by Dennis Coyne
(note that an Open Thread has recently been published for Petroleum)
I expect World Fossil fuel output to peak in 2025. If the World economy continues to grow in the future, a gap between Energy produced from all sources (including non-fossil fuels) and the demand for Energy will grow over time. If the gap between energy demand and energy supply is not filled by growth in non-fossil fuel energy sources there must be lower demand for energy due to reduced economic growth rates.
The analysis that follows attempts to determine demand for primary energy under the assumption that energy supply is plentiful so that supply is constrained by demand. The consumption of primary energy is related closely to Real GDP (constant US dollars at market exchange rates).
In 1988 Real GDP was about 35 trillion (T) 2010 US $ and in 2000 Real GDP was about 50 T 2010$. The energy intensity of Real GDP measured in exajoules (EJ) per T 2010 US$ of Real GDP has decreased over time as shown in the chart that follows.
To find the annual rate of decrease in the Energy intensity of real GDP from 1970 to 2015, I look at the slope of the natural log of energy intensity vs. year, the average annual rate of decrease was 0.93% per year from 1970 to 2015, chart below.
There was a period with a faster annual rate of decrease in energy intensity of 1.3% per year from 1987 to 2000 as shown in the chart below.
It is not likely that a 1.3% annual rate of decrease could be maintained for very long and might not be possible due to diminishing returns. Recently, energy intensity has decreased at an annual rate of 1% from 2010 to 2015, this rate might continue for some time as thermal losses from electric power generation are reduced as wind, solar, and hydro replace some power generation, as heat pumps replace furnaces and boilers, and as batteries replace some of the fuel used for transportation.
The projection of future energy intensity of real GDP assumes a 0.93% annual rate of decrease until 2050 and a 0.0% annual rate of decrease from 2051 to 2100. This is shown in the chart that follows.
The chart below shows the natural log of Energy Intensity of real GDP from 1970 to 2050, the slope of the curve is the average annual rate of decrease of energy intensity. The trend line is for the 1970 to 2015 data only.
World real GDP per capita has grown relatively steadily since 1971 at an annual rate of about 1.45% per year. The slope of the trend line for the natural log of real GDP per capita is the average growth rate shown in the chart below for 1971 to 2015.
I assume this 1.45%/year average growth rate continues until 2020, falls to 1.4%/a from 2021 to 2025, to 1.35%/a from 2026-2030, and to 1.3% from 2031-2039. I assume real GDP per capita annual growth rates gradually increase as the economy adapts. Real GDP per capita grows at 1.35%/a from 2040-2045 and then remains at 1.4%/a from 2046 to 2100. Note that lower growth rates imply lower energy demand, if all else is held equal.
I use the UN medium fertility population scenarios to estimate future population growth, but my expectation is that population will grow more slowly than this projection.
By combining the population scenario above with the GDP per capita growth scenario from 2016 to 2100 we can project future GDP.
From 2010 to 2015 World Real GDP grew at an average rate of 2.5% per year, below I show the World Real GDP growth rate for the scenario above from 2017 to 2100.
From 1980 to 2015 the average annual real GDP growth rate was 2.9%/year. Slower population growth is the primary cause of the slowdown in real GDP growth in this model.
Primary Energy Demand is just the Energy intensity of Real GDP times the Real GDP, the projection is shown below. The sharp increase in projected demand after 2050 is due to the (unrealistic) assumption that energy intensity will stop decreasing in 2050. A more realistic scenario would have energy intensity continue to decrease from 2051 to 2100, but at a continually smaller rate of decrease. This “unrealistic” scenario was chosen in anticipation of objections that my projection might be too optimistic.
I have covered fossil fuel supply on many occasions and the scenario below is based on my medium oil, natural gas, and coal scenarios (these will be covered briefly at the end of the post.)
The difference between primary energy demand and fossil fuel supply (where we assume fossil fuel demand is equal to fossil fuel supply) is simply the non-fossil fuel demand.
In order to meet this demand, 2015 non-fossil fuel supply (77 EJ) would need to grow at the rates shown in the chart below. Note that the 2002-2015 average rate of non-fossil fuel growth was about 2.9% per year. Petroleum output grew by 6.5%/year on average from 1900-1972.
For comparison to past growth rates of energy output, the chart below compares oil and natural gas output (petroleum output) from 1921 to 1970 to the growth of non-fossil fuel energy in the scenario above from 2016 to 2065. The scenario is conservative relative to past history.
If the World wanted to reduce carbon emissions, a continued growth rate of non-fossil fuels from 2038 to 2063 of 5%/year could potentially reduce all fossil fuel use as an energy source to zero. In practice, fossil fuel use is unlikely to be reduced to zero until growth slows to a level where all steel can be produced from recycled materials, this is not likely before 2100. Lower population growth rates would reduce the need for economic growth and the need for steel output, in addition bio char is a possible substitute for coal in steel making which would require research and development.
A future transition to an economy using fewer fossil fuels will be necessary due to peak fossil fuels and such an energy transition may be possible, but is far from certain. High growth rates (5%/year or more) of non-fossil fuel energy after 2035 could reduce most oil and natural gas use by 2065, which would reduce carbon emissions to the atmosphere. A reduction of coal consumption might depend on increased steel recycling, lower economic growth rates, reduced use of steel in general, and the potential use of biochar as a substitute for coal where new steel is needed.
In a future post I will consider how variations on this basic transition scenario might influence future climate change by applying simple climate models such as CSALT and MAGICC.
An introduction to the oil shock model can be found here.
Medium oil scenario – URR=17,600 EJ through 2100
Carbon emissions are 330 Pg (or Gt) from 1870 to 2100. URR in barrels is 3100 Gb.
Medium Natural Gas scenario– URR =16,300 EJ through 2100
Carbon emissions 230 Pg from 1870-2100. URR is 15,200 TCF (trillion cubic feet) through 2100.
Medium Coal scenario– URR=16,400 EJ through 2100
Carbon emissions 410 Pg from 1770-2100, URR=390 Gtoe or 800 Gt from 1770-2100.
Other carbon emissions from land use change, natural gas flaring, and cement production from 1770 to 2100 estimated as 230 Pg C, with total carbon emissions of 1200 Pg from 1770 to 2100.
An estimate in 2009 by Allen et al suggests 1000 Pg of total carbon emissions has about a 50% chance of keeping warming below 2 C above preindustrial temperatures. If that estimate is accurate we would need to reduce fossil fuel emissions and cement emissions to zero by 2060 to have a 50% probability of remaining below 2C above pre-industrial temperatures for global surface temperatures, a difficult task at best.