Thursday, May 25, 2006

Texas and US Lower 48 Oil Production as a Model for Saudi Arabia and the World

by Jeffrey J. Brown & "Khebab"

As many people now know, 50 years ago this March M. King Hubbert predicted that US Lower 48 and Texas oil production would peak, and enter a terminal decline, somewhere between 1966 and 1971. Dr. Hubbert also predicted that world oil production would peak, and enter a terminal decline within 50 years, i.e., by 2006. To be clear, despite what is either a profound misunderstanding of or a misrepresentation of Dr. Hubber'ts work in some quarters, Dr. Hubbert was not predicting the end of world oil production by 2006; he was predicting that production peaks when producing regions have consumed about half of their recoverable
conventional oil reserves.

In our previous article, "M. King Hubbert's Lower 48 Prediction Revisited," we outlined a simplified way of predicting what Kenneth Deffeyes designated as Qt, ortotal recoverable conventional oil production for a region. The method has been designated Hubbert Linearization, or HL, by Stuart Staniford, with The Oil Drum blog.

Using the HL technique, the purpose of this paper is to use historical Texas and Lower 48 oil production as a model for future oil production in Saudi Arabia and the world. Figures One and Two show HL plots for Texas the Lower 48. Texas peaked at 56.5% of Qt. The Lower 48 peaked at 51.9% of Qt.

Fig 1. Hubbert linearization applied on the Texas oil production.




Fig 2. Hubbert linearization applied on the Lower-48 oil production



Note that prior to its peak, Texas was the "swing producer," i.e., its production was regulated by the Texas Railroad Commission in order to keep oil prices within a certain range. Perhaps because of its swing producer role, Texas peaked later than the Lower 48, relative to their respective Qt's. However, Texas oil production, now down about 75% from its peak, has fallen much more sharply than has the Lower 48 overall, now down about 50% from its peak.


Saudi Arabia succeeded Texas as the swing producer, at least until recently. The emerging "swing producer" is the release of petroleum from emergency reserves. The problem of course will be replenishing the emergency reserves.


Figures Three and Four show HL plots for Saudi Arabia, now at 58.1% of Qt, and the world, now at 48.2% of Qt. In other words, Saudi Arabia and the world are now approximately where Texas the Lower 48 were at in 1972 and 1970 respectively.

Fig 3. Hubbert linearization applied on the Saudi Arabia oil production




Fig 4. Hubbert linearization applied on the world oil production


Figure Five shows superimposed production graphs for Texas and the Saudi Arabia, with Texas production in 1972 lined up with Saudi Arabia production in 2005. Note the difference in the vertical scales.
Fig 5. Saudi-Arabia and Texas
oil production (in million of barrels per day)


Figure Six shows superimposed production graphs for the Lower 48 and the world, with Lower 48 production in 1970 lined up with world production in 2005. Again, note the difference in vertical scales.

Fig 6. World and Lower-48 oil production (in million of barrels per day)


Note that both Russia and the North Sea also peaked in the vicinity of 50% of Qt. Russia is a complex case, because of the post-Soviet collapse in production, but the post-1984 cumulative production falls within the predicted HL limits. North Sea production has been falling steadily since peaking in 1999 at 52% of Qt. Notably, the top 10 major oil companies working the North Sea in the late Nineties, using the best data, best engineers and best technology available, were predicting that North Sea production would not peak until 2010 at the earliest. Many of these same companies are now saying the world peak oil production is decades away.


In summary, based on the HL method and based on our historical models, we believe that Saudi Arabia and the world are now on the verge of irreversible declines in conventional oil production. While there will be massive efforts directed toward unconventional sources of oil, we predict that unconventional sources of oil will only serve to slow and not reverse the decline in total world oil production.


As we stated in our previous article, in order to speed the adjustment to the realities of declining world oil production, we recommend that the United States abolish the Payroll (Social Security + Medicare) Tax and replace it with either a liquid transportation (petroleum) fuel tax or an overall (nonrenewable) energy tax. In effect we would tax energy consumption to fund Social Security and Medicare. The primary reason for implementing the proposal is that it would cause an immediate and massive across the board push for greater energy efficiency and it would unleash enormous free market forces against profligate energy use. This proposal would also find favor with those concerned about Global Warming.


Data sources:

Jeffrey J. Brown is an independent petroleum geologist in Addison, Texas.

E-mail: westexas@aol.com



Wednesday, May 10, 2006

Weekly Petroleum Status Report Update

The graphs in the "Weekly Petroleum Status Report" post have been updated from the last EIA numbers. Gasoline consumption seems to have increased. Stock coverages are down both for gasoline and crude oil.

Monday, May 08, 2006

Some Detailed Views on Norway's Oil Production

The other day on theoildrum, Roberto has produced some beautiful graphs on Norway production per field. I thought that was a bright idea and decided to repeat the same exercise but on different field characteristics. The gross wellhead production (Norwegian share) per field is available in ascii format on the NPD website. I used only the gross oil production figures (I excluded NGL and condensate) resulting in 64 oil fields.
Fig 3. Ratio of the current production (2005) to the maximum production (in %).
Fig 4. Field maximum output (mbpd).
Fig 5. Field peak year.
Fig 6. Number of production wells per fields.
Fig 7. Number of injection wells per fields.
Fig 8. Average wellbore depth (kilometers).
Fig 9. ratio of the number of wellbores to the peak production in mbpd.
Fig 10. Estimated logistic growth rate per fields.
Fig 11. Oil revenue per fields (inflation adjusted, US$).

Fig 1. Cumulative production per field (2005, Gb).


Fig 2. First year of production per field.


Fig 3. Ratio of the current production (2005) to the maximum production (in %).


Fig 4. Field maximum output (mbpd).


Fig 5. Field peak year.


Fig 6. Number of production wells per fields.


Fig 7. Number of injection wells per fields.


Fig 8. Average wellbore depth (kilometers).


Fig 9. ratio of the number of wellbores to the peak production in mbpd.


Fig 10. Estimated logistic growth rate per fields.


Fig 11. Oil revenue per fields (inflation adjusted, US$).

Wednesday, May 03, 2006

Weekly Petroleum Status Report

This post is based on the inventory data reported every week by the EIA here and here. I decided to update a serie of graphs every week based on the current data. The idea is to try to give more context and enhance any eventual outliers or events in the data.

Last Update: April 26, 11:00

List of Figures:
  1. US Gasoline Stocks.
  2. US retial gasoline prices.
  3. Gasoline consumption projections
  4. Gasoline consumption projections (with a 30 day Moving average)
  5. Observed gasoline stock coverage values (NDFC)
  6. Observed U.S. gasoline imports (4-weeks average).
  7. U.S. crude oil stocks
  8. Crude oil spot prices
  9. U.S. Crude oil imports
  10. U.S. Total Crude Oil and Petroleum Products Net Imports
  11. Crude oil input to refineries
  12. Observed crude oil stock coverage values
  13. U.S. Weekly Percent Utilization of Refinery Operable Capacity (UROC)
Notes:
  1. For some figures (Fig. 3, 4, 9, 10 and 11), the year-to-year increase is adjusted assuming a linear trend (see Are Prices Affecting the Gasoline Consumption in the US? for details).
  2. data points are interpolated for missing days.
Gasoline
Fig 1. US Gasoline Stocks (EIA).


Fig 2. US Retail Gasoline Prices (EIA).


Fig 3. Gasoline consumption projections: The gray level image in the background is the observed seasonal fluctuations (darker areas mean more frequent values). The red curve is the observed data for 2007. The * means that the data for the year 2006 and 2005 have been adjusted to match the yearly consumption for 2007 given by a linear growth model. The dark dotted line is the average fluctuation.


Fig 4. Same as Fig. 3 but after application of a 30 days moving average on the data in order to reduce noise.


Fig 5. Observed stock coverage values (NDFC): the shaded gray values are the observed distribution of NDFC values for each day of the year from all the available years (1991 to 2006) (dark means high probability, white means low probability). The lines are for the last 5 years.


Fig 6. Observed U.S. gasoline imports (4-weeks average).


Crude Oil

Fig 7. US Crude Oil Stocks (EIA).


Fig 8. Cushing, OK WTI Spot Price FOB (Dollars per Barrel).


Fig 9. U.S. Crude oil imports (4-weeks average): The gray level image in the background is the observed seasonal fluctuations (darker areas mean more frequent values). The red curve is the observed data for 2007. The * means that the data for the year 2006 and 2005 have been adjusted to match the yearly import level for 2007 given by a linear growth model. The dark dotted line is the average fluctuation.


Fig 10. U.S. Total Crude Oil and Petroleum Products Net Imports (4-weeks average): The gray level image in the background is the observed seasonal fluctuations (darker areas mean more frequent values). The red curve is the observed data for 2007. The * means that the data for the year 2006 and 2005 have been adjusted to match the yearly import level for 2007 given by a linear growth model. The dark dotted line is the average fluctuation.


Fig. 11 Crude oil input to refineries: The gray level image in the background is the observed seasonal fluctuations (darker areas mean more frequent values). The red curve is the observed data for 2007. The * means that the data for the year 2006 and 2005 have been adjusted to match the yearly consumption for 2007 given by a linear growth model. The dark dotted line is the average fluctuation.


Fig. 12 Observed crude oil stock coverage values (NDFC): the shaded gray values are the observed distribution of NDFC values for each day of the year from all the available years (1991 to 2006) (dark means high probability, white means low probability). The lines are for the last 5 years.


Fig. 13 U.S. Weekly Percent Utilization of Refinery Operable Capacity (UROC): the shaded gray values are the observed distribution of the UROC values for each day of the year from all the available years (1991 to 2006) (dark means high probability, white means low probability). The dark dotted line is the average fluctuation.
.

Tuesday, May 02, 2006

Are Prices Affecting the Gasoline Consumption in the US?

Now that prices in the U.S. have nearly doubled since 2001 (see Fig. 1), I was wondering if gasoline consumption has been affected.

Fig 1. US Retail Gasoline Prices (EIA, April 21, 2006).

We can observe two patterns in the gasoline consumption fluctuations:
  1. A year-to-year steady rise that can be fairly modeled by a linear trend (the estimated slope is 143.4 kbpd/year) as shown on Fig. 2;
  2. an intra-year seasonal fluctuations (i.e. higher consumption in the summer);
Fig 2. US Gasoline Supplied (EIA, April 21, 2006).

Once the linear trend is removed (Fig. 3), it's then easier to deal with the seasonal pattern.
Fig 3. Residuals of US Gasoline supplied after the trend has been removed.

From the above residuals, we compute the frequencies of the observed residual values (histogram) for each day of the year as shown on Fig. 4. We can clearly observed the increase in gasoline consumption during the driving season. To have a rough idea of what we should expect for 2006, we use the linear trend line to adjust the seasonal pattern and the 2004 and 2005 consumption values.

Fig 4. The gray level image in the background is the observed seasonal fluctuations derived form the residuals shown on Fig. 3 (darker areas mean more frequent values). The red curve is the observed data for 2006. The * means that the data for the year 2005 and 2004 have been adjusted to match the yearly consumption for 2006 given by the linear model (red line on Fig. 1). The dark dotted line is the average fluctuation.


So far, the gasoline consumption levels don't seem to be out of the range of what have been observed in the past despite being below the average consumption for that time of the year. Notice also how significant was the big drop in consumption last year after the hurricanes Katrina and Rita.