Manufacturing 40 30 20 10 0 -10 -20 -30 1965 1970 1975 1980

Figure 4: Discovery rate of oil trend, 1965-2002 Source: Heinberg (2004)

facilities is to find productive uses for the heat. This strategy, called co-generation or combined heat and power (CHP), is applicable in urban areas, industrial parks and in buildings generally, but its widespread application requires a major change in the structure of the electric power grid. Other industrial water uses include quenching of hot coke or red hot steel ingots, wood pulping, washing, rinsing and dyeing of textiles, tanning of leather and surface finishing of metals (including electroplating). These uses leave polluted and sometimes toxic waste streams that need treatment (which uses even more water), and whose costs in many instances are not reflected in the cost of production.

Reserves of easily recoverable oil are diminishing, stimulating technological innovation to extract oil from deep ocean underwater reservoirs and non-conventional sources, such as oil and tar sands, and natural gas from shale, as a close substitute for many uses of petroleum. Since the early 1980s the amount of new oil discovered each year has been less than the amount extracted and used (Figure 4). The overall peak is only a question of time. However, market forces including high prices may reduce demand and increase the use of substitutes, causing demand to peak before supply. Some think peak oil may still be 20 years in the future. Others think it has happened already (Campbell and Laherrère 1998; Campbell 2004; Heinberg 2004; Strahan 2007).

The energy and other costs of replacing oil exploration and development are rising. The energy return on investments in energy (EROIE) of oil discovered in the

1930s and 1940s was about 110, but for the oil produced in the 1970s it has been estimated at 23, while for new oil discovered in that decade it was only 8 (Cleveland et al. 1984). Decades ago, only 1 per cent of the energy in oil discovered was needed to drill, refine and distribute it, but since then the EROIE has declined drastically. In the case of deep-water oil, the EROIE is not above 10. For Canadian tar sands the EROIE appears to be only about 3, which means that a quarter of all the useful energy extracted is needed for the extraction itself. These costs are reflected in the rising price of oil (and gas, which is a partial substitute) and are a sign of increasing oil scarcity.

High quality metal ores are also gradually being depleted (OECD 2008). While absolute scarcity is not yet perceived as an immediate problem for most metals, the indicators on the life expectancy of reserves (cf Tables 1 and 2) show that lower grade ores must be used. However, in order to do so more energy is needed to extract the useful metal content, adding marginally to GHG emissions. And whilst metals appear above ground in our economies in increasing quantities, a UNEP Resource Panel report on metals has shown the opportunity for much improved recycling rates (UNEP 2010a). Metals such as iron and steel, copper, aluminum, lead and tin enjoy recycling rates that vary between 25 and 75 per cent globally, with much lower rates in some developing economies. Improved recovery and recycling rates are also important for high-tech specialty metals that are needed in manufacturing to make key components for products that range from wind turbines and photovoltaic panels to the battery packs of hybrid

253 1985 1990 1995 2000

Gigabarrels annually