|
|
| Line 1: |
Line 1: |
| {{ComponentTemplate2 | | {{ComponentTemplate2 |
| |Status=On hold | | |Status=On hold |
| |Description=1 Drivers | | |FrameworkElementType=driver |
| | |
| 1. Introduction
| |
| In order for the IMAGE model to explore future scenarios, exogenous assumptions need to be made for a range of factors that shape the direction and rate of change in key model variables and results. Together with the endogenous functional relationships and endogenous model parameters that jointly typify the behaviour of the model, these exogenous assumptions drive the outcome of any particular model calculation. In other words, these assumptions are the drivers that determine the model results subject to the assumed external conditions.
| |
| | |
| For IMAGE, six groups of assumptions are distinguished that make up the scenario drivers. In this chapter these six scenario drivers will be described. These drivers are the basis of any scenario and are generally rooted in a scenario narrative or storyline. This includes cases where current trends and dynamics are assumed to continue into the future; commonly referred to as reference or ‘business-as-usual’ cases. As a rule, scenario drivers do not constitute numerical model inputs, but they do govern in qualitative or semi-quantitative terms the more elaborate set of exogenous assumptions in terms of model input files for the various parts of the model framework. In other words, numerical model drivers for any specific scenario will need to be established on the basis of the six more generic scenario drivers. The model drivers for the various modules will be explained in more detail in the following Chapters 4 through 8. The scenario drivers and underlying narrative, together with the concrete quantitative model drivers form a scenario, inextricably linked with the results from the IMAGE scenario run
| |
| | |
| 2. Scenario drivers
| |
| When considering how the world might unfold in the longer term, six key scenario drivers are distinguished: demographics, economics, culture and lifestyle, natural resources, technological development and the policy and governance environment. To a large degree these scenario drivers are interdependent, and their future direction is often inferred from a simple to very elaborate ‘storyline’ or narrative. Such storylines describe the type and function of the scenario at hand: a reference projection with no new policies, a single ‘best-guess’ projection combining trends from the past with assumptions about how they might unfold in the future, multiple contrasting scenarios that span a range of uncertainty about the future, or a specific or broad policy scenario aiming to improve future outcomes. For examples, see (Brink, 2010; OECD, 2008; OECD, 2012; UNEP, 2011; Westhoek et al., 2011)
| |
| | |
| Demographics
| |
| The future state of the world will depend on how many people are expected to populate the world, as all –average- per capita activities scales with the number of people to obtain regional and global demands for goods and services. Typically the population projections that are used as input to drive the IMAGE model are adopted from published sources such as UN and IIASA. Behind these numerical projections are economic, technical, educational and policy assumptions that influence the estimated future population as the net outcome of fertility and mortality, adjusted for migration flows. This provides internally consistent, or at least plausible, overall population scenarios on the basis of underlying demographic trends. At the level of model drivers, besides the total population size the breakdown over gender, income classes, urban/ rural split, and the level of education matter for such issues as consumption preferences and patterns, and access to goods and services. With use of a downscaling procedure, the population at country or regional level is also available at the grid level; see Chapter 2. Population data are used in the energy- and agricultural economics modeling, but also directly in various other modules of IMAGE such as water stress, nutrients, flood risks and human health.
| |
| | |
| Figure 3.1: Scenario drivers (left) and groups of model drivers (right) are inferred from the scenario narratives, and model driver quantifications are informed by external data sources, such as time series and cross-sector data, and literature sources.
| |
| | |
| Economics. At the most aggregate level, the level of economic activity is typically described in terms of GDP (Gross Domestic Product) per capita. Models outside the IMAGE 3.0 framework estimate long-term GDP growth perspectives, following from development of the key production factors (e.g. capital, labour, natural resources) and the sectoral composition of the economy; such as the ENV-Growth model of the OECD. The overall GDP is composed of Value Added (VA) per relevant sector and private consumption, estimated with more detailed models accounting for intersectoral linkages, responses to prices, and others (Chateau, 2013).. Economic variables are used as model drivers for the energy demand model (chapter 4.1.1), and the non-agricultural water demand contributing to water stress (Chapter 6.1.3). For the household energy demand model, the average income is broken down into urban and rural population, and for each into quintiles of income levels. The latter is derived from the assumed, uneven income distribution by means of the so-called GINI factor. The GINI factor is a measure of income disparity in a population. If all enjoy the same income, GINI equals unity. The lower the GINI is, the wider is the gap between the lowest and highest income groups. The agricultural economy model MAGNET is in fact an economy-wide Computable General Equilibrium (CGE) model, that is able to reproduce the exogenous GDP growth projections made with the less complex economic growth models; see section 4.2.1. The macro indicator GDP/capita is also used directly in various other IMAGE modules, such as nutrients in urban wastewater, human health, and flood risks.
| |
| | |
| Culture and lifestyle. For comparable levels of affluence, the observed behaviour of societies with respect to consumption differs strongly between countries and regions, and to a lesser extent also within countries. The modal split of passenger transport between walking, cycling, cars, buses, trains, boats and planes depend on income, but also on traditions and habits engrained in social groups. Preference for different foodstuffs depends on availability and affordability, but also strongly on cultural factors. These may be inspired by religion (no pork for Jewish and Islamic households; no beef or no meat at all for Hindus; etc.), but also by tradition, public moral and health concerns. In addition, concerns over environmental degradation, animal welfare, inter-regional and intergenerational equity and other issues can influence actual behavior, in accordance with dominant societal norms and values. Consumer preferences and lifestyles can change over time, as norms and values can change. The direction and rate of such changes can be inferred from the underlying scenario storyline. Policies may be put in place to enable, encourage or even enforce such changes, given sufficient public support.
| |
| | |
| Natural resource availability
| |
| Ultimate natural resources are considered a given and are not subject to adjustment by human activities, other than depletion of non-renewable resources by transforming them. For example extracting coal from underground deposits to burn them to generate thermal energy and a range of -mostly gaseous-substances- including carbon dioxide. Similarly, the total land surface is fixed, except for relatively limited man-made expansion of land into shallow coastal waters. Human activities such as agriculture and forestry, however, transform the land-cover and land-use and thereby alter the quantity of suitable land for agriculture..
| |
| | |
| What can be assumed as exogenous scenario drivers are assumptions about how much of the ultimate resource is available in future. This depends on assumptions about future technological capabilities and assumed policies reflecting human preferences. Technological optimism is associated with higher end estimates of non-conventional fossil and nuclear fuel reserves. Nature conservation and other concerns can limit the potential for conversion of natural land for agriculture, but may also impose limits to hydro-electricity generation.
| |
| | |
| In IMAGE 3.0 renewable and non-renewable energy resource by volume and price are explicitly modeled; see section 4.1.3. Similarly the potential for land to be used for agriculture, ranked by suitability and subject to nature conservation policies, sets boundaries on future land conversion; see section 4.2.3.
| |
| | |
| Technological change
| |
| At the overall scenario level, assumptions about technological progress are the all-important drivers of economic growth. Given an effective labour force, the increase in labour productivity delineates the potential for economic growth. It is generally assumed that the degree of technological progress at the scenario level is bound to be reflected in all instances where technology plays a role. In other words: rapid technological growth assumptions behind high economic growth rates imply that technologies in specific sectors (energy, agriculture) will also improve relatively quickly. Across and within sectors the directions of technological change may differ, however. For example, renewable energy technologies may improve much more rapidly than fossil fuel based technology, an uncertain factor.
| |
| | |
| In the energy sector, see section 4.1, the improvement in technologies over time is largely governed by an endogenous mechanism that links technology cost to the cumulative installed capacity: learning-by-doing. In agriculture technological factors are estimated exogenously, based on historical data and published projections of crop and livestock productivity, water and fertilizer use efficiency, irrigation systems performance, etc.
| |
| | |
| Policy and governance
| |
| At the scenario level it can make a big difference if –implicit or explicit- policies are assumed that reflect alternative directions that shape future development of human and natural systems. And it matters also how governance structures and institutional settings are assumed to evolve. Overall policy thinking can vary in many ways, but in the context of IMAGE important scenario bifurcations include more focus on the shorter term and/or on material wealth, versus longer term sustainability concerns. But also inter-regional and/or intergenerational equity may gain more or less weight as underlying future trend. As mentioned under culture and lifestyle, such assumed overall policy directions have the potential to influence pretty much all relevant scenario drivers, and model drivers alike.
| |
| | |
| In addition to the alternative policy directions, it matters how the governance structures and associated institutional arrangements develop over time in different regions of the world, or different groups of regions which share certain characteristics, such as high-income industrialized, medium income emerging or low income developing countries.
| |
| | |
| Besides the structural elements, the assumed policy and governance environment can also make certain policy measures and instruments more or less plausible. For example, a concerted and jointly implemented global climate mitigation strategy can hardly be imagined in a world of diverging regions primarily occupied with shorter term domestic interests, including barriers to international trade and weak intergovernmental bodies.
| |
|
| |
| Relations between scenario drivers
| |
| As mentioned in the descriptions of the six scenario drivers, assumptions for each tend to depend to a lesser or greater extent on what is assumed for one or several of the other scenario drivers. Plausibility of the set of drivers, over and above the plausibility of each driver in its own right, therefore hinges on a careful consideration of the nature and direction of these relationships. An overarching story or narrative is proven to be helpful in choosing meaningful combinations across each of the six domains (IPCC, 2000); (Millennium_Ecosystem_Assessment, 2006).
| |
| | |
| 3. Model drivers
| |
| Direct model drivers are inferred from the scenario drivers, and are used to set parameter values used in IMAGE 3.0. Their start values are estimated from literature and data, and future changes in values are inferred from the ultimate drivers. The resulting parameter values are used as input for different parts of IMAGE 3.0. A selection of key model drivers is given in Table 3.1, for further details and other model drivers not shown here; see Chapters 4 through 8.
| |
| | |
|
| |
| | |
|
| |
| Table 3.1: Model drivers
| |
| Model driver groups / model drivers Used for (chapter) Source
| |
| Population projections
| |
| Population per region Many chapters UN, IIASA
| |
| Gridded population Water stress (6.1.3); River Flooding (7.3); Human development (7.6) Own downscaling
| |
| Urban/rural split of population Human development (7.6) UN, IIASA
| |
| Economic development pathways
| |
| GDP/capita Agriculture economy and forestry (4.2.1); Water stress (6.1.3); River Flooding (7.3); Human development (7.6); Climate policy (8.1) OECD, CPB
| |
| Value Added per region and major sector (Industry, Agriculture, Services, other); Household consumption (also per income quintile for energy demand) Energy demand (4.1.1); Water stress (6.1.3); Human development (7.6) OECD, CPB
| |
| GINI coefficient Energy demand (4.1.1); Human Development (7.6) WorldBank
| |
|
| |
| Trade regimes, tariffs, etc.
| |
| Agricultural trade policies Agriculture economy and forestry (4.2.1) GTAP
| |
| Trade restrictions Energy supply (4.1.3) Own assumptions
| |
| Environmental and other policies
| |
| Protected areas Agriculture economy and forestry (4.2.1), Land use allocation (4.2.3) Own assumptions
| |
| Climate policy Climate Policy (8.1) Own assumptions
| |
| Air pollution policies Energy conversion (4.1.2) Own assumptions
| |
| Biofuel policies Agriculture economy and forestry (4.2.1), ?? which of 4.1.1.-4.1.3? Own assumptions
| |
| Energy security Energy security policies (8.2) Own assumptions
| |
| Technological change in the energy system
| |
| Energy intensity parameters Energy demand (4.1.1) IEA
| |
| Technology Energy demand (4.1.1)
| |
| Technological development Energy conversion (4.1.2) Many
| |
| Rules on use technologies
| |
| Energy conversion (4.1.2) Own assumptions
| |
| Learning rates Energy supply (4.1.3) Own estimates
| |
| Technological change in agricultural and forestry
| |
| Yield increase Agriculture economy and forestry (4.2.1) Own estimates, based on FAO
| |
| Cropping intensity Agriculture economy and forestry (4.2.1) Own estimates, based on FAO
| |
| Irrigation project and conveyance efficiency Hydrology (6.1.3) PIK
| |
| Expansion of irrigated areas Hydrology (6.1.3) IIASA, FAO, others
| |
| Livestock production parameters (systems mix, food conversion efficiency, ration, carcass weight, manure management) Livestock systems (4.2.4) Own estimates, based on ???
| |
| Forestry production parameters (systems mix, harvest of timber, board, pulp per km2) Forest management (4.2.2) Own estimates
| |
| Lifestyle parameters
| |
| Income, own- and cross- price elasticities of agricultural commodities demand Agriculture economy and forestry (4.2.1) GTAP
| |
| Lifestyle parameters Energy demand (4.1.1) Own estimate
| |
| Preferences Energy demand (4.1.1) Own estimate
| |
| | |
| 4. Drivers and types of scenarios
| |
| Under baseline conditions scenario drivers are typically assumed to develop either along a pathway that is considered a best guess translation of current trends into the future. Alternatively, multiple contrasting scenarios can be considered to explore the range of plausible future trends. The former approach is chosen in many studies, such as the OECD Environmental Outlook (OECD, 2012), with the specific aim to serve as starting point for policy interventions aiming to improve the baseline outcomes within and across sectors and domains. The latter approach recognizes structural uncertainties in how the world might unfold, and aims to explore how such uncertainties would play out in future ranges of outcomes. They also serve to investigate how robust policy interventions play out under different future conditions. Examples of multiple baseline studies are (IPCC, 2000), (Millennium_Ecosystem_Assessment, 2006). See Figures 3.2 and 3.3 for illustrative results from the so-called Shared Socio-economic Pathways (SSPs), recently developed to support climate change research across different research communities. (Moss et al.; Vuuren et al., 2012). An important element of the SSPs are the qualitative narratives or storylines characterizing alternative futures. From there, assumptions are made about internally coherent sets of scenario drivers, and finally key model drivers such as population and GDP growth factors.
| |
| | |
| Figure 3.2 illustrates the wide range in the population size projected for the long term: by 2100 the world could see around the same number of people as today, or about twice as many. The projections are made by IIASA, using a population (Millennium_Ecosystem_Assessment, 2006)modeling approach that links aggregate education levels to fertility and mortality rates per country, and together with migration flows these determine the future population size.
| |
| | |
| Taking the population projections and the underlying educational attainment per age cohort, long term economic growth models are used to project the associated economic development expressed as GDP per capita. Together with the population the total GDP is calculated. This was done by three different teams at OECD, IIASA and PIK, each using their own models. Figure 3.3a shows the GDP projections until 2100 from the OECD model ENV-Growth, differing by a factor 3.7 in 2100. As the most populous scenarios are at the lower end of the GDP range, the differences in GDP per capita will be even bigger than that. Differences in population and economic growth rates between different countries and regions make that the distribution of total economic assets over different parts of the world are bound to shift, with Asia and later Africa and to a lesser extent Latin America becoming dominant in the world; see Figure 3.3b.
| |
| | |
| As mentioned, different models were used by teams at different institutes, allowing to investigate how different model structures and assumptions, but also different interpretations of the qualitative scenario story lines in model parameters, lead to quite different results (Figure 3.3c). The projections for the SSP3 scenario not only differ with respect to the level in 2100, but also in the pathway over the century.
| |
| | |
| For more information on how Baseline scenarios play out in more detail in economic, social and ecological terms, see the results per module of the IMAGE 3.0 framework in chapters 4 through 8.
| |
| Once Baseline scenarios are implemented, modifications at various levels can be made that reflect policy interventions aiming to diverge from the trends emerging under baseline conditions. This can take many different shapes and forms, depending on the subject, scale, timeframe and policy levers under consideration. Reducing climate change impacts is one obvious example, but also reducing nutrient loading of coastal sees, slowing down the rate of biodiversity loss, reducing water stress, and many other options to alleviate anticipated future problems have been explored with IMAGE during the last decade. Again, for more information on policies, instruments and goals pursued, see Chapters 4 through 8 in this book.
| |
| | |
| 5. Key references
| |
| (IPCC, 2000) Special Report on Emissions Scenarios.
| |
| (OECD, 2012) OECD Environmental Outlook to 2050: The Consequences of Inaction
| |
| (Millennium_Ecosystem_Assessment, 2006)
| |
|
| |
| | |
| | |
| Figure 3.2: The total global population is projected to peak and decline in the coming century, except under the high end assumptions (SSP3). By 2100 the number of people may range between the current level and two times more than in 2000 in the SSPs. The OECD Outlook assumes an intermediate population growth trajectory, close to the medium population SSP scenarios.
| |
| https://secure.iiasa.ac.at/web-apps/ene/SspDb/dsd?Action=htmlpage&page=welcome
| |
| | |
| | |
| | |
| Figure 3.3a: Projected total GDPppp for the world in the SSP scenarios (GDP = Gross Domestic Product; ppp = purchasing power parity) show different levels and different dynamics over time, in particular in the second half of the century. Up to 2050 the projection used in the OECD Outlook is in the middle of the SSP range, but in the second half it is in between the middle (SSP2) and high-end (SSP5) of the more recent projections
| |
| https://secure.iiasa.ac.at/web-apps/ene/SspDb/dsd?Action=htmlpage&page=welcome
| |
| | |
| | |
| Figure 3.3b: The projected total GDPppp for five aggregate world regions in the SSP2 scenario confirms the decreasing future role for the historically dominant OECD in the future global economy {OECD (OECD member states in 1990; REF (Reforming economies; LAM (Latin America); MAF (= Middle-East and Africa) and Asia
| |
| https://secure.iiasa.ac.at/web-apps/ene/SspDb/dsd?Action=htmlpage&page=welcome
| |
| | |
| | |
| Figure 3.3c: The nature and inherent properties, and the parameter settings of the economic models (OECD, IIASA and PIK) used to produce the SSP projections are different. The resulting pathways for the low-end scenario SSP3 from illustrates this, but also shows a commonality in the main trend of declining growth on the longer term despite the high population growth assumed in SSP3
| |
| https://secure.iiasa.ac.at/web-apps/ene/SspDb/dsd?Action=htmlpage&page=welcome
| |
| | |
| Brink, B.t., 2010. Rethinking Global Biodiversity Strategies. 500197001, PBL
| |
| Bruinsma, J.E., 2003. World agriculture: towards 2015/2030. An FAO perspective. Earthscan, London, 432 pp.
| |
| Chateau, J., R. Dellink, E. Lanzi and B. Magné 2013. An overview of the OECD ENV-Linkages model – version 3, OECD, Paris ( www.oecd.org/environment/modelling).
| |
| FAO, 2006. World Agriculture: towards 2030/2050, FAO, Rome
| |
| FAO, 2013. FAOSTAT database collections. Food and Agriculture Organization of the United Nations, Rome.
| |
| IPCC, 2000. Special Report on Emissions Scenarios.
| |
| Millennium_Ecosystem_Assessment (Ed.), 2006. Ecosystems and human well-being: scenarios (edited by Carpenter S.R., Pingali P.L., Bennett, E.M. and Zurek, M.B.), Volume 2. Island Press, Washington, D.C.
| |
| Moss, R.H. et al., The next generation of scenarios for climate change research and assessment. Nature, 463: 747-756,doi:10.1038/nature08823.
| |
| OECD, 2008. Environmental outlook to 2030. OECD Publications, Paris.
| |
| OECD, 2012. OECD Environmental Outlook to 2050: The Consequences of Inaction OECD Publishing, Paris.
| |
| UNEP, 2011. UNEP Bridging the Gap Report, http://www.unep.org/publications/ebooks/bridgingemissionsgap/
| |
| Vuuren, D.P.v. et al., 2012. A proposal for a new scenario framework to support research and assessment in different climate research communities. Global Environmental Change
| |
| Pages 21–35, Volume 22(Issue 1),http://dx.doi.org/10.1016/j.gloenvcha.2011.08.002.
| |
| Westhoek, H. et al., 2011. The Protein Puzzle PBL Netherlands Environmental Assessment Agency, Bilthoven
| |
| | |
| | |
| | |
| }} | | }} |
