Minnesotans For Sustainability


Sustainable:  A society that balances the environment, other life forms, and human interactions over an indefinite time period.




Olduvai Cliff Revisited

The Olduvai Cliff Event: ca. 2007

Richard C. Duncan
March 5, 2001

United States Natural Gas
Canada Natural Gas
Mexico Natural Gas
North America Natural Gas and Electricity
The Olduvai Cliff Event: An Epidemic of Blackouts -- Worldwide


The Olduvai theory states that the life-expectancy of Industrial Civilization, defined in terms of world energy use per capita ("e"), is less than or equal to 100 years. History: We know that the peak of "e" occurred in 1979 and that "e" declined from 1979 to 1999 (the 'slope'). Future: The Olduvai theory predicts that "e" will decline even faster from 2000 to the so-named 'cliff event' (the 'slide'). A previous study put the 'cliff event' in year 2012 (Duncan, 2001). However, it no appears that 2012 was too optimistic. The following study indicates that the 'cliff event' will occur about 5 years earlier than 2012 due an epidemic of 'rolling blackouts' that have already begun in the US. This 'electrical epidemic' spreads nationwide, then worldwide, and by ca. 2007 most of the blackouts are permanent. The 'modern way of life' is history by ca. 2025.


'NA' means North America. 'NG' means natural gas. 'Tcf' means trillion (10^12) cubic feet. 'bcfd' means billion cubic feet per day. 'BP' means 'British Petroleum Statistical Review of World Energy' (an ongoing annual series). 'USGS 2000' means 'US Geological Survey World Petroleum Assessment 2000' (a 5-year study issued in March 2000). 'EUR' means 'expected ultimate recovery' in Tcf. 'GCPE' (a USGS term) means 'grown conventional petroleum endowment' in Tcf. In this discussion, 'GCPE' and 'EUR' mean the same thing.


The Olduvai theory has evolved over 22 years with four major advances to date. The first in 1979 was that industrial civilization can be described by a single pulse waveform (i.e. overshoot and collapse). The second in 1989 was a robust definition of the Olduvai theory (i.e. per K. Popper, a scientific theory must be falsifiable). The third in 1993 used data and mathematics to demonstrate that the peak of world energy use per capita occurred in 1979. The fourth in 2000 identified electrical energy (specifically: communication, computation and control: the vital 'C^3 functions') as the sin-quo-non of 'the modern way of life' (not fossil fuels, per se). A fifth advance is proposed in this study. It forecasts that electrical blackouts in the US -- due to a shortage of natural gas in North America -- will precipitate the Olduvai 'cliff event' worldwide (i.e. 'avalanche' or 'trigger') in ca. 2007.


In November 2000 I gave a talk to the Pardee Keynote Symposia, Geological Society of America, Summit 2000, "The Peak of World Oil Production and the Road to the Olduvai Gorge". It was soon posted on the web and generated many responses: pro, con, skeptical, hostile, uninformed, and even one personal threat. Nonetheless, all responses proved 'useful' in one way or another. Figure 4 in my GSA paper shows the Olduvai 'schema' -- a graph that depicts the 'cliff event' in year 2012 (soon to be published in Duncan, 2001). This schema depicts US and worldwide blackouts as the 'trigger' of the Olduvai cliff. However, at least two perceptive reviewers, namely Perry Arnett and Bruce Thomson, surprised me by saying that they thought the Olduvai 'cliff event' in 2012 was TOO OPTIMISTIC because (to paraphrase) "things just can't hold together that long."

Was I TOO OPTIMISTIC? If so, what could cause the blackouts (i.e. the 'cliff event') before 2012? It wasn't likely to be directly caused by the world oil peak because very little oil is burned in power plants (too expensive). Nuclear power has its problems (i.e. $$$, hot wastes, etc.), but overall it's reliable. And there's lots of coal (albeit dirty, it's cheap). But Whoa! Could it be a shortage of natural gas (NG)? So I called a colleague and suggested that we forecast world NG production because of the skyrocketing demand for gas-turbine power plants. But he quickly observed, "We don't have enough reliable data to forecast WORLD natural gas production." Thus I dropped the idea -- momentarily.

In addition to the BP data, I now have the USGS Assessment 2000 'reserve estimates' for the US, Canada, and Mexico for NG (F95, F50, F5, and Mean). Using both sets of data, I built separate models to forecast the NG production for the US, Canada, and Mexico from 2000 to 2040. These curves were then summed to get the curve for NA historic production (1960-1999) and the NA NG production forecast (2000-2040). The results of these models are the basis for this study.

United States Natural Gas

Model Results: US NG production in 1999 comprised a remarkable 73.5% of the total North America production. US production peaked in 1971 at 22.0 Tcf (i.e. one year after the US oil peak) and from 1971 to 1999 US gas production declined by an average of 0.50 %/year. A secondary US peak is forecast to occur in 2007 at 20.1 Tcf. Then from 2007 to 2040, US production falls by some 41% -- an average decline of 1.5 %/year during 33 years. My US model forecasts the US EUR = 1,840 Tcf. The USGS report forecasts the US Mean GCPE = 1,910 Tcf. Note carefully that it is the USGS Mean value for US gas (i.e. not the USGS F95 value!) that is in good agreement with my US model.

"The U.S. reserve level is barely more than half what it was when it peaked at 293 Tcf in 1967, before starting a steady decline. ... offshore Gulf of Mexico reserves are being depleted at a rate of 25-30% per year of remaining reserves." (Parent, 2001)

"US consumption of all primary sources of energy except nuclear power will increase this year. The biggest gainer will be natural gas, demand for which is expected to rise 2.4% to 23.45 quads. ... This year's energy from natural gas will set a record high level due to increased demand in the electric utility and industrial sectors, and because of near-normal winter weather." (Radler, 2001b, p. 67)

The US Energy Information Agency (EIA) forecasts: "For 2020, ... pushed by expected demand growth for gas, primarily for electricity generation ... [sic] US natural gas demand will increase by 62% during 1999-2020, rising to 34.7 Tcf from 21.4 Tcf. Natural gas demand for electricity generation ... will triple over that period, as 89% of the generation capacity built over the next 2 decades will be gas-fired." (True, 2001)

But President George W. Bush isn't so confident as the US EIA. "To develop a national energy policy is a matter of high concern for this administration, because it's a matter of high concern for our nation. It's becoming very clear to the country that demand is outstripping supply, that there are more users of electricity and natural gas than there is new units being found, and we've got to do something about that." (quoted in Crow, 2001)

Canada Natural Gas

Model Results: Canada NG production in 1999 comprised a sizeable 21.5% of NA production. From 1983 to 1995, Canada production grew by an astounding 114%, i.e. an average growth rate of 6.0 %/year during 12 years. Then from 1995 to 1999 growth slowed to 2.1 %/year. Canada NG production is forecast to peak in 2005 at 6.1 Tcf. Then from 2005 to 2040, Canada production plunges by 86% -- an average drop of 4.3 %/year during 35 years. My model forecasts Canada EUR = 290 Tcf, in exact agreement with the USGS Canada Mean GCPE = 290 Tcf.

"Canadian gas marketing has experienced strong growth in recent years, to the extent that production has outstripped reserve additions for the past several years, resulting in a continuing decline in remaining reserves." (Parent, 2001)

"Despite Canada's drilling of a record number of wells, deliverability has increased only marginally. The per-well average has been declining, due in part to the drilling of an increasing number of low-deliverability shallow wells. To offset the annual decline in production from existing wells, production from new wells must amount to 20% of current production, a formidable barrier to increasing production." (Parent, 2001)

Mexico Natural Gas

Model Results: Mexico NG production in 1999 comprised a mere 5.4% of NA production. Mexico production shows an early (i.e. 'local') peak in 1982 at 1.3 Tcf, followed by an overall decline through 1995. However, from 1995 to 1999 production grew by a strong 29% -- a notable average of 6.4 %/year. Mexico NG production is forecast to peak in 2011 at 1.5 Tcf. Then from 2011 to 2040, production falls by some 56% -- an average decline of 2.7 %/year for 29 years.

But now arises an apparent disagreement. Namely: My model forecasts Mexico EUR = 85 Tcf, whereas the USGS forecasts the Mean GCPE = 150 Tcf. I think this difference can be explained in that the USGS 2000 report used data normalized to 1995. However, in 1999 Mexico (Pemex) revised downward its "proved reserves" of NG to 30.1 Tcf from 63.5 Tcf. This was a decrease of 53% in one fell swoop. (Pemex now admits that it exaggerated its reserves.) Thus, if we now multiply the USGS value of 150 Tcf by 53% we get 80 Tcf -- in good agreement with my 85 Tcf.

We will see, however, that the "true" value of Mexico's NG reserves is of little consequence to this study because Mexico (1) is a minor player in NA NG, (2) its now a net importer of NG from the US, and (3) its domestic demand is fast outstripping its production.

"In 1994-99 demand for natural gas in Mexico had increased by 6% annually. Electric power demand had increased at an even higher rate, 9%, based in large part on the increased use of combined-cycle turbine technology that requires natural gas." (Baker, 2001)

"At present, Pemex's ability to supply gas for the domestic market is in question. ... it is doubtful that any significant quantity of gas would become available for long-term contracts to the US." (Baker, 2001)

"Gas production in Mexico dipped last year to average 4.69 bcfd from 4.79 bcfd in 1999, while demand is estimated to have grown as much as 10%. ... With demand for natural gas in the US as strong as it is, it is hard to believe that Mexico can continue to rely on imports from the north much longer." (Radler, 2001a, p. 17)

"Pemex has acknowledged the alarming rate of growth in Mexican gas demand growth in recent years. ... From 1995-99, the demand for gas grew at a rate of 4.4 %/year. ... In some scenarios of prospective use, projections indicate expectations of an increase in demand for natural gas of 9 %/year over the next decade. ... Just one change in current practice, the switch from heating oil to natural gas in combined-cycle turbine power plants, would be capable of increasing the annual rate of growth in gas demand to 16%." (Barbosa, 2001)

"Mexico is expected to be a net importer of gas from the U.S. for the foreseeable future." (Parent, 2001)

North America Natural Gas and Electricity

Model Results: North America NG production in 1999 was 26.23 Tcf. The NA NG production curve, as previously mentioned, is the sum of the production for the US, Canada, and Mexico. An early (i.e. 'local') NA peak occurred in 1972 at 25.0 Tcf. Then from 1972 to 1986 NA production declined by some 20%. Next, the trend quickly reversed and from 1986 to 1999 NA production increased by a solid 2.0 %/year for 13 years. Looking ahead however, North America NG production is forecast to peak in 2007 at 28.5 Tcf. Then from 2007 to 2040 NA production falls by 51%, i.e. an average decrease of 2.1 %/year during 33 years.

"The shortage of natural gas is not limited to the US but has also become a problem in Mexico. ... volumes coming to the US from Mexico fell from a total of more than 54 bcf in 1999 to just 4.71 bcf in 2000 and than to nothing. Mexican domestic demand for gas no longer allowed for exports." (Radler, 2001a, p. 17)

"Canada will be stepping up its resource development program. Mexico will be moving up the curve of energy development and utilization, but will need gas from the U.S. for the foreseeable future. Look for more imports, including LNG, to bolster U.S. supply alternatives. Look for energy efficiency and conservation to come back into vogue, as consumers seek ways to deal with higher energy costs. There may not be a 30 Tcf [North American] market out there." (Parent, 2001)

"And the California crisis poses concerns for the US natural gas industry in particular. The power sector is expected to account for the fastest-growing area of natural gas demand in the years to come. The vast majority of the more than 200 power plants slated to come on stream in the US in the early part of this decade are expected to be fueled by gas." (Rouffignac, 2001)

"The crisis in California arose out of a political system that discourages energy development. At the core of the state's ruinous electric shortage lies a chronic deficiency of generation capacity. ... The US must not follow its bellwether state down the same costly path." (O&GJ, 2001. p. 19)

The Olduvai Cliff Event: An Epidemic of Blackouts -- Worldwide

It was MIT Professor Jay Forrester, I think, who first related 'technological fixes' to the hierarchy of physical limits to growth (a paraphrase follows): "We can use technology to remove physical limits to growth, but each time we do we always bump into a new, more recalcitrant, limit to growth. Ultimately all technological fixes fail."

Recall that the US EIA forecasts that "US natural gas demand will increase by 62% ... rising to 34.7 Tcf in 2020." In contrast, it is widely recognized (by both the analysts and my forecasts) that the US cannot rely on future exports from either Mexico or Canada. And US production itself is in doubt. Meanwhile, a voracious 'increase in demand is expected' from the NA residential, commercial, industrial, and electric utility sectors. For example: More that 200 power plants are slated to come on line in the US alone(!) the vast majority are expected (i.e. designed) to be fueled by natural gas.

With due respect to the analysts quoted in this study, none of them seem to understand the serious consequences of the looming 'shortages' of natural gas in North America. For that we must turn to my forecasting models and the Olduvai theory.

It is true, given sufficient capital, that we can build a lot of new gas-turbine power plants and transmission lines and thereby remove the 'ruinous shortage of electrical generation and transmission capacity'. However per Jay Forrester we immediately run into a higher and stiffer limit to growth: natural gas. Then we can (again) try nuclear power....... And after that coal......... And finally wood....... And that puts us right back into the 18th century.

The Golden State leads the way down the road to the Olduvai Gorge.


Baker, G. (2001). Mexico's natural gas pricing crisis: Will confrontation lead to policy liberalization? O&GJ, 1-22-01, p. 18-25.
Barbosa, F. (2001). Pemex targets major increase in natural gas production to meet soaring domestic demand. O&GJ, 1-22-01, p. 70-74.
Crow, P. (2001). US oil and gas industry has high expectations from Bush administration on energy issues. O&GJ, 2-12-01, p. 66-77.
Duncan, R.C. (2001). World energy production, population growth, and the road to the Olduvai Gorge. Population and Environment. Forthcoming in the May-June issue.
Parent, L. (2001). Natural Gas: There's work to be done to meet increasing demand. World Oil, February, p. 94-98.
Radler, M. (2001a). More gas for Mexico. O&GJ, 1-29-01, p. 17.
Radler, M. (2001b). US demand growth in 2001 to respond to slowing economy. O&GJ, 1-29-01, p. 66-78.
Rouffignac, A. de (2001). California power crisis energy policy flashpoint for Bush. O&GJ, 2-12-01, P. 70-71.
True, W. R. (2001). Major gas projects fuel surge in long-term plans. O&GJ, 2-5-01, p. 61-68.
Used with permission of the author.


Please send mail to webmaster@mnforsustain.org with questions or comments about this web site. Minnesotans For Sustainability (MFS) is not affiliated with any government body, private, or corporate entity. Copyright 2002 Minnesotans For Sustainability