The scenarios cover a wide range of driving forces, from demographic to social and economic developments. This section summarizes the assumptions on important scenario drivers. For simplicity, only three of these are presented separately here following the exposition in Chapters 2, 3, and 4. Nonetheless, it is important to keep in mind that the future evolution of these and other main driving forces is interrelated in the SRES scenarios (see Tables TS-2 and TS-3 for a summary of the ranges of the main driving forces across the scenario groups in 2100).

The SRES scenarios span a wide range of assumptions for the most salient scenario drivers, and thus reflect the uncertainty of the future. Evidently, views of the future are a time-specific phenomenon, and this report and its scenarios are no exception. However, it is important to emphasize that this is an explicit part of the Terms of Reference for the SRES writing team - to reflect a range of views, based on current knowledge and the most recently available literature (see Appendix I). The scenario quantification results reflect well the literature range, except for extreme scenarios.

Three different population trajectories were chosen for SRES scenarios to reflect future demographic uncertainties based on published population projections (Lutz, 1996; UN, 1998; see Chapter 3). The population projections are exogenous input to all the models used to develop the SRES scenarios. The models used do not develop population from other assumptions within the model. Figure TS-3 shows the three population projections in comparison with the three population projections used in the IS92 scenarios. Global population ranges between seven and 15 billion people by 2100 across the scenarios, depending on the rate and extent of the demographic transition. The insert in Figure TS-3 shows population development in the developed (industrialized) regions. The range of future populations is smaller than in the IS92 scenarios, particularly in the developed (industrialized) regions, for which the lowest scenario indicates a very modest population decline compared to IS92 scenarios. The greatest uncertainty about future growth lies in the developing regions across all scenarios in the literature. An equally pervasive trend across all scenarios is urbanization (see Chapter 3). Altogether, three different population projections were used in the 26 harmonized scenarios. Other scenarios explored alternative population projections consistent with the storylines.

The lowest population trajectory is assumed for the A1 and B1 scenario families and is based on the low population projection in Lutz (1996), which combines low fertility with low mortality and central migration rate assumptions. After peaking at 8.7 billion in the middle of the 21st century, world population declines to 7.1 billion by the year 2100. As discussed in Chapters 3 and 4, this population development is somewhat higher than the previous low population used in the IS92 scenarios. The B2 scenario family is based on the UN median 1998 population projection (UN, 1998). The global population increases to about 9.4 billion people by 2050 and to about 10.4 billion by 2100. This population scenario is characteristic of recent median global population projections, which describe a continuation of historical trends toward a completion of the demographic transition that would lead to a constant global population, and is consistent with recent faster fertility declines in the world together with declining mortality rates. Hence, the population is somewhat lower than previous UN median projections, as used in the IS92 scenarios. This median scenario projects very low population growth in today's industrialized countries, with stabilization of growth in Asia in the second half of the 21st century and in the rest of the world towards the end of the 21st century. The A2 scenario family is based on the high population growth of 15 billion by 2100 reported in Lutz (1996), which assumes a significant decline in fertility for most regions and a stabilization at above replacement levels. It falls below the long-term 1998 UN high projection of 18 billion. It is also lower than in the highest IS92 scenario (17.6 billion by 2100). Nevertheless, this scenario represents very high population growth compared with that in current demographic literature. Some demographers attach a probability of more than 90% that actual population will be lower than the trajectory adopted in the A2 scenario family (Lutz et al., 1997). A more detailed discussion of the population projections used to quantify the four scenario families is given in Chapters 3 and 4.

The SRES scenarios span a wide range of future levels of economic activity (expressed in gross world product). The A1 scenario family with a "harmonized" gross world product of US$529 trillion (all values in 1990 US dollars unless otherwise indicated) in 2100 delineates the SRES upper bound, whereas B2 with "harmonized" US$235 trillion in 2100 represents its lower bound. The range of gross world product across all scenarios is even higher, from US$197 to 550 by 2100.

Although the SRES scenarios span a wide range, still lower and higher gross world product levels can be found in the literature (see Chapters 2, 3, and 4). Uncertainties in future gross world product levels are governed by the pace of future productivity growth and population growth, especially in developing regions. Different assumptions on conditions and possibilities for development "catch-up" and for narrowing per capita income gaps in particular explain the wide range in projected future gross world product levels. Given a qualitatively negative relationship between population growth and per capita income growth discussed in Chapters 2 and 3, uncertainties in future population growth rates tend to narrow the range of associated gross world product projections. High population growth would, ceteris paribus, lower per capita income growth, whereas low population growth would tend to increase it. This relationship is evident in empiric data - high per capita income countries are generally also those that have completed their demographic transition. The affluent live long and generally have few children. (Exceptions are some countries with small populations, high birth rates, and significant income from commodity exports.) This relationship between affluence and longevity again identifies development as one of the most important indicators of human well being. Yet even assuming this relationship holds for an extended time into the future, its quantification is subject to considerable theoretic and empiric uncertainties (Alcamo et al., 1995).

Two of the SRES scenario families, A1 and B1, explicitly explore alternative pathways to gradually close existing income gaps. As a reflection of uncertainty, development catch-up diverges in terms of geographically distinct economic growth patterns across the four SRES scenario families. Table TS-1 summarizes per capita income for SRES and IS92 scenarios for the four SRES world regions. SRES scenarios indicate a smaller difference between the now industrialized and developing countries compared with the IS92 scenarios. This tendency toward a substantially narrower income "gap" compared with the IS92 scenarios overcomes one of the major shortcomings of the previous IPCC scenarios cited in the literature (Parikh, 1992).

IPCC Homepage