Socio-ecological indicators for sustainability

ECOLOGICAL ECONOMICS ELSEVIER Ecological Economics 18 (1996) 89-112 Methodological and Ideological Options Socio-ecological indicators for sustainability Christian Azar, John Holmberg *, Kristian Lindgren Institute of Physical Resource Theo~', Chalmers Universio, of Technology and UniversiO; of GOteborg, S-412 96 GOteborg, Sweden Received 20 February 1995; accepted 12 February 1996 Abstract A systematic framework of indicators for sustainability is presented. In our approach there is an emphasis on societal activities that affect nature and on the internal societal resource use, as opposed to environmental quality indicators. In this way the indicators may give a warning signal to an unsustainable use of resources early in the chain from causes in societal activities to environmental effects. The aim is that these socio-ecological indicators shall serve as a tool in planning and decision-making processes at various administrative levels in society. The formulation of the indicators is made with respect to four principles of sustainability, which lead to four complementary sets of indicators. The first deals with the societal use of lithospheric material. The second deals with emissions of compounds produced in society. The third set of indicators concerns societal manipulation of nature and the long-term productivity of ecosystems. Finally, the fourth set deals with the efficiency of the internal societal resource use, which includes indicators for a just distribution of resources. Keywords: Indicators;Sustainability 1. Introduction The publication of the Brundlandt report 'Our C o m m o n Future' (WCED, 1987) and the Rio Declaration (United Nations, 1992a) put the challenge of sustainable development on the agenda for planners, decision makers and politicians at all administrative and institutional levels of the global society. Since then, much effort has been made to define and operationalise the concept of sustainability. Many researchers have suggested various types of non-monetary measures to indicate to what extent environmental states and functions, material flows, or societal activities can be regarded as sustainable (see, e.g., Vos et al., 1985; Liverman et al., 1988; Kuik and Verbruggen, 1991; Opschoor and Reijnders, 1991; Holmberg and Karlsson, 1992; Adriaanse, 1993; Alfsen and Sa~b¢, 1993; Bergstrrm, 1993; Gilbert and Feenstra, 1994; Moffatt, 1994; OECD, 1994). Most sets of indicators developed so far have focused on the state of the environment rather than on the relation between society and ecosystems. In the present paper we formulate indicators based on four socio-ecological principles for sustainability (Holmberg et al., 1996). The principles * Corresponding author, e-mail: [email protected] 0921-8009/96/$15.00 Copyright © 1996 Elsevier Science B.V. All rights reserved. PH S0921-8009(96)00028-6 90 C. Azar et al. / Ecological Economics 18 (1996) 89-112 and, hence, the indicators focus early in the causal chain--i.e., in the chain of causes in society to effects in the environment. In this way socio-ecological indicators may give an earlier warning than would environmental quality indicators. There are two aspects that are important in the construction of our indicators: (i) There are in many cases long time delays between a specific activity and the corresponding environmental damage. This means that indicators based on the environmental state may give a warning too late, and in many cases only indicate whether past societal activities were sustainable or not. (ii) The complexity of the ecosystems makes it impossible to predict all possible effects of a certain societal activity. Some damages are well-known, but others have not yet been identified. Most of the sustainability indicators suggested so far are formulated with respect to known effects in the environment. We suggest that indicators of sustainability should be formulated with respect to general principles or conditions of sustainability. The socio-ecological principles that form the basis of the socio-ecological indicators focus on the societal activities and interactions with nature and internal societal resource use. The first principle deals with societal use of elements from the lithosphere. The second principle deals with the necessary restrictions on emissions of anthropogenically produced substances. The third principle concerns the anthropogenic manipulation of nature. Finally, the fourth principle deals with the efficiency of the societal resource use. These principles have in common that they are formulated in terms of societal activities. We use these four principles as a systematic framework for developing indicators. Our aim is then to define indicators that are based on data that reflect societal activities rather than the state of the environment. In the present paper we exemplify the four different types of indicators by calculating their values using mainly global data. In Section 2, we discuss various approaches that have been made to indicate sustainability. In Section 3, the four socio-ecological principles for sustainability are reviewed. Then, in Section 4, we develop a set of indicators for each of the four principles. Numerical estimates of these indicators are also presented. Section 5 is devoted to a discussion of the results and on areas for future research. 2. Indicators for sustainability There are both monetary and physical approaches to indicating sustainability. In this paper we focus on physical indicators. Such indicators can be divided into three (main) groups: (i) societal activity indicators (that indicate activities occurring within society--the use of extracted minerals, the production of toxic chemicals, recycling of material), (ii) environmental pressure indicators (that indicate human activities that will directly influence the state of the environment--e.g., emission rates of toxic substances) and (iii) indicators of the state of the environment or environmental quality indicators (that indicate the state of the environment--e.g., the concentration of heavy metals in soils and pH levels in lakes). It should be noted that most indicators for sustainability developed and used so far belong either to the group of environmental pressure indicators or to the state of the environment indicators. This is shown in Table 1 where we have illustrated where some authors have put their focus when developing and evaluating indicators for sustainability. 3. Principles of sustainability In our formulation of indicators for sustainability we use a framework of principles that should be fulfilled in a sustainable society (see Holmberg et al., 1996). The principles are presented below. 9l C. Azar et al. / Ecological Economics 18 (1996) 89-112 Table 1 The focus of some indicators for sustainability Reference Indicated area Societal activities Environmental pressure Adriaanse (1993) Alfsen and Sa~b0 (1993) Ayres (1995) ten Brink ( 1991 ) Brown et aL (1994) Carlson (1994) ECE (1985) Environment Canada (1991) Gilbert and Feenstra (1994) Holdren (1990) Holmberg and Karlsson (1992) Miljoministeriet ( 1991) Nilsson and Bergstr~m (1995) OECD (1994) Opschoor and Reijnders (1991) SNV (1994) Haes et al. (1991) Vos et al. (1985) This paper The Netherlands Norway Mainly USA Specific ecosystem The world Sweden ECE member countries Canada Specific ecosystem The world Not specific Denmark Municipality and company OECD countries Not specific Sweden Specific ecosystem The Netherlands The world x x x x x x (x) (x) x x x (x) x x x x x x State of the environment x x (x) x (x) x x x x (x) x x x (x) x x (x) x x x x The symbol 'x' indicates the main focus of the work, while (x) means that such indicators are included, but only play a minor role in the work. 3.1. Principle 1: substances extracted f r o m the lithosphere must not systematically accumulate in the ecosphere. E l e m e n t s f r o m the lithosphere must not be spread at a rate w h i c h will g i v e rise to a systematic increase in the ecosphere. Such an increase will o c c u r if the sum of the a n t h r o p o g e n i c e m i s s i o n s and the natural f l o w s f r o m the lithosphere to the e c o s p h e r e ( w e a t h e r i n g processes and v o l c a n i c eruptions) exceeds the sedimentation rate and the rate of final disposal in the lithosphere. B e c a u s e o f the c o m p l e x i t y and delay m e c h a n i s m s o f processes in the ecosphere, it is e x t r e m e l y hard to say what level of a c c u m u l a t i o n will cause an effect. In fact, e v e r y substance has a limit (often unknown), a b o v e w h i c h d a m a g e occurs in the ecosphere. Increasing amounts o f carbon dioxide in the atmosphere, o f sulphur d i o x i d e leading to acid rain, of phosphorus in lakes and o f h e a v y metals in soils and in our bodies are all e x a m p l e s of such accumulation. In practice, this principle implies restrictions on the extraction rate of metals and fossil fuels in c o m b i n a t i o n with increased r e c y c l i n g of material and decreased dissipative use of scarce elements. It also implies substitution of abundant e l e m e n t s for scarce elements. 3.2. Principle 2: socie~-produced substances must not systematically accumulate in the ecosphere. In the technosphere, m o l e c u l e s and atomic nuclei o f different kinds are produced, s o m e o f them long-lived, in a m o u n t s p r e v i o u s l y u n k n o w n to the ecosphere. If they are emitted faster than they are d e g r a d e d into m o l e c u l e s or nuclides that can be integrated in the e c o s p h e r i c cycles, a n d / o r faster than they are r e m o v e d to the lithosphere, such substances will a c c u m u l a t e in the ecosphere. In order to reduce the a m o u n t emitted, one can degrade substances within the technosphere or deposit t h e m in final disposals. C F C m o l e c u l e s destroying the o z o n e layer, increasing a m o u n t s o f D D T and P C B in biota, and radioactive inert gases in the a t m o s p h e r e are all e x a m p l e s o f such accumulation. T h e principle also implies that persistence is a very important aspect of substances that are foreign to nature, and therefore there should be strong restrictions on the use of persistent substances foreign to nature. 92 c. Azar et al. / Ecological Economics 18 (1996) 89-112 Finally, we note that higher concentrations in the ecosphere may lead to increased sedimentation rates (for Principles I and 2) a n d / o r higher degradation rates (for Principle 2) so that a new equilibrium concentration may be established. 3.3. Principle 3: the physical conditions for production and diuersi~ within the ecosphere must not become systematically deteriorated. A sustainable society must not systematically reduce the physical conditions for the long-term production capacity in the ecosphere or the diversity of the biosphere. Society must neither take more resources from the ecosphere than are regenerated nor systematically reduce natural productivity or diversity by manipulating natural systems. Deforestation, soil erosion, land degradation with desertification as an extreme form, extinction of species of plants or animals, exploitation of productive land for asphalt roads and refuse dumps, and destruction of freshwater supplies are examples of such reduction, Society is dependent on the long-term functions of the ecosystems. Even if Principle 1 and Principle 2 are fulfilled, society must be careful with its manipulation of the resource base in order to avoid a loss of the productive capacity for the supply of food, raw materials and fuel, This dependence will become more obvious when the use of fossil fuels is reduced (in accordance with Principle 1). 3.4. Principle 4: the use of resources must be efficient and just with respect to meeting human needs. Principles 1, 2 and 3 constitute the external conditions for a sustainable metabolism of a society. The assimilative capacity as well as the available resource flows are limited. In order to fulfil human needs for a growing global population, the resources and services obtained from nature must be used efficiently within the society. Socially, efficiency means that resources should be used where they are needed most. This also leads to the requirement of a just distribution of resources among human societies and human beings. 4. Socio-ecological indicators for sustainability For each of the socio-ecological principles reviewed in the previous section we define a set of socio-ecological indicators. These indicators are formulated so that they reflect to what extent (a certain aspect of) a societal activity violates the corresponding principle. 4.1. Indicators for Principle 1 The basic idea behind the first principle is that the total flow of an element from the lithosphere to the ecosphere--i.e., societal emissions of an element extracted from the lithosphere as well as weathering and volcanic processes, should not exceed the return flow of the same element from the ecosphere to the lithosphere, by sedimentation processes and by flows to final deposits in the lithosphere. As a starting point for the formulation of indicators for Principle l, we use Fig. 1, which shows some basic features of the cycle of a specific element between the lithosphere, the technosphere and the ecosphere. The variables used are X T and X~ for the total contents of the element in the technosphere and the ecosphere, respectively, and X~ for the total resources. Furthermore, kex is the annual rate of extraction, k w is the total natural contribution to ecosphere (i.e., weathering and volcanic eruptions), kem represents the emissions of the element from the technosphere and k s is the rate of sedimentation from the ecosphere to the lithosphere. It should be observed that in general k~ is an increasing function of the total content in the ecosphere. Society can prevent the accumulation of lithospheric materials in the ecosphere by (i) limiting the extraction rate, (ii) limiting the leakage from the technosphere--i.e., by using high degree of recycling and by avoiding 93 C. Azar et al. / Ecological Economics 18 (1996) 89-112 ufinal n * I deposit[ I "'--~ [ I XR I ~'~ kex XT [ Technosphere[ l,w Lithosphere -~ j I XE Ecosphere ks Fig. 1. An element is extracted from the lithosphere and used in the technosphere. Eventually it will be emitted to the ecosphere, where it will remain until sedimentation processes once again bury it in the lithosphere. The figure shows that the total contents of the element will increase in the ecosphere if the total emissions from the technosphere (kern) plus the natural flow from the lithosphere (k w) exceed the rate of sedimentation (ks). The dashed line represents direct flows from the technosphere to the lithosphere for final deposition in repositories. At present, this option is only planned for radioactive waste. dissipative use, (iii) returning the material to underground repositories (dashed line in Fig. 1) and (iv) increasing the sedimentation rate (e.g., by guiding the emissions from the technosphere to areas in the ecosphere where the sedimentation rate is high). There are several reasons for putting the focus early in the causal chain when indicating Principle 1. The elements that are extracted do not disappear and as long as we do not have clever strategies for preventing accumulation in the ecosphere, a plausible first approximation is therefore that the elements extracted will eventually leak to the ecosphere. Furthermore, the diffuse emissions from the consumption sector of the economy now exceed the more easily detected emissions from the production sector for many elements. Bergb~ick (1992) has shown that this is the case for many metals used in Sweden. The residential time is often longer in the consumption sector than in the production sector. 4.1.1. Indicator no. 1,1: lithospheric extraction rates If the rate of extraction is high compared to the natural sedimentation process, accumulation will occur in the technosphere, and sooner or later in the ecosphere. The material accumulated in the technosphere will eventually leak to the ecosphere. Thus, it is important to indicate the extraction rate of each element. We define the first indicator 11,, for Principle 1 as the extraction rate divided by the rate of natural supply ' from the lithosphere to the ecosphere by weathering and volcanic eruptions, i.e.: kex 11'1 = k--w-' (4.1) where we use the convention that the first digit in the subscript of I refers to the principle and the second is an index for the different indicators within this group. This first indicator, the Lithospheric Extraction Indicator (LED, has the advantage that it is straightforward and easy to understand, and that data are relatively easy to obtain. This type of indicator has been proposed as a measure of anthropogenic disruption of natural cycles by Benjamin and Honeyman (1992). If I1,1 = 1, the present anthropogenic extraction is equal to the natural supply to the ecosphere. This implies that anthropogenic perturbations of the natural biogeochemical cycles are considerable and we can expect significant changes in the concentration of the element in question in the ecosphere. According to Principle 1, sustainability requires that the concentration of each of the elements must not increase systematically in the ecosphere. One could say that anthropogenic perturbations of the natural i The natural supply to the ecosphere is balanced by the natural sedimentation back to the lithosphere. 94 c. Azar et al. / Ecological E~onomics 18 (1996) 89-112 biogeochemical cycles must be small, which implies values of I~.~ (much) less than unity. This should be viewed as a rule of thumb, and in specific cases one may argue for more exact acceptable values for this indicator. Even if there is no obvious critical value for the indicator, one can make a first comparison by looking at the order of magnitude of the indicator for different elements. In this way one can get an indication of, for instance, which metals could possibly substitute for others. If we focus on the use of a specific element, the indicator should be calculated for a time series to reveal the trend towards or away from sustainability. 4.1.2. Indicator no. 1,2: accumulated lithospheric extraction It is also of interest to indicate the accumulated amount of a specific element in the technosphere in relation to the ecospheric content of that element. W e are especially interested in the total pre-industrial content in the human area soils, 2 which we denote by X~ and we define: I~=.. l o xx ~ E , f _ k ~ x ( t ) d t = XE'' (4.2) where X A is the accumulated extraction from the lithosphere. For many metals most of the mining has been carried out during this century and most of the extracted amount still remains in the technosphere. In this case the indicator can be approximately written I~.z ~ X T / X ~ , where X~ is the amount in the technosphere. W e refer to this ratio as the Accumulated Extraction Indicator. ~" The indicator IL2 c a n be used as an early warning signal. If its value is high, then there are reasons for analysing the flows of that element in more detail (e.g., by comparing flows in natural sedimentation processes with the leakage from the technosphere) and possibly considering final deposit of the element. Cadmium has been discussed in this context by van der Voet and Kleijn (1992). 4.1.3. Indicator no. 1,3: non-renewable energy suppl.v Several environmental problems are related to the use of non-renewable energy. Hence, we have chosen to indicate the ratio between non-renewable energy and total primary energy supply. Non-renewable primary energy supply Ii ,3 = Total primary energy supply (4.3) In Fig. 2, we present a time series for this indicator. 4.1.4. Other indicators There are also other indicators that can be formulated from the variables of Fig. l, Fig. and Fig. , and that reflect important aspects of societal resource use (e.g., k e m / k w o r indicators for emissions to specific parts of the ecosphere), such as anthropogenic emissions of metals to the atmosphere, aquatic systems and soils (Lantzy and Mackenzie, 1979; Nriagu and Pacyna, 1988). The efficiency in the use of elements within society (e.g., the degree of recycling) is dealt with in connection with Principle 4. Also, indicators focusing on the problem of the limited amount of lithospheric resources belong to Principle 4 since they deal with intergenerational justice. It has been argued that for most elements the assimilation capacity of the ecosphere is a more restrictive constraint on the use of these elements than the amount of the resources available in the lithosphere (see, e.g., Holmberg et al., 1996). A possible exception may be phosphorus, which is a non-substitutable macro-nutrient. 2 The human area is the top soil layer (to 0.2 m depth) of land used for the technosphere. Vitousek et al. (1986) call this the "human area" and assign it a value of 2 million km2. It is possible to choose a different reference area, but it will not affect the comparison between different substances in any significant way. Wallgren (1992) has called this ratio the Future Contt~mination Index (FCI) when calculating its value for some metals used in Sweden. 95 C. Azar et al. / Ecological Economics 18 (1996) 89-112 Global non-renewable energy input [% of total input] 100 80 40 20 0 I I I I I I I Fig. 2. The ratio between non-renewable energy and total energy supply. Data based on BP Statistical Review. 4.1.5. Numerical examples In Table 2 the indicators I1. ~ and Ii, 2 are given for some of the elements that are extracted from the lithosphere. If we look at I H, in Table 2, we see that for many elements the anthropogenic flows dominate over the natural flows. The discussion of the environmental implications for metals such as Pb, Hg and Cd has already begun. But we see from the table that this discussion must soon include other elements as well (e.g., Cu, which has even higher values for the indicator I~.1). A plausible reason why the use of Cu is not yet discussed can be that this metal has a longer residence time within society and that we have traditionally put our focus on effects in nature. A high degree of recirculation of elements within society and careful handling of the elements (non-dissipative use) can reduce the risks associated with high values for the indicator. Note that I~.j indicates a flow while Ii, 2 indicates a state. A high value of 11.~ and a low value of It, 2 is a warning that a rapid accumulation is taking place and I1, 2 will increase. Therefore, I~,~ lies earlier than lj, 2 in the causal chain. The values for the indicator I~.~ for the elements C, S and P are considerably larger than unity (see Table 2). The environmental effects that are known from the emissions of these elements--the greenhouse effect, acidification and eutrophication--clearly show that in all three cases we have exceeded a critical value for this indicator. 4.2. Socio-ecological indicators based on Principle 2 According to the second socio-ecological principle for sustainability, substances that are produced in society must not systematically accumulate in the ecosphere. Here, we formulate socio-ecological indicators for substances that are naturally existing (Section 4.2.1) and for substances that are foreign to nature (Section 4.2.2). 4.2.1. Indicators for man-made substances that are naturally existing The main idea behind the two first principles for a sustainable society is that sustainability requires that human disruption of the natural cycles and flows of substances are small enough to avoid a systematic accumulation. 4.2.1.1. Indicator no. 2,1: anthropogenic flows compared to natural.flows. As the first indicator for substances that are naturally occurring, we suggest the ratio: Ea 12'L = E--~" (4.4) 96 C. Azar et al. / Ecological Economics 18 (1996) 89-112 Table 2 Indicators for elements extracted from the lithosphere Element Metals AI Fe K Mg Ti Mn Zr V Zn Cr Cu Li Ni Pb Ga Nb U Sn Mo Be Cd Hg Ag Conc. in soils ( m g / k g ) 72000 26 000 15000 9 000 2 900 550 230 80 60 54 25 24 19 19 17 l1 2.7 1.3 0.97 0.92 0.35 0.09 0.05 Semi-metals Si 310 000 B 33 As 7.2 Ge 1.2 Sb 0.66 Non-metals C S F P Se 25 000 1 600 950 430 0.39 Weathering a and volcanic (kton) Mining (kton) Fossil fuels ~' (kton) lr '~ c 11._~ d l 100000 390 000 230000 140 000 44 000 8 300 3 500 I 200 910 830 380 360 300 290 260 170 41 20 15 14 5.3 1.4 0.75 18000 540 000 24000 3 100 2 500 8 600 880 32 7 300 3 800 9000 9.9 880 3 300 0.037 14 47 210 110 0.34 20 5.2 15 34000 34 000 340 690 I 700 170 140 350 260 34 55 220 570 85 24 14 3.4 5.7 17 10 3.4 10 1.7 0.048 1.4 0.11 0.028 0.096 I. 1 0.3 0.32 8.3 4.6 24 0.64 4.8 12 0.092 0.17 1.2 II 8.5 0.76 3.9 6.5 22 0.01 1 4 700 000 500 110 18 9.9 780000 33 000 14000 6500 5.9 4 600 0.37 19 0.27 54 58 000 2 300 21 000 2.1 95 000 250 18 17 10 0.021 0.52 0.33 0.96 6 5 400000 100 000 240 1 700 12 6.4 3.7 0.17 3.5 2 0.01 0.02 6.9 2.6 23 2 19 4.2 3 17 Weathering mobilization is calculated using average concentration in soils (column 1) and suspended sediment flux of 1.5.1016 g per year in rivers (Nriagu, 1990). b Data for contents of trace elements in crude oil are from USA (Yen, 1975). However, the only elements for which flow associated with crude oil are considerable compared with the flow associated with coal are V, Ni and Hg. The flows of elements associated with fossil fuels predominate over the amount that is mined for several elements: e.g., V. Li, Ga, Be, Hg, Si, B, Ge, S, Se and, of course, for C. Surprisingly, this is also true for aluminium. Indicator I U is calculated as anthropogenic flows from the lithosphere to the ecosphere divided by the natural flows. The anthropogenic flows are mining and flows associated with fossil fuels and the natural flows are weathering and volcanic processes. Data from 1990. d Indicator Ii. 2 is calculated as the accumulated mining since 1900 divided by the amount in the top soil layer in the human area (see the text). If flows from fossil fuels had been included, the indicator value would increase substantially for a number of elements. Source: BP Statistical Review, Yen (1975), Valkovid (1983), Nriagu (1989), Sposito (1989), Crowson (1992), Speight (1992), Wallgren (1992), Walker and Kastings (1992), Sigenthaler and Sarmiento (1993), Holmberg et al. (1996) and Karlsson el al. (1994). C. Azar et al. / Ecological Economics 18 (1996) 89-112 97 Table 3 Indicators for Principle 2 Substance 12.i 12,2 CO 2 CH 4 N2(g) ~ N(active) N20 NH~ and NH~ NO x SO, C t£ 0.07 1.4-4.1 1.3-3 0.5-1 0.02-0.09 1.9-2.6 0.7-6 0.1 1.8 a.t, 2.7 b 1.5 b 1.1 The value for the second indicator for CO~ gives the atmospheric concentration in the year 2100 divided by the pre-industrial concentration, given that the present global emissions are kept constant. In order to estimate the long-term atmospheric concentration of CO,, one has to make assumptions about emissions scenarios and fossil fuels reserves, Maier-Reimer and Hasselman (1987) estimate that for a stock of fossil fuels equal to 5000 Gton C, the atmospheric concentration of CO 2 would reach peak concentration as high as 4.5-6.4 times the pre-industrial level (all depending on the rate at which the stock of fossil fuels is combusted). No net-sink in biomass is assumed. b Values from IPCC (1995). Sources: Bolin (1979), SiSderlund and Rosswall (1982), IPCC (1990), IPCC (1992), IPCC (1995), Jaffe (1992), Bates et al. (1992), Graedel and Crutzen (1993), UNSCEAR (1993). Here E a is the anthropogenic rate of production of a given substance and E n is the natural rate of production. This indicator is analogue to indicator Ii, l (see Eq. 4.1). 4.2.1.2. Indicator no. 2,2: the long-term implications of present emissions. It is not clear from the value of indicator IzA, what the long-term content will be in the ecosphere. An analysis of the long-term content requires specific models of how, where and when the substance is released to the ecosphere. In some cases, however, this is easily done. For some substances that are released to the atmosphere, and whose life-times are long enough to guarantee a relatively homogenous distribution throughout the atmosphere, a simple model can be used in order to get a rough picture of the total content M(t) of the substance in the atmosphere at time t. The time derivative of M(t) is given by: 1V1( t ) -~ E a q- E , - k M ( t ) , (4.5) where k is the rate of degradation (or removal) from the atmosphere. Whenever this formula is applicable, it is straightforward to evaluate the long-term stationary atmospheric content for the present rate of emissions divided by the pre-industrial atmospheric c o n t e n t , Mprein~1. This ratio is our second indicator for Principle 2, and it is given by: ga+E 12,2 n kMvreind Ea 1 +--. E, (4.6) In Table 3, we present values for the indicators /2, ~ and 12,2 for some common substances. If 12.~ = 1, then the anthropogenic flows equal natural rates of production, which implies a considerable disruption of the natural cycle of that substance. The implication of the indicator value 12,~ = 1 is in general not clear due to uncertainties about the decay processes for different substances. This is, however, not the case for those substances that can be described by Eq. 4.5. For these substances, the long-term atmospheric concentrations will increase by a factor of two (which is easily seen in Eq. 4.6). It should be noted that also rather small values for the indicator Iz,~ could be unsustainable. Consider, for instance, emissions of CO 2. Anthropogenic emissions amount to 7.1 ___ 1.1 Gton C / y e a r (IPCC, 1995). But at the same time, respiration by biota gives rise to an annual production of 100 Gton C. Thus, we have 12.~ = 0.07, which may appear rather small. But since CO 2 is a stable compound, the present rate of emission will cause a 98 C. Azar et aL /Ecological Economics 18 (1996) 89-112 systematic increase in the atmospheric concentrations despite the fact that the value for the indicator is low. At present, the atmospheric concentration of CO 2 is 25% higher than its pre-industrial level, and the concentration is increasing at approximately 0.5% per year. This increase is mainly due to the flow from the lithosphere which is reflected in indicator 1~.~. Here emissions due to deforestation are also included in the indicator values. However, when an environmental problem is known and detailed modelling studies exist relating emissions to atmospheric concentrations, it could be more useful to develop indicators that directly measure how far the present rate of emissions is from the level which does not give rise to any increase in the ecosphere. In the case of CO 2 there are, of course, several such studies. IPCC (1995) estimates that in order to stabilise atmospheric concentrations at 350 ppm (slightly below the present level) by the year 2100, the accumulated emissions from 1990 to the end of the 21st century must not exceed some 3 0 0 - 4 3 0 Gton C. This implies an annual average rate of carbon emissions equal to 3.3 _+ 0.6 Gton C / y e a r . 4 This value could be referred to as a sustainable level of CO 2 emissions (given that the climatic changes associated with the present atmospheric concentration of CO 2 are acceptable) and a sustainability indicator for CO 2 emissions could be defined as the ratio between the present and the sustainable emission rates. If this ratio is equal to unity, then the present emissions are sustainable. At present, we estimate it to lie in the range 1.5 to 3.0. 4.2.2. Indicators f o r substances that are foreign to nature The number of chemicals used in society has virtually exploded. In the present industrial society tens of thousands of chemicals are used regularly. There are, for example, over 70 000 chemicals on the U.S. T S C A (Toxic Substances Control Act) Inventory (Clements et al., 1994). W e know that the links between emissions, ecospheric concentrations and damage are often characterised by long time lags and a high degree of complexity. This means that it is extremely difficult to predict the environmental consequences of an emission of a specific substance to the ecosphere. But how should we determine which restrictions are needed? According to Principle 2, we should not emit substances to nature at a rate that is faster than they are degraded into naturally existing substances that can be incorporated in the cycles of nature. 5 This in turn implies strong restrictions on the use of persistent 6 substances. Sustainability also implies strong restrictions on the use o f substances that have long-term i m p a c t s - - i . e . , substances that might degrade rapidly, but have the potential to reduce biodiversity, that are mutagenic or that affect reproductive systems. In order to develop indicators that reflect to what extent the present use of artificially produced chemicals is sustainable or not, we first must identify the chemicals that are persistent (including metabolites) or that have long-term impacts, and then we need to find statistics on their emission rates. Ideally, one would like to compare the emission rates with the degradation rates, but this is not possible for the vast majority of chemicals since the decay processes are very complex and the rate of degradation depends on the specific environment. When there is a lack of emission data, estimates can be made using mass-balance calculations of industrial processes (see, e,g., Ayres et al., 1989 and Ayres et al., 1995). 4.2.2.1. Indicator no. 2,3: production ~'olumes o f persistent chemicals foreign to nature. Hence, we suggest that the trend in the production volume of persistent substances should be continuously monitored. The U.S. Environmental Protection Agency has produced a list containing 80 persistent bioaccumulators produced in the U.S. in quantities over 5 tons annually. 7 The OECD has produced a Representative List of High Production 4 For the 22nd century even stronger reductions are needed in order to avoid a renewed increase of CO2 in the atmosphere. 5 The International Joint Commission (1994) writes that "persistent toxic substances are too dangerous to the biosphere and humans to permit their release in any quantity." 6 Here, persistent substances also include substances that are easily degradeable, but whose metabolites are persistent. 7 A later version of this list (Clements et al., 199~-) included only those chemicals that are produced in quantities over 500 tons/year. This criterion reduced the number of chemicals on the list to 33. C. Azar et al. / Ecological Economics 18 (1996) 89-112 99 Table 4 The productionvolumes of common persistentbioaccumulatorsin the OECD CAS-number Chemical name 12,3 = Prod. vol. (kton/year) Di-pentabromobenzeneether 1,5,9-Cyclododecatriene p-Chiorobenzontrichloride Ethane, hexachloro Hexachlorocyclopentadiene Tetrabromo-bisphenol Benzene-l,2,4,5-tetrachloro 1163-19-5 4904-61-4 5216-25-1 67-72-1 77-47-4 79-94-7 95-94-3 US 10, EU 1, JP 1 US 10, EU 10 US 10, EU 1 US 1, EU 1 US 1, EU 1 US 10, EU 1, CH 1, JP 1 US l, EU 1 Source: Smith (1994). Here, EU stands for the European Union, CH for Switzerlandand JP for Japan. Volume Chemicals containing the 1363 most commonly used chemicals in the OECD area. 8 Since all major industrialised countries except the Eastern European countries (including the former Soviet Union) are members of the OECD, data from the OECD could be used as a first indication of the trends in the worldwide production of these chemicals. The combination of the OECD list and the EPA list yielded 7 chemicals (Pettersson, 1994). But, even for these common chemicals the OECD had not available any time series over the production volumes. Data were only available for a single year and only to an order of magnitude accuracy (see Table 4). This means that there is an urgent need for compiling data on the global production volumes of persistent substances, so that the trend in the use of such chemicals can be established. We have asked three major international organisations 9 whether they have the overall picture of this trend. But none of them could answer the question positively. Hence, it could be useful to develop an indicator that measures the societal knowledge of the global production of persistent substances. 4.2.2.2. Indicator no. 2,4: long-term implication o f emissions o f substances that are foreign to nature. Here we present an indicator for substances that are foreign to nature, but whose degradation processes can be described by a simple mathematical equation. For such substances, mainly gases emitted to the atmosphere, we will compare the production volumes with the rate of decay. Assume, as in Eq. 4.5, that the decay process of a specific gas can be described by: /kl(t) = E a - k M ( t ) . (4.7) This is, for instance, possible for radioactive gases and CFCs. Using this representation, we can define a sustainability indicator as the ratio between long-term content of the substance for the present rate of emissions in relation to the present content in the atmosphere. We have: •2,4 G kMpresent (4.8) It should be noted that the indicator 12,4 is normalised by the present content in the atmosphere, whereas I2,2 is normalised by the pre-industrial content in the atmosphere. This means that the interpretation of the indicator value is different for the two types of indicators. For 12,2 = 1 the anthropogenic emissions are equal to zero. For 12,4 = 1 the present anthropogenic emissions will stabilise the present atmospheric concentrations. This could be 8 The OECD list contains substances that occur on two or more national high volume lists with a production volume of over a 1000 tons/year or on one national list, but with a production volume exceeding 10000 tons/year (Freij, 1994). 9 The OECD, the European Commissionand the U.N. EconomicCommissionfor Europe. I00 C. Azar et al./ Ecological Economics 18 (1996) 89-112 Table 5 Indicator 12,4 for substances that are foreign to nature Substance 1_9.5 CFC 11 CFC 12 CFC 113 CFC 114 CFC 115 HCFC-22 Methylchloroform Carbon tetrachloride Halon- 1211 Halon- 1301 85Kr 4.4 7.5 14 15 39 5.3 1.4 14 4.2 190 2 All compounds except 85Kr are ozone depleters. For these substances we have used global production data from the year 1985 (German Bundestag, 1989). Since 1985, aggregate production of CFCs has been halved (Brown et al., 1994). For the last compound, SSKr, a radioactive gas that is emitted by nuclear fuel reprocessing plants, we have used emission data from (UNSCEAR, 1993). sustainable, but only if the present content in the atmosphere is low enough. This is, for example, not the case for CFCs. Thus, much effort is made to phase out the production of CFCs. If 12,4 > 1 the present rate of emissions will cause the atmospheric content to increase. The long-term stationary state content of the gas for a specific production rate is given by the value of the indicator multiplied by the present content. When a time series is based on this indicator, it is possible to fix the normalisation coefficient, Mpresent, at a specific year, so that the trend of the emissions is reflected by the indicator. For the ozone-depleting substances (see Table 5), we have used production data. Emission data for ozone depleters are more uncertain than production data since emissions are due to leakage from the technosphere. This means that our indicator value is only valid if the entire production volume eventually leaks out. The present global destruction rate of ozone depleters is negligible in comparison to production rates, but this will hopefully change in the near future. 4.2.2.3. Some other desirable indicators f o r Principle 2. So far we have mainly discussed substances that are intentionally produced, but significant problems also arise from the unintentional production of substances that are foreign to nature. Such production occurs in chemical industries, in incinerators, at waste deposits where substances might react, etc. Several such substances have been identified as very persistent, bioaccumulating and extremely toxic (e.g., dioxin). However, there is a significant lack of data on the production volumes of unintentionally produced substances and many of them even remain unidentified. In order to make any indication early in the causal chain, we suggest that focus is put on processes that are known to give rise to unintentional production of substances that are foreign to nature. Finally, an indicator for the production of substances that are not themselves persistent, but that have long-term impacts (e.g., that are mutagenic), that affect reproductive systems or cause losses in biodiversity, would be desirable. 4.3. Indicators f o r Principle 3 The global population is expected to nearly double by the year 2050 (United Nations, 1992b). In order to provide a sustainable supply of biomass for food, material and energy for the growing population, we need to maintain the services of the ecosystems. These services include, for example, generation and maintenance of C. Az,ar et al. / Ecological Economics 18 (1996) 89-112 101 soils, disposal of wastes and cycling of nutrients, pest control and pollination (Ehrlich and Ehrlich, 1992). This means that the productivity of lands and the biodiversity of ecosystems must not worsen. This is the essence of Principle 3. Human activities threaten ecosystem productivity and biodiversity in two ways. The exchange of substances between society and nature is dealt with by the indicators for Principles 1 and 2. In this section we focus on societal activities that by manipulation or harvesting of ecosystems may threaten sustainability and, in particular, biodiversity and ecosystem productivity. 4.3.1.1. Manipulation and harL,esting. Principle 3 deals with harvesting and manipulation of ecosystems. Harvesting of funds includes activities such as hunting wildlife, catching fish, harvesting trees, groundwater extraction, etc. Holmberg and Karlsson (1992) discuss three ways of manipulation: (i) we can displace nature (by forcing away or disturbing ecological systems or geophysical systems--e.g., by construction of cities, airports, etc.), (ii) we can reshape the structures of nature (by damming of rivers, ploughing, reforestation, deforestation, etc.), (iii) we can guide processes and flows (by gene manipulation, animal and plant breeding, etc.). Since our ambition is to develop indicators that focus on the societal activities that give rise to changes in the state of the environment, indicators for Principle 3 should focus on those aspects of societal manipulation and harvesting of nature that are not sustainable rather than focusing on the environmental consequences of the unsustainable methods. A large set of indicators can be found in the work by Vitousek et al. (1986), in which the anthropogenic use of biomass is compared to net primary production (NPP) in various ecosystems. For example, it is reported that 40% of terrestrial potential NPP is used directly, co-opted, or foregone due to anthropogenic manipulation or harvesting of the ecosystems. Below, we will present a number of other indicators that are relevant for Principle 3. 4.3.2. Large-scale transformation of lands Since the beginning of the 18th century, humanity has carried out a large scale transformation of the Earth's ecosystems and productive surfaces (see Fig. 3). The area used for crops and grass lands has increased dramatically at the expense of huge losses of primary forests. In the long run this trend is obviously not 16 r,ops 14 8188 12 tural grass 10 condary forests mary forests 6 ~er lands 4 exploitable forests 2 0 1700 1975 Fig, 3. Global land use. It is clearly shown that crops and grass lands have increased at the expense of primary forests. Source: Buringh and Dudal (1987). 102 c. Azar et al. / Ecological Economics 18 (1996) 89-112 sustainable due to limited land area. Furthermore, it violates both the second principle 10 and the third (e.g., systematic losses of biodiversity). 4.3.2.1. Indicator no. 3,1: transformation o f lands. This means that it is of primary interest to focus on the amount of land diverted to different purposes. It should be stressed that such an indicator is useful for global, regional, national as well as local planners and decision makers. It could be presented as in Fig. 3. The present total annual loss of crop land is estimated at more than 10 Mha, of which soil erosion is responsible for 5 - 7 M h a / y e a r , urban expansion 2 - 4 M h a / y e a r and salinization and water logging 2 - 3 M h a / y e a r (Kendall and Pimentel, 1994). There is, however, a considerable crop land gain in the order of 16 M h a / y e a r which is achieved by deforestation. Below we present a few examples of indicators for the anthropogenic use of the different types of l a n d s - - i . e . , agricultural and forest areas, as well as oceans and lakes. These indicators should focus early in the causal chain. 4.3.3. Indicating use o f land areas Agriculture affects both the productivity of soils and biological diversity. Losses in soil productivity occur as a result of a number of changes in soil quality: e.g., soil erosion, nutrient runoff, water logging, desertification, compaction, crusting, organic matter loss, salinization, nutrient depletion by leaching, toxicant accumulation and acidification. Losses in biodiversity could also affect the productivity of soils, and changes in soil quality can have a negative impact on biodiversity. Toxicant accumulation and acidification are related to Principles 1 and 2 and, thus, indicators early in the causal chain for substances causing these problems have already been discussed in Sections 4.1 and 4.2, respectively. Here, we will focus on agricultural practices that cause some of the other problems listed above. 4.3.3.1. Indicator no. 3,2: soil co~'er. Soil erosion is the single most serious cause of degradation of land (Kendall and Pimentel, 1994). It is caused by both water and wind. It is estimated that water erosion affects 560 Mha of lands whereas wind erosion affects 200 Mha (Craswell, 1993). Erosion exists naturally, but it can be enhanced by anthropogenic mismanagement of lands. Although erosion is a complicated process, there is a general consensus that the most important reason for the widespread existence of erosion is agricultural practices that leave the soil without cover (Kendall and Pimentel, 1994). An indicator early in the causal chain for soil erosion would then be the ratio between crop land which is sufficiently covered 11 L~ and the total amount of crop land Lv: L~ 13'2 = L--~- (4.9) There are, of course, other measures that can be taken in order to prevent soil erosion, but soil cover is the most important. Soil erosion also creates a number of other problems further downstream (e.g., flooding, silting of dams and damage to coral reefs). Indicators for such problems will not be given here since these problems are indicated by environmental quality indicators. Other indicators early in the causal chain that need to be developed for the use of agricultural lands and agricultural practices are indicators that focus on irrigation (malpractised irrigation gives rise to salinization), 10Deforestation is presently estimated to give rise to net emissions of CO 2 amounting to approximately 1.9 Gton C/year (Sigenthaler and Sarmiento, 1993). ~ Sufficiently covered is a vague concept, but it implies that the degree of cover ensures that the rate of erosion does not exceed the rate of soil renewal. This means that the degree of cover will differ from region to region depending on the natural circumstances. 103 C. Azar et al. / Ecological Economics 18 (1996) 89-112 Table 6 Base cation losses from Swedish forest soil due to actual stem harvesting and possible future whole-tree harvesting Actual stem harvesting Possible future whole-tree harvesting Mg 2+ K+ Ca 2+ 84% 91% 24% 79% 97% 100% The figures in the table indicate the percentage of the Swedish forest area that is losing base cations. Source: Based on Olsson et al, (1993). nutrient balance in agricultural soils, agricultural practices affecting species and genetic diversity (e.g., monoculture, animal and plant breeding and genetic modification of plants and animals). 4.3.3.2. Indicator no. 3.3: nutrient balance in soils. A basic principle for the long-term sustainability of land use is that harvest rates do not systematically exceed growth rates. This principle also applies to the nutrients in the soil. In the long run, the sustainability of the ecosystem cannot be maintained if nutrient export exceeds nutrient input to the soils. With respect to nutrient decline in soils, increased harvesting has similar effects on the ecosystems as acidification resulting from air pollutants. Increased biomass harvesting leads to increased export of mineral nutrients from the soil. It is therefore of importance to indicate the long-term nutrient balance in soils. A long-term base cation balance can be written: •3,3 = ABCs = B f w + BCo + B C r - BC1 - BCu, (4.10) where ABC~ is the change in the stock of a specific base cation in the soil, BCw the weathering rate, BC d the deposition rate, BC~ anthropogenic return rate, BC l the leaching rate and BC, the net uptake of base cations in biomass. The stock of base cation in Swedish forest soils has decreased by about 1% per year in the past 30-50 years (Falkengren-Grerup et al., 1987; Hallb~icken, 1992). Of this decrease, about half is due to biomass uptake and the other half is due to leakage caused by acidification (Berdrn et al., 1987). In Table 6 we have indicated the share of the Swedish forest area where the stock of base cations are decreasing (the share of the Swedish forest area where ABC~ < 0). 12 The table also includes information about the share of the Swedish forest area that would have a decreasing stock of base cations if increased biomass utilisation, in the form of whole-tree harvesting, would be practised in the future. Leaching of base cations exceeds atmospheric input for almost all Swedish field experiments where nutrient fluxes have been investigated (Falkengren-Grerup et al., 1987). However, even if BC a and BC 1 were excluded from Eq. 4.10, ABC~ for Ca would be negative for a large share of Swedish forest area, if stem harvesting is practised, and for all three base cations if whole-tree harvesting would be practised. Even if we can reduce the acidification caused by air pollutants, a sustainable long-term forestry will require a supply of nutrients to the forest soils--e.g., by returning ashes (i.e., by increasing the anthropogenic return rate). 4.3.3.3. Indicating biological diversity of forest land. There are mainly two groups of socio-ecological indicators for the biological diversity of forest lands: (i) indicators that focus on preserved zones (e.g., by comparing the area that is preserved with the total area) and (ii) indicators that focus on methods of forestry. When indicating the methods of forestry, there are different aspects that are important (e.g., the share of old, large a n d / o r dead trees, drainage by ditches, use of pesticides and introduction of foreign species). The Forest ~2 The figures in the table are only valid for forest soils deeper than 70 cm. If thinner soils were included, the share of areas with ABC~ < 0 would be somewhat greater. 104 c. Azar et al./ Ecological Economics 18 (1996) 89-112 Stewardship Council (an international NGO) is developing criteria and principles that take these aspects into account. 4.3.4. Indicating use o f marine and lake resources The status o f marine resources is affected by pollution as well as harvest rates and harvest methods. Methods for indicating the sources of pollution are developed in Sections 4.1 and 4.2, respectively. The intensity o f the use of marine resources is a serious problem, and indicators for this societal activity is needed. The total world fish catch has increased from 22 million tons in 1950 to a level of 100 million tons (1988-1992). In all the major fishing areas in the world this catch relies on a yield that is at or beyond the limit for sustainability (Brown et al., 1994). A growing world population implies that the catch per capita must decrease. In 1988 the annual per capita catch peaked at 19.4 kg, but according to preliminary data for 1993, it has decreased to 17.6 kg. 4.3.4.1. Indicator no. 3,4: har~,esting o f funds. A typical indicator for harvesting (e.g., of a fish population) is the yield divided by the growth: h /3.4 = - - , g (4.11) where h is the harvest per year and g is the annual growth. The maximum sustainable yield (MSY) is the maximal yield that can be achieved by harvesting a population at the same time as a stationary population size is maintained. Since the M S Y concept focuses on the long-term sustainable yield from an ecosystem, it can serve as a basis for discussions on management of harvesting (May et al., 1979). The concept has been criticized, though, for not taking into account the stability of the system. Results from studying model systems suggest that when harvesting at the MSY, there is a risk of overexploita- ill iii:ilili;iJZ;i : !!:ili!JJiYi1:ii!i! I Service E .,..L, i " 4-, D Nature Fig. 4. A schematic illustration of the physical flows within the technosphere. Substances extracted from nature (E) and substances recycled within in the technosphere (R) are converted into products P and losses L caused by the preconsumption conversion. (Some substances--e.g., the elements--may be turned into products without conversion.) P delivers services to the human sphere when used in the consumption conversion. After consumption, used products P together with L are possibly converted, and thereafter sent into the recirculated fraction R or are discharged to the ecosphere (D) (Holmberg and Karlsson, 1992). c. Azar et al. / Ecological Economics 18 (1996) 89-112 105 tion leading to the extinction of the harvested species (see, e.g., Beddington and Cooke, 1982), This is especially important if the harvest rate is controlled by fixed catch quota. Uncertainties about the fish stocks may lead to harvest yields that destabilize the ecosystem. In a recent study of a three-species model (Azar et al., 1995), we have compared the stability of two harvesting s t r a t e g i e s - - ( i ) constant quota and (ii) constant e f f o r t - - s h o w i n g that constant catch quota can lead to both oscillations (and chaos) and an increased risk of overexploitation. 4.4. Indicators f o r Principle 4 Principles l, 2 and 3 constitute the framework for a sustainable influence on nature. Principle 4 states that if we want a prosperous society within this framework, the societal metabolism must be efficient and just. This principle covers four aspects: overall efficiency, inter- and intragenerational equity, and basic human needs. Below, we develop indicators for some of these aspects. 4.4.1. Indicator no. 4,1: ocerall efficiency A simple schematic description of the societal metabolism is given in Fig. 4. The overall efficiency indicators are measures of the productivity in the technosphere. They indicate how much service ~3 that is delivered for a certain amount of resources extracted from nature, normalized with respect to the situation a certain year y: Service E ServiCey " 14.1 (4.12) Ey It is also possible to complement the overall efficiency indicators with specific efficiency indicators that focus on the internal conversion in the t e c h n o s p h e r e - - e . g . , the flow P per unit of total input E + R, or the recirculation R compared to the total input E + R, P ~ql = E + R ' rl2 R E+R (4.13) The recirculation flow R can also be compared to the total output P * + L: 713 R P* + L" (4.14) In a stationary state P equals P * and E equals D. It is possible to normalize the efficiencies with normalization values "qn determined according to various principles: a normalization to the maximum possible theoretical value, to the best available technology (BAT), or to a desirable value. W e get complementary efficiency indicators: r/i Ii = --, (4.15) rTn where i = 1, 2, 3. In Table 7 some examples of overall efficiency indicators are shown. W e can see, for instance, that the supply of food per hectare of land has increased, whereas the supply of food per phosphate input has decreased since 1970. 13 The service flows can be the temperature of a room, the light from a bulb, etc. The physical flows can be, for instance, energy, exergy, different kinds of materials or products. It is important to note that through increased efficiency it is possible to increase the rate of service flow without increasing the exchange with nature (i.e., inputs of energy and materials and emissions of waste). 106 C. Azar et al./Ecological Economics 18 (1996) 89-112 Table 7 Overall efficiency indicators Efficiency indicators 1970 1980 (l 987) 1990 1 1 1 1 1 1 0.67 0.66 0.69 1.03 1.01 1.07 (0.61) (0.60) (0.67) (1.08) (1.07) (1.18) 1 1 1 1 1 1.07 1.06 1.11 0.91 1.06 1.21 1.12 1.14 0.99 0.93 Food Calories in food/Phosphate input (World) Proteins in food/Phosphate input (World) Fats in food/Phosphate input (World) Calories in food/Area input (World) Proteins in food/Area input (World) Fats in food/Area input (World) Energy GDP/Primary energy input to the World (USD/J) GDP/Primary energy input to Sweden(USD/J) Dwelling area/Primary energy input to the dwellingsector in Sweden (m2/J) Personal transport/Primary energy input to personal transport in Sweden (Pers. km/J) Goods transport/Primary energy input to goods transport in Sweden (ton km/J) The values are normalizedto the year 1970. Sources: Bumb (1989), FAO Food Balance Sheet (FAO, 1991), Schipper et al. (1994) and Brown et al. (1994). As an example of a complementary specific efficiency indicator we have chosen to indicate the efficiency of the use of nutrients in the Swedish agricultural system• This indicator is defined as; Nutrients in provisions from the Swedish agricultural system r/= The supply of nutrients to the Swedish agricultural system (4.16) The supply of the nutrients N, P and K in fertilizers (excluding manure), imported cattle feed, deposited nitrogen and biologically fixed nitrogen are 111, 15 and 23 k g / h a , respectively (Granstedt and Westberg, 1992). The content of nutrients in provisions (vegetable and animal food) harvested in Sweden are 29, 6 and 6 k g / h a (Granstedt and Westberg, 1992). The differences between these figures are losses due to leakage, sorption, volatilization, etc. The efficiency indicators for the use of nutrients in the Swedish agricultural system in 1990 are then: r/N = ~ = 26% rh, = 6 = 40% "qK = 6 = 26% 4.4.2. I n d i c a t o r no. 4,2: i n t r a g e n e r a t i o n a l j u s t i c e The intragenerational justice indicators are measures of an uneven distribution of services and resource use. These indicators can be defined as the amount of services delivered per capita in a specific region (e.g., a country) compared to the amount of services delivered per capita in a reference region (e.g., the world): 14 z = Supply to region A per capita • Supply to reference region per capita . (4.17) Obviously this indicator will not reflect unequal distributions of income within the specific region. Table 8 gives some examples of intragenerational justice indicators, where we have compared the use of energy and food per capita in Sweden with the global per capita use. The global per capita use of primary energy approaches the Swedish use, but the Swedish per capita use is still much higher than the global use. 107 C. Azar et al. / Ecological Economics 18 (1996) 89-112 Table 8 Intragenerational justice indicators Intragenerational justice indicators 1970 1980 (1987) 1990 Food Calories per capita in Sweden/Calories per capita in the World Proteins per capita in Sweden/Proteins per capita in the World Fats per capita in Sweden/Fats per capita in the World 1.18 1.35 2.11 1.17 1.43 2.14 (1.13) (1.37) (1.98) Energy (Primary energy per capita in Sweden)/(Primary energy per capita in the World) 7.3 5.8 3.8 Sources: FAO Food Balance Sheet (FAO, 1991) and Schipper et al. (1994). 4.4.3. Indicator no. 4,3: intergenerational justice These indicators should cover aspects related to the use of non-renewable resources over time. It is important to note that non-renewable resources must not necessarily be kept in the lithosphere, but may be equally or even more valuable to future generations if they are kept in " c l o s e d " cycles within the technosphere. One rough indicator along these lines could be: XL (4.18) 14"3 -~- X R -t.- X ~T ' where X L is the annual loss of the resource due to dissipative use, and X R and X T represent the available amount of the resource in the lithosphere and the technosphere, respectively (see Fig. 1). The inverse of 14,3 gives the time it would take before the resource would be exhausted (given constant annual losses and no new discoveries). If X c represents the accumulated loss, this indicator would show how much we have used in comparison to how much we have left. 4.4.4. Indicator no. 4,4: basic human needs The human needs indicators aim at indicating to what extent basic human needs are satisfied. Streeten et al. (1981) suggest that one could choose a few basic indicators that together reflect the fulfilment of basic needs (see Table 9). The exact formulation of needs and corresponding indicators should be left to the experts of the different sectors. Many indicators of this kind are already calculated by the United Nations (e.g., literacy, infant mortality rate, life expectancy, calorie supply, etc.). A possible basic human needs indicator is given by: /4.4 = The share of the population that doe s not get their basic needs fulfilled. Table 9 Basic needs and corresponding typical indicators Basic needs Typical indicator Food Per capita daily calorie intake as a percentage of requirements Percentage of population with adequate food intake Percentage of population with access to potable water Percentage of population with access to sanitation facilities Infant mortality per thousand births Life expectancy at birth Literacy Primary school enrollment as a percentage of the population aged 5 to 14 Water and sanitation Health Education Source: Based on Streeten et al. (1981). (4.19) 108 C. Azar et al. / Ecological Economics 18 (1996) 89-112 N u m b e r of u n d e r n o u r i s h e d p e o p l e in m i l l i o n s 1000 800 600 400 200 0 : 1970 36% i 1975 33% , 1980 26% 1990 20% year and percentage of population In developing countries Fig. 5. The number of people in developing countries who are chronically undernourished, using data from Uvin et al. (1994). It should be noted that undernourishment has decreased since 1970, in both absolute and relative terms. A typical basic needs indicator is the number of chronically undernourished people: i.e., people who on average during the course of a year do not consume enough food to maintain their weight and engage in light activity (according to FAO): see Fig. 5 (Uvin et al., 1994). 5. Discussion and conclusions The main result of this paper is the m e t h o d for developing socio-ecological indicators for sustainability. Such indicators should be (i) based on a framework for sustainability (the four socio-ecological principles for sustainability) and (ii) focus early in the causal chain. Practical experience from Swedish companies and local authorities has shown that the socio-ecological principles that we have used as a systematic framework, function well when making strategic decisions. Today 40 Swedish municipalities and 20 larger Swedish companies use these principles in their strategic planning processes (Holmberg et al., 1996; RobSrt, 1994). It is also our experience that the focus for future work with environmental issues is shifting from discussion and investigations of environmental effects (environmental pathology) to strategical planning of the societal metabolism (societal prophylaxis). This implies that there is a need for indicators that focus early in the causal chain--i.e., that put focus on the activities in society. This does not mean that the socio-ecological indicators can replace the traditional environmental quality indicators. But we think that they will make a necessary complement to indicators that focus later in the causal chain. The main reason for focusing early in the causal chain is that we want to capture also societal activities that are potentially unsustainable. In cases where the causal chain and the environmental effects are known, and the time lags are small, the reason for focusing early in the causal chain is not that strong. In these cases it is possible to aggregate emissions into an overall index (see, for instance, the so-called theme indicators developed by Adriaanse (1993) for climate change, depletion of the ozone layer, acidification and eutrophication). There are no exact limits defining sustainability. Instead, the border between sustainability and unsustainability is not sharp. This means that it is not possible to determine exact reference values for sustainability. However, many of the societal activities today are far from sustainable, which means that time series of the indicators would enable us to say whether sustainability is approached or not. 109 C. Azar et a l . / Ecological Economics 18 (1996) 89-112 Table I0 Socio-ecological indicators based on socio-ecological principles Principle 1: Substances extracted from the lithosphere must not systematically accumulate in the ecosphere I u : Lithospheric extraction compared to natural flows IL.2: Accumulated lithospheric extraction ll,~: Non-renewable energy supply Principle 2: Society-produced sub- Principle 3: The physical Principle 4: The use of re- stances must not systematically accumulate in the ecosphere conditions for production and diversity within the ecosphere must not systematically be deteriorated sources must be efficient and just with respect to meeting human needs 12,1: Anthropogenic flows compared to natural flows 12.2: Long-term implication of emissions of naturally existing substances 12,3: Production volumes of persistent chemicals 12,4: Long-term implication of emissions of substances that are foreign to nature 13.1: Transformation of lands 14.1: Overall efficiency 13,2: Soil cover /4.2: Intragenerational justice I~3: Nutrient balance in soils la.~: Intergenerational justice 13.4: Harvesting of funds •4.4: Basic human needs W e have suggested specific indicators for each of the four principles (see Table 10). This broad approach implies that some of the indicators should be considered as p r e l i m i n a r y (especially the indicators for Principles 3 and 4). Still, one advantage is that it offers a point of departure for future work on indicators which takes an integrated approach to sustainability. It is our hope that our formulation of indicators will inspire future research in this area. In future work we shall develop the indicators further. For policy purposes it is important that the n u m b e r of indicators is relatively small. Therefore, we will investigate the possibility of aggregating some indicators: e.g., the indicators for lithospheric extraction for each e l e m e n t into an index for all elements or groups of elements. It is also of importance to display time series of the indicators. W e shall also apply socio-ecological indicators to regional and local situations, and make the m e t h o d accessible to planners and decision makers at regional and local administrative levels of society. In an o n g o i n g project we are studying the resource use of the island of G o t l a n d in Sweden. Here, special emphasis is given to p r o b l e m s related to the responsibility of material flows b e t w e e n regions (Carlson et al., 1995). Acknowledgements W e express our gratitude to Ulrika Carlson for discussions on indicators and for contributions to the empirical b a c k g r o u n d on several indicators. W e also thank BjiSrn Andersson, Curtis C. Bohlen, H e r m a n Daly, Karl-Erik Eriksson, Sten Karlsson, Karl-Henrik Robert, Johan Swahn, Stefan W i r s e n i u s and an a n o n y m o u s reviewer for their critical reading of the manuscript. This research was supported by the Swedish Council for B u i l d i n g Research. References Adriaanse, A., 1993. Environmental Policy Performance Indicators. Sdu, Den Haag. Alfsen, K.H. and S~eb¢, H.V., 1993, Environmental quality indicators: background principles and examples from Norway. Environ. Resource Econ., 3: 415-435. Ayres, R.U., 1995. Statistical Measures of Sustainability. Working paper 95/34/EPS. 1NSEAD, Fontainebleau. Ayres, R.U., Norberg-Bohm, V., Prince, J., Stigliani, W.M. and Yanowitz, J., 1989. Industrial Metabolism, the Environment, and Application of Materials-Balance Principles for Selected Chemicals. International Institute for Applied Systems Analysis (IIASA), Research Report 89-11. 1 I0 C. Azar et aL/Ecological Economics 18 (1996) 89-112 Ayres, R.U., Ayres, L.W. and Hahn, A., 1995. Industrial Metabolism of Selected Toxic Chemicals, Centre for the Management of Environmental Resources of INSEAD, Fontainebleau. Azar, C., Holmberg, J. and Lindgren, K., 1995. Stability analysis of harvesting in a predator-prey model. J. Theor. Biol., 174: 13-19. Bates, T,S., Lamb, B.K., Guenther, A., Dignon, J. and Stoiber, R.E., 1992. Sulphur emissions to the atmosphere from natural sources. J. Atmosph. Chem., 14: 314-337. Beddington, J.R. and Cooke, J.K. 1982. Harvesting from a prey-predator complex. Ecol. Model., 14: 155-177. Benjamin, M.M. and Honeyman, B.D., 1992. Trace metals. In: S.S. Butcher, R.J. Charlson, G.H. Orians and G.V. Wolfe (Editors), Global Biogeochemical Cycles. Academic Press, London, pp. 317-352. Berd~n, M., Nilsson, S., Ros~n, K. and Tyler, G., 1987, Soil Acidification Extent, Causes and Consequences. An Evaluation of Literature Information and Current Research. National Swedish Environmental Protection Board, Report No. 3929. Bergb~ick, B., 1992. Industrial Metabolism. The Emerging Landscape of Heavy Metals Immission in Sweden. PhD Thesis, Link~ping University, Link~ping. Bergstr6m, S., 1993. Value standards in sub-sustainable development. On limits of ecological economics. Ecol. Econ., 7: 1-18. Bolin, B., 1979. On the role of the atmosphere in biogeochemical cycles. Q. J. R. Meteorol. Soc., 105: 25-42. Brown, L.R., Kane, H. and Roodman, D.M., 1994. Vital Signs. Worldwatch Institute, USA. Bumb, B.L., 1989. Global Fertiliser Perspective, 1960-1995: the Dynamics of Growth and Structural Change. Alabama, USA. Buringh, P. and Dudal, R., 1987. Agricultural land use in space and time. In: M.G. Wolman and F.G.A. Fournier (Editors), SCOPE 32: Land Transformation in Agriculture, pp. 9-44. Carlson, U., 1994. Socio-ecological Indicators for Sweden, A First Evaluation for Sweden. Institute of Physical Resource Theory, Chalmers University of Technology, G~Steborg (in Swedish). Carlson, U., Azar, C., Holmberg, J. and Lindgren, K., 1995. Socio-ecological Indicators for the Energy Sector of Gotland. Institute of Physical Resource Theory, Chalmers University of Technology, G~Steborg. Clements, R.G., Boethling, R.S., Zeeman, M. and Auer, C.M., 1994. Persistent bioaccumulative chemicals: screening the TSCA Inventory. Paper presented at the SETAC Foundation Workshop: Environmental Risk Assessment for Organochlorine Chemicals, July 24-29, 1994, Nottawasaga Inn, Toronto. Craswelk E.T., 1993. Management of sustainable agriculture. In: D. Pimentel (Editor), World Soil Erosion and Conservation. Cambridge University Press, Cambridge, pp. 257-276. Crowson, P., 1992. Minerals Handbook 1992-93. Statistics and Analyses of the Worlds Minerals Industry. Stockton Press, New York, NY. ECE, 1985. Draft set of ECE environmental indicators, C E S / 5 4 8 / A d d . 6 / R e v . 1 . Ehrlich, P.R. and Ehrlich, A.H., 1992. The value of biodiversity. Ambio, 21:219-226. Environment Canada, 1991. A Report on Canada's Progress towards a National Set of Environmental Indicators. State of Environmental Report No. 91-1, Falkengren-Grerup, U., Linnermark, N. and Tyler, G., 1987. Changes in acidity and cation pools of south Swedish soils between 1949 and 1985. Chemosphere, 16: 2239-2248. FAO, 1991. Food Balance Sheet. Food and Agricultural Organisation, Rome. Freij, L., 1994. Chemical Substances Lists. Report No. 10/94 National Chemical Inspectorate, Stockholm. German Bundestag, 1989. Protecting the Earth's Atmosphere. Bonn. Gilbert, A.J. and Feenstra, J.F., 1994. A sustainability indicator for the Dutch environmental policy theme. Ecol. Econ., 9: 253-266. Graedel, T.E. and Crutzen P.J., 1993, Atmospheric Change: An Earth System Perspective. W.H. Freeman, New York, NY. Granstedt, A. and Westberg, L. 1992. The Nutrient Flows and Efficiency in the Swedish Agricultural System. Department of Crop Production Science, Ecological Agriculture, Swedish University of Agricultural Sciences (in Swedish). Haes, H., U., de, Nip, M. and Klijn, F., 1991. Towards sustainability: indicators of environmental quality. In: O. Kuik and H. Verbruggen (Editors), In Search of Indicators of Sustainable Development. Kluwer Academic Publishers, Dordrecht. Hallb~icken, L., 1992. Long term changes of base cation pools in soil and biomass in a beech and spruce forest in Southern Sweden. Z. Pflanzenern~:ihr. Bodenkd., 155: 51-60. Holdren, J.P., 1990. Energy in transition. Sci. Am., 263(3): 1(18-115. Holmberg, J. and Karlsson, S., 1992. On designing socio-ecological indicators. In: U. Svedin and B. H~igerh~ill-Aniansson (Editors), Society and Environment: A Swedish Research Perspective. Kluwer Academic Publishers, Dordrecht. Holmberg, J., Robert, K.-H. and Eriksson K.-E., 1996. Socio-ecological Principles for a Sustainable Society - Scientific Background and Swedish Experience. In: R. Costanza (Editor), Getting down to Earth: Practical Applications for Ecological Economics. Island Press, Washington, DC, pp. 17-48. International Joint Commission, 1994. Seventh Biennial Report on Great Lakes Water Quality. International Joint Commission, Washington, DC. IPCC, 1990. Climate Change: The IPCC Scientific Assessment. J.T. Houghton, G.J. Jenkins and J.J. Ephraums (Editors), Cambridge University Press, Cambridge. IPCC, 1992, Climate Change 1992: The Supplementary Report to the 1PCC Scientific Assesment. J.T. Houghton, B.A. Callander and S.K. Varney (Editors), Cambridge University Press, Cambridge. C. Azar et al. / Ecological Economics 18 (1996) 89-112 I 11 IPCC, 1995, Climate Change 1994: Radiative Forcing of Climate Change and An Evaluation of the IPCC IS 92 Emission Scenarios. J.T. Houghton, L.G, Meira Filho, J. Bruce, Hoesung Lee, B.A. Callander, E. Haites, N. Harris and K. Maskell (Editors), Cambridge University Press, Cambridge. Jaffe, D.A., 1992. The nitrogen cycle. In: S.S. Butcher, R.J. Charlson, G.H. Orians and G.V. Wolfe (Editors), Global Biogeochemical Cycles. Academic Press, London. Karlsson, S,, Azar, C., Eriksson K.-E., Holmberg, J., Lindgren, K. and Wirsenius, S., 1994. Materials Flows. On the Metabolism of the Industrial Society in a Sustainability Perspective. AFR-Report No. 38, Swedish Waste Research Council, Stockholm, Sweden. Kendall. H.W. and Pimentel, D., 1994. Constraints on the expansion of the global food supply. Ambio, 23: 198-205. Kuik, O. and Verbruggen, H. (Editors), 1991. In Search of Indicators of Sustainable Development. Kluwer Academic Publishers, Dordrecht. Lantzy, R.J. and Mackenzie, F.T., 1979. Atmospheric trace metals: global cycles and assessment of man's impact. Geochim. Cosmochim. Acta, 44:411-425. Liverman. D.M., Hanson, M.E., Brown, B.J. and Merideth Jr., R.W., 1988. Global sustainability: towards measurement. Environ. Manag., 12: 133-143. Maier-Reimer E. and Hasselman K., 1987. Transport and storage of CO 2 in the ocean - an inorganic ocean-circulation carbon cycle model. Climate Dyn., 2: 63-90. May, R.M., Beddington, J,R., Clark, C.W., Holt, S.J. and Laws, R.M. 1979. Management of multispecies fisheries. Science, 205: 267-277. Miljcminesteriet, 1991. Milj~Sindikatorer 1991 [Environmental Indicators 1991]. Ministry of Environment, Copenhagen. Moffatt, 1., 1994. On measuring sustainable development indicators. Int. J. Sustain. Dev. World Ecol., 1: 97-109. Nilsson, J. and BergstriSm, S.. 1995. Indicators for the assessment of ecological and economic consequences of municipal policies for resource use. Ecol. Econ., 14: 175-184. Nriagu, J.O., 1989. A global assessment of natural sources of atmospheric trace metals. Nature, 238: 47-49. Nriagu, J.O., 1990. Global metal pollution. Environment, 32: 7-32. Nriagu, J.O. and Pacyna, J.M., 1988. Quantitative assessment of worldwide contamination of air, water and soils by trace metals. Nature. 333: 134-139. OECD, 1994. Environmental Indicators. A Core Set. OECD, Paris. Olsson, M., Ros6n, K. and Melkerud, P.-A., 1993. Regional modelling of base cation losses from Swedish forest soils due to whole-tree harvesting. Appl. Geochem., Supply Issue No. 2, 189-194. Opschoor, H. and Reijnders, L., 1991. Towards sustainable development indicators, In: O. Kuik and H. Verbruggen (Editors), In Search of Indicators of Sustainable Development. Kluwer Academic Publishers, Dordrecht. Pettersson, I., 1994. National Chemical lnspectorate, Sweden. Personal communication, October 10. Robert, K.-H., 1994. The Natural Challenge. Ekerlids F~Srlag, Falun (in Swedish). Schipper, L., Johansson, F., Howarth, R., Andersson, Bj~Srn, Andersson, Bo and Price, L., 1994. The Swedish Use of Energy in an International Perspective. NUTEK, The Swedish National Board for Industrial and Technical Development, Report No. 1994,10 (in Swedish). Sigenthaler, U. and Sarmiento, J.L., 1993. Atmospheric carbon dioxide, and the ocean. Nature, 365:119-125. SNV, 1994. A Swedish System of Environmental Quality Indices, A Report to the Swedish Government. Swedish Environmental Protection Board, Solna, October, 20 (in Swedish). Smith, A.M., 1994. OECD, Paris. Personal communication, October 11. SiSderlund, R. and Rosswall, T., 1982. The nitrogen cycles. In: The Natural Environment and the Biogeochemical Cycles. Springer Verlag, Berlin. Speight, J.G., 1992. Fuel Science and Technology Handbook. Marcel Dekker, New York, NY. Sposito, G., 1989. The Chemistry of Soils. Oxford University Press, Oxford. Streeten, P.S., Burki, J., ul Haq, M., Hicks, N. and Stewart, F., 1981. First Things First - Meeting Basic Needs in Developing Countries. Oxford University Press, New York, NY. ten Brink, B., 1991. The AMOEBA Approach as a Useful tool for Establishing Sustainable Development? In: O. Kuik and H. Verbruggen (Editors), In Search of Indicators of Sustainable Development. Kluwer Academic Publishers, Dordrecht. United Nations, 1992a. The Rio Declaration and Agenda 21. New York, NY. United Nations, 1992b. Long-range World Population Projections. Two Centuries of Population Growth 1950-2150. Department of International and Social Affairs, United Nations, New York, NY. UNSCEAR, 1993. Sources and Effects of Ionising Radiation, United Nations Scientific Committee on the Effects of Atomic Radiation. UNSCEAR 1993 Report to the General Assembly. New York, NY. Uvin, P., Cohen, M.J., Snyder, A.C., Hoehn, R,A. and Harris, M., 1994. Overview of world hunger. In: Hunger 1995 - Causes of Hunger. Bread for the World Institute, pp. 10-20. Valkovi6, V., 1983. Trace Elements in Coal. CRC Press, Boca Raton, FL. van der Voet, E. and Kleijn, R., 1992. Cadmium recycling: for better or worse? Paper presented at the International Society for Ecological Economics Conference, August 3-6, 1992, Stockhholm. 112 C. Azar et al. / Ecological Economics 18 (1996) 89-112 Vitousek, P.M., Ehrlich, P.R., Ehrlich, A.H. and Matson. P.A.. 1986. Human appropriation of the products of photosynthesis. Bioscience, 36: 368-373. Vos, J.B., Feenstra, LF., de Boer, J., Braat, L.C. and van Baalen, 1985. Indicators for the State of the Environment. Institute for Environmental Studies, Free University, Amsterdam. Walker J.C.G. and Kastings, J.F., 1992. Effects of fuel and forest conservation on future levels of atmospheric carbon dioxide. Palaeogeogr. Palaeoclimatol. Palaeoecol., 97:151 - 189. Wallgren, B., 1992. Basis of the Ecocycle Society. The Environment Advisory Council, Ministry of Environment and Natural Resources, Stockholm (in Swedish). WCED, 1987. Our Common Future. The World Commissions on Environment and Development, Oxford University Press, Oxford. Yen, T.F., 1975. The Role of Trace Metals in Petroleum. Ann Arbor Science Publishers, Ann Arbor, MI.