic produced the sharpest single-year decline in recorded history in 2020. But in both cases, emissions bounced back within a year or two, and the long-run trajectory resumed.
What is genuinely surprising is that the rate of growth has slowed in the 2010s compared to the 2000s, when China's rapid industrialisation pushed global figures sharply upward year after year. This is not the same as a peak or a decline, but it does suggest that the energy transition, however incomplete, is beginning to appear in the aggregate numbers. The question the data cannot answer on its own is whether that slowdown is enough, and whether it will persist.
Why This Dataset Matters Beyond the Headlines
Climate policy debates are full of contested claims, and raw data is one of the few things that can cut through the noise. This dataset is particularly valuable because it goes back far enough to establish a genuine baseline, before industrialisation, before mass energy consumption, before the atmospheric accumulation that now drives policy negotiations. That long view matters when evaluating whether any recent trend represents meaningful progress or just a small deviation from a steep climb.
It is also structured in a way that makes it accessible. Researchers can pull figures by fuel type to model substitution effects. Journalists can chart long-run trends without specialised software. Policymakers can benchmark national commitments against global totals. The granularity by fuel type is especially useful because it shows that coal, oil, and gas have different trajectories and will require different policy responses. A carbon tax that works for one may not be sufficient for another.
For anyone who wants to engage seriously with the numbers behind climate change, this is a foundational resource. You can explore the full dataset, download the underlying CSV files, and embed the interactive charts directly at https://datahub.io/core/co2-fossil-global.