Global crude storage exists for exactly the situation the market is in right now: since a US-Israel military campaign against Iran opened in late February 2026, the Strait of Hormuz — the transit point for roughly a quarter of the world's seaborne crude — has been under sustained attack, and the world's storage and shipping system has been absorbing the shock in real time. This piece lays out where crude actually sits when it isn't moving — onshore tanks and caverns, strategic reserves, and floating storage at sea — and what the 2026 disruption has actually done to each of those pools, as distinct from the steady-state numbers usually quoted for them.
Energy · Global Oil Markets · Strategic Reserves
Where the World's Oil Sits When It Isn't Moving: Storage, Floating Barrels, and the Hormuz Shock of 2026
1. Why this isn't a steady-state moment
Most explainers on global oil storage describe a stable system: roughly 25% of the world's seaborne crude and products trade normally transits the Strait of Hormuz, floating storage sits in a fairly narrow band of around 80–120 million barrels worldwide in calm markets, and strategic reserves sit largely untouched, existing as insurance rather than as working inventory. None of that describes August 2026. A US-Israel military campaign against Iran that opened in late February 2026 triggered Iranian attacks on shipping in and around the Strait; a mid-year truce broke down, Iranian forces resumed attacking vessels, and the United States reinstated a blockade on Iranian ports. As of early August 2026, diplomatic efforts to restore normal transit had stalled, with Iran and the US trading incompatible conditions for reopening the Strait.
This matters for a piece on "where oil sits when it isn't moving," because the crisis has visibly reshaped every pool of storage discussed below: strategic reserves are being drawn down and rebuilt at different rates depending on how exposed each country is; floating storage has become a genuine holding pattern for cargoes that can't reach buyers, not just routine transit-time inventory; and the gap between authorised storage capacity and what's actually sitting in the tank has widened sharply in at least one major country. Numbers quoted below are dated as precisely as the sourcing allows, because in a fast-moving disruption a figure that was accurate in March is not necessarily accurate in August.
2. Onshore capacity: the difference between "capacity" and "what's actually in the tank"
Onshore storage is the largest single pool of crude that isn't in transit, and it splits into two very different categories that get conflated in casual reporting: strategic reserves, held by governments as an emergency buffer, and commercial inventory, held by refiners and traders as ordinary working stock. The distinction matters because strategic reserves are drawn down deliberately as policy, while commercial stock moves with day-to-day supply and demand — and reporting on "how much oil a country has" routinely blurs the two, or quotes a country's authorised tank capacity as if it were the current fill level.
The clearest example of that confusion, live right now: the US Strategic Petroleum Reserve has an authorised capacity of 714 million barrels across its Gulf Coast salt-cavern sites — a figure widely (and correctly) cited as the SPR's size. But the actual level in the reserve fell below 300 million barrels in early August 2026, the lowest since 1983, after a 172-million-barrel emergency release authorised in March in response to the Iran conflict and the Hormuz disruption. That leaves the SPR at roughly 42% of its authorised capacity — a materially different fact from "the US holds 714 million barrels," and the two numbers are not interchangeable. Refill activity was slated to begin in November 2026 via exchange mechanisms rather than outright purchases.
| Country | Reported level / stock (M bbl) | Authorised or estimated capacity (M bbl) | As of |
|---|---|---|---|
| United States (SPR) | 298.7 | 714 | 10 Aug 2026 |
| China (govt + commercial, all estimates) | ~1,400 (govt ~360 + commercial ~1,000) | >1,800 total tank capacity (2024 estimate) | Dec 2025 stock / mid-2024 capacity |
| Japan (govt + private) | ~470 combined, historically cited as 220–250+ days of import cover | Widely-cited baseline, not independently re-verified for this piece | |
| India (strategic underground only) | 39 designed capacity across Mangaluru, Visakhapatnam, Padur (Phase I) | Standing figure, ISPRL | |
3. Floating storage: oil that has nowhere to go
Floating storage means crude or product sitting on a tanker that has been stationary, or moving without a fixed destination, for an extended period — commonly defined as seven days or more idle. In a functioning market it is a small, fairly stable pool, typically estimated in the range of 80–120 million barrels worldwide, and it reflects ordinary frictions: ships waiting for berth space, cargoes in transit longer than usual, or traders holding oil for a better price. A standard Very Large Crude Carrier (VLCC) holds around 2 million barrels, so the global floating-storage pool at any given time is on the order of a few dozen VLCC-equivalents spread across the world's shipping lanes.
What is happening in mid-2026 is a different phenomenon: floating storage as a forced holding pattern for oil that has a buyer but cannot currently reach one. Vortexa's tracking, cited in early-August 2026 reporting, put Iranian crude specifically held in floating storage at roughly 135 million barrels — up 14% over the preceding month — with a growing flotilla of laden Iranian tankers idling off the country's own coastline, and additional clusters reported off Malaysia and near Sri Lanka. That figure is Iran-specific, not a global total, and it is larger on its own than the entire global floating-storage pool usually cited for calm markets, which is the clearest single indicator of how abnormal the current situation is. The mechanism reported behind it is a US blockade aimed at Iranian ports, which restricts not just outbound crude exports but the return of empty vessels needed for the next loading — compounding the buildup independently of demand for the oil itself.
Separately, Bloomberg reported in August 2026 that oil tankers have begun conducting ship-to-ship transfers outside the Strait of Hormuz as a workaround — cargo is moved from one vessel to another in safer waters nearby, allowing some volume to keep moving even as direct transit through the Strait remains contested. That is a distinct adaptation from floating storage proper (the oil is still moving, just via a longer, split-cargo route), and its existence is itself evidence that Hormuz has not been a hard, total blockade for all traffic throughout the crisis, even during its most acute phases.
4. What "closure" of the Strait of Hormuz actually means here
Some secondary reporting on the 2026 crisis describes the Strait as having been fully closed since the conflict opened in February. The more consistently and credibly sourced picture, drawn from ongoing coverage through August 2026, is narrower but still serious: sustained attacks on shipping, a US blockade of Iranian ports, a collapsed mid-year truce, and a diplomatic standoff over the terms for restoring normal transit — alongside continued, if constrained and adapted, movement of some cargo (including via the ship-to-ship transfers noted above). This piece treats "severe, sustained disruption with intermittent full stoppages" as the better-supported characterisation than "total closure throughout," while flagging that the situation has evidently moved through different phases of severity since February and any single-sentence summary risks flattening that.
5. The consumption backdrop, and why it's moving too
Global oil demand is not holding steady through this either. The IEA's February 2026 Oil Market Report had global demand tracking toward roughly 104.9 million barrels a day for the year. By its August 2026 report, the IEA had revised that trajectory down substantially — projecting a demand contraction attributed to the Hormuz disruption and elevated fuel prices, with the pace of the year-on-year decline expected to ease from around 4.9 million barrels a day in the second quarter of 2026 toward roughly 2.8 million barrels a day in the third quarter, before a return to growth projected for the final quarter. That is a live, moving forecast rather than a settled number, and it should be read as the IEA's estimate as of its most recent report, not as an outcome that has already occurred.
The IEA's own emergency-stockholding rule for member countries — net reserves equivalent to at least 90 days of net oil imports — exists precisely to give countries room to absorb a shock like this one without an immediate scramble. Whether individual members are currently within that threshold, given the US drawdown described in Section 2, is a country-by-country question this piece has not attempted to answer for every member state.
6. How the shore-to-ship and ship-to-shore flow actually moves prices
Storage isn't a passive backdrop to the price of oil — the direction crude is moving between tank and tanker, and how fast, is itself a price signal, and it runs through a fairly specific mechanism: the shape of the futures curve.
When the market is in contango — futures for delivery months from now are priced higher than the spot price today — it becomes profitable to buy crude now, pay to store it (on land or on a chartered tanker), and sell it forward at the higher price, pocketing the difference net of storage and financing cost. That is what pulls oil off the market and into storage, onshore or floating: producers and traders have a financial reason to hold rather than sell immediately. The 2020 demand collapse is the textbook case — contango got so steep that traders chartered tankers purely to sit on crude offshore, because storing it was more profitable than selling it into a market with nowhere to put it onshore.
When the market flips to backwardation — spot trades above the forward price, usually a sign of current tightness or a supply scare — that arbitrage reverses. Holding oil in a tank or on a ship now costs money in foregone value, because the same barrel is worth more today than it will be worth later on the forward curve. That pulls stored crude out of tanks and off ships and toward refineries and consumers as fast as logistics allow, because sitting on it is now a losing trade. Onshore draws (Cushing, SPR, commercial tank farms all falling) and a shrinking floating-storage pool are both symptoms of the same backwardated condition, not separate phenomena.
The August 2026 disruption complicates this in a specific way this piece can show directly, from live pricing, without needing the full futures curve: as of 14 August 2026, Brent (the seaborne, internationally-traded benchmark) was priced at $87.23/bbl against WTI (priced at the landlocked Cushing, Oklahoma hub) at $81.27/bbl — a $5.96/bbl discount for WTI. That gap is itself a shore-vs-ship signal: WTI's price is set at a hub with pipeline and limited storage access rather than direct port access, so when seaborne routes tighten (as Hormuz has), a barrel that can reach open water commands a premium over one that's landlocked at Cushing, widening the discount. Cushing's own stock level — 22.6 million barrels for the week ending 7 August 2026, per the EIA's own weekly series — sits well down from the hub's typical working range, which is consistent with a market pulling inland storage down rather than building it, the pattern backwardation would predict.
Iran's floating-storage buildup, covered in Section 3, is the ship-to-shore version of this same logic running in reverse for a specific, blockade-driven reason rather than a price signal: those cargoes are laden and waiting, not because holding them is profitable, but because the US port blockade is physically preventing the ship-to-shore transfer that would normally happen on a commercial timetable. That distinction matters for reading the price impact correctly. Ordinary contango-driven floating storage is a voluntary bet that later prices will be higher; forced floating storage of the kind Iran is holding is involuntary and, if it eventually clears (Hormuz reopens, the blockade lifts), represents a supply overhang that could land on the market all at once rather than being drawn down gradually — a materially different price risk than storage building for ordinary commercial reasons.
7. The other release valve: pipelines being expanded, restarted, or stalled right now
Storage and tankers aren't the only way the system is absorbing the Hormuz shock. A less visible shift is happening in overland pipeline capacity — routes that don't touch the Strait at all, either because they were built that way or because they're being expanded specifically to reduce exposure to maritime chokepoints. Six are worth naming here, because they're at very different stages and moving in different directions, and putting them against a global baseline matters for reading them correctly.
That baseline, from Global Energy Monitor's own oil-and-gas infrastructure trackers: roughly 250,000 kilometres of oil, gas and NGL pipeline is currently in development worldwide — enough to circle the planet more than six times over. That figure is not a 2026 crisis response; it reflects a pipeline build-out trend that predates the Hormuz disruption by years. Against that backdrop, the six routes below are a small, visible slice of a much larger baseline expansion — some of them genuinely accelerated or redirected by the 2026 crisis (Kazakhstan's expansion and Iraq's restart both explicitly cited chokepoint/attack exposure as the reason), and some moving on a timeline the crisis hasn't touched at all (Power of Siberia 2's stall is a pure pricing dispute; ESPO's 2025 capacity expansion predates the crisis by months). The honest read is that the crisis is redirecting a handful of specific decisions inside a much bigger, longer-running structural trend — not causing a wholesale pivot away from seaborne trade, which remains structurally necessary for the large majority of crude that has no pipeline alternative to reach.
| Pipeline | Route | Status, mid-Aug 2026 | Scale |
|---|---|---|---|
| Kazakhstan–China oil pipeline | Kenkiyak–Atyrau & Kenkiyak–Kumkol lines, into China | Active expansion, agreed between KazMunayGas and CNPC | Kenkiyak–Atyrau: 6→15 Mt/yr; Kenkiyak–Kumkol: 10→20 Mt/yr |
| Kirkuk–Ceyhan pipeline | Northern Iraq to Turkey's Mediterranean coast | Reopened 18 Mar 2026 after a 12-year shutdown (halted 2014); bilateral Iraq-Turkey agreement expired and is under a 12-month extension negotiation as of Aug 2026 | Restarted ~250,000 bpd, targeting up to 770,000 bpd |
| Druzhba ("Friendship") pipeline | Russia to Hungary & Slovakia via Ukraine/Belarus | Southern branch disrupted by a drone strike on Ukraine's Brody pumping station, 27 Jan 2026; repaired and flows restored 23 Apr 2026; the disruption itself became the flashpoint for a stalled EU sanctions package | Normal southern flow ~300,000–400,000 bpd; Croatia's Adria pipeline offered as a non-Russian alternative able to carry up to 280,000 bpd |
| Power of Siberia 2 | Western Siberia through Mongolia to China | Legally-binding route memorandum signed Sep 2025; commercial terms deadlocked over gas pricing since; no construction start, no timeline, as of Jul 2026 | Proposed 2,600km line, 50 billion cubic metres/yr capacity (natural gas, not crude) |
| ESPO (Eastern Siberia–Pacific Ocean) | Taishet, eastern Siberia, to Kozmino Bay on the Pacific, with a spur direct to Daqing, China | Already the operating Russia-China crude artery (unlike the still-stalled Power of Siberia 2 above); design capacity reached 80 million tonnes/yr via expansions completed in 2025; Kozmino terminal and Far East facilities reported handling increased volumes into early 2026 | Kozmino terminal: 720,000 bpd tanker-loading capacity; China spur: ~600,000–700,000 bpd; contracted China deliveries ~30 Mt/yr, reportedly running above that (~35 Mt/yr) |
| China–Myanmar oil & gas pipelines | Kyaukphyu, on Myanmar's Indian Ocean coast, to Yunnan province, China | Operational since 2017; reaffirmed as one of three priority China-Myanmar infrastructure projects during Myanmar's president's Beijing visit, Jun 2026; but coup-era security problems inside Myanmar persist, and reporting as recent as Aug 2026 describes the line as running well under its designed throughput — a figure this piece could not independently confirm behind a paywalled primary source | Bypasses the Strait of Malacca entirely — shortens the route from the Middle East/Africa by a reported ~1,820 nautical miles versus the standard Malacca-South China Sea path |
The pattern across several of these is the same: build or expand a pipeline route that physically cannot be affected by a chokepoint blockade, drone strike on a port terminal, or tanker-targeting campaign, because the oil never goes to sea in the first place. Kazakhstan's pipeline expansion is explicitly a response to drone attacks on the Caspian Pipeline Consortium's Novorossiysk export terminal — a direct substitution of overland Chinese-bound capacity for exposed seaborne export. Iraq's Kirkuk-Ceyhan restart, after standing idle for over a decade, gives northern Iraqi crude a Mediterranean outlet that never transits the Gulf or the Strait of Hormuz at all. China's Myanmar pipeline exists for the same reason on a different chokepoint — bypassing Malacca rather than Hormuz — though it predates this crisis by nearly a decade and current reporting suggests it is running under capacity, not being newly leaned on. Even Druzhba, disrupted rather than expanded this year, illustrates the same structural point in reverse: a pipeline is not immune to attack (the Brody strike proved that), but the 86-day outage and its resolution both played out entirely on land, through repair and diplomacy, with none of the tanker-routing, floating-storage, or chokepoint dynamics described in Sections 3 and 6.
Power of Siberia 2 is the clearest outlier — it's gas, not crude, and it is stalled on commercial terms rather than accelerating in response to the crisis. It's included here because it is routinely mentioned in the same breath as these other Russia- and China-linked routes, and because "signed a memorandum" and "under construction" are being conflated in some coverage; as of the most recent reporting this piece could find, they remain two very different things for this project specifically. ESPO is the useful contrast to it: where Power of Siberia 2 is a proposed, deadlocked gas line, ESPO is the crude pipeline actually moving Russian oil to China today, at scale, and its most recent capacity expansion (to 80 million tonnes/yr) was completed in 2025 — before the Hormuz crisis began, not because of it.
Related on this blog
See also: Thermal Parity Is Dead: What $85 Oil and India's Gas-Price Formula Actually Mean, The FCC Lever: Trading Petrol Yield for LPG, and Where India Is Building Room to Store the Difference
Sources
- Upstream Online / Informat, "Attacks on the Caspian pipeline force Kazakhstan to increase its exports to China"; Jamestown Foundation, "China Enters Kashagan Oil Project, Will Boost Kazakhstan-China Pipeline Capacity" (Kenkiyak-Atyrau/Kumkol expansion figures)
- The National, "Iraq resumes crude oil exports to Turkey's Ceyhan port through pipeline" (18 Mar 2026); Iraq Business News, "Iraq-Turkey Pipeline 'Remains Operational'" (1 Aug 2026, agreement-extension status)
- 2026 Druzhba pipeline dispute, contemporaneous reporting via Baird Maritime, "Druzhba pipeline: Why it complicates EU sanctions on Russia", and Insights on India, "Druzhba Pipeline Explained" (23 Apr 2026, Brody strike/repair timeline, Adria alternative)
- Al Jazeera, "What is the Power of Siberia 2 pipeline that Russia, China are planning?" (20 May 2026); Asia Times, "Power of Siberia 2 deadlock belies Russia-China 'no-limits' pact" (Jul 2026)
- Natural Gas Intelligence, "Some in Oil, Natural Gas Sector Expected Quick Hormuz Reopening Before 'Project Freedom'" (US Gulf vessel-escort initiative, not a pipeline); Global Energy Monitor, Freedom Oil Pipeline entry (shelved ~2015 Permian-to-California project, unrelated and inactive)
- Global Energy Monitor, Oil & Gas overview (~250,000km of oil/gas/NGL pipeline in global development, cross-checked against the org's individual project trackers, incl. its ESPO pipeline entry and its Sino-Myanmar pipeline entry)
- S&P Global, "Russia crude oil pipeline capabilities to mainland China — The ESPO crude oil pipeline" (design/terminal/spur capacity figures)
- The Diplomat, "Myanmar's Depleted Oil Pipeline" (Aug 2026, cited via search summary; this piece could not open the article directly to confirm its specific throughput figures)
- OilPriceAPI, live spot prices (Brent $87.23/bbl, WTI $81.27/bbl, Brent-WTI spread $5.96/bbl), queried 14 Aug 2026, 01:33 UTC
- US Energy Information Administration, Weekly Cushing, OK Ending Stocks of Crude Oil (22.566M bbl, week ending 7 Aug 2026, released 12 Aug 2026)
- CNBC, "Oil in U.S. Strategic Petroleum Reserve falls below 300 million barrels, lowest since 1983" (10 Aug 2026)
- Bipartisan Policy Center, "How the U.S. Strategic Petroleum Reserve Works" (SPR authorised capacity, 714M bbl)
- CNBC, "U.S. Strategic Petroleum Reserve faces stress as emergency releases strain old infrastructure" (28 Jul 2026)
- Bloomberg, "Oil Tankers Conduct Ship-to-Ship Transfers Outside Hormuz as Attacks Persist" (11 Aug 2026)
- Bloomberg, "Strait of Hormuz: What Have Iran, US Said as Efforts to Restore Traffic Stall" (11 Aug 2026)
- Congress.gov / Congressional Research Service, "The Strait of Hormuz: Security Developments and Impacts on Oil, Gas, and Other Commodities" (~25% of world seaborne crude/products trade baseline)
- World Oil, "Dozens of Iranian tankers idle as U.S. blockade restricts exports" (5 Aug 2026), citing Vortexa data on Iranian floating storage (~135M bbl, +14% m/m)
- US Energy Information Administration, "China, the United States, and Japan hold most strategic oil inventories" (China ~1.4bn bbl combined stock estimate, Dec 2025)
- Kayrros (via secondary reporting), China total crude tank-farm capacity estimate, >1.8bn bbl (mid-2024)
- IEA, Oil Market Report — August 2026; IEA, Oil Market Report — February 2026 (global demand trajectory figures)
- ISPRL (Indian Strategic Petroleum Reserves Limited), standing capacity figures for Mangaluru, Visakhapatnam and Padur
This piece deliberately separates figures this article has reasonable confidence in (the US SPR numbers, the Iran-specific floating-storage figure, the IEA demand-trajectory revisions, all tied to a specific named source and date) from figures that are widely repeated but harder to independently verify (China's total reserves, the precise current Hormuz throughput, and any single global floating-storage total for the current period). Where a figure could not be traced to a clear primary source, or where different outlets reported materially different numbers, that is flagged in the text rather than resolved by picking one. This is a genuinely fast-moving situation as of publication (August 2026); figures here should be treated as dated to the sources cited, not as a live picture.
About this article: Researched, written and edited by Umashankar Triplicane Dwarakanathan, with AI research assistance; every figure is meant to trace to the primary source cited. See the Editorial Policy for how sourcing, AI use and corrections work.
