Oil — from well to every tank

Exploration, extraction, refining, transport, combustion, spills and plastics.

The tank is almost the end of the story

Finding and extracting: from seismic surveys to a producing well

Before a single well is in production, oil companies carry out seismic exploration surveys: they generate sound waves at the surface or from a ship and analyze how they bounce off different underground geological layers, to infer where a pocket of crude might be trapped under impermeable rock. Once the likely existence of a reservoir is confirmed, extraction can happen from onshore wells or from offshore platforms — some at water depths of more than a thousand meters, with steel and concrete structures able to operate continuously for decades in extreme ocean conditions. The crude oil coming out of the well is almost always mixed with water and associated natural gas, which have to be separated before it can be transported.

From crude to products: the refinery's work

Crude oil, as it comes out of the well, is not directly usable for almost any final purpose: it is a complex mixture of thousands of different hydrocarbon compounds. The refinery separates that mixture through distillation — heating the crude and collecting separately the fractions that boil at different temperatures — and then transforms those fractions using processes that add heat, hydrogen and chemical catalysts, turning heavy, less useful molecules into petrol, diesel, aviation kerosene, lubricants and a wide family of feedstocks for the chemical industry. Every tonne of crude entering a refinery comes out converted into dozens of different products, not a single fuel — one of the reasons replacing oil is not as simple as replacing a single end use.

Transport: the phase with the highest spill risk

Between the well and the refinery, and between the refinery and the final point of consumption, oil and its derivatives travel enormous distances by pipeline, rail or oil tanker. Each of these transport modes has its own history of incidents: small, repeated spills during routine loading and unloading, and severe accidents — much less frequent, but with a disproportionate environmental and media impact when they occur — that can contaminate coastal soils, marshes and sensitive marine ecosystems for years. Cleaning up a large spill is never complete or instantaneous: much of the spilled oil persists in sediments for decades.

Burning it is not its only destiny

Unlike coal, whose destiny is almost exclusively combustion, oil branches out into uses that go far beyond generating heat or powering an engine. A significant fraction of refined crude becomes plastics, solvents, synthetic fibers, asphalt and an enormous variety of industrial chemical products that are part of everyday life without being immediately associated with oil. Those products "store" oil's carbon for a while — sometimes years, sometimes only the lifespan of a single-use container — but require their own manufacturing chain, and at the end of their useful life need collection, recycling, incineration or landfill, each with its own environmental balance.

When fuel demand falls, the refinery does not just disappear

A refinery is a massive industrial facility, designed to operate for decades and hard to quickly convert to another use. If demand for oil-derived fuels falls in a sustained way — through transport electrification, for example — the facility does not simply "switch off": it needs an industrial reconversion plan, decontamination of soil where it has operated for decades, and in many cases, a workforce transition for the specialized staff who depended on it. Ignoring this closure cost when comparing oil with other energy sources is as incomplete as ignoring the cost of extracting it.

Reduce before replacing

Oil's energy density and the enormous infrastructure already built around it — refineries, petrol stations, vehicle fleets — explain why it still dominates sectors like long-haul transport and aviation, where alternatives remain more limited or more expensive. Energy prudence here does not mean eliminating oil overnight, but reducing avoidable travel and heat first, electrifying uses where the full life-cycle balance genuinely improves, and reserving oil-derived fuels for uses where no mature alternative yet exists, while that alternative finishes developing.

What makes it hard to replace, and what does not justify not trying

In favor: very high energy density, easy to store and transport, and a worldwide infrastructure already built over more than a century around it.

To watch: spill risk at every transport stage, air pollution from combustion in engines and boilers, and a branching into plastics and chemical products that means "stop burning oil" alone does not solve all of its environmental impact.

Why it is not one fuel but a family of products: deciding to replace oil requires asking which specific use — light transport, heavy transport, aviation, heating, chemical feedstock — because each has alternatives at very different levels of maturity. Electrifying an urban car is far simpler today than electrifying a transatlantic flight or replacing the naphtha used to make plastics.

The cost that rarely enters the fuel price: the expense of cleaning up a large spill, restoring an affected coastal ecosystem, and decontaminating and reconverting a refinery at the end of its life is almost never reflected in the price paid by whoever filled up while the infrastructure was in full use.

The numbers, with their source

Two spills that define the scale of the risk

The Exxon Valdez spill (Alaska, 1989) released about 257,000 barrels (roughly 11 million gallons) of crude into the sea, and remains, decades later, a standard reference for measuring the magnitude of any later spill. The Deepwater Horizon platform spill (Gulf of Mexico, 2010) released, according to estimates from the technical group tasked with measuring the leak rate, about 4.9 million barrels over 87 days — roughly 12 times more oil than the Exxon Valdez, and the largest spill in US history. Put another way: the entire volume of the Exxon Valdez was released, in the Deepwater Horizon case, every few days over the three months it took to control the well.

Sources: Wikipedia — Volume and extent of the Deepwater Horizon oil spill; Wikipedia — Exxon Valdez oil spill.

The comparative life-cycle table

Why big spills, however severe, are not the dominant part of the impact

It is easy to think of oil only through its most publicized accidents, precisely because they are visually striking and concentrated in time and space. But oil's mortality figure (~18 deaths per TWh) is dominated, according to the studies that calculate it, mostly by accumulated air pollution from burning its derivatives in engines and boilers over decades, not by spills — a pattern very similar to coal's. This does not minimize the severe and lasting ecological damage of a spill like Deepwater Horizon, but it does explain why reducing everyday use of oil-derived fuels has, in aggregate public-health terms, more impact than preventing only the next major accident.

What this table is missing

None of the figures on this page include the cost of manufacturing, using and end-of-life managing the plastics and other chemical products derived from oil that are never burned as fuel — a completely different life cycle developed in this platform's "Materials and Waste" section.

Sources

• The IEA maintains a full profile of oil as an energy source.
• UNEP documents the impact of hydrocarbon pollution on oceans and coasts.
• On the scale of major historical spills: Wikipedia — Volume and extent of the Deepwater Horizon oil spill and Wikipedia — Exxon Valdez oil spill.