Docstoc

Executive Summary

Document Sample
Executive Summary Powered By Docstoc
					SECOND DRAFT 1 December, 2003               Chapter 9 Provisioning and Regulating Ecosystems Services



  Chapter 9. Changes in Provisioning and Regulating Ecosystem
   Goods and Services and their Drivers Across the Scenarios
               MA Scenarios Report             Second Draft: 1 December, 2003


[NOTE TO READERS: This is a rough draft documentation of quantitative/qualitative
estimates of future provisioning/regulating ecosystem services. The text still needs to be
edited for language, re-organized, streamlined, and shortened, and redundancies need to
be removed. Overall results have not been synthesized, although in the longer sections I
have tried to summarize results for individual services. – JA]


Chapter 9. Changes in Provisioning and Regulating Ecosystem Goods and Services and their Drivers Across
the Scenarios ____________________________________________________________________________ 1
  9.1     INTRODUCTION ______________________________________________________________ 2
  9.2      INDIRECT DRIVERS OF ECOSYSTEM SERVICES ________________________________                      2
     9.2.1     Population ________________________________________________________________              2
     9.2.2     Economic Development ______________________________________________________              5
     9.2.3     Technological Change _______________________________________________________             7
     9.2.4     Socio-Political Drivers _______________________________________________________          8
  9.3      DIRECT DRIVERS OF ECOSYSTEM SERVICES __________________________________ 9
     9.3.1     Energy Use and Production ___________________________________________________ 9
     9.3.2     Emissions ________________________________________________________________ 10
     9.3.3     Climate Change and its Consequences _________________________________________ 11
     9.3.4     Land Use/Cover Change ____________________________________________________ 13
     9.3.5     Irrigated Area _____________________________________________________________ 14
  9.4      ECOSYSTEM SERVICES ______________________________________________________                    15
     9.4.1    Food ____________________________________________________________________                 15
     9.4.2    Fuel ____________________________________________________________________                 22
     9.4.3    Genetic Resources, Biochemicals, Ornamental Resources __________________________          24
     9.4.4    Freshwater Resources ______________________________________________________               25
     9.4.5    Air Quality Regulation ______________________________________________________             30
     9.4.6    Climate Regulation/Carbon Uptake ____________________________________________             33
     9.4.7    Water Regulation __________________________________________________________               34
     9.4.8    Erosion Control ___________________________________________________________               34
     9.4.9    Water Purification and Waste Treatment ________________________________________           34
     9.4.10   Biological Control _________________________________________________________              35
     9.4.11   Regulation of Human Diseases _______________________________________________              35
     9.4.12   Pollination _______________________________________________________________               39
     9.4.13   Storm Protection __________________________________________________________               39
  9.5     Linkage with Well-being ________________________________________________________ 40
  9.6     Review of Uncertainty __________________________________________________________ 40
  9.7     Cross-Cutting Synthesis ________________________________________________________ 41
SECOND DRAFT 1 December, 2003                      Chapter 9 Provisioning and Regulating Ecosystems Services



9.1      INTRODUCTION

The capacity of ecosystems to provide services is determined by many different direct and in-
direct driving forces operating at the local to global level (see Chapter 8). If these driving
forces significantly change in the future they will also catalyze changes in the provision of
ecosystem goods and services. In this chapter we present estimates of changing ecosystem
services from both the quantitative and qualitative viewpoint.

The quantitative conjectures come from a modeling exercise described in Chapter 4. The
modeling exercise had the following basic steps – First we derive a set of quantitative as-
sumptions for the indirect drivers of ecosystem changes (e.g. population and economic
growth). Second, information about indirect drivers are used to derive assumptions about the
direct drivers of ecosystem change such as energy use and irrigated area. In some cases (e.g.
land cover change) models are used to derive these direct drivers. Next the direct drivers are
input to a suite of numerical simulation models. These models generate first estimates of tem-
poral and spatial changes in a wide range of ecosystem services.

While quantitative information in this chapter is derived from the modeling exercise, the qua-
litative conjectures are deduced from both the storylines of the scenarios (Chapters 6 and 7)
and modeling results.

We begin this chapter with a discussion of the assumed changes in indirect and direct drivers
and then describe estimates for each of the provisioning and regulating ecosystem services in
turn. In this chapter we focus on results for 2050, which is a compromise between the shorter
time horizon of a typical agricultural or urban prospective study, and the longer time horizon
of climate impact studies. The year 2050 also gives us a long term perspective on the ecologi-
cal consequences of current actions and policies. Nevertheless, where appropriate we also
provide information about the year 2100 and temporal trends throughout the 21st century.


9.2      INDIRECT DRIVERS OF ECOSYSTEM SERVICES

The key indirect driving forces of the MA scenarios include population, income, technologi-
cal development, and changes in human behavior. (See Chapter 8 for background information
about driving forces.) Assumptions for the future evolution of these driving forces cover a
wide range of assumptions, consistent with the storylines of the four scenarios.

9.2.1 Population 1

9.2.1.1 Methodology and Assumptions

Change in population is important because it will influence the number of consumers of future
ecosystem services. Furthermore, it will directly affect the amount of energy used, the magni-
tude of air and water pollutant emissions, the amount of required land, and the other direct
drivers of ecosystem change.

Four population projections were newly developed for the Millennium Assessment scenarios.
All four projections are based on the IIASA 2001 probabilistic projections for the world (Lutz
et al., 2001), but are designed to be consistent with the four MA storylines. (Table 9-1).
1
    Quantitative results in this section are based on the IIASA population model.

                                                          2
SECOND DRAFT 1 December, 2003            Chapter 9 Provisioning and Regulating Ecosystems Services




The first step in deriving the projections was to make qualitative judgments about the magni-
tude of fertility, mortality, and migration in 13 world regions for each of the MA storylines.
Next the qualitative assumptions were converted into quantitative assumptions using an ap-
proach based on conditional probabilistic projections (See Chapter 8) Using this approach, the
high/medium/low categories were mapped to three evenly divided quartiles of the uncondi-
tional probability distributions, as defined in the IIASA projections, for each component of
population change. Single, deterministic scenarios for fertility, mortality, and migration in
each of 13 regions were derived for each storyline, defined as the medians of the conditional
distributions for these variables. Population projections for each MA scenario were then pro-
duced based on the deterministic scenarios for each component of population change. Re-
gional population projections were then downscaled to the country level. More information on
the methodology for deriving population projections is given in Chapter 8.

Table 9-1 lists the qualitative assumptions about fertility, mortality and migration for each
storyline. These assumptions are expressed qualitatively as High/Medium/Low and in relative
rather than absolute terms. That is, a high fertility assumption for a given region means that
fertility is assumed to be high relative to the median of the probability distribution for future
fertility in the IIASA projections. Since the storylines describe events unfolding through
2050, the demographic assumptions specified here apply through 2050 as well. For the period
2050-2100 assumptions were assumed to remain the same, in order to gauge the consequences
of trends through 2050 for the longer term. This is not intended to reflect any judgment re-
garding the plausibility of trends beyond 2050.

Assumed fertility and mortality in currently high fertility countries: Trends in these rates were
based on demographic transition reasoning. In Global Orchestration (GO), higher invest-
ments in human capital (especially education and health) and greater economic growth rates
are assumed to be associated with a relatively fast transition, implying lower fertility and mor-
tality than in a central estimate. In Order from Strength (OS), lower investments in human
capital and slower economic growth lead to a slower transition (i.e., higher fertility and mor-
tality). Techno Garden (TG), with more moderate investments and economic growth assump-
tions, is assumed to undergo a moderate pace of change in both fertility and mortality. The
Adapting Mosaic (AM) storyline begins similarly to Order from Strength but diverges later
because large investments in education payoff in an acceleration of economic growth and
technological development in all regions. Demographic trends are therefore specified to fol-
low Order from Strength for 10 years, and then diverge to ―medium assumptions‖ of mortality
and fertility by mid-century.

Assumed fertility in currently low fertility countries: The determinants of long-term fertility
trends are poorly known, and therefore there is little basis for preferring one set of assump-
tions over another for a given storyline. In the face of this uncertainty the overarching ratio-
nale for specifying trends for given storylines was chosen to be the scope of convergence in
fertility across low fertility countries. Since Order from Strength describes a regionalized, di-
vergent world, and Global Orchestration a globalizing, convergent world, these characteris-
tics were applied to future fertility. Thus the low fertility countries were divided into two
groups (one with ―Very Low Fertility‖, one with ―Low Fertility‖, see note to Table 1), and
fertility assumptions were adopted such that fertility in these two groups would tend to con-
verge in the Global Orchestration scenario and diverge in the Order from Strength scenario.
In the Adapting Mosaic scenario, fertility initially follows the Order from Strength assump-
tions, then diverges toward medium levels. In the Techno Garden scenario, medium fertility
is assumed.

                                               3
SECOND DRAFT 1 December, 2003             Chapter 9 Provisioning and Regulating Ecosystems Services




Mortality in industrialized country regions: Mortality was assumed to be lowest in the Global
Orchestration scenario, consistent with its high economic growth rates. Technological
progress is assumed to occur in the health sector as well, and reduction in inequality within
the region. In contrast, Order from Strength, which assumes growing inequality within the
industrialized countries and even the potential for re-emergence of some diseases, is assumed
to have the highest mortality. Techno Garden assumes a medium pace of mortality change,
and Adapting Mosaic follows the Order from Strength assumptions for 10 years before di-
verging to medium levels in 2050.

Migration: Net migration rates are assumed to be low in the regionally oriented scenarios
(Adapting Mosaic and Order from Strength), consistent with higher barriers between regions.
In Global Orchestration, permeable borders and high rates of exchange of capital, technology,
and ideas are assumed to be associated with high migration. Techno Garden assumes a more
moderate migration level.

9.2.1.2 Population Projections

Population size

Table 9-2 shows results for global population size for 2050. The range between the lowest
(GO) and the highest (OS) scenario is 8.1 to 9.6 billion in 2050, and 6.8 to 10.5 billion in
2100. These ranges cover 50-60 percent of the full uncertainty distribution for population size
in the IIASA projections. The primary reason that these scenarios do not fall closer to the ex-
tremes of the full uncertainty distribution is that they correlate fertility and mortality: the Or-
der from Strength scenario generally assumes high fertility and high mortality, and the Global
Orchestration scenario generally assumes low fertility and low mortality. Both of these pairs
of assumptions lead to more moderate population size outcomes.

The Adapting Mosaic scenario is nearly identical to the Order from Strength scenario at the
global level over most of the century, even though it is designed to follow the Order from
Strength scenario only for 10 years and then diverge from it. The reason for this outcome is
that (a) the effects of deviations in fertility in the Adapting Mosaic scenario do not become
apparent in population size for many decades due to population momentum, and (b) both fer-
tility and mortality trends diverge. Thus, although fertility declines in Adapting Mosaic rela-
tive to Order from Strength after 2010, tending (eventually) toward a smaller population size,
mortality declines relative to Order from Strength as well, tending toward a larger population
size. The net result is little difference, especially in the short to medium term.

The relationship across scenarios differs by region. While in developing country regions the
ranking is the same as in the global results (i.e., the Global Orchestration scenario produces
the lowest population size, and the Order from Strength scenario the highest), this ranking is
reversed in many of the industrialized country regions (Western Europe, Eastern Europe, So-
viet Europe, and Pacific OECD). The main reason is that the Order from Strength scenario is
assumed to have divergent fertility trends among the industrialized country regions coupled
with low migration. Thus, those regions with currently very low fertility rates (less than 1.5
births per woman) are projected to see little change in fertility levels in the future, maintaining
the fertility difference between these regions and North America and China, where fertility
remains around replacement level of about 2 births per woman. These assumptions, in the ab-
sence of countervailing increases in net migration into the region, produce substantial popula-
tion declines in the very low fertility regions. For example, in Western Europe population de-

                                                4
SECOND DRAFT 1 December, 2003            Chapter 9 Provisioning and Regulating Ecosystems Services


clines by nearly 20 percent by 2050, and by more than 50 percent by 2100. Declines are even
greater in other European regions. In contrast, in the Global Orchestration scenario, fertility
rates are assumed to converge across the industrialized country regions, leading to increases
in those regions where fertility is currently very low. In addition, migration into the region is
assumed to be high in this scenario. The combined effect is to make Global Orchestration the
highest population scenario for the industrialized country regions.

The range of outcomes for one region, North America, is particularly small over all four sce-
narios, despite widely differing sets of assumptions about input variables. The reason is that
assumptions about the different components of population change, as dictated by the story-
lines, tend to offset each other. When fertility is assumed to be low, mortality is low as well,
and migration (which has a substantial influence on population growth in this region) is high.
A similar situation holds, in reverse, when fertility is high. Thus the range of population size
outcomes is only 426 to 439 million in 2050, and 420 to 540 million in 2100.

In Sub-Saharan Africa, the HIV/AIDS epidemic takes a heavy toll in all scenarios. Life ex-
pectancy for the region as a whole is assumed to decrease, and not to return to current levels
for 15 to 25 years, depending on the scenario. In individual countries where HIV prevalence
rates are highest, population is projected to decline. However the population of the region as
a whole is projected to grow in all scenarios, driven by the large countries of the region whose
prevalence rates are estimated to be relatively low and either past or near their peaks (UN,
2003), the momentum inherent in the young age structure of the region, and by relatively high
fertility.

Aging

The age distribution of the population will have an important influence on the vulnerability
and adaptive capacity of society. This is reflected, for example, in the computation of the
number of malnourished children in Section 9.4.1.

In all scenarios, substantial aging of the population occurs. The least amount of aging occurs
in the Order from Strength scenario, due to its high fertility and mortality assumptions in the
developing country regions, but even in this case the proportion of the population above age
65 more than doubles from about 7 to 17 percent by 2100 (Figure 2). In the Global Orchestra-
tion scenario, the proportion above age 65 triples by 2050 (to 22 percent) and increases by a
factor of 6 (to 42 percent) by 2100. This result is driven by low fertility assumptions in de-
veloping country regions, along with low mortality assumptions for all regions of the world.
Within these general trends at the global level, results vary by region. In all industrialized
country regions, the proportion over 65 doubles to at least 30 percent by 2100 in all scenarios
(the only exception is the Order from Strength scenario in North America). In contrast, while
aging is extraordinarily fast in developing country regions in most scenarios – the proportion
over 65 increases, for example, from 5 percent currently to over 40 percent by the end of the
century in the Global Orchestration scenario – in the Order from Strength scenario the older
age group never accounts for more than 20 percent of the population in any of these regions.
In fact in Sub-Saharan Africa, where fertility and mortality are the highest, little aging occurs
over the first half of the century in any scenario, and even by the end of the century, the pro-
portion over 65 reaches only 22 percent in the most extreme outcome (the Global Orchestra-
tion scenario).

9.2.2 Economic Development


                                               5
SECOND DRAFT 1 December, 2003            Chapter 9 Provisioning and Regulating Ecosystems Services


9.2.2.1 Methodology and Assumptions

Assumptions about future economic development influence the future of ecosystem services
by affecting the direct drivers of ecosystem changes such as energy use and food consump-
tion, and indirect drivers such as technological progress. The relationship between income de-
velopment and drivers differs greatly between ecosystem services. For several services, model
calculations assume that the higher the income, the greater the per capita consumption of
commodities up to some saturation level (e.g. energy consumption per sector). For other ser-
vices high income may lead to a decrease in consumption (e.g. domestic water use, or fuel
wood consumption).

The MA scenarios for income cover a range of economic growth rates consistent with the
scenario storylines described earlier in this report. Table 9-3 shows the qualitative assump-
tions for economic variables fitting to these storylines. Using these assumptions together with
the World Bank’s economic prospects to 2015 and IPCC's SRES scenarios as starting points,
we have selected economic growth rates for each scenario (Table 9-4). Assumptions range
from high economic growth for Global Orchestration and low economic growth for Order
from Strength, with Techno Garden and Adapting Mosaic falling between (and partly branch-
ing off of these).

9.2.2.2 Economic Projections

In Global Orchestration economic growth is spurred above historic averages in many parts of
the world by a combination of further trade liberalization, economic cooperation and rapid
spread of new technologies. The OECD regions have a per capita growth rate of about 2.4
percent per year in the 2000-2025 period, slowing down to a level around 1.8 percent per year
afterwards (Table 9-4). The Asian economies return to their rapid growth during the 1980s
and 1990s during most of the this period (with growth rates between 5 to 6 percent per year).
The Latin American region overcomes its debt and balance of trade problems – and finds it-
self back on track with strong economic growth. Africa goes carries out institutional reforms
which form the basis for strong economic growth after 2025 when it finally is able to exploit
its rich natural and human capital. After 2025 Africa achieves growth rates similar to Asian
economies in the 1980s and 1990s. As low income countries grow much faster than other
countries, the income gap between richer and poorer regions closes in relative terms – but
hardly in absolute terms (Table 9-5). In all scenarios, growth rates for the FSU are relatively
high because this region uses its highly skilled labor force to recover from the economic
downturn of the 1990s.

Economic development in Techno Garden follows a similar pattern to Global Orchestration,
but with lower growth rates from 2000 to 2050 (Table 9-4). By the end of the period, howev-
er, earlier investments in technology and education pay off with higher economic growth rates
similar to Global Orchestration.

Under the Order from Strength scenario, global economic growth (Table 9-4) is sluggish
(staying below historic rates) because of the low level of international trade (except for food
staples) and limited exchange of technology. The high income countries manage to maintain a
growth rate of 1.9 percent per year during the first half of the century, but this drops to 1.2
percent per year in the second half. The income gap between rich and poor regions widens be-
tween 2000-2025 (Table 9-5). Despite the sluggish economy, average GDP per person in-
creases by a factor of two between now and 2050 (Table 9-5).


                                               6
SECOND DRAFT 1 December, 2003             Chapter 9 Provisioning and Regulating Ecosystems Services


The Adapting Mosaic scenario initially follows the pattern of the Order from Strength scena-
rio, but because of large investments in education and health care, economic growth rates pick
up over time and approach those of the Techno Garden scenario in the last half of the century
(Table 9-4).

Uncertainty of Income Assumptions. Assumptions made for 17 large regions and then down-
scaled to all countries … Consistent with other global studies. But this is also obviously a
simplification, but no strong basis for making an assumption about a particular country.

9.2.3 Technological Change

The rate of technological change is an indirect driver of changes in ecosystem services be-
cause it affects the efficiency by which ecosystem services are produced or used. For exam-
ple, a higher rate of technological change leads to more rapid increase in the yield of crops per
hectare or in the efficiency of water use by power plants. The rate of technological change is
assumed to be highest under the Global Orchestration scenario and lowest under the Order
from Strength scenario. The rate of change accelerates under the Techno Garden and Adapt-
ing Mosaic scenarios but slows under the Global Orchestration and Order from Strength sce-
narios.

9.2.3.1 Energy

Energy Efficiency Improvements
Short piece to be written

9.2.3.2 Water

Improvements in Efficiency of Water Use
Short piece to be written

9.2.3.3 Agriculture

a. Crop yield improvements
Short piece to be written

b. Basin-level Irrigation Efficiency (too long)

Under the Order from Strength scenario government budgetary problems are assumed to wor-
sen resulting in dramatic government cuts on irrigation system expenditures. Water users
fight price increases, and a high degree of conflicts results in lack of local water-user coopera-
tion for cost-sharing arrangements. The turnover of irrigation systems to farmers and farmer
groups is accelerated but not accompanied by the necessary reform of water rights and neces-
sary funding. Rapidly deteriorating infrastructure and poor management combined reduce
system- and basin-level water use efficiency under this scenario. As a result, efficiency le-
vels, which are already quite low in most of Asia, drop by 23-28 percent to reach levels of on-
ly 0.25-0.30 by 2050. In East and South Africa, levels decline slightly less, by about 20 per-
cent, to reach 0.44 by 2050. Declines are similar in Latin America, to reach 0.32-0.34 by
2050. In MENA, where efficiency levels are very high at 0.6-0.7 in 2000, levels are expected
to decline to 0.56 by 2050. Efficiencies are more resilient in industrialized countries as elites
concentrate resources on some systems to maintain minimum food self-sufficiency levels. As


                                                  7
SECOND DRAFT 1 December, 2003             Chapter 9 Provisioning and Regulating Ecosystems Services


a result, levels decline by about 8 percent in the OECD, and slightly more, about 15 percent in
the FSU.

Under the Global Orchestration scenario, careful market-oriented reform in the water sector
and more comprehensive and coordinated government action will lead to greater water man-
agement investments in efficiency-enhancing water and agricultural technology. The effec-
tive price of water for the agricultural sector is assumed to increase more rapidly to induce
water conservation as well as to free up agricultural water for other environmental, domestic,
and industrial uses. Large investments in the countries of the South lead to rapid increases in
efficiency levels in Asia and Sub-Saharan Africa, where levels increase to as high as 0.4-0.5
and 0.56-0.74, respectively. Selected—economically viable—investments also enhance effi-
ciency levels in those countries and regions, where relatively high efficiency had been
achieved by 2000, including the OECD and MENA, although increases are small. The high-
est irrigation efficiency level is achieved in the region facing greatest water scarcity, North
Africa at 0.8.

Under the Techno Garden scenario, technological innovations for on-farm efficiency increas-
es help boost irrigation efficiency levels across the world to previously unseen levels. More-
over, river basins make progress toward more integrated basin management through real-time
measuring and management of water resources. Gradual introduction of water price increases
in some agricultural areas induce farmers in these regions to use water more efficiently. As a
result, high efficiency levels are reached, particularly in those regions, where no or little fur-
ther improvement had been expected, like the OECD and MENA. There, levels increase to
0.75-0.9 by 2050. Advances are also significant in Asia, reaching 0.5 by 2050, and in Sub-
Saharan Africa, where levels of 0.75 can be reached.

Under the Adapting Mosaic scenario, local adaptations, including expansion of water-
harvesting and other water conservation technologies as well as the increased application of
agroecological approaches help boost efficiency levels in some regions and countries. Effi-
ciency increases achieved—although important—remain scattered in areas and regions within
countries, and the global and regional impacts are smaller than under the Techno Garden and
Global Orchestration scenarios. In Asia, the results are mixed with increases in efficiency in
East Asia balanced by declines in South Asia. In Sub-Saharan Africa, the outcomes are more
favorable, with conservation strategies boosting efficiency levels by 2-11 percent. The FSU is
less successful with efficiency-enhancing methods, experiencing a slight decline in efficiency
levels. The OECD region, as a whole, does successfully apply locally developed irrigation ef-
ficiency enhancing methodologies, with increased efficiency levels in the some countries
more than balancing declines in other countries.

9.2.4 Socio-Political Drivers

9.2.4.1 Energy

Tendency towards energy conservation
Short piece to be written

9.2.4.2 Water

Structural Changes of Water Use in Domestic and Industrial Sectors
Short piece to be written


                                                8
SECOND DRAFT 1 December, 2003                     Chapter 9 Provisioning and Regulating Ecosystems Services


9.2.4.3 Agriculture

Food Preferences
Short piece to be written


9.3      DIRECT DRIVERS OF ECOSYSTEM SERVICES

Here we estimate future trends of some of the most important driving forces of ecosystem
change – energy use and production, air pollutant emissions, climate change, and the change
in land use and land cover. Changes in these aspects of the earth system have a direct effect
on ecosystem processes, and they are therefore termed ―direct drivers‖.

9.3.1 Energy Use and Production 2

The amount and kind of energy used in future will have many direct effects on ecosystem ser-
vices. In particular, the use of fossil fuel use will determine the rate of air pollutant emissions
and therefore air quality regulation (Section 9.5.5). The level of biofuel use will affect the
type and distribution of land cover, and the services provided by forest and other land cover
types (Section 9.4.2), while the magnitude of thermal-generated electricity will influence wa-
ter withdrawals (Section 9.4.4). Energy use is also one of the principal determinants of cli-
mate change, which itself is a direct driver of changes in ecosystem services (Section 9.4.3).

The amount of future energy use in the different scenarios will be influenced by the future
demand for energy (driven mostly by economic and population growth), by continuing im-
provements in the technological efficiency of energy use (Section 9.2.3), and by human beha-
vior, as in the motivation of energy consumers to conserve energy.

Global Orchestration. This scenario assumes relatively fast technological progress within the
energy sector driven by economic growth and the goal to provide low-cost energy for all
people with a high level of reliability. There is also optimism that environmental factors will
not have a serious impact on humans or ecosystems. Because of this optimism and the desire
to keep down fuel costs, there is no attempt to broaden international climate policy. Hence,
greenhouse gas and other emissions are not controlled except in OECD countries. Neverthe-
less, modernization of industry and households leads to some reductions in air pollutant emis-
sions. After finally being confronted with climate change, the decision is made that adaptation
is a much more cost-effective strategy (as change has become unavoidable). Thus, fossil fuel
use rapidly expands, in particular to produce more convenient end-use energy carriers such as
gaseous and liquid fuels and electricity. Total energy use increases up to 1200 exajoules by
2050 (as compared to a current level of 400 exajoules) and levels off towards the end of the
century. (Figure 9-1). In the second half of the century, new (non-fossil) fuel options rapidly
penetrate the market.

Techno Garden. In Techno Garden, society is convinced that environmental degradation de-
creases human well being, and therefore it supports long term reductions of greenhouse and
other air pollutant emissions. To mitigate climate change, the international community adopts
a goal of limiting the atmospheric concentration of total greenhouse gases (carbon dioxide and
equivalent gases) to a concentration of 550 parts per million by 2150 (carbon-dioxide equiva-
lent). Since emissions stem mostly from energy use, this atmospheric goal leads to a much

2
    Quantitative results in this section are based on the IMAGE model.

                                                         9
SECOND DRAFT 1 December, 2003                     Chapter 9 Provisioning and Regulating Ecosystems Services


slower growth in total energy use and the extensive use of non-fossil fuels (modern biofuels
and solar and wind energy, as examples) and lower carbon fuels (principally natural gas).
(Figure 9-1). Because of these shifts in the energy system, total energy use only reaches 510
exajoules in 2050, and slowly increases thereafter, despite relatively high economic growth
rates.

Order from Strength. In this scenario, global energy use at mid-century is 810 exajoules.
This is much lower than Global Orchestration because the Order from Strength scenario has
lower economic growth rates especially in developing countries. The use of fossil fuels is sub-
stantially higher than in the Techno Garden scenario because reducing greenhouse gas emis-
sions is not a goal of society in this scenario. Total energy use increases almost linearly
throughout the entire century. At the same time, the slow dispersion of new technologies, and
increased barriers for global energy trade (particularly important for natural gas and oil) result
in reliance on domestic energy sources which mean continued intensive use of fossil fuels in
many countries. For China and India this implies a continued reliance on coal.

Adapting Mosaic. This scenario is similar to the Order from Strength scenario in that it has
lower economic growth rates than Global Orchestration and lacks global climate policies.
Global energy use in 2050 (880 exajoules) is between Global Orchestration and Techno Gar-
den. But it differs from the Order from Strength scenario because local approaches are
adopted for improving efficiency of energy use and for exploiting environmentally-friendly
fuels. As a result total energy use stabilizes soon after mid-century and non-fossil fuels play
an increasing role in the energy economy.

9.3.2 Emissions

9.3.2.1 Greenhouse Gas Emissions3

The rate and intensity of climate change will be largely determined by the change in future
greenhouse gas emissions. Energy use, agricultural activity, industrial processes and defore-
station will be the main sources of these emissions. Emissions of methane and nitrous oxide
gases stemming almost entirely from agricultural sources will account for about one-fifth of
total global greenhouse gas emissions (in units of equivalent carbon dioxide) at mid-century
under the Global Orchestration scenario (Figure 9-2). Energy-related emissions will account
for almost all of the remaining emissions, and will determine the overall trend of greenhouse
gas emissions for all of the scenarios.

Under the Global Orchestration scenario greenhouse gas emissions peak at mid century at
24.5 gigatons, as compared to 10.5 gigatons in 2000. Emissions decline afterwards because
total energy use stabilizes and a greater percentage of low carbon fuels are used (Figure 9-2).

The strong climate policies in the Techno Garden scenario limit the increase in total energy
use, and in particular fossil fuel use, as noted above. Hence emissions grow much more slow-
ly, and already peak around 2020 at 11 gigatons. There is also a slower increase in agricul-
ture-related emissions because of more efficient use of technology in agriculture.

Meanwhile, the trend in emissions under the Order from Strength scenario follows the linear
increase in total global energy use of this scenario. Emissions almost double between 2000
and 2050, and again between 2050 and 2100.

3
    Quantitative results in this section are based on the IMAGE model.

                                                        10
SECOND DRAFT 1 December, 2003                     Chapter 9 Provisioning and Regulating Ecosystems Services




Global emissions peak at 17 gigatons around mid-century under the Adapting Mosaic scenario
and then gradually decline as energy growth slows and more low-carbon fuels are used.

9.3.2.2 Air Pollution Emissions

The emissions of sulfur dioxide, nitrogen oxides and other air pollutants lead to air pollutant
concentrations and deposition that can interfere with the ecosystem service of ―air quality
regulation‖. In Section 9.4.5 of this chapter we use them as indicators of this ecosystem ser-
vice.

9.3.3 Climate Change and its Consequences 4

9.3.3.1 Climate Change

The future rate of change of temperature and precipitation will be affected by the rate of
changing greenhouse gases in the atmosphere and other factors (for example, long-term cli-
matological driving forces such as variations in solar radiation). Although trends in emissions
vary considerably between the scenarios, the differences in global temperatures in 2050 are
not very large (Figure 9-3). By 2050 the most extreme scenarios show an increase in global
temperature (relative to current climate) of between 1.6 to 2.0 degrees. This relatively small
difference between scenarios is because of the lag time between the build-up of emissions in
the atmosphere and the response of the climate system to this build-up. But differences be-
tween the scenarios are sharper by the end of the century – under Techno Garden the increase
in global average surface temperature is slightly over 2 degrees Celsius ( oC) and nearly 3.5 oC
under the vigorous economic growth and high emissions of Global Orchestration (Figure 9-
3).

Although differences in total temperature change are appreciable only at the end of the cen-
tury, there are much sharper differences in their decadal rate of temperature change (Figure 9-
4). This is of particular importance from the point of view of climate impacts because it is
presumed that the faster the rate of climate change, the more difficult the adaptation of society
and nature to climate change. As a reference point, studies of past climate change suggest that
risks to ecosystems grow significantly when the increase in temperature is faster than about
0.1 oC per decade (Reference).

The rate of temperature change under the Techno Garden scenario becomes slower and slow-
er, reaching about 0.15 oC per decade in the middle of the scenario period. Meanwhile the rate
sharply increases under the Global Orchestration scenario until mid-century when it reaches
more than 0.35 oC per decade and then declines. Between these scenarios, the Adapting Mo-
saic scenario levels off at mid-century at around 0.28 oC per decade and then declines. Mean-
while at mid-century the Order from Strength scenario is lower (around 0.25) than Adapting
Mosaic but is still increasing so that it has the highest value of all scenarios (0.30) at the end
of the century.

Although these values may be uncertain, we expect (with medium certainty) that three of the
four scenarios will have rising temperature up to at least mid-century. Likewise, it is likely
that three out of four will have a declining rate of temperature increase after mid-century.



4
    Quantitative results in this section are based on the IMAGE model.

                                                        11
SECOND DRAFT 1 December, 2003                     Chapter 9 Provisioning and Regulating Ecosystems Services


While the computation of future regional temperature is uncertain, still more uncertain are the
computations of precipitation patterns within regions. Climate models can provide insight into
overall global and regional trends, but cannot provide accurate estimates of future precipita-
tion patterns when the landscape plays an important role (as in the case of mountainous or hil-
ly areas). Recognizing this uncertainty, we use a standard integrated assessment approach to
estimate uncertain, but plausible, future changes in precipitation.5 Figure 9-5 shows a typical
spatial pattern of changes in precipitation up to 2050. Note that approximately three-quarters
of the land surface has increasing precipitation, consistent with the expectation that a warmer
atmosphere will stimulate evaporation of surface water, increase the humidity of the atmos-
phere and lead to higher rates of precipitation overall. Nevertheless some arid areas become
even drier according to Figure 9-5, including the Middle East, parts of China, Southern Eu-
rope, the Northeast of Brazil, and west of the Andes in Latin America. This will increase wa-
ter stress in these areas as we will see in Section 9.5.4.

Uncertainty of Climate Change. The estimates of climate change have particularly high un-
certainty because different climate models give different intensities and patterns of climate
change. Generally the models give a more consistent picture for temperature change than for
precipitation. [Better -- global T global precp, more uncertain regional T, very uncertain re-
gional P. Uncertain change in CO2 uptake??]

Sea Level Rise 6

One of the major impacts of climate change (with high certainty) will be a rise in average
global sea level. Sea level will rise because warmer temperatures will melt permanent ice and
snow and cause a thermal expansion of ocean water. Furthermore, climate change may cause
stronger and more persistent winds in the landward direction along some parts of the coastline
and this will also contribute to rising sea level at these locations.

We have made a first order estimate of the expected (global average) sea level based on the
climate change scenarios corresponding to the four MA scenarios. The average rise up to
2100 ranges from 50 cm (Techno Garden) to 70 cm (Global Orchestration) (Figure 9-5). The
actual increase in different regions might be higher or lower, depending on changes in ocean
currents, prevailing winds, and land subsidence rates.

Note in Figure 9-6 that sea level still has a rising tendency at the end of the century although
Figure 9-3 indicates that air temperatures stabilize under three of the four scenarios. The in-
crease in sea level lags decades behind the increase in temperature because there is a long de-
lay in heating the enormous volume of the world’s oceans. This means that temperature could
stabilize over the course of the scenario period while sea level continues to rise. For example,
the trends shown in Figure 9-6 imply that sea level could further rise (with low to medium
certainty) by at least an additional 1 m during the course of the 22nd century.




5
  Precipitation is computed on a 0.5 x 0.5o latitude x longitude global grid by scaling precipitation patterns from
general circulation climate models to changes in global temperature. These climate models show both increasing
and decreasing precipitation over the long term, depending on location and the change in global greenhouse gas
emissions, According to the method used here, the higher the global temperature increase in a scenario, the
stronger the ―local‖ increases or decreases in precipitation.
6
    Quantitative estimates in this section are based primarily on the IMAGE model.


                                                        12
SECOND DRAFT 1 December, 2003                     Chapter 9 Provisioning and Regulating Ecosystems Services


9.3.3.3 Change in Potential Vegetation
Contribution to be edited

9.3.4 Land Use/Cover Change 7

A wide variety of factors determine present and future changes in land use and cover. Global
models aggregate these many factors into a few major categories, namely changes in extent of
agricultural and timberland (expansion or contraction) and climate change. When agricultural
land expands it does so at the expense of natural forest or grassland, and if it contracts it is re-
placed by natural or urban land. As climate changes it will change the natural vegetation at a
particular location as we noted above. Another important factor, the expansion of urban land,
is only estimated in a crude way by global models because of their rough spatial resolution.
Here we focus on changes in forest land and agricultural land (pasture land plus cropland for
food, feed, and biofuel crops).

Up to 2020

Although the scenarios have different rates of driving forces, three out of four scenarios have
similar global trends of land use/cover change over the next two decades (Global Orchestra-
tion, Techno Garden, Adapting Mosaic) (Table 9-6). We expect (with medium certainty) that
major differences between MA scenarios will only become apparent after these decades.

In the first decades of the scenario period all four scenarios have ongoing expansion of agri-
cultural land which occupies current forest and grassland. The demand for agricultural land is
partly for growing grains, feeds, and other foods (Table 9-6). But this is not the principal rea-
son because, as we note in Section 9.4.1, growing production is achieved mostly by intensify-
ing the use of existing farmland rather than occupying new cropland. A more important factor
is the increasing demand for pasture and rangeland to satisfy increasing meat demands in each
region and scenario. (Although livestock and other animals are partly fed by the feed grown
on cropland, they are still expected to require extensive hectares of pasture and rangeland.)
Another important factor is the increasing demand over the long run for land to grow crops
for biofuels (Table 9-6).

As a result, the rate of loss of forests in the early decades of the Twentieth Century is compa-
rable in most scenarios to that experienced during the last 30 years, despite ongoing and even
strengthened efforts for conserving natural areas in some of these scenarios. (We note that the
models used here do not explicitly account for conservation policies) (Table 9-5).

The Order from Strength scenario is an exception to the above trends and exhibits a faster rate
of deforestation at the beginning of the scenario period. (Table 9-7). This comes from the
faster expansion of agricultural land, resulting mainly from rapidly growing population com-
bined with slow improvements in crop yield in low-income regions. Since crop yield remains
low compared to increasing demand for food products, more agricultural land is needed (al-
though most increases here in crop production are also achieved through intensification of ex-
isting agricultural land).

After 2020




7
    Quantitative results in this section are based on the IMAGE model.

                                                        13
SECOND DRAFT 1 December, 2003             Chapter 9 Provisioning and Regulating Ecosystems Services


Order from Strength. After 2020 the Order from Strength scenario continues its rapid deple-
tion of forest area until mid-century when it slows because of slowing population growth and
increasing crop yield (Table 9-7). According to this scenario forest land in Sub-Saharan Afri-
ca disappears at a rate of 2.2 percent per year from 2020 to 2050 (compared to about 0.7 per-
cent per year from 1970 to 2000). Two-thirds of the central African forest present in 1995 dis-
appears by 2050 while about 40 percent of Asia’s forests and 25 percent of Latin America’s
forests are depleted (Table 9-6). During this period, forests disappear in Latin America at a
slightly higher rate than between 1970 and 2000. The deforestation rate is much slower in
FSU and OECD countries because their stabilizing domestic food demand and increasing crop
yields lead to the abandonment of some agricultural land.

Global Orchestration. By 2020 the Global Orchestration scenario begins to depart from the
other scenarios. Agricultural area expands at a fast rate, similar to Order from Strength, but
for different reasons. Here, rapid income growth and stronger preferences for meat result in
growing demand for food and feed, and this leads to a rapid expansion of crop area in all re-
gions. Crop land expands in the FSU and OECD to cover the export of food to other regions.
Forests disappear at a slower rate than in the Order from Strength scenario, but still higher
than current global rates (Table 9-7). About 50 percent of the forests in Sub-Saharan Africa
disappear between 2000 and 2050. By 2100, the area remaining of global forests is only
slightly above that of the Order from Strength scenario.

Techno Garden. The Techno Garden scenario results in the lowest conversion of natural land
to agricultural land. The assumption of active intervention in managing and protecting ecosys-
tems leads to lower meat consumption (since less land is usually needed for producing meat-
versus non-meat products). This is partly offset, however, by strong economic development in
lower-income regions resulting in a strong increase in food demand. The improvement of
yields (as a result of a technology dissemination) plays an important role in slowing the ex-
pansion of agricultural land in the first decades of this scenario period. It should be noted that
growing crops for biofuels could become an important factor in driving future land use in all
four scenarios. Although this scenario has the lowest rates of land conversion, the depletion of
forest land is still significant in Africa and Southeast Asia. In the second half of the century,
deforestation rates in this scenario are far lower than the other scenarios (Table 9-7).

Adapting Mosaic. The Adapting Mosaic scenario, like Order from Strength, also assumes rel-
atively slow yield improvement in the first decades. One consequence is that not only Africa
but also South Asia experience a virtual depletion of their forest areas up to 2100. However, a
lower increase in population – and locally successful experiments in innovative agricultural
systems (translated into an increasing the rate of improvement of crop yields), mitigate a fur-
ther expansion of agricultural land in other regions after 2040. The long term deforestation
rates in this scenario are slightly above those of Techno Garden.

9.3.5 Irrigated Area

Short piece to be written




                                               14
SECOND DRAFT 1 December, 2003                     Chapter 9 Provisioning and Regulating Ecosystems Services


9.4      ECOSYSTEM SERVICES


9.4.1 Food 8

9.4.1.1 Introduction and Overview of Results

Under this category of ecosystem services falls the many food and material products derived
from vegetation. The ecosystem services provided by agriculture are assessed in two ways –
First, we assess the total volume of services delivered by agriculture and use total food pro-
duction as a measure of these services. Another way is to assess the services ―delivered‖ to
each person or the outcomes of these services, and to use per capita food consumption or the
number of malnourished children as a measure of these services. Both are equally important –
Total food production is related to the amount of agricultural land, water, and other resources
required to deliver food services, while the per capita food consumption establishes a connec-
tion between ecosystem services and human well being. In the following paragraphs we dis-
cuss both measures of ecosystem services.

All four MA scenarios result in increased average food production, both total and per capita,
by 2050 compared to the base year (Figures 9-7, 9-8, 9-9, 9-10). Yet different means are used
to achieve this production, and—most importantly—with differing outcomes for the poor.
Global Orchestration has the largest production increases, whereas increases are lower, but
relatively similar for Techno Garden, Adapting Mosaic and Order from Strength.

Under the Global Orchestration scenario, rapid income growth in the North and South, in-
creasing trade liberalization, and urbanization fuel growth in food demand. Global cereal,
meat, and fish demand grow fastest, with cereals being used increasingly as livestock feed.
Grain production growth is driven by growth in yield, making large crop area expansion un-
necessary; rapid growth in food demand is also met through increased trade. By 2050, inter-
national food prices are lower for livestock products and rice, whereas pressure on maize from
demands for animal feed and wheat as direct food item leads to increased prices for these
crops. Calorie consumption levels increase faster than for any other scenario, and the number
of malnourished children by 2050 drops to a third of current levels.

Under the Order from Strength scenario, growth in developed countries is somewhat lower,
and much reduced in the group of developing countries, protectionist trade policies prevail,
and populations increase at a rapid rate. Per capita food availability by 2050 is also enhanced.
However, production growth is achieved through significant expansion in crop area, as re-
duced investments in yield improvement are insufficient to kept up with demand levels. A
second reason for crop area expansion lies in remaining protection levels implemented by
trading partners, which increase the cost of procuring food, particularly for poor countries of
the South, at the same time that elites in countries of the North and South continue to expand
and diversify their diets. As food production levels cannot keep pace with (albeit somewhat
depressed) food demand, international food prices for major crops increase significantly (re-
duced livestock demand results in reduced livestock prices, on the other hand). As high levels
of food prices surpass the cost of protection, food-deficit countries resort to food imports. As
a result, trade levels are not much reduced under the Order from Strength scenario, compared
to Global Orchestration, but the cost of procuring food is much higher. Calorie consumption



8
    Quantitative results in this section are based on the IMPACT model.

                                                        15
SECOND DRAFT 1 December, 2003                       Chapter 9 Provisioning and Regulating Ecosystems Services


levels improve only very slowly up to 2050, and the number of malnourished children barely
declines.

Under the Techno Garden scenario, growing incomes in both developed and developing coun-
tries are combined with medium population growth, increasing trade liberalization, and a
drive for innovations in all sectors, including food production. The Techno Garden scenario
operates somewhat similar to the Global Orchestration scenario, with substantial improve-
ments in crop yields and lower preference for meaty diets making crop are expansion unne-
cessary. Increased food demand is also met through exchange of goods, and technologies.
Both calorie consumption levels and the reduction in the number of malnourished children are
similar, albeit somewhat lower, compared to the Global Orchestration scenario.

Under the Adapting Mosaic scenario, finally, the focus is on the adaptation of local approach-
es to the improvement of ecosystem services. Incomes grow slowly, and populations continue
to grow. On the other hand, this scenario achieves increased food production in ways similar
to the Order from Strength scenario. Food production is produced locally, on expanded crop
areas, but insufficient achievements in production increases the demand for trade and raises
international food prices, higher than among all other scenarios. Calorie consumption levels
improve, but the increase is the lowest among the four scenarios. The number of malnou-
rished children is reduced slightly more than under OS, however, due to a focus on social in-
vestments under this scenario.

Although at the global level, outcomes for food supply, demand, and food trade are similar
across MA scenarios, there are substantial differences across the various regions in the South
and North, as will be detailed in the following sections (Figures 9-7 through 9-10).

9.4.1.2 Food Production and Consumption to 2050

Under the Global Orchestration scenario, food crops 9 production has the largest increases of
all scenarios. Production grows by 3,321 megatons to 7,227 megatons in 2050; cereal produc-
tion alone is expected to increase by 72 percent by 2050, the largest increase across the four
scenarios. (Figure 9-7) Global production of fish is expected to increase by 69 megatons by
2050, and demand for livestock products by 357 megatons (Figure 9-9). Globally, average per
capita demand of cereals as food source is projected to increase slightly by 10 kg to reach 172
kg in 2050. (Figure 9-8). As a benchmark for cereal consumption we can use the 1997 value
of 139 kg per person per year from the OECD region, a region with highly diversified food
consumption patterns. In 2050 Asia and MENA are substantially above the benchmark, indi-
cating a far less diversified diet compared to the OECD region, whereas LAC and SSA re-
main below this value across scenarios, indicating gaps in food production, (but also more
root and tubers oriented diets). Per capita demand for livestock products is likely (with high
certainty) to increase much more rapidly worldwide, driven by strong income growth and in-
creasing preference for livestock products. Globally, annual per capita consumption is ex-
pected to increase from 36 kg in 1997 to 70 kg by 2050; with large increases in Asia, FSU,
and OECD (Figure 9-10). If we again take the OECD base year value as a benchmark, by
2050, only Latin America will have achieved similar per capita meat consumption levels as
the OECD did in 1997, with Sub-Saharan Africa particularly lagging behind.

Under the Order from Strength scenario, assumptions how the world responds to growing
food production challenges play out in depressed demand for meat, fish, and grains in devel-

9
    Food crops include cereals, roots and tubers, soybean, sugar crops, vegetables, and fruit crops.

                                                          16
SECOND DRAFT 1 December, 2003             Chapter 9 Provisioning and Regulating Ecosystems Services


oping countries. Food production in all categories shows large increases compared to today,
but grain production is somewhat below Global Orchestration, and fish and meat production
are substantially below the Global Orchestration scenario (Figures 9-7 and 9-9). Average per
capita cereal food consumption will drop slightly in different regions (Figure 9-8). The de-
clines in per capita cereal consumption as food are most pronounced in Southeast Asia, fol-
lowed by East Asia, with an average Asian decline in per capita cereal demand of 16 kg. Sub-
Saharan Africa, where per capita food supplies are already quite low in 2000, will however
maintain current (low) consumption levels (Figure 9-8). Per capita consumption for meat
products will continue to rise in the North and the South, albeit at smaller rates than under the
Global Orchestration scenario. Globally, annual per capita meat demand is expected to in-
crease by 5 kg to reach 41 kg by 2050 (Figure 9-10). By 2050, still none of the MA regions
can reach per capita meat consumption levels of OECD in the base year, with Sub-Saharan
Africa only achieving 20 percent of OECD base year meat consumption by 2050, and even
Latin America 2050 per capita consumption levels still 23 kg below OECD’s base year levels.

Under the Techno Garden scenario, total food crop demand increases by 3,017 megatons up
to 2050; cereal demand increases by 1,070 megatons, fish products by 56 megatons, and meat
products by 166 megatons. Per capita cereal demand is expected to increase by 9 kg overall,
with the largest increases in South Asia (23 kg), the OECD and FSU regions (10 kg) (Figure
9-8). Preference for meat products is lower under Techno Garden compared to GO. As a re-
sult, per capita demand for livestock products grows by only 6 kg globally during 1997-2050.
The increase is largest in Asia, at 12 kg, followed by Latin America, with 11 kg (Figure 9-10).
By 2050, MA regions have yet to catch up between 24-71 kg of meat consumption to reach
OECD base year levels.

Under the Adapting Mosaic scenario, by 2050, demand for all food products is somewhat de-
pressed as people cannot afford higher-value foods and focus on local production for staple
crops. Total food crop demand grows by 2,797 megatons to reach 6,704 megatons by 2050.
Cereal demand expands by 994 megatons, fish demand by 46 megatons, and demand for meat
products by 178 megatons. Average per capita cereal food demand contracts by 10 kg to 151
kg in 2050. The regions of FSU and Asia experience the sharpest decline at 15 kg and 14 kg,
respectively (Figure 9-8). Per capita meat consumption levels under Adapting Mosaic only
increase significantly for the OECD region, by 24 kg, whereas developing regions cannot
reach OECD base year levels (Figure 9-10).

9.4.1.3 Extensification versus Intensification of Agriculture

Under the Global Orchestration scenario, the rapid rate of technology development and in-
vestments in agricultural research will lead to substantial yield increases rendering large ex-
pansion in new crop areas unnecessary (Figure 9-11). Globally, harvested area for grains is
projected to expand at 0.01 percent annually during 1997-2050 and then to contract at 0.28
percent annually up to 2100. Only in SSA, will a large expansion of cropland be necessary
for increasing production (Figure 9-11).

Although total cropland will not greatly expand, much more of it will be irrigated in 2050 as
compared to now. Irrigated area will grow from 239 to 273 million ha (the largest among all
four MA scenarios) spurred by large investments in irrigation systems (Figure 9-12). The
growth of irrigation is one of the main factors explaining productivity increases. Furthermore,
total agricultural land will grow because of the demand for pasture land and biofuels (See
Section 9.3.4).


                                               17
SECOND DRAFT 1 December, 2003            Chapter 9 Provisioning and Regulating Ecosystems Services


Under the Order from Strength scenario, society invests relatively little in crop technology
and supporting infrastructure. As a result, expansion in area will need to carry the brunt of
food supply increases. Globally, crop area is projected to increase by 137 million ha to 823
million ha to supply future food needs, equivalent to an annual rate of 0.34 percent, before
slowing to 0.25 percent per year during 2050-2100 (Table 9-6). Area expansion for cereals
will be spread out among the developing regions, with Latin America, Sub-Saharan Africa,
and MENA all experiencing harvested area expansion in the order of more than 40 percent.
Expansion will be slightly lower in the FSU and OECD regions and lowest in Asia (Table 9-
6). At the same time irrigated area is expected to contract, by 1 million ha during 1997-2050
and a further 16 million ha during 2050-2100, with area declines in Asia and the FSU more
than offsetting net increases in the other regions (Figure 9-12).

The Techno Garden scenario, characterized by innovations in agricultural technology and
crop productivity, but also a less meat-based diet, requires even less area expansion than the
Global Orchestration scenario. Up to 2050, irrigated area grows substantially, but less so than
in Global Orchestration (Figure 9-12). By 2100, irrigated area is expected to decline as the
drive for increased innovations and investments is slowed due to slowing pressure on ecosys-
tem services and food production (Figure 9-12). Globally, cereal harvested area contracts by
0.01 percent annually during 1997-2050 and a further 0.14 percent annually during 2050-2100
to 637 million ha (Table 9-6). However, total food crop area is expected to increase by 0.11
percent annually during 1997-2050 before contracting by a similar rate during 2050-2100.
Although most regions will achieve production growth by intensification of existing cropland,
expansion of cultivated land will still be important in SSA, (accounting for 30 percent of total
production growth up to 2050) and Latin America and MENA (accounting for about 11 per-
cent of total production growth)(Figure 9-11).

The Adapting Mosaic scenario postulates a combination of slow growth in food demand, low
investments in food production technologies, and no breakthroughs in yield-enhancing tech-
nologies. Globally, irrigated area is expected to grow very slowly up to 2050, and then decline
(Figure 9-12). It will increase however particularly in Sub-Saharan Africa and Latin America.
Depressed food demand under the Adapting Mosaic scenario will not be able to compensate
for stagnant crop yields. As a result, crop harvested area is expected to increase at 0.16 per-
cent per year for cereals and at 0.23 percent annually for all food crops, during 1997-2050, be-
fore contracting at –0.06 percent and –0.04 percent annually, respectively. Similar to the oth-
er MA scenarios, most cereal harvested area will be added in Sub-Saharan Africa, at 39 mil-
lion hectares, followed by Latin America, at 10 million hectares, and MENA, at 7 million hec-
tares.




                                              18
SECOND DRAFT 1 December, 2003            Chapter 9 Provisioning and Regulating Ecosystems Services


9.4.1.4 The Role of Trade and International Food Prices

Under the Global Orchestration scenario, trade liberalization and economic opening helps
fuel rapid increases in food trade. Total trade in grains, livestock and fish products increases
from 208 megatons to 692 megatons by 2050, the largest increase among the MA scenarios
(Figure 9-13). Net grain trade increases more than 200 percent from 1997 to 2050. The OECD
region, in particular, responds to the increasing cereal demands in Asia and MENA, with an
increase in net cereal exports of 89 megatons. Moreover, the very rapid yield and area in-
creases projected for the SSA region (Table 9-6) turn the region from net cereal importer at
present to net grain exporter by 2050. Net trade in meat products increases 674 percent, albeit
from low levels. Net exports will increase particularly in Latin America, by 23 megatons,
while the OECD region and Asia are projected to increase net imports by 15 megatons and 10
megatons, respectively. Net fish exports increase from 12 megatons in 1997 to 23 megatons
by 2050, an increase of 83 percent. Sub-Saharan Africa and the FSU region slightly increase
net exports, while net imports increase in the MENA region. Latin America remains the larg-
est net exporter of fish products.

Large investments in agricultural research and infrastructure, particularly in developing coun-
tries, help bring down international food prices for livestock products and rice. Over the
1997-2050 period, livestock prices decline by 9-13 percent, and rice prices by 31 percent,
whereas maize and wheat prices increase by 14 percent and 39 percent, respectively because
of demand for animal feed (Figure 9-14).

Under the Order from Strength scenario, countries maintain current protection levels. At the
same time, food production stalls because of low investments in technology and infrastruc-
ture, and this puts pressure on countries to import food. Low income growth, finally, dampens
food demand somewhat in developing countries. Hence, even though this a scenario in which
trade is not encouraged, total trade in food commodities multiplies by a factor of 3 relative to
1997 (Figure 9-13). The combination of population growth and lagging food production leads
Asia to import from OECD, despite existing trade barriers. Meanwhile Sub-Saharan Africa is
a net importer, albeit at reduced levels, due to higher trade barriers and depressed demand.
Net trade in meat products is much lower compared to the Global Orchestration scenario,
reaching 41 megatons by 2050, but most trade is carried out intra-regional. Regarding trade in
fish products, Asia, the FSU, and the OECD region have slight export positions, while Sub-
Saharan Africa and MENA are net importers by 2050.

Depressed demand from lower income levels cannot compensate for even lower investments
in food production and supporting infrastructure, and high population growth. As a result,
prices for all cereals are projected to increase over the coming decades: with price increases
ranging from 19 percent (maize) to 46 percent (rice) (Figure 9-13). Meat prices, on the other
hand, continue to decline by between 3-12 percent.

Under the Techno Garden scenario, trade liberalization continues apace. Pressure on trade is
somewhat reduced due to the preference for a diet with less meat, relatively good production
conditions in the various countries and regions, and somewhat lower income growth com-
pared to the Global Orchestration scenario. Total trade for grains, meat, and fish products
grows to 565 megatons by 2050 (Figure 9-13). Net cereal trade is dominated by Asian net im-
ports of 124 megatons and OECD net exports of 159 megatons, followed by net imports in
MENA of 70 megatons. Net meat trade is dominated by net imports in the OECD region (17
megatons in 2050), supplied through net exports from Latin America, Sub-Saharan Africa,
and Asia. Net fish trade increases globally by 77 percent to 22 megatons, with Latin America

                                              19
SECOND DRAFT 1 December, 2003            Chapter 9 Provisioning and Regulating Ecosystems Services


remaining the major fish exporter, followed by the FSU. Growth in production and trade will
more than compensate increased demand, resulting in declines for international food prices
across the board. By 2050, prices for wheat, rice, and maize are projected to decline by 11-26
percent and prices for beef, pork, and poultry by between 6-23 percent (Figures 9-14).

Under the Adapting Mosaic scenario, the focus is on local food production and conservation
strategies, with limited exchange of goods and services. Low income growth depresses food
demand, but the large increase in population puts upward pressure on food, which is being
produced without technological breakthroughs or enhancements due to lack of investment in
this area. Total grain, meat, and fish trade increases to 582 megatons by 2050 (Figure 9-13).
Cereal trade increases by 175 percent over 1997 levels, most of which is accounted for by in-
creased net imports in Asia and MENA, and increased net exports of the OECD region. Simi-
larly to the other MA scenarios, the FSU can improve its net export position. Appropriate
technologies and conservation strategies help Sub-Saharan Africa to become a small net ce-
real exporter by 2050. Total net meat trade increases by 31 megatons, the smallest increase
among the MA scenarios. By 2050, Asia is projected to supply about 20 megatons of lives-
tock products to all other regions, save Latin America. Moreover, the OECD region is be-
coming a slight net exporter of fish products, increasing with a total increase in net trade of 6
megatons over the 1997-2050 period.

Insufficient food production causes international cereal prices to increase by 52-56 percent for
wheat, maize, and rice, whereas livestock prices decline by 2 percent (beef, pork) and 15 per-
cent (poultry) (Figures 9-14).

9.4.1.5 Outcomes for Calorie Availability and Child Malnutrition

Although food production levels by 2050 are similar across scenarios, outcomes for calorie
availability and child malnutrition levels in developing countries vary considerably. The in-
crease in global average caloric availability is largest under the Global Orchestration scena-
rio, at 818 kilocalories per capita per day, followed by the Techno Garden scenario, at 507
kcal/cap/day, whereas increases are only 207 kcal/cap/day and 250 kcal/cap/day under the
Adapting Mosaic and Order from Strength scenarios, respectively (Figure 13). Under the
Global Orchestration scenario, all regions experience large increases in calorie availability,
led by Asia with an increase in 1,035 kcal/cap/day. Moreover, if we consider average OECD
per capita kilocalorie availability of 3,374 kilocalories today as a benchmark for other regions
in the future, by 2050, all regions but Sub-Saharan Africa are projected to reach this level.

Under the Order from Strength scenario, on the other hand, the increase in per capita calorie
consumption is largest in the OECD region, at 616 kcal/cap/day. Under this scenario, none of
the MA regions can reach the base year kilocalorie availability enjoyed by the OECD over the
next 50 years, with lowest kilocalorie availability in Sub-Saharan Africa, followed by Asia.

Under the Techno Garden scenario, per capita consumption increases at similar rates for Asia,
Latin America, the OECD region, and Sub-Saharan Africa. Under this scenario, only Latin
America can reach base year OECD kilocalorie availability levels by 2050, although the ME-
NA region comes close. Moreover, per capita calorie levels per day in all regions but Sub-
Saharan Africa surpass 3,000 kilocalories.

Under the Adapting Mosaic scenario, finally, increases in calorie availability are very low.
Similarly to the Order from Strength scenario, by 2050 none of the MA regions can achieve
per capita kilocalorie levels enjoyed by the OECD region today. The kilocalorie availability

                                               20
SECOND DRAFT 1 December, 2003            Chapter 9 Provisioning and Regulating Ecosystems Services


remains particularly low in Sub-Saharan Africa, at less than 2,500 kilocalories/cap/day, and
only reaches 3,000 kilocalories/cap/day in two of the developing regions, Latin America and
MENA.

Food consumption together with the quality of maternal and child care and health and sanita-
tion are important determinants for child malnutrition outcomes (Figure 9-15). All four MA
scenarios result in reduced child malnutrition by 2050. However, under the Adapting Mosaic
scenarios the number increases up to 2020 before declining up to 2050 (Figure 9-15). The in-
crease is even larger for the Order from Strength scenario, before the number finally declines
to 151 million children by 2050, only 15 million children less than today (Figure 9-15).

Today, South Asia accounts for slightly more than half of all malnourished children in devel-
oping countries, followed by Sub-Saharan Africa, where 20 percent of all malnourished child-
ren are located. Under the Global Orchestration scenario, the number of malnourished child-
ren will decline by 57 million children in South Asia and by 19 million in Sub-Saharan Afri-
ca. Under the Order from Strength scenario, on the other hand, the number of malnourished
children is expected to increase by 11 million in Sub-Saharan Africa and to decline by only 9
million in South Asia. Under the Adapting Mosaic scenario, the number of malnourished
children will still increase by almost half a million children in Sub-Saharan Africa and decline
by only 27 million in South Asia. Under the Techno Garden scenario, finally, the number of
malnourished children declines by 11 million in Sub-Saharan Africa and 43 million children
in South Asia, levels similar, albeit somewhat lower, to those achieved under the Global Or-
chestration scenario (Figure 9-15).

9.4.1.6 Summing Up

Although the outcomes for food production do not vary significantly across scenario at the
global level, outcomes do vary by region, particularly for the poor. Moreover, the ways and
means that increased food production levels are achieved, vary significantly by scenario, with
some focusing on area expansion and local production, whereas others rely on yield im-
provements and enhanced trade. In the Order from Strength and Adapting Mosaic scenarios
protectionist policies, together with lack of investments in agricultural research and agricul-
ture-related infrastructure result in increased food prices, depressed food demands, and slow
improvements in food consumption on a caloric basis. Moreover, the number of malnourished
children will decline only slowly if such a storyline plays out. The outcome is different for
the Global Orchestration and Techno Garden scenarios where more food is produced by
boosting crop yield, and increasing the international exchange of goods, services, and know-
ledge. In this case crop area can be conserved, food prices increase much less, per capita food
consumption goes up faster, and the number of malnourished children declines more quickly.
The outcomes for food demand and supply, the ways that food is being produced, and traded,
and the distribution of food among regions varies substantially across scenario.

Uncertainty of Agricultural Estimates and Ecological Feedbacks to Agriculture

As a whole, the quantified scenarios show a confident picture of the future – Total food con-
sumption increases into the future along with economic development, while global food trade
smoothes out the differences in food-growing ability between nations. Production keeps up
with increasing demand. A basic question is whether this is a feasible view of the future from
the standpoint of ecological sustainability.




                                              21
SECOND DRAFT 1 December, 2003                      Chapter 9 Provisioning and Regulating Ecosystems Services


On one hand this confident view of the future is not unlike our experience over the last hun-
dred years in which food production and consumption have steadily increased as countries
have gotten richer, despite temporary setbacks due to political crises, poor planning or the oc-
casional drought. On the other hand we should not assume that the global agricultural system
will remain as robust as it now apparently is. In particular the following factors could pose in-
creasing risks to the agricultural production computed in these scenarios:
 Scarcity of water – Section 9.5.4 shows that many of the areas where crop and fish pro-
    duction will intensify and/or expand are also areas currently in the ―severe water stress
    category‖ and will have an increasing level of water stress across all scenarios (e.g. the
    Middle East, Sub-Saharan Africa, parts of China, India). This is particularly important be-
    cause irrigation is and will continue to play an important role in the agriculture of these
    regions. It is also shown that wastewater discharges are likely to double over much of this
    area, also endangering the source of freshwater fish not coming from aquaculture. Of
    course solutions may be found for water scarcity (and the water calculations themselves
    are uncertain), but in the meantime the possible limiting role of water resources should be
    kept in mind when interpreting the food production scenarios.
 Intensification of agricultural inputs. Nearly all scenarios assume improvements in agri-
    cultural efficiency based not only on irrigating arid lands, but also by increasing the inputs
    of fertilizers and pest controls. Unfortunately the required inputs of fertilizer, pesticides,
    mechanical equipment, etc. were not quantified. Nor have we evaluated the long term risks
    of intensive agricultural inputs on pest outbreaks, groundwater contamination, soil degra-
    dation, and other ecological impacts. With low to medium certainty we expect these risks
    to be of greatest concern in the Global Orchestration scenario which has the highest level
    of agricultural inputs and a low level of environmental protection. Next in line could be ei-
    ther Order from Strength because of its low environmental consciousness. Perhaps the
    Adapting Mosaic scenario would have the lowest level of risk because of its lower level of
    agricultural inputs and higher level of actions to protect the environment.
 Sustainability of marine fishery. The scenarios show a medium to large increase in fish
    production and consumption in all regions of the world. But we have not yet analyzed in
    detail the ability of the world’s marine fishery to sustain the computed fish production.
 Affordability of food and food insecurity. The procedure for calculating food consumption
    and production assumes that increasing demand will be satisfied by the world food market
    (with various levels of trading). In these calculations prices reach an appropriate level for
    balancing supply with demand. As of yet we have not analyzed in detail the affordability
    of food prices to all income groups. We have noted, however, that lack of technological
    development under the Order from Strength and Adapting Mosaic scenarios will force
    lower income countries to import cereal at prices substantially above today. Cereals at
    these prices may not be within easy reach of lower income groups.


9.4.2 Fuel 10

The living biosphere provides humanity with both traditional fuel wood, and so-called ―mod-
ern biofuels‖. Among the modern biofuels are: (i) alcohol derived from fermenting maize and
sugar cane, (ii) fuel oil coming from rapeseed, and (iii) fast growing tree species that provide
fuel for power generating turbines, (iv) agricultural wastes, also burned to generate power.
While use of fuel wood has been steadily replaced by other energy carriers in the world, it still
accounts for a large percentage of total energy use in some places. At the same time the cur-
rent use of modern biofuels is quite modest, but could greatly expand according to some ener-

10
     Quantitative results in this section are based on the AIM model.

                                                          22
SECOND DRAFT 1 December, 2003             Chapter 9 Provisioning and Regulating Ecosystems Services


gy scenarios (see Section 9.3.1). An important advantage of these fuels is that they produce
only small amounts of greenhouse gas emissions (and other air pollutant emissions) and hence
can be part of a long term strategy of climate protection. For this reason they play a signifi-
cant role in the Techno Garden energy scenario where climate policy is given high priority
(Section 9.3.1.). In the MA scenarios it is assumed that biofuels are in the form of woody
plants and agricultural wastes and are used mostly for electricity production.

Global Orchestration. Here the global production of biofuels increases from its current level
by a factor of 6 (Figure 9-16). The regions making the biggest contribution to this increase are
Asia (factor of 8), followed by the MENA countries (nearly a factor of 6) and Sub-Saharan
Africa (about a factor of 4.5). There are two main factors leading to this large expansion in
biofuel use. First, good land is available for biofuel production because competition from food
production is low – Food crops are grown very efficiently on existing crop areas in most re-
gions (because of the high crop yield achieved from investments in agricultural research and
fertilizer and other inputs). Secondly, the demand for electricity is high because of strong
economic growth. Hence there is a large demand in general for energy and for biofuel elec-
tricity in particular because biofuels can be grown on relatively cheap and productive land.
However, one unwelcome consequence of this intense use of biofuels is a high rate of defore-
station in the regions committed to biofuel production (Section 9.3.4).

Techno Garden. Global production of biofuels increase by about a factor of four (Figure 9-
16) for reasons similar to the Global Orchestration scenario. Production lags behind the
Global Orchestration scenario because income (and therefore energy demand) is lower in the
Techno Garden scenario.

Order from Strength. The level of investment in agricultural technology is low in this scenario
and as a result crop productivity is also relatively low. At the same time, population growth is
larger than the other scenarios, and food demand is proportionately large. Since productivity
is low on existing cropland, the increased demand for food has to come at least partly from
new croplands. Energy crops must compete with food crops for land and this makes land and
biofuels more expensive. In addition, slower economic growth leads to lower growth in ener-
gy demands. These factors result in the slowest growth of biofuel production among all the
MA scenarios. Nevertheless, global biofuel production still grows by more than a factor of 2
to fulfill the needs of the growing population (Figure 9-16).

Adapting Mosaic. This scenario is an intermediate case compared to the other scenarios.
Economic growth and crop productivity are higher than Order from Strength scenario but
lower than the others. As a result, energy demand is somewhat higher and competition with
food production somewhat lower than in the Order from Strength scenario, and biofuel pro-
duction is also somewhat higher. Globally, biofuel production increases by a factor of 2.8
over today (Figure 9-16) led by the MENA countries (factor of 6), Sub-Saharan Africa (factor
of 4) and Asia (factor of 3).

Major Uncertainties. Although calculation of the land requirement of energy crops takes into
account current productivity of soils, they do not factor in the degradation that will result from
these crops. Biofuel crops tend to degrade soils faster than many other crop varieties because
they have high productivities and require large amounts of fertilizer and other inputs. There-
fore, it is important to keep in mind that soil degradation may make energy cropping less eco-
nomical and ecologically damaging over the long run.



                                               23
SECOND DRAFT 1 December, 2003                    Chapter 9 Provisioning and Regulating Ecosystems Services


9.4.3 Genetic Resources, Biochemicals, Ornamental Resources 11

Although the future states of these services were not directly evaluated by model calculations,
here we examine the trends of some related indicators. These include: (1) the extent of natural
land versus agricultural land since these resources usually require undisturbed habitat, (2) the
rate of change of this habitat as indicated by the rate of deforestation and the rate of climate
change since the faster the change, the more doubtful that plants and animals can adapt to
these changes, (3) the level of water stress in freshwater resources which indicates the pres-
sure on aquatic and riparian species. By examining the trend of these variables we can make
some preliminary judgments about the future trends of genetic resources, biochemicals, and
ornamental resources. 12

Under the Global Orchestration scenario, pressures grow on remaining undisturbed terrestrial
and aquatic ecosystems – Existing forests through the 21st century disappear at rates compara-
ble to the last few decades. The decadal rate of temperature change is much higher than at
present, and ranks as the highest among the MA scenarios through most of the scenario pe-
riod. As a result of changing temperature (and precipitation), the type and viability of current
vegetation also changes over extensive areas. To meet growing food demand due to higher in-
comes and population, the level of agricultural production on existing cultivated land is inten-
sified by increasing the application of fertilizer and other inputs, and these chemicals also
contaminate nearby protected natural areas. In freshwater ecosystems, the level of water stress
increases over wide areas because of rapidly increasing withdrawals. In addition to these in-
creasing pressures, society is also not particularly mindful of the connection between its activ-
ities and the state of ecosystem services. In sum, we expect (with low certainty) that genetic
resources, biochemicals and ornamental resources may severely decline under this scenario.

Under the Techno Garden scenario the rate of deforestation is high, but eventually drops be-
low current rates. Climate protection policies lead to lower rates of temperature change (as
compared to the first decades of the 21st century), but vegetation areas still change extensive-
ly. Efficient water use leads to lower growth in water withdrawals and slower increases in
stress on freshwater ecosystems. However, high levels of fertilizer and pesticides are used on
agricultural land to boost crop yields which leads to contamination of natural areas. On the
whole we expect (with low certainty) that this scenario will pose lower (but still significant)
risks to genetic resources, biochemicals and ornamental resources compared to the Global
Orchestration scenario.

Under the Order from Strength scenario the rate of forest disappearance is even greater than
under Global Orchestration (because of more inefficient agricultural production). Also grow-
ing population and inefficient water use leads to rapid growth in water withdrawals and stress
on freshwater ecosystems. On the other hand, climate change is not as great therefore vegeta-
tion changes are not as extensive as in Global Orchestration. Also, a side-effect of the lower
level of wealth in this scenario is that farmers cannot afford to apply as many pesticides and
fertilizer to cropland land, meaning that the loading of these chemicals onto nearby natural
areas is somewhat lower than under Global Orchestration. Society in this scenario also gives
low priority to environmental protection. Summing up the different factors, we expect (with
low certainty) that this scenario has about the same risk to genetic resources, biochemicals
and ornamental resources as the Global Orchestration scenario.

11
  Some results presented in this section stem from the AIM, IMAGE; IMPACT and WaterGAP models.
1212
    We note that these are only a few of the many important factors that will influence the state of genetic re-
sources, biochemicals, and ornamental resources in the future. For example, these do not include an indicator for
the marine environment.

                                                       24
SECOND DRAFT 1 December, 2003                 Chapter 9 Provisioning and Regulating Ecosystems Services




The rate of forest disappearance under the Adapting Mosaic scenario drop below current rates
but are still high. The rate of climate change is not as high as Global Orchestration, nor as
low as Techno Garden, and therefore the extent of area with changed vegetation is also be-
tween these two scenarios. Water withdrawals significantly increase but not as much as under
Order from Strength because water is used more efficiently under Adapting Mosaic. The use
of fertilizer and other inputs on agricultural land is similar to Order from Strength. Under this
scenario society is mindful of the connection between its activities and ecosystem services.
Considering the different factors, we expect (with low certainty) that the risk to genetic re-
sources, biochemicals and ornamental resources is between that of the Global Orchestration
and Techno Garden scenarios.


9.4.4 Freshwater Resources 13

9.5.4.1 Water Availability

The ecosystem services provided by freshwater systems have many dimensions. In this sec-
tion we focus especially on the services of water supply for households, industry and agricul-
ture, and habitat for freshwater ecosystems including fisheries. As indicators of these services
we describe the changing state of water availability, water withdrawals, water stress and re-
turn flows.

We begin by describing changing ―water availability‖ which is used here to mean the sum of
annual surface runoff and groundwater recharge. This is the total volume of water that is an-
nually renewed by precipitation and theoretically available to support society’s water uses and
the needs of freshwater ecosystems. In reality society can exploit only a small fraction of this
volume because water-rich areas are not necessarily near high population areas, or because
water is ―unusable‖ as it rushes past cities in the form of floods, or because society cannot af-
ford adequate water storage facilities. One estimate is that only about 30 to 60 percent of typi-
cal river basin water resources is ―mobilizable‖. (Reference, Falkenmark). Nevertheless we
believe that the concept of water availability reflects the average quantity of water available to
meet the freshwater needs of society and ecosystems.

Since estimates of current water availability vary greatly, we present two independent esti-
mates in Figure 9-17. Present availability is estimated to be from 42.6 to 55.3 thousand cubic
kilometers per year. By 2050 global water availability increases by 4 to 5 percent (depending
on the scenario). On a regional scale Latin America experiences the smallest average increase
in water availability (1 to 2 percent, depending on the scenario), and FSU the largest (8 to 12
percent). The changes in availability are small up to 2050 because of two compensating ef-
fects – increasing precipitation tends to increase runoff while warmer temperatures intensify
evaporation and transpiration which tends to decrease runoff. Hence, the direction of change
of runoff does not correspond exactly to the direction of change of precipitation shown in
Figure 9-5.

By 2100 the effect of increasing precipitation becomes more important and runoff increases
over most land areas. Large areas on each continent have 25 percent or more runoff by 2100
(relative to the current climate period). Although availability increases in most areas, there are

13
 Quantitative estimates in this section are based primarily on the WaterGAP model and secondarily on the AIM
water resources model.


                                                    25
SECOND DRAFT 1 December, 2003            Chapter 9 Provisioning and Regulating Ecosystems Services


important arid regions where availability drops 50 percent or more under all scenarios includ-
ing Southern Europe, parts of the Middle East, and Southern Africa. In general there are very
large differences between regions (Figure 9-17).

The differences between scenarios are not as large as the differences between regions in 2100
(Figure 9-17). The smallest changes in water availability (between 2100 and the climate nor-
mal period) occur under the Techno Garden scenario because it has the lowest rate of climate
change, and the largest changes take place under the Global Orchestration scenario which has
the fastest rate of climate change through most of the scenario period. For these two scenarios,
world-wide water availability has a net increase of from 5 to 12 percent between 2100 and
current climate (The low estimate is from Techno Garden and the high from Global Orches-
tration). Availability increases 7 to 15 percent in OECD countries, 9 to 23 percent in Sub-
Saharan Africa, 5 to 13 percent in Asia, 11 to 24 percent in FSU, and 2 to 5 percent in Latin
America. For these scenarios, the water availability in the already-arid MENA countries sinks
by about 6 percent. Results for the Order from Strength and Adapting Mosaic scenarios fall
between these figures (except for the MENA countries in which the decrease is 7 percent for
these scenarios.)

While the increase in availability makes more water available for water supply, an increase in
runoff can also correspond to more frequent flooding. We estimate (with medium certainty)
that regions with large increases in water availability will also have more frequent high runoff
events. We did not analyze this effect because no validated model is currently available in the
literature for making world-wide calculations of flooding.

9.5.4.2 Water Withdrawals and Use

While water availability indicates the amount of water theoretically exploitable, water with-
drawals give an estimate of the water abstracted by society to fulfill its domestic, industrial
and agricultural needs. Hence it is a useful indicator of ecosystem services. (The water re-
quirements for supporting a freshwater fishery are discussed later.) As compared to availabili-
ty, water withdrawals show large changes over time and between scenarios up to 2050. Cur-
rent world-wide withdrawals (1995) are estimated to be between about 3600 and 3700 thou-
sand cubic kilometers per year, or about 7 to 8 percent of estimated water availability. While
this does not seem like much, the intensity of withdrawals is high relative to water availability
in several regions of the world (see next section).

Global Orchestration. Strong economic growth coupled with an increase of population leads
to a world-wide increase in withdrawals of around 40 percent (Figure 9-18). But the changes
are only slight in OECD and FSU countries because of compensating effects – continuing im-
provements in water efficiency and stabilization and decrease of irrigated land tend to de-
crease water use while economic and population growth tend to increase water use. In MENA
countries water withdrawals decline in all scenarios because of strong water conservation. Al-
though the efficiency of water use also improves over time in other regions, the effect of in-
creasing population and economic growth leads to fulfillment of pent-up demands in the do-
mestic and industrial sectors. Hence large increases are seen – withdrawals increase by more
than a factor of 2.5 in Sub-Saharan Africa, a factor of 1.7 in Latin America and a 50 percent
increase in Asia.

According to this scenario, many more people gain access to water supply, as domestic water
use increases in nearly all regions (a factor of 6 in Sub-Saharan Africa, 55 percent in Asia, 35
percent in FSU, 80 percent in MENA and 41 percent in Latin America) (Figure 9-19). The on-

                                               26
SECOND DRAFT 1 December, 2003             Chapter 9 Provisioning and Regulating Ecosystems Services


ly exception is OECD where water use in the domestic sector continues its current declining
trend.

Techno Garden. Here strong structural changes in the domestic and industrial sectors and
improvements in the efficiency of water use in all sectors lead to decreases in water with-
drawals in OECD and FSU countries (11 and 24 percent, respectively). The same factors lead
to a slow down in the growth of withdrawals in the rest of the world (Figure 9-18). Neverthe-
less, water withdrawals grow by a factor of 2.4 in Sub-Saharan Africa because of pent-up de-
mand for household water use and growing industrial water requirements.

Order from Strength and Adapting Mosaic. Although these scenarios do not have the largest
economic growth, they have the largest withdrawals because of their slower improvement of
the efficiency of water use and faster population growth. World-wide withdrawals increase by
51 to 80 percent and in OECD 5 to 32 percent (Order from Strength and Adapting Mosaic, re-
spectively). In the FSU countries, withdrawals level off under the Order from Strength scena-
rio and decrease by 9 percent under Adapting Mosaic. In Sub-Saharan Africa withdrawals
increase by a factor of 4 in both scenarios. In Latin America, the increase is a factor of 3 to
3.5, in MENA 25 to 40 percent, and in Asia 60 to 90 percent (Figure 9-18).

9.5.4.3 Water Scarcity / Water Stress

After considering changes in water availability and withdrawals in the previous paragraphs,
here we evaluate the consequences of these changes on water stress in freshwater systems.
The concept of ―water stress‖ is used in many water assessments to obtain a first estimate of
the changing pressure of society on water resources (References). It is assumed that the higher
the level of water stress, the greater the limitations to freshwater ecosystems, and the more
likely that chronic or acute shortages of water supply will occur. A common indicator of wa-
ter stress is the withdrawals-to-availability ratio (wta). This indicator implies that future water
stress will tend to decrease in general because of growing water availability, but increase be-
cause of increased withdrawals. An often used approximate threshold of ―severe water stress‖
is a wta of 0.4 (References). River basins exceeding this threshold, especially in developing
countries, are presumed to have a higher risk of chronic water shortages and risk to freshwater
ecosystems.

Figure 9-20 (a) depicts the current area of the world in the ―severe water stress‖ category.
Much of northern and southern Africa, as well as central and southern Asia are included. In
total about 18 percent of the world’s river basin area falls into this category. About 2.3 billion
people live in these areas.

Figure 9-20 (b) shows the area in the severe water stress category under the Global Orches-
tration scenario in 2050. Some areas, especially in OECD and FSU fall out of the severe
stress category because of stabilizing withdrawals and increasing water availability due to
higher precipitation under climate change. The areas in the rest of the world slightly expand.
A total of about 4.3 billion people live in these areas. Over most of these areas increasing
withdrawals tend to increase the level of water stress as compared to today.

Under Techno Garden, water withdrawals (up to 2050) drop in OECD and FSU, and grow
more slowly in other regions. Water stress follows these trends and declines in many parts of
the OECD and FSU, and increases more slowly than the other scenarios in other parts of the
world.


                                                27
SECOND DRAFT 1 December, 2003                     Chapter 9 Provisioning and Regulating Ecosystems Services


Under the Adapting Mosaic and Order from Strength scenarios water withdrawals increase
sharply as discussed above. The area under severe water stress increases in 2050 up to about
22 percent and 23 percent of total watershed area for the Adapting Mosaic and Order from
Strength scenarios, respectively. Water stress increases over all of these areas. 5.5 to 5.7 bil-
lion people live in river basins with severe water stress, about 60 percent of the world’s popu-
lation.

9.5.4.4 Water Quality / Return Flow

While water stress is a measure of changes in the quantity of water relative to water use, here
we use the concept of ―return flows‖ to assess water quality. Return flows are the difference
between withdrawals and consumption and therefore provide a rough estimate of the magni-
tude of wastewater discharged into the receiving water in a watershed.14 Depending on the
type of return flows, high rates of these flows could correspond (with medium certainty) to
low water quality and high levels of water contamination and pressure on freshwater ecosys-
tems. We use return flows as a surrogate variable for water quality because it is not possible at
this time to compute world-wide changes in water quality for the different scenarios.

Since irrigation usually consumes more water than domestic or industrial uses, the return
flows for irrigation are also usually a smaller fraction of its withdrawals. The quality of the re-
turned water is also sector-dependent – the water discharged after cooling a power plant tur-
bine is hot but relatively clean compared to that of the raw wastewater discharged by many ci-
ties or factories. Irrigation return flows are normally not returned to rivers as municipal or in-
dustrial ―point sources‖ but enter the river in a diffuse way along many kilometers of its
length. Hence, their impact on river quality will not be the same as the impact of return flows
from a city or industry. Perhaps the most important factor to take into account in assessing the
impact of return flows is whether they will be treated or not. In OECD the trend is towards at
least partially treating all return flows. Already almost all domestic and industrial wastewater
is treated, although most agricultural return flows are untreated. For the rest of the world
wastewater treatment is very uncommon except in some cities. 15

Global Orchestration. By 2050 world-wide return flows increase by 40 percent. The magni-
tude of return flows follows that of withdrawals, meaning the larger (or smaller the volume of
withdrawn water), the larger (or smaller) the size of wastewater discharges. Return flows de-
crease on the average in OECD and FSU countries because of leveling off population, de-
creasing irrigated area, and improving efficiency of water use. These factors tend to decrease
withdrawals and hence return flows. Furthermore, even though low priority is given to envi-
ronmental protection, the richer societies in this scenario maintain their current efforts at envi-
ronmental management. Hence it is reasonable to assume that the level of wastewater treat-
ment in OECD countries will remain at least at its current level.

Because of the booming water withdrawals, return flows increase by a factor of 3.6 in Sub-
Saharan Africa (between now and 2050), a factor of 2 in Latin America, and more moderately
in the MENA countries (22 percent) and Asia (48 percent). Figure 9-15 illustrates the large

14
  Since consumption is assumed to be a constant fraction of withdrawals the magnitude of return flows will mir-
ror that of withdrawals. Nevertheless the relationship between return flows and withdrawals will change over
time because the different water use sectors (domestic, agriculture, etc.) have different ratios of consumption to
withdrawals.
15
     Although disinfection of water supplies is common throughout the world.


                                                        28
SECOND DRAFT 1 December, 2003            Chapter 9 Provisioning and Regulating Ecosystems Services


area where return flows are estimated to at least double under the Global Orchestration scena-
rio (between now and 2050). Over 88 percent of the watershed area of Sub-Saharan Africa is
in this category, 46 percent of the MENA countries, 35 percent of Asia, and 33 percent of Lat-
in America (Table 9-8). Consistent with the storylines of this scenario, low priority is given to
environmental management in the world’s poorer regions. Therefore, it is likely that wastewa-
ter will remain untreated in many areas, and that the level of water contamination and degra-
dation of freshwater ecosystems may increase (low to medium certainty).

We estimate that 5.1 billion people or nearly 60 percent of the world will live in these areas in
2050 (Table 9-9). We emphasize, however, that return flows will cause major problems only
if they remain untreated.

Techno Garden. The trends for this scenario up to 2050 are in the same direction as Global
Orchestration, but the stronger emphasis on improving water efficiency and somewhat lower
economic growth rates lead to a stronger decrease in return flows (between now and 2050) in
OECD and FSU (18 and 42 percent, respectively). The same factors lead to slower growth of
return flows in Sub-Saharan Africa (factor of 3.6), MENA (16 percent), and Asia (nearly 20
percent) as compared to Global Orchestration. The change in Latin America is the same as in
Global Orchestration (increase by a factor of 2). Similarly large areas will have increases of
100 percent or more return flows (Table 9-8), and a total of 4.8 billion people will live in
these areas (Table 9-9). Since the emphasis in this scenario is on environmental management,
and since return flows do not increase too much in MENA or Asia, it may be that most of the
wastewater flows in these regions will be treated. It is less likely that the enormously increas-
ing return flows of Sub-Saharan Africa and Latin America will be fully treated.

Adapting Mosaic. In this scenario, return flows decrease in FSU (between now and 2050) be-
cause withdrawals decrease. In all other regions return flows increase much more than in
Global Orchestration and Techno Garden because of the lower level of water use efficiency
and larger population which leads to higher withdrawals and more return flows. The increase
of return flows in Sub-Saharan Africa is a factor of 5.5, in Latin America a factor of 3.6, in
Asia 75 percent, in the MENA countries 55 percent, and in OECD 3 percent. The area of wa-
tersheds with at least a 100 percent increase in return flows between now and 2050 is consi-
derably larger than in Global Orchestration and Techno Garden, and 6.1 billion people (65
percent of the world’s population in 2050) live in these areas (Tables 9-8 and 9-9). Since this
scenario puts a strong emphasis on local environmental protection, and since wastewater
treatment technology is simple and can be applied easily on the local level, we expect (with
medium certainty) a high level of wastewater treatment.

Order from Strength. Above we noted that this scenario has the largest withdrawals because
of its slower improvement of the efficiency of water use and faster population growth. Accor-
dingly it also has the largest return flows, with a doubling of world-wide total flows (between
now and 2050). The smallest increase is in FSU countries with 10 percent, followed by OECD
with a nearly 40 percent increase. All other regions experience much larger increases – Asia
and the MENA countries with approximately a doubling, Latin America more than a factor of
4 and Sub-Saharan Africa a factor of 5.6. The area with a doubling of return flows is some-
what larger than in the Adapting Mosaic scenario (Table 9-8) and 6.8 billion people live in
these areas (73 percent of global population) (Table 9-9). As compared to the Adapting Mo-
saic scenario, the level of environmental concern here is much lower, and therefore the ex-
pected level of wastewater treatment is also much lower. The combination of exploding
wastewater discharges and negligence of the environment could lead to large risks to freshwa-
ter ecosystems and water contamination. An additional dimension of this scenario is that re-

                                               29
SECOND DRAFT 1 December, 2003                      Chapter 9 Provisioning and Regulating Ecosystems Services


turn flows continue to rapidly increase after 2050. For example return flows increase in Sub-
Saharan Africa by a factor of 5.6 between 1995 and 2050, and double again between 2050 and
2100.

9.5.4.5 Certainty/Uncertainty of Freshwater Estimates

 Return flows are used as an indicator of water quality, but it would be more desirable to
  have a direct indicator of future water quality so that we can have a more certain connec-
  tion with the state of freshwater ecosystems and risk of water contamination. This is not
  yet possible globally.
 Tools used to estimate indicators are too rough to factor in local policies and management.

9.5.4.6 Summing Up Freshwater Services …

      Changing climate will tend to increase the availability of freshwater over most of the
       world’s river basin areas. Nevertheless, already arid areas such as the Middle East and
       Northeast Brazil may (with low to medium certainty) experience a significant drop in wa-
       ter availability under climate change.
      Water use in the domestic sector strongly grows in developing country regions implying
       that a greater number of people will have access to freshwater.
      Most regions across all scenarios will experience a large growth in water withdrawals be-
       cause of growing population and fulfillment of pent-up water demands. As an exception,
       the OECD and FSU may experience just small changes in withdrawals under the Global
       Orchestration and Techno Garden scenarios because of stabilizing or declining popula-
       tion, saturation of water demands, and improving efficiency in water use.
      The total area affected world-wide by severe water stress may not change very much over
       the scenario period (with low to medium certainty) but the intensity of water stress will
       grow in these areas and many more people will be potentially affected by severe water
       stress, especially in developing countries (medium to high certainty). Globally, roughly
       2.3 billion people now live in river basins with severe water stress. Under Global Orches-
       tration this will increase to 4.3 billion, under Techno Garden to X.X billion, under Adapt-
       ing Mosaic 5.5 billion and Order from Strength 5.7 billion.
      Because of increasing water withdrawals, return flows will increase the greatest under
       Order from Strength and the least under Techno Garden. The amount of untreated waste-
       water may increase in Sub-Saharan Africa (between now and 2050) by more than a factor
       of 3 under the Global Orchestration and Techno Garden scenarios, and more than a factor
       of 5 under the Order from Strength and Adapting Mosaic scenarios.
      The scenarios can be ranked according to the level of pressure on water resources from
       greatest to smallest: Order from Strength, Adapting Mosaic, Global Orchestration, Techno
       Garden.


9.4.5 Air Quality Regulation16

The ecosystem service of ―air quality regulation‖ is the service provided by the atmosphere in
maintaining acceptable air quality for humans, animals, and plants. The atmosphere regulates
itself by mixing and diluting air pollutants with horizontal and vertical air masses, as well as
by scouring itself of pollutants via precipitation and diffusion. In principle air quality regula-
tion depends on two main factors: the load of air pollutants and the dilution and scouring ca-

16
     Quantitative results in this section are based on the AIM model.

                                                          30
SECOND DRAFT 1 December, 2003              Chapter 9 Provisioning and Regulating Ecosystems Services


pabilities of the atmosphere. Because it is difficult to characterize the dilution and scouring
capabilities of the atmosphere world-wide and over the entire scenario period, we instead es-
timate changes of the load or emissions of air pollutants and use this as an indirect indicator of
air quality regulation. The reasoning is that the higher the level of air pollutant emissions, the
more difficult it is for the atmosphere to self-regulate its air quality. Hence, the higher the lev-
el of emissions, the lower the ecosystem service of air quality regulation.

For each of the scenarios we have computed the emissions of two key air pollutants, sulfur
dioxide and nitrogen oxides. These emissions lead to problems both near to and far from their
source – In the vicinity of pollution sources a combination of high emissions and unfavorable
meteorological conditions can lead to the buildup of high concentrations of sulfur dioxide,
ozone and other gases lasting several hours and posing a threat to human health. These so-
called ―air pollution episodes‖ are especially dangerous to the very young and old members of
the population. Under more typical conditions the local level of sulfur dioxide can be high
enough to cause long-term damage to vegetation and building materials. Sulfur dioxide and
nitrogen oxides emissions are also transported hundreds of kilometers from their source and
are then deposited to vegetation and soils via precipitation and diffusion where they cause
acidification of soils and freshwater systems (as well as direct impacts on vegetation). Be-
cause of their important role in many key air pollution related problems, sulfur dioxide and ni-
trogen oxides are good indicators of air quality regulation.

9.5.5.1 Global Orchestration

Sulfur Dioxide. Emissions of sulfur dioxide sink by about 45 percent world-wide (between
now and 2050) and in most regions (Figure 9-22). The reduction is 75 percent in OECD coun-
tries and 70 percent in FSU countries. This stems from the assumption that there is a continua-
tion of current controls of sulfur dioxide emissions and a major shift to lower sulfur fuels. Sul-
fur dioxide emissions sink (up to 2050) by 15 percent in Asia, 33 percent in MENA, and 64
percent in Latin America where strong economic growth leads to new sulfur emission controls
(Figure 9-22). However in Sub-Saharan Africa, emissions more than double because the eco-
nomic level is still not high enough to support sulfur emission controls.

Nitrogen Dioxide. Emissions here follow an opposite trend to sulfur dioxide (Figure 9-23).
World-wide emissions (between now and 2050) increase by over 50 percent although emis-
sions decrease by nearly 60 percent in OECD countries because of pollution controls. Else-
where emissions are driven upwards by the expansion of energy use for transportation and
power generation. The biggest increase is a factor of 3.6 in Asia and 2.7 in FSU, owing to
their high economic growth rates. The increase in emissions is 26 percent in Sub-Saharan
Africa and 22 percent in MENA because of lower economic growth here which leads to low-
er energy use.

9.5.5.2 Techno Garden

Sulfur Dioxide. Emission sink by over 60 percent world-wide (up to 2050) and decrease in
almost all regions (80 percent in OECD, more than 50 percent in Asia, nearly 75 percent in
FSU, 38 percent in MENA, and over 75 percent in Latin America.)(Figure 9-22). These re-
ductions are driven by two key assumptions (1) that the goal to reduce greenhouse gases leads
to a strong shift away from fossil fuels and this has the indirect effect of decreasing the use of
sulfur-containing fuels (coal, oil), (2) remaining emissions of sulfur dioxide are more strictly
controlled. Again emissions almost double in Sub-Saharan Africa because it lags economical-
ly behind other regions.

                                                31
SECOND DRAFT 1 December, 2003            Chapter 9 Provisioning and Regulating Ecosystems Services




Nitrogen Dioxide. Emissions increase world-wide by one-third (between now and 2050) be-
cause of the pollutants produced by expanding transportation energy use (despite some pollu-
tion controls) and extensive combustion of biofuels (Figure 9-23). NOx emissions drop by
two-thirds in OECD countries and by half in Latin America because of pollution controls, and
because the lower economic growth rate compared to Global Orchestration produces a lower
rate of emissions. Elsewhere emissions increase as in Global Orchestration because of expan-
sion of energy use for transportation and power generation. The biggest increase is a factor of
3 in Asia and FSU, owing to their high economic growth rates. Increases in Sub-Saharan
Africa and MENA are about the same as in Global Orchestration.

9.5.5.3 Adapting Mosaic

Sulfur Dioxide. This scenario shows a 28 percent reduction in emission world-wide but very
different trends in different regions (Figure 9-22). Reductions are still strong in OECD (68
percent), Latin America (more than 50 percent) and FSU countries (nearly 50 percent), but
lower in MENA (7 percent). Emissions increase in Asia (8 percent) because its energy system
is still dependent on fossil fuels and sulfur controls are minimal. Emissions increase by 73
percent in Sub-Saharan Africa because of the absence of sulfur controls. Note this increase is
not as great as in the previous scenarios because economic growth is slower and therefore
proportionally less fossil fuels are used.

Nitrogen Dioxide. The world-wide increase in emissions is lowest (28 percent) among the
MA scenarios. NOx emissions drop by two-thirds in OECD countries and by over 40 percent
in Latin America because of pollution controls. Increases in other regions are substantial be-
cause of expanded transportation energy use (a factor of 2.8 increase in Asia and FSU.) In-
creases in Sub-Saharan Africa and MENA are about the same as in Global Orchestration.

9.5.5.4 Order from Strength

Sulfur Dioxide. Under this scenario there is a small net change in global emissions (Figure 9-
22). Emissions are reduced by 64 percent in OECD countries, 48 percent in Latin America,
and 32 percent in FSU. Elsewhere, emissions increase. The increase is largest in Asia (58 per-
cent) where the reliance on fossil fuels and absence of controls leads to high emissions. The
increase is 70 percent in Sub-Saharan Africa and 21 percent in MENA because of their rela-
tively low rate of economic growth, reliance on fossil fuels and absence of pollution controls.

Nitrogen Dioxide. Emissions increase world-wide by 38 percent (between now and 2050).
The lower economic growth (compared to Global Orchestration) leads to a lower rate of
emissions, but the lack of pollution controls in most regions leads to a higher rate. The net ef-
fect is a world-wide increase in emissions that falls between the Global Orchestration and
Techno Garden scenarios. The emissions decline significantly in OECD countries because of
pollution controls, and slightly in Sub-Saharan Africa because (???). The growth of emissions
is substantial in other regions – about 50 percent in MENA and Latin America, and about a
factor of 2.6 in Asia and FSU.

9.5.5.5 Summing Up Air Quality Regulation

It is difficult to generalize about the overall trends of air quality regulation because of the
wide range in trends of future emissions of SO2 and NOx.


                                               32
SECOND DRAFT 1 December, 2003                     Chapter 9 Provisioning and Regulating Ecosystems Services


 Under Global Orchestration we expect lower levels of sulfur-related air pollution in all
  regions except Sub-Saharan Africa. For NOx, however, the trends are favorable only for
  OECD and Latin America. Elsewhere (with low to medium certainty) NOx-related pollu-
  tion increases substantially (Asia and FSU) or moderately (Sub-Saharan Africa and ME-
  NA).
 Under Techno Garden, the reductions in sulfur-related pollution are even greater (except
  for Sub-Saharan Africa) than in Global Orchestration. Meanwhile NOx pollution declines
  substantially only in OECD and Latin America, while elsewhere it continues to increase
  about the same as in Global Orchestration.
 Under Adapting Mosaic, sulfur-related pollution declines in all regions except Asia, where
  it has a slight net increase. Trends for NOx are similar to the Global Orchestration scena-
  rio.
 The level of sulfur-related air pollution declines only slightly world-wide under the Order
  from Strength scenario, because of declines in OECD, FSU and Latin America. Asia and
  MENA have the largest emission increases of all scenarios. There is a significant decline
  in NOx-related pollution in OECD countries, and a major increase elsewhere.

All in all, there is an improvement in air quality regulation across all scenarios in the OECD
region. Meanwhile there is a substantial deterioration in Sub-Saharan Africa. For other re-
gions there are mixed results for the scenarios.


9.4.6 Climate Regulation/Carbon Uptake 17

The biosphere plays a key role in the climate system, for example, by respiring and taking up
CO2 , by emitting CH4, and by reflecting or absorbing solar energy. While these processes no
doubt affect the climate system, they do not all regulate the climate system. ―Regulate‖ from
the systems point of view means an action that brings a system departing from equilibrium
back to equilibrium, or a corrective action that returns a system behaving unstably back to
stability. Obviously the emission of CH4 from the biosphere is not a regulatory process since
it will enhance the greenhouse effect. On the other hand, vegetation cover has an important in-
fluence on local climate through its retention and fluxes of moisture and by absorbing and re-
emitting solar radiation. In this way vegetation cover can play a decisive role in regulating lo-
cal climate. On the global basis, the biosphere helps regulate climate by capturing greenhouse
gases, thereby reducing the concentration of these gases in the atmosphere and slowing down
climate change. The most important gas captured is CO2. When the biosphere and atmosphere
are in equilibrium, the biosphere takes up as much CO2 for plant growth as it emits by plant
respiration. But under certain circumstances, the biosphere can take up more than it emits
such as when there is a net increase in the area of forests and other dense vegetation, or when
plant productivity is stimulated by increasing temperature or atmospheric CO2.

While it is outside the scope of our analysis to assess the effect of vegetation on local climate,
we do estimate here the effectiveness of the global biosphere in taking up CO2 from the at-
mosphere. The indicator used to describe this process is the net primary productivity (NPP) of
the biosphere in units of gigatons carbon per year. NPP is currently (2000) about 61.4 giga-
tons carbon per year and will increase across all scenarios and regions because of increasing
temperature and atmospheric CO2. Global estimates for 2050 range from 70.4 to 74.6 giga-
tons. Global Orchestration has the largest increase because it has the fastest pace of increas-

17
     Quantitative estimates in this section are based primarily on the IMAGE model.


                                                        33
SECOND DRAFT 1 December, 2003                  Chapter 9 Provisioning and Regulating Ecosystems Services


ing temperature and atmospheric CO2. Conversely, the Techno Garden scenario has the smal-
lest carbon uptake because it has the lowest temperature and CO2 levels. The largest uptakes
of CO2 occur in regions with extensive forests such as Russia and Canada.

On one hand these estimates give a realistic representation of the future climate regulation
function of the biosphere because they take into account the effect of deforestation in reduc-
ing the area of the biosphere, and shifts in vegetation zones caused by climate change. On the
other hand they may be over optimistic because they do not factor in possible changes in soil
processes that may lead to a net release rather than uptake of CO2 by the biosphere. Moreover,
the processes by which higher CO2 stimulates greater carbon uptake by plants is not very well
understood and may be incorrectly represented in current models. Finally, the estimates of
CO2 uptake presented here do take into account the future establishment of large-scale forest
plantations for storing CO2 from the atmosphere.


9.4.7 Water Regulation
To be written


9.4.8 Erosion Control
Contribution to be edited


9.4.9 Water Purification and Waste Treatment 18

―Water regulation‖ refers to the function of wetlands and other terrestrial ecosystems in help-
ing to remove substances harmful to humans and aquatic ecosystems from wastewater. Al-
though we have not directly addressed the future condition of wetlands in the quantitative
scenario analysis, we can estimate the trends of various indirect factors that will affect the ca-
pacity of wetlands to cope with wastewater. For example:
 Water availability – A large enough reduction in runoff can reduce the area and effective-
   ness of wetlands for processing wastes. The larger the reduction in runoff, the higher the
   risk to wetlands.
 Land encroachment – Wetlands are drained and occupied because of the expansion of
   agricultural or urban land. The larger the expansion of agricultural land and population,
   the greater the risk of disappearing wetlands.
 Magnitude of wastewater load – Large loads of wastewater may overload the capacity of
   wetlands to process wastes. The higher the loads of wastewater, the higher the risk of
   overload.

Global Orchestration. Up to 2050, this scenario has the largest reductions in runoff due to
climate change, but also the largest increases in runoff in other areas. (As discussed in Section
9.3.3. we expect most river basins to have increasing runoff due to climate change but an im-
portant smaller area will experience reductions in runoff.) The expansion of population and
agricultural land is large, but small compared to the other scenarios. Under this scenario re-
turn flows increase by 40 percent, the second lowest increase among the scenarios. In conclu-
sion, Global Orchestration is likely (with medium certainty) to lead to a reduction in the abili-
ty of wetlands to handle wastewater loadings, but this reduction may be lower (with low cer-
tainty) than the Order from Strength and Adapting Mosaic scenarios.

18
     Some results presented in this section stem from the AIM, IMAGE, IMPACT and WaterGAP models.

                                                    34
SECOND DRAFT 1 December, 2003                  Chapter 9 Provisioning and Regulating Ecosystems Services




Techno Garden. Under the Techno Garden scenario, runoff declines by the smallest amount
of all scenarios. The expansion of population and agricultural land is also the smallest, as is
the increase in return flows. An additional factor is that society in this scenario is oriented to-
wards environmental protection and therefore wetlands conservation may be a priority. (Al-
though one can argue whether wastewater discharges would be allowed into future protected
wetlands.) While we expect some reduction in the ability of wetlands to process wastes, the
risk is the lowest (with medium certainty) for this scenario.

Order from Strength. Under this scenario runoff does not decline very significantly up to
2050. However, the expansion of agricultural land and population is the largest among the
scenarios. Likewise, the magnitude of return flows is the largest among the scenarios. Under
Order from Strength we expect (with low certainty) that wetlands will be under the largest
risk and may experience a large reduction in their ability to process wastes

Adapting Mosaic. This scenario has a reduction in runoff comparable to Order from Strength.
The expansion of agricultural land and population is large, but not as large as Order from
Strength. The magnitude of return flows is second largest among the scenarios. Although
these factors would tend to reduce the viability of wetlands, society under this scenario is also
environmentally-oriented and may be inclined to preserve wetlands. Overall, the risk to wet-
lands and their ability to treat wastes may be less than Order from Strength but greater than
Global Orchestration.


9.4.10 Biological Control

To be written


9.4.11 Regulation of Human Diseases 19

9.4.11.1 Introduction

We interpret the regulation of human diseases to mean the ability of ecosystems to dampen
the spread of human disease. Here we consider two major aspects of regulating disease (1)
factors that lead to the spread of disease such as population density or the climate-related
range of disease vectors. (2) the average robustness of health, that is, the ability of people to
resist or recover from disease. In order to make a first judgment about the trend of disease
regulation under the MA scenarios we have examined various direct and indirect indicators of
these two aspects:

(a) Indicators Related to the Spread of Disease and Disease Vectors.
 Population density – The density of population has an important influence on the spread
    of disease in that the higher the density the faster the possible spread of disease. As surro-
    gate variables for population density we use total population and level of urbanization
    (since we have not estimated future population density).
 Climate-related range of vectors – It is known that many infectious diseases (vector-,
    food-, or water-borne) are sensitive to changes in climatic conditions. The IPCC has cited
    results of model-based studies to claim (with ―medium to high confidence‖) that under a

19
     Some results presented in this section stem from the AIM, IMAGE; IMPACT and WaterGAP models.

                                                    35
SECOND DRAFT 1 December, 2003             Chapter 9 Provisioning and Regulating Ecosystems Services


  range of climate scenarios there would be a net increase in the geographic range of poten-
  tial transmission of malaria and dengue vector-borne infections. Although we have not
  carried out calculations on the change of the range of disease vectors, here we assume
  (with medium certainty) that the larger the increase in temperature and precipitation, the
  greater the geographic range of important disease vectors.
 Amount of wastewater – The presence of quantities of untreated wastewater (especially
  from municipalities) obviously increases the risk of communicating disease. As an indica-
  tor of this risk we use the number of people living in river basins where return flows in-
  crease at least 100 percent between now and 2050.
 Area of irrigation – A well-known health problem in developing countries is the transmis-
  sion of Schistomiasis through disease vectors living in or near irrigation canals. Hence the
  future extent of irrigated areas in developing countries may be correlated (with low cer-
  tainty) to the spread of disease.

(b) Indicators Related to Average Robustness of Health
 Air pollution – The higher the level of air pollution, the higher the stress on human health
    (conversely, the lower its robustness). In reality thresholds of impact exist for most signif-
    icant air pollutants, but we cannot assess where these thresholds will be exceeded under
    the scenarios. As surrogate variables for air pollution we use the magnitudes of SO2 and
    NOx emissions.
 Number of malnourished children – As part of our assessment of world food production
    we have estimated the future number of malnourished children (see Section 9.4.1). There
    is an obvious inverse relationship between inadequate nutrition and robustness of health.
 Access to domestic water supply – Access to adequate freshwater is essential for main-
    taining health and resisting disease. Here we use the per capita domestic water use as
    rough indicator of access to an adequate water supply.
 Access to health care – Access to health care generally increases with income. Here we
    use average GDP per person per year as a rough indicator of access to health care.

(We should note that some of these factors are correlated with each other, and therefore con-
sidering them as independent factors results in some double-counting.)

9.4.11.2 Scenario Results

Global Orchestration.
(a) Spread of Disease – Population is significantly higher in 2050 than today in all regions,
except in the OECD and FSU where it does not markedly change. The percentage of popula-
tion living in urban versus rural areas increases by about 10 percent in almost all regions and
all scenarios up to 2050. The exceptions are the FSU countries where the level of urbanization
is much lower under the Order from Strength and Adapting Mosaic scenarios. With regards to
climate change, there is about a 2 0C increase in temperature between now and 2050, and an
overall increase in precipitation as discussed in Section 9.3.3.. We note that the scenarios do
not differ too much from each other up to 2050, although these differences become larger by
2100. (But by 2050 there are already larger differences in the rate of temperature change as
noted in Section 9.3.3). With regards to increasing wastewater, over 5 billion people will live
in river basins where return flows will at least double between now and 2050 (Table 9-9).
Consistent with the storyline of Global Orchestration, we can expect (with low certainty) that
much of this wastewater may be untreated. Concerning the extent of irrigated area, we assume
under this scenario that this area will increase world-wide over 10 percent between now and
2050, with particularly large growth in Latin America and Sub-Saharan Africa.


                                               36
SECOND DRAFT 1 December, 2003              Chapter 9 Provisioning and Regulating Ecosystems Services


To sum up, the trends of population, urbanization, climate change, wastewater, and irrigated
area are all moving in the direction of higher risk of spread of disease. The trends are more
unfavorable in Latin America and Sub-Saharan Africa than in other regions. Unfortunately it
is difficult to quantify this increased risk. At best we can compare it qualitatively to the risk of
other scenarios, which we do in the following paragraphs.

(b) Robustness of People – With regards to the level of air pollution, the emissions of SO2 be-
tween now and 2050 decline in all regions (except in Sub-Saharan Africa), while NOx emis-
sions strongly increase on average in every region (except in MENA and Sub-Saharan Africa
where the increase is modest). Up to 2050 we expect the number of malnourished children
under this scenario to decline from 166 to 49 million children. Average domestic water use
grows strongly everywhere except in OECD countries where it is already near saturation,
while average income increases globally by more than a factor of 13. In total, most factors
point to an increase in robustness of health of most of the world’s population.

Techno Garden.
(a) Spread of Disease – OECD and FSU countries have about the same population in 2050 as
in Global Orchestration. The population outside of these regions is slightly higher than under
Global Orchestration. Urbanization trends are as noted for Global Orchestration. The tem-
perature increase up to 2050 (about 1.6 0C) is the lowest for all scenarios. About 4.8 billion
people will live in river basins where return flows at least double between now and 2050, but
the storyline of the Techno Garden scenario suggests that most wastewater (at least from mu-
nicipalities and industries) will be treated. Concerning the extent of irrigated area, we assume
under this scenario that this area will increase world-wide over 5 percent between now and
2050. As in the Global Orchestration scenario, growth is particularly large in Latin America
and Sub-Saharan Africa.

To sum up, while the preceding trends are all moving in the direction of higher risk of spread-
ing disease, this risk may be lower than under the Global Orchestration scenario – Population
is slightly higher, the degree of climate change is lower, fewer people are likely to be exposed
to untreated wastewater, and irrigated land does not expand as much.

(b) Robustness of People – Concerning the trend of air pollution, the emissions of SO2 decline
even stronger than in the Global Orchestration scenario in all regions (except in Sub-Saharan
Africa where they also double). Meanwhile NOx pollution declines substantially in OECD and
Latin America, and increases everywhere else about the same as in Global Orchestration. Up
to 2050 we expect the number of malnourished children under this scenario to decline from
166 to 79 million children (this means more malnourished children than under the Global Or-
chestration scenario.) Domestic water use grows almost as strongly as in Global Orchestra-
tion, and global average income increases by a factor of 10. Considering the trends of these
factors, the overall robustness of health is expected to improve to nearly the same level as in
Global Orchestration.

Order from Strength
(a) Spread of Disease – Population in 2050 is slightly lower in OECD and 10 percent lower in
the FSU under this scenario. Other regions have the highest increases of all scenarios. The
level of urbanization is comparable to Global Orchestration except in the FSU where it de-
creases. The degree of temperature increase is between the Global Orchestration and Techno
Garden scenarios. As compared to the Global Orchestration scenario, many more people will
live in river basins where return flows at least double by 2050 (6.8 billion people). Consistent
with the storyline of Order from Strength, much of this wastewater outside of OECD may re-

                                                37
SECOND DRAFT 1 December, 2003            Chapter 9 Provisioning and Regulating Ecosystems Services


main untreated. With regards to irrigated area, there is a small net world-wide decline be-
tween now and 2050, especially due to a large decline in FSU and Asia. However the extent
of irrigated are increases in Latin America and Sub-Saharan Africa.

Of all the scenarios, Order from Strength shows the most unfavorable trends (with regards to
the spread of disease) for population and wastewater. However, its trends for urbanization,
climate change, and irrigated area are not the most unfavorable. Hence, the risk of spread of
disease may be (with low certainty) close to that of Global Orchestration and Techno Garden
scenarios. However, we may expect it to be closer to the Global Orchestration scenario be-
cause of the low attention given to environmental protection in this scenario. As in all of the
scenarios, the regions under highest risk may be (with low to medium certainty) Latin Ameri-
ca and Sub-Saharan Africa (because of the trends in almost all variables).

(b) Robustness of Human Health – With regards to the trend of air pollution, the level of sul-
fur-related air pollution declines only slightly world-wide under the Order from Strength sce-
nario, because of declines in OECD, FSU and Latin America. Asia and MENA have the larg-
est emission increases of all scenarios. There is a significant decline in NO x-related pollution
in OECD countries, and a major increase elsewhere. Up to 2050 we expect the number of
malnourished children under this scenario to decline from 166 to 151 million children, a much
smaller decline than under Global Orchestration. Domestic water use increases more than in
the other scenarios (except in Sub-Saharan Africa where it has a large increase over today, but
the smallest relative to the other scenarios.) Global average income grows by a factor of 3.6,
the smallest growth by far among the scenarios. In sum, most indicators point to an increase
in robustness, but of the smallest magnitude of all the scenarios.

Adapting Mosaic
(a) Spread of Disease – The population of OECD and FSU countries is about the same as to-
day. Other regions have large increases over today’s population, but the increases are not as
large as under the Order from Strength scenario. The degree of temperature increase is be-
tween the Global Orchestration and Techno Garden scenarios. Many people will live in river
basins where return flows will double (6.1 billion) but this is not as many as under the Order
from Strength scenario. According to the storyline of Adapting Mosaic, we expect (with low
certainty) that much of the wastewater in 2050 will be treated. Concerning the extent of irri-
gated land, there is a slight world-wide increase between now and 2050 which reflects the dif-
ference between a big decline in FSU and Asia, and large increases in Latin America and Sub-
Saharan Africa.

To sum up, this scenario shows intermediate trends for all variables. While the trends of these
indicators indicate strongly increasing risk over time, this risk may fall (with low certainty)
between that of the Global Orchestration and Techno Garden scenarios. It may (with low cer-
tainty) be closer to Techno Garden because society finds local solutions to environmental
problems.

(b) Robustness of People – Concerning air pollution, sulfur-related pollution declines in all
regions except Asia, where it has a slight net increase. Trends for NO x are similar to the
Global Orchestration scenario. Up to 2050 we expect the number of malnourished children
under this scenario to decline from 166 to 116 million children which is a better result than
Order from Strength, but not as good as the achievements of Global Orchestration or Techno
Garden. Domestic water use increases strongly here in all scenarios. Global income increases
by over a factor of 6, somewhat higher than Order from Strength but lower than the other sce-


                                               38
SECOND DRAFT 1 December, 2003                   Chapter 9 Provisioning and Regulating Ecosystems Services


narios. To sum up, robustness also increases under this scenario but to an intermediate degree
compared to the other scenarios.

9.4.11.3 Summing Up Regulation of Human Diseases

While the risk of spreading disease is becoming greater across the scenarios, so is the robust-
ness of the population to resist disease. It is not possible here to judge which of these tenden-
cies will be more important.

The Techno Garden seems to have a lower risk of spreading disease than the Global Orches-
tration scenario but the robustness of the population may also be lower. Hence, these two sce-
narios (with low certainty) may have the same overall ability to regulate human diseases. The
other two scenarios, Order from Strength and Adapting Mosaic are also close to each other,
and have a lower ability to regulate human disease. Trends in regulating diseases are most fa-
vorable for OECD and FSU, and least favorable for Latin America and Sub-Saharan Africa.


9.4.12 Pollination 20

The effectiveness of pollination is related to a wide variety of factors, including the condition
of pollinators and their efficiency in pollinating. The condition of pollinators is related to the
load of pesticides used to control pests on agricultural land since pesticide also affects pollina-
tor reproduction. The application of pesticide (total mass per unit agricultural area) is ex-
pected to be heaviest under the Global Orchestration where yields and overall food produc-
tion are highest and there is small concern about the environmental impact of intensive agri-
culture. Next heaviest could be the Order from Strength scenario which has lower yields and
production, but also a laisse faire attitude towards environmental protection. Third in line in
pesticide application could be the Techno Garden scenario with high yields and production
but also a high level of environmental consciousness. The lowest level of pesticide application
is likely to be the Adapting Mosaic scenario because food production and yield is low com-
pared to most other scenarios and the level of environmental consciousness high.

Another important factor affecting the effectiveness of pollination is the level of landscape
fragmentation, that is, the degree to which natural landscapes are broken up by different land
uses of society. (The higher the fragmentation level, the more difficult it is for pollinators to
reach plants). It is difficult to estimate which scenario would have the highest rate of frag-
mentation. On one hand this might be Order from Strength which has the largest expansion of
agricultural area and gives low priority to environmental protection. On the other hand Global
Orchestration has the most intensive agricultural production and a low level of environmental
consciousness. Either of these two scenarios are likely to have the highest level of fragmenta-
tion. To sum up, with low certainty we can order the scenarios according to their risk of inter-
ference with pollination, from highest to lowest: Order from Strength, Global Orchestration,
Techno Garden and Adapting Mosaic.


                                21
9.4.13 Storm Protection

―Storm protection‖ is the ecosystem service that describes the role of ecosystems in protecting
society from storm damage. Here we focus on coastal storm protection, and especially on how
20
     Results presented in this section are based on the IMAGE and IMPACT models.
21
     Estimates of sea level rise are provided by the IMAGE model.

                                                      39
SECOND DRAFT 1 December, 2003                     Chapter 9 Provisioning and Regulating Ecosystems Services


the level of protection may be different under the different scenarios because of changes in
adaptive capacity and sea level. In Section 9.3.3 we describe the sea level rise that is expected
to accompany the climate change scenarios used for our scenario analysis. Sea level will rise
because warmer temperatures will melt permanent ice and snow and cause a thermal expan-
sion of ocean water. Furthermore, climate change may cause stronger and more persistent
winds in the landward direction along some parts of the coastline and this will also contribute
to rising sea level at these locations. In the four scenarios (given a medium climate sensitivi-
ty) the global-mean sea level is expected to increase from 50 cm (Techno Garden) to 70 cm
(Global Orchestration) in 2100 (Figure 9-6).

While the precise impact of sea level rise on reducing storm protection is difficult to assess,
we estimate (with high certainty) that populated coastal areas under all scenarios will require
new storm protection measures such as stronger and higher dikes and/or flood gates in estu-
aries. These measures are all expensive undertakings and they might be affordable only in the
world’s richer countries. Unfortunately, we could not analyze the financial capabilities of in-
dividual nations for storm protection. However, as a first crude approach, we assume that the
adaptive capacity of nations is proportional to their average income (Table 9-5). With this rea-
soning, nations under the Global Orchestration have the highest adaptive capacity, followed
by Techno Garden, Adapting Mosaic, and Order from Strength. Using the same rationale, the
OECD countries should have the highest level of adaptive capacity throughout the scenario
period for all scenarios, followed by Latin America, FSU, Asia, MENA and Sub-Saharan
Africa.

We now combine this rough estimate of adaptive capacity with the computed rates of sea lev-
el rise shown in Figure 9-6. We assume the higher the adaptive capacity and the slower the
rate of sea level rise, the higher the level of storm protection. Under these assumptions, either
Techno Garden or Global Orchestration will have the highest level of protection, and either
the Adapting Mosaic or Order from Strength scenarios will have the lowest level.

9.5      Linkage with Well-being

This section can build especially on Sections 9.4.1 and 9.4.11.
 Per capita caloric food consumption
 Domestic water use
 Energy use
 Linkage with health regulation above
 Income

9.6      Review of Uncertainty

How certain are the model results used to quantify the scenarios?

As discussed in Chapter 4, model results are highly uncertain since they make assertions
about the enormously complex ecosystems of the world over several decades into the future.
Another problem is that in some cases the output from one model is used as input for another
model, and therefore it can be argued that the uncertainty of the models propagate and multip-
ly. However, this problem can be lessened by interpreting modeling results in a conservative
way.22 A further drawback of the modeling approach is their lack of connections and feed-
backs between human and environmental systems, as discussed below.

22
     For example, results from a simple climate model are input to a global water model to compute changes in ru-

                                                        40
SECOND DRAFT 1 December, 2003                     Chapter 9 Provisioning and Regulating Ecosystems Services




But the main point is that despite their uncertainties, models allow us to combine complex
ideas and data from the social and natural sciences together in a consistent way that provides
useful information to supplement the storylines presented in Chapters 5 and 6. Indeed, the
modeling results show certain tendencies as discussed in the previous paragraphs that can help
us anticipate coming risks to ecosystem services. Moreover they can provide information that
is useful for developing policies to lessen these risks.

Do the scenarios cover the entire range of possible futures?

The quantitative scenarios do not cover the entire range of possible futures because they do
not contain major surprises which we know from history will have a profound effect on eco-
system services (breakthroughs in technology, unexpected migration movements, major in-
dustrial accidents). The models used to quantify the scenarios also do not generate ―breaking
points‖ in which ecological thresholds are exceeded (e.g. rapid changes in water quality or
pest outbreaks over large agricultural areas). Exceeding these thresholds could have important
consequences on the future of the world’s ecosystems. Models do not generate breaking
points because they poorly represent global feedbacks and linkages. This unfortunately re-
flects the state of the art of global modeling which needs to be improved to address urgent
MA-relevant questions (see Chapter XX). On the other hand the scenario storylines do in-
clude many examples of surprises and breaking points.

Which ecosystem services are best represented, which the least?

One measure of the certainty/uncertainty of estimates in this chapter is the availability of
quantitative estimates for ecosystem services. As shown in Table 9-10, only a small number
of ecosystem services could be quantified for this chapter, and many of the indicators were
only indirectly related to the ecosystem service. On the other hand, quantitative estimates
were available for two of the most comprehensive ecosystem services – food and freshwater.

9.7 Cross-Cutting Synthesis
(Some Preliminary Notes for Jon and Wolfgang)

1. The scenarios clearly illustrate tradeoffs between ecosystem services

Although scenarios were computed out to 2100, the following results refer to year 2050 which
is given particular attention in the MA scenarios. A robust preliminary conclusion is that the
scenarios in general depict an intensification of the tradeoffs already observed between differ-
ent ecosystem services.

(i) Possible Gains in “Provisioning Services”.
 World total production of grains increases around 50 percent for all scenarios, with larger
    differences between scenarios for the poorer world regions. However per capita consump-

noff due to climate change. In this case the uncertainties of the climate model are propagated to the water model.
This problem can be minimized by recognizing that the uncertainty of the climate model is relatively high for
computed spatial patterns of temperature and precipitation, but not so high for the magnitude and direction of
these changes. Therefore, statements about the changes in runoff at particular locations will be highly uncertain
and should be avoided, whereas statements about the size of the area in a large region affected by increasing or
decreasing runoff have a lower level of uncertainty and are appropriate for the MA scenario analysis. The key is
to aggregate results either spatially or temporally because uncertainties that are important on the fine scale partly
cancel out when data are aggregated.


                                                        41
SECOND DRAFT 1 December, 2003             Chapter 9 Provisioning and Regulating Ecosystems Services


  tion of grain (for food and feed) remains near its current level of around 300 kilograms per
  year. Consumption in the Sub-Saharan region does not substantially increase under all
  scenarios.
 Domestic water use per person per year grows in the Sub-Saharan and other poorer re-
  gions by a factor of five or more (depending on the scenario and region), and this implies
  an increased access of the population in these regions to freshwater. In OECD countries,
  there is a decline in domestic water use per person because of more efficient water use.
  Because of stabilization of food consumption and gains in access to water supply and oth-
  er factors, the percentage of malnourished children falls by 40 percent in Sub-Saharan
  Africa under the Global Orchestration scenario. The decline is much smaller under the
  Order from Strength scenario.
 The amount of wood extracted from remaining forests for fuel wood and fuel products is
  likely to greatly increase up to 2050 (despite the loss of land noted below). We have not,
  however, analyzed the sustainability of this wood extraction.

(ii) Possible Losses in “Provisioning Services”.
 Some of the gains in agriculture will be achieved through expansion of agricultural land,
     and at the expense of uncultivated natural land. A very first rough estimate is that, up to
     2050, 10 to 20 percent of current grassland and forest land will be lost, mainly due to the
     expansion of agriculture (and secondarily, because of the expansion of cities and infra-
     structure). The provisioning services associated with this land (genetic resources, wood
     production, habitat for terrestrial biota and fauna) will also be lost. (We noted above that
     the loss of wood production on this land might be compensated by more intensive produc-
     tion elsewhere.)
 Although gains are made in access to freshwater, the scenarios also indicate a likely in-
     crease in the volume of polluted freshwater (especially in developing countries if the ca-
     pacity of wastewater treatment is not greatly expanded). Moreover, the expansion of irri-
     gated land (which contributes to the increased production of grains) leads to substantial
     increases in the volume of water consumed in arid regions of Africa and Asia. These and
     other changes in the freshwater system are likely to cause a reduction in the provisioning
     services now provided by freshwater systems in developing countries (e.g. genetic re-
     sources, fish production, habitat for other aquatic and riparian animals)

(iii) Uncertain Changes in Regulating Services

It is not clear whether climate regulation will be increased or decreased under the scenarios.
On one hand, the warmer, moister climate will, on average, increase primary productivity and
the uptake of CO2 in the atmosphere. On the other hand the depletion of natural forest and
grassland may lead to a decrease in standing biomass on the earth. This question must be fur-
ther analyzed.

(iv) Possible Losses in Regulating Services
 The scenarios all assume an increase in per capita income and imply an increase in ma-
    terial well-being. This is likely to lead to higher consumption electricity and fuel for trans-
    port, as well as a higher production of industrial products. The result will be a decline in
    air quality maintenance, as indicated by a substantial rise in the emissions of sulfur dio-
    xide and nitrogen dioxides, especially in developing countries. Whereas richer countries
    are expected to maintain or expand their control of local and regional air pollution, the
    same is not expected for developing regions.
 The loss of natural land, discussed above, will also affect the regulating services provided
    by this land (erosion control, regulation of human diseases, water regulation).

                                                42
SECOND DRAFT 1 December, 2003              Chapter 9 Provisioning and Regulating Ecosystems Services




2. The rates of temperature and precip change are significantly different between scenarios

Rate of T & precip change
To 2050
GO – fast up
TG -- medium down
OfS, AM – slow up

After 2050
GO – fast down
TG medium down
OfS – slow up
AM – slow down

3. Different scenarios have different likelihood of surprises and breaking points

Order from Strength
 Pressure on the environment is greatest from the Order from Strength scenario because of
   the combination of slow, unrelenting growth of population, slower economic growth, and
   lack of interest in environmental management. Consequences are a linear increases in
   global energy use throughout the century (Figure 9-XX), an acceleration of current defore-
   station rates with near depletion of forests in Africa and parts of Asia by the end of the
   century…
 Under these circumstances, society is most likely to be surprised by thresholds being ex-
   ceeded. Disappearance of important species and products?

Global Orchestration may also be the world most confronted with unknown consequences of
climate change, since here the average rate of change of temperature and precipitation is like-
ly to be the fastest in the first half of the century. Of all the scenarios, this world could see the
greatest consequences of climate change on water resources, changing natural vegetation and
crop yield. Also intensification of agriculture is extreme.

Techno Garden has high level of agricultural intensification.

4. Disruption of Landscape and their Ecosystem Services by Mineral Exploitation (visual,
biodiversity, tourism).

One factor affecting the degree of disruption of landscape will be the intensity and type of
mineral exploitation. From the scenarios we can deduce that the biggest disruption by far will
be caused by the Order from Strength scenario where total fossil fuel use increases by more
than a factor of 2.5 by 2100 as compared to 2000. Not only is the magnitude of fossil fuel use
large, but in this scenario society gives environmental protection low priority. This combina-
tion of factors suggests that mineral exploitation will have the largest impacts on the land-
scape under this scenario. Following in impact is the Global Orchestration scenario where
fossil use increases about a factor of two over the same period, and where environmental
management is also largely neglected. The impact is likely to be the smallest under the Tech-
no Garden scenario because fossil fuel substantially declines up to 2100 (Figure 9-XX), and
because environmental management is given high priority. An intermediate case is the Adapt-
ing Mosaic scenario which also gives priority to environmental protection, but fossil fuel use
nearly doubles up to 2100.

                                                 43
SECOND DRAFT 1 December, 2003           Chapter 9 Provisioning and Regulating Ecosystems Services




5. Scenarios indicate hot spot regions of particularly rapid changes in ecosystem services

 Central part of Africa – Rapid increase in withdrawals, return flows, and water stress; de-
  forestation; expansion of modern agriculture and inputs to agricultural land. Possible grain
  exporter?
 Middle East – Rising income, greater meat demand, more import dependency, decreasing
  water availability, higher water stress because of decreased water availability and in-
  creased withdrawals
 South Asia – Continuing deforestation, intensive industrial inputs to agriculture, rapidly
  increasing return flows and water stress.

6. Changes in ecosystems will be as big as over the last decades. Dynamic changes in coastal
areas, forests, agricultural land. Ag land further intensified.




                                             44

				
DOCUMENT INFO