Chapter 10
139
that determine our choices for energy sources? And do we have enough fossil fuel available for all of this to make any difference in terms of climatic consequences?
Between 1980 and 2002 the world’s energy consumption increased by 44% from about 10 TW (285 × 1015 Btu) to 14 TW (411 × 1015 Btu). The percentage of fossil fuels in the total energy mix over this period is shown in Figure 10.1: human consumption has shifted less than 5% away from fossil fuels. At least over this time period, the corresponding term in the IPAT equation described in the last chapter can be regarded as a constant.
The last term in the IPAT equation is CO2/fossil fuel— that is, how much CO2 we emit by burning a given fuel in order to get a unit of energy. The term is often referred to as carbon dioxide intensity, and similar to energy intensity, we want to keep it low. The term carbon intensity is also used to reflect CO2/gross domestic product (GDP), similar to the definition of energy intensity. In this book I will use it to measure CO2/energy exclusively. This is the only term in the equation that changes with the particular fossil fuel that we choose to use. Th e main fossil fuels currently in use are natural gas, petroleum, and coal. The distribution of the use of these fuels among the developing countries and the rest of the world is shown in Figure 10.2. In Box 10.1 I show how to calculate the CO2 intensities of the various fossil fuels.
Figure 10.1. Changes in the fraction of fossil fuels in the world’s energy consumption Source: US Energy Information Administration.1
Figure 10.2. Mix of consumed fossil fuels in 2002 for countries in the Organisation for Economic Co- operation and Development and the rest of the world Source: US Energy Information Administration.1
Box 10.1
In Box 2.1 of Chapter 2 I compared the two main life- supporting processes of respiration and photosynthesis with the burning of fossil fuels. For simplicity, the fuel that I used there was methane, the main constituent of natural gas.We burn methane to get energy through the following chemical reaction:
CH + 2O→ CO + 2HO + energy. [10.1]
42 22
The amount of energy that we get is 210 Cal/mole.
One mole of methane is 16 g (12 g from the carbon atom and 4 g from the four hydrogen atoms).
One mole of methane is responsible for the emission of 1 mole of CO2. One mole of CO2 is 44 g (12 g for the carbon atom and 32 g for two oxygen atoms). From these two relations we calculate the emission of CO2 per unit of energy:
44g CO2 0.21g of CO2
=
210Cal 1Cal of energ y
or, converting to Btus, – 0.053 g of CO2/1 Btu of energy (1 Btu = 0.252 Cal).
Let us now compare this with coal.The main constituent of coal is carbon.We burn carbon according to
C + O2 → CO2 + energy. [10.2]
The amount of energy is 94.4 Cal/mole.
Burning of 1 mole of carbon is responsible for the emission of 1 mole of CO2 that weighs 44 g, so the CO2 emission per unit of energy is
44g CO2 0.47g of CO2
=
94.4Cal 1Cal of energ y .
For a given energy output, we emit more than twice the amount of CO2 by burning coal as compared to natural gas. Petroleum is a mixture of carbon- hydrogen compounds of various lengths with carbon dioxide intensity intermediate between coal and natural gas.
These numbers approximately agree with published numbers (US Department of Energy) based on the actual composition of the fuels and are often referred to as “carbon coeffi cients.”1
Table 10.1 summarizes the CO2 intensities of the three main fossil fuels together with their 2004 percentage use (out of the total energy mix) and the 2004 price for constant energy (million Btu). The price comparison is complicated by many factors such as transportation, taxes, subsidies, producing capacity, regulation, and so forth. For comparison I have selected average US prices for industrial use. Some of these issues will be discussed in Chapter 12. If the only driving force in choosing among fuels is the price per unit of energy, then the incentive will be to use coal, the fuel with the highest CO2 intensity and thus the greatest climatic consequences. Coal is also the fuel with the most adverse environmental consequences, yet it is the least expensive fuel and, as we will see shortly, the fuel with the largest reserves. It is highly inconvenient to drive automobiles with coal as the fuel source. However, one can use technology, to be explored in the next chapter, that enables us indirectly to accomplish this. Such technology comes with a price tag that changes the price comparison. The rest of this chapter will attempt to assess how far into the future we can rely on fossil fuels.
Table 10.1.
CO2 intensities, price, and use of fossil fuels in 2004
| Fuel | CO2 intensity (gCO2/Btu) | 2004 price ($/MBtu) | 2004 use (%) |
|---|---|---|---|
| Petroleum | 0.078 | 9.5 | 39 |
| Natural gas | 0.053 | 7.2 | 23 |
| Coal | 0.12 | 2.5 | 24 |
We have already encountered the “magic” of exponential growth in the previous chapter and in Chapter 2. A simple, familiar example is a bank deposit. If the bank pays a constant 5% interest, in 14 years the deposit of $100 will double to $200. In 14 more years the deposit will double again to $400 and so on. In principle, this is an easy way to accumulate money, but, as usual, if things seem too good to be true, then they usually are. Usually these “unlimited growth” mechanisms hit a wall. In the case of the bank deposit, we have to pay taxes on the interest, and if the inflation rate happens to be a bit larger than the interest we receive, then after 28 years we will still get the $400 but our purchasing power with these dollars will be considerably lower than our original $100, and we will end up with a loss. This is the reason that the data on the standard of living presented in the previous chapter were adjusted to the rate of inflation by presenting them in constant US dollars. Items grow exponentially if the growth depends on the size of present quantity; items grow linearly if the additions are a constant quantity independent of the existing pool. Populations grow exponentially because babies are born to existing couples: the more couples, the more babies, provided that each couple has on average the same number of babies. Mathematically, it is relatively simple to construct a model based on such dynamics to predict a future critically dependent on population growth. The person credited with starting this kind of modeling is the British economist Robert Malthus (1766– 1834). According to Malthus, populations grow exponentially whereas the food to supply them grows linearly. W henever the food supply grows, the population growth accelerates to match. On the other hand, the only accommodating mechanism for a population that grows faster than its food supply is a catastrophic collapse through hunger, war, or disease. Such a model is relatively simple to compute, and, indeed, as soon as digital computers started to play a major role in economic modeling, these kinds of “limit to growth” projections became very popular. The best known of the early computer models was a report titled Limits to Growth published by an organization called the Club of Rome in 1972.2 They extended the Malthusian limit of food supply to natural resources, but the conclusions were very similar. The report’s main conclusion was that finding additional natural resources was not going to solve the problem because the much faster exponential population growth would soon outpace it. The only solution was to limit population growth.
However, around 1956 an American geologist by the name of M. King Hubbert, who was working in the oil industry, did a similar calculation concentrating on American oil reserves. In his case the exponential growth was not the population but the related growth in energy demand. Based on such modeling he predicted that the US oil supply will have a bell-shaped curve, such as the one shown in Figure 10.3, with a peak around 1970.
Hubbert’s work did not attract much attention at first. However in 1973 the oil- producing Arab countries declared an embargo on the shipment of oil to the West, the result of which were major shortages in supply, and not long after the price of oil increased by a factor of 10 (from around $3/barrel in 1973 to $38/barrel in 1982, back to around $10/barrel in 1997, climbing to $57/barrel in 2005, when this chapter is writt en, without an embargo). People have started to pay att ention.3
The Malthusian Club of Rome scenarios have strong opponents among economists and other social scientists and are still topics of major debates. Their predictions of the timing of various collapse scenarios were never accurate enough to be directly tested. However, Hubbert’s predictions about the US oil supply, after minor adjustments, were found to accurately reflect the observations of available supply. Presently there are attempts to extend Hubbert’s analysis to global availability of fossil fuels. The climatic consequences of such limits will be discussed next.
Oil annual production (normalized)
1.0
0.8
0.6
0.4
0.2
0.0
−100 −50 0 50 100 t − t m (years)
Figure 10.3. Schematic presentation of the Hubbert peak
The hidden term that does not show explicitly in the IPAT equation (previous chapter) but has a profound effect on the energy (i.e., “technology ”) terms in the equation is the supply of fossil fuel. Do we have enough fossil fuel to raise the global GDP in a sustainable way, keeping the other terms in the equation constant? Table 10.2 provides us with yearly energy use and emission data that will serve as a base line for calculating the available supply and corresponding potential CO2 emissions.
Table 10.3 shows the proven reser ves of conventional fossil fuels (oil, natural gas, and coal) in 2002. The last column in this table is marked R/P (reserves/yearly production). This ratio approximately indicates the number of years that the reser ves will last at current production levels.
We expect that oil and natural gas will last about 50 years and coal about 200 years. As are many of the statistical indicators that we use here, the numbers are “soft .” The data for the proven oil reserves are probably softer than other data. A good measure of the “softness” of the data is to list the changes in the estimates over time. Table 10.4 provides such tabulation for oil for different geographic regions and for the world at large. One can observe some interesting
Table 10.2.
Global energy use and emission in 2005
| Energy use (Btu) | 4.7 × 1017 |
| Carbon dioxide emission | 7.9 Gt-C |
| Change in the atmospheric concentration due to addition of carbon dioxide | 1 ppmv = 1.5 × 1014 moles of carbon = 1.8 Gt-C |
Source: World Bank.4
Proven “conventional” world fossil fuel reserves in 2002
| Fossil fuel | Proven reser ves | R/P (years) |
|---|---|---|
| Oil | 16 | 40.6 |
| Natural gas | 14.7 | 60.7 |
| Coal | 710 | 204 |
Note: In units of multiples of 1999 energy use. Source: British Petroleum.5
Proven oil reserves in billions of barrels
| Region | 1982 | 1992 | 2001 | 2002 | R/P |
|---|---|---|---|---|---|
| North America | 1.4 | 1.4 | 1.0 | 0.76 | 10.3 |
| South and Central America | 0.46 | 1.1 | 1.5 | 1.5 | 42 |
| Europe and Eurasia | 1.3 | 1.1 | 1.3 | 1.5 | 17 |
| Middle East | 5.6 | 10.1 | 10.4 | 10.4 | 92 |
| Africa | 0.9 | 0.9 | 1.2 | 1.2 | 27.3 |
| Asia- Pacific | 0.6 | 0.7 | 0.7 | 0.6 | 13.7 |
| World | 10.3 | 15.3 | 16 | 16 | 40.6 |
Note: In units of multiples of 1999 energy use. Source: British Petroleum. 5
patterns: About 60% of the world’s proven reserves are in the Middle East, so the world’s estimate of proven reserves is critically dependent on the estimate of Middle East reserves. Between 1982 and 1992, there is a jump by a factor of almost 2 in the proven oil reserves in the Middle East. If one normalizes for that jump (by extrapolating the 1992– 2002 average backward to 1982), then the data seem very consistent. Nevertheless, “proven reserves” is more of an economic term than a scientific term. It enters into the asset allocation of oil companies and is thus regulated by rules that include the probability of tapping the reserves in the near term. The probability of using the reserves will in turn depend on the market prices of the fuels and the extraction costs. Data are also supplied by some of the countries that own the reserves. High estimates increase the nominal wealth of the country and thus enable it to obtain a better credit rating, A different measure, of somewhat bett er objectivity, is the amount of fuel in the ground. Table 10.5 shows an estimate of ultimate fossil fuel energy reserves. Again, the estimate of “ultimate” should be regarded with some suspicion mainly because it does not take into account some of the factors the “proven reserves” estimate does. If the fuel’s extraction from the ground takes more energy than the amount of energy expected to be generated from the fuel, then it can hardly be counted as fuel reserve. But the cost and amount of energy needed to extract a fuel source depend on the extraction technology, which improves with time.
The values in column 2 of Table 10.5 are adjusted to common energy units, and column 3 shows the amount of the released carbon. The last column in the table estimates the increase in the CO2 atmospheric concentration that will result from burning the fuel, assuming that only 50% of the released CO2 will stay in the atmosphere. The conversion to atmospheric
Table 10.5.
Ultimate energy reserves
| Fossil fuel | Energy reserves (units of 1999 energy use) | Carbon (1011 tons C) | Estimated increase in atmospheric concentration (ppmv) |
|---|---|---|---|
| Oil | 31.6 | 2.7 | 75 |
| Natural gas | 26.3 | 1.5 | 42 |
| Coal | 552 | 80 | 2222 |
| Oil shale | 9.7 | 3 | 83 |
| Tar sand | 5.5 | 1 | 28 |
| Methane hydrate | 660,000 | 37,000 | 1,025,000 |
Source: US Geological Survey.6
concentrations is based on the technique explored in Box 2.5 in Chapter 2. Th e conclusion from these numbers is very clear and seems robust enough to withstand the softness of the numbers in the table: oil and natural gas will add only 120 ppmv to the atmosphere. Following our discussion in Chapter 8, such an increase is currently modeled to raise the average global temperature by about 1°C. If coal, oil shale, and tar sand are added to the mix, then the added carbon dioxide could increase atmospheric concentrations by as much as 2500 ppmv. Such an increase could raise the average global temperature by as much 10°C and will open the possibility of a Venusian fate. The use of methane hydrate, which a number of sites are already developing, opens the possibility for a much faster time line to reach such a fate. Th e R/P values given in Table 10.3 provide us with a time of about two generations to reach a policy decision to prevent such an occurrence. This is the fork at the end of now. Th e next chapter will explore alternative energy scenarios, and the following two chapters will explore available mechanisms to manage such a feeding transition.
