Dataset: Gauge Fuel Prices with Oil Inventory Index


As people take to the roads and skies for the holiday, it’s that time of the year when the cost of travel is hard to ignore.

Though fuel prices have fallen since the summer, they remain stubbornly high, keeping the focus on what comes next. Are high prices here to stay?

In this article, we describe how an Ursa dataset generated from satellite imagery can help answer this question about the direction oil prices are likely headed.

First – what goes into an oil price forecast? One authority on the matter – the US Energy Information Administration (EIA) – highlights the importance of inventories to oil price forecasts.

According to the EIA, petroleum inventories “reflect changing market pressures on crude oil prices, and thus provide a good barometer of crude oil price change in the short run.”

In short, the basic nature of this relationship is straight-forward. Rising inventories place downward pressure on prices, and falling inventories place upward pressure on prices.

However, a deeper understanding comes from viewing inventories relative to “normal” levels in order to gain context.

Oil Inventory Index

To provide that context, Ursa created a dataset called the Oil Inventory Index. The index provides a daily update of global oil inventories, which is also broken down into an OECD & non-OECD basis.

A score above 100 means inventories are “above normal” and a score less than 100 signifies “below normal.” The calculation of normal is based on a multi-year average (pre-pandemic) for a given month.

So how much insight does this dataset provide regarding the price of oil?

For starters, the correlation between the Oil Inventory Index and ICE Brent futures (the global oil price benchmark) is equal to 0.86, a strong piece of evidence that bolsters the case for the dataset.

The time period covered – early 2018 to present – includes both the pandemic and war in Ukraine, each major events that respectively caused wild gyrations in the oil market.

Prices vs. Inventories

Let’s look back at 2021 when there was much uncertainty whether the nascent recovery at the time would last or not.

By January 2021, oil prices were approaching pre-pandemic levels, after clawing their way back from the depths of the year before.

There was a sense that economies had begun to reopen, pushing oil prices higher, but the next phase would hinge on fundamentals.

If inventories fell in 2021, this would support the case for higher oil prices, but if not, then oil prices would likely reverse course and head lower.

The graph below shows how this unfolded:

Inflation data

Another pivotal moment has been the upheaval caused by the invasion of Ukraine.

Oil prices skyrocketed above $120 per barrel in June, as traders wagered the conflict would erase Russian oil supply from the market.

Again, the key piece of information needed to assess the next move in prices (higher or lower) comes from fundamentals, and specifically inventories, which is what the Oil Inventory Index was designed to capture.


Energy prices have been a significant contributor to higher living costs that led central banks to hike interest rates, creating upheaval in virtually every market imaginable (e.g., stocks, bonds, currencies, commodities).

The release of new inflation data has become a market-moving event, especially when the figures suggest the path of future interest rate hikes.

With all the attention on inflation data, this also raises the importance of any information on the typical basket of goods and services that people spend money on, including fuel purchases.


The Oil Inventories Index dataset will be available for purchase as a subscription on AWS Data Exchange in December 2022.

Sign up for the Ursa Space Newsletter for updates.






Live count of Ursa Space’s SAR + Optical imagery/data catalog

Learn more about our imagery as a service.