(landscape scale) methods and models for operationalizing sustainable ...... Within this manual, the foundations of ... this and other forest management success, Roosevelt published a book entitled: The ...... Kates RW, Clark WC, Corell R, Hall JM, Jaeger CC, Lowe I, McCarthy JJ, ...... Illustrator, Acrobat, Dreamweaver.
AN INVESTIGATION OF SUSTAINABLE URBANIZATION AND LANDSCAPE FORM: THE REPUBLIC OF MOLDOVA CASE STUDY by Richard R. Shaker A Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in Geography at The University of Wisconsin-‐Milwaukee August 2011
AN INVESTIGATION OF SUSTAINABLE URBANIZATION AND LANDSCAPE FORM: THE REPUBLIC OF MOLDOVA CASE STUDY by Richard R. Shaker A Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in Geography at The University of Wisconsin-‐Milwaukee August 2011 Major Professor
Date
Graduate School Approval
Date
ii
ABSTRACT AN INVESTIGATION OF SUSTAINABLE URBANIZATION AND LANDSCAPE FORM: THE REPUBLIC OF MOLDOVA CASE STUDY by Richard R. Shaker The University of Wisconsin-‐Milwaukee, 2011 Under the Supervision of Christopher A. De Sousa, Ph.D. As we embark through the beginning stages of the 21st century, we have already seen the earth stretched beyond its bio-‐capacity. The future of the earth and humanity is, and will continue to be, directly affected by the way we interact and behave within our landscapes. To combat the ills humanity has caused, and projected to continue to cause, sustainable development (SD) has been created as a focal remedy, despite discrepancies. To show the applicability and usefulness of SD this dissertation research, through a two article approach, provides: 1) a better understanding of SD indicators through a survey of composite SD indices, and the creation of two multimetric SD indices for local spatial application; 2) local (landscape scale) methods and models for operationalizing sustainable development for regional planning purposes; 3) a better understanding on spatial models and geographic information science (GIScience) techniques for sustainable development; 4) research on landscape design and human behavior for investigating the feasibility of sustainability; 5) ultimately, a work that links sustainable development theory with an applied application.
iii
To date, there exist no ‘ideal’ instrument for achieving sustainability neither on the regional nor the local scale. Article one addresses this problem by reviewing common SD indices, and developing an assessment system for analyzing and evaluating SD for use at the local scales of urban planning and regional development. For SD spatial assessment, two composite indices (Ecological Demand Index and Sustainability Demand Index) are created using a demographic and health survey for the Republic of Moldova. Article two uses both mean Ecological Demand Index (EDI) and mean Sustainability Demand Index (SDI) from article one as evaluation indices of 66 landscapes in the Republic of Moldova. In this article, non-‐spatial, spatial, and nonlinear multivariate statistical models were used to investigate relationships found between-‐ landscape form and population characteristics-‐ and the local indicators of sustainable development condition. Significant statistical relationships were found through this exploration of SD indicators for assessing landscape functions for operationalizing SD. Future research involves evolutionary analysis of human behavior in relation to urban and landscape form using Internet 2.0 technology. Major Professor
iv
Date
© Copyright by Richard R. Shaker, 2011 All Rights Reserved
v
TABLE OF CONTENTS List of figures................................................................................................................................................x List of tables................................................................................................................................................xii Acknowledgements................................................................................................................................xiv 1. Introduction............................................................................................................................................1 2. Background..............................................................................................................................................8 2.1. Sustainability and sustainable development...................................................................8 2.2. Measuring sustainable development...............................................................................13 2.3. Landscape ecology, planning, and sustainability........................................................16 2.4. Human behavior and sustainability..................................................................................21 3. Problem statement............................................................................................................................22 4. Article 1: Assessing sustainable development using additive household community and property composition indices...........................................................................25 4.1. Abstract.........................................................................................................................................25 4.2. Introduction................................................................................................................................26 4.3. Sustainable development assessment and monitoring............................................29 4.3.1. Survey of common sustainability indices..........................................................31 4.3.1.1. Gross domestic product (GDP)..............................................................31 4.3.1.2. Human development index (HDI)........................................................31 4.3.1.3. Wellbeing (WB)............................................................................................32 4.3.1.4. Ecological footprint (EF)..........................................................................32 4.3.1.5. Environmental sustainability index (ESI).........................................33 4.3.1.6. Environmental performance index (EPI)..........................................34
vi
4.3.1.7. Living planet index (LPI)..........................................................................34 4.3.1.8. Green net national product (EDP) and SEEA...................................35 4.3.1.9. City development index (CDI)................................................................36 4.3.1.10. Environmental vulnerability index (EVI).......................................36
4.4. Why evaluate sustainability at the household scale?................................................37 4.5. The assessment system..........................................................................................................40 4.6. The assessment criteria.........................................................................................................40 4.6.1. Household community information.....................................................................42 4.6.2. Property composition factors.................................................................................44 4.6.3. The geographic factor................................................................................................47 4.7. The multimetric assessment process...............................................................................48 4.8. Examples from the Republic of Moldova........................................................................50 4.8.1. Study area and historical context.........................................................................50 4.8.2. Official demographic and health survey data..................................................53 4.8.3. Method..............................................................................................................................57 4.8.3.1. Ecological demand index (EDI).............................................................59 4.8.3.2. Sustainability demand index (SDI)......................................................64
4.9. Results and discussion............................................................................................................68 4.10. Conclusion.................................................................................................................................71 5. Article 2: Regional application of two indices of household demand for assessing sustainable development: a spatial analysis through Republic of Moldova landscapes................................................................................................................................................................72 5.1. Abstract.........................................................................................................................................72 5.2. Introduction................................................................................................................................74
vii
5.3. Methods.........................................................................................................................................80
5.3.1. Study site selection......................................................................................................80 5.3.2. Moldovan “landscape unit”......................................................................................85 5.3.3. Household ecological demand index (EDI) and household sustainability demand index (SDI)....................................................................................88 5.3.4. Landscape unit selection and independent variables..................................99 5.3.5. Data analysis................................................................................................................104 5.4. Mean ecological demand index (EDI) results............................................................112 5.4.1. Exploratory spatial data analysis ......................................................................112 5.4.2. Non-‐spatial, spatial, and nonlinear statistical models..............................115 5.5. Mean sustainability demand index (SDI) results......................................................123 5.5.1. Exploratory spatial data analysis ......................................................................123 5.5.2. Non-‐spatial, spatial, and nonlinear statistical models..............................126 5.6. Mean ecological demand index (EDI) discussion.....................................................134 5.7. Mean sustainability demand index (SDI) discussion..............................................136 5.8. Conclusion.................................................................................................................................139
6. Concluding remarks.......................................................................................................................141 6.1. Dissertation recap..................................................................................................................145 6.1.1. Article one....................................................................................................................146 6.1.2. Article two....................................................................................................................148 6.2. Future research paradigm..................................................................................................151 7. Literature cited.................................................................................................................................155 8. Appendix A: MDHS (2005:259-‐273) household questionnaire..................................177 9. Appendix B: EDI and SDI construction notes......................................................................195
viii
10. Curriculum vita..............................................................................................................................206
ix
LIST OF FIGURES Fig. 1. Global urban and rural population trend, 1950-‐2030.................................................3 Fig. 2. Hierarchical and pluralistic view of landscape ecology and its relationship to sustainability.............................................................................................................................................19 Fig. 3. Map of the demographic and health survey geographic distribution within the Republic of Moldova (47°24’N, 28°22’E).......................................................................................56 Fig. 4. Spatial distribution of the 10 individual mean SD indicators chosen for the composite ecological demand index (EDI) developed in the Republic of Moldova....62 Fig. 5. Spatial distribution of the composite ecological demand index (EDI) developed in the Republic of Moldova...........................................................................................63 Fig. 6. Spatial distribution of the 15 individual mean SD indicators chosen for the composite sustainability demand index (SDI) developed in the Republic of Moldova........................................................................................................................................................66 Fig. 7. Spatial distribution of the composite sustainability demand index (SDI) developed in the Republic of Moldova...........................................................................................67 Fig. 8. Frequency distribution and box plot showing quartiles of the (A) composite ecological demand index (EDI), and (B) composite sustainability demand index (SDI) calculated from 11,066 households surveyed in the Moldova Demographic and Health Survey (2005). Explaining the box plot: the box marks the lower and upper quartiles of a distribution, and the length of the box is the interquartile range. The whiskers mark observations within the quartiles ± 1.5 * (interquartile range). Observations beyond the whiskers are considered outliers................................................69 Fig. 9. Republic of Moldova location map with 2004 land cover (47°24’N,28°22’E)...................................................................................................................................81 Fig. 10. Original multi-‐hierarchical land management structure systematically divided the Republic of Moldova into four scales: (A) 120 elementary landscape features; (B) 74 landscape units, 5 regions, and 2 zones.......................................................88 Fig. 11. Map of 399 demographic and health survey cluster locations with spatial representation in 66 of 74 landscape units within the Republic of Moldova................91 Fig. 12. (A) Moran’s I scatterplot and (B) spatial correlogram displaying spatial autocorrelation for mean ecological demand index (EDI)..................................................114
x
Fig. 13. (A) Spatial distribution of mean ecological demand index from 11,066 households averaged to 399 Thiessen polygons; (B) Local Anselin Moran’s I displaying rendered z-‐score spatial clustering of mean ecological demand index (EDI)............................................................................................................................................................115 Fig. 14. (1) Actual versus predicted scatterplot of mean ecological demand index (EDI) for (A) standard least squares regression and (B) artificial neural networks; and, (2) their corresponding response profile for multivariate SD landscape form and population characteristics model..........................................................................................119 Fig. 15. (A) Actual versus predicted scatterplot of mean ecological demand index (EDI) and estimated simulations autoregressive (SAR) model error; (B) spatial distribution of actual mean ecological demand index (EDI), estimated mean EDI, SAR model residuals, and SAR model errors; (C) frequency distribution displaying normal distribution of SAR model residuals; and (D) spatial correlogram displaying actual mean EDI, estimated mean EDI, SAR model residuals, and SAR model errors..........................................................................................................................................................122 Fig. 16. (A) Moran’s I scatterplot and (B) spatial correlogram displaying spatial autocorrelation for mean sustainability demand index (SDI)...........................................125 Fig. 17. (A) Spatial distribution of mean ecological demand index from 11,066 households averaged to 399 Thiessen polygons; (B) Local Anselin Moran’s I displaying rendered z-‐score spatial clustering of mean sustainability demand index (SDI)............................................................................................................................................................126 Fig. 18. (1) Actual versus predicted scatterplot of mean sustainability demand index (SDI) for (A) standard least squares regression and (B) artificial neural networks; and, (2) their corresponding response profile for multivariate SD landscape form and population characteristics model..........................................................................................130 Fig. 19. (A) Actual versus predicted scatterplot of mean sustainability demand index (SDI) and estimated conditional autoregressive (CAR) model error; (B) spatial distribution of actual mean sustainability demand index (SDI), estimated mean SDI, CAR model residuals, and CAR model errors; (C) frequency distribution displaying normal distribution of CAR model residuals; and (D) spatial correlogram displaying actual mean SDI, estimated mean SDI, CAR model residuals, and CAR model errors..........................................................................................................................................................133
xi
LIST OF TABLES Table 1. Republic of Moldova household ecological demand index (EDI) metrics and relative scoring criteria................................................................................................................58 Table 2. Republic of Moldova household sustainability demand index (SDI) metrics and relative scoring criteria...............................................................................................59 Table 3. Pearson product-‐moment correlation coefficients for the 10 standardized indices used in computing the Republic of Moldova household ecological demand index (EDI)..................................................................................................................................................70 Table 4. Pearson product-‐moment correlation coefficients for the 15 standardized indices used in computing the Republic of Moldova household sustainability demand index (SDI)................................................................................................................................71 Table 5. Republic of Moldova household ecological demand index (EDI) metrics displaying discrete scoring criteria..................................................................................................93 Table 6. Republic of Moldova household sustainability demand index (SDI) metrics displaying discrete scoring criteria..................................................................................................96 Table 7. Independent landscape variable summary used in indicator-‐based statistical assessment of SD function............................................................................................100 Table 8. Results of stepwise multiple regression (ordinary least squares) model for mean household ecological demand index (EDI) as a function of Republic of Moldova landscape form and population characteristics. (A) Final regression model showing standardized coefficients; (B) analysis of variance (ANOVA) for overall significance model..........................................................................................................................................................116 Table 9. Results of simultaneous autoregression (SAR) multiple regression model for mean household ecological demand index (EDI) as a function of Republic of Moldova landscape form and population characteristics. (A) Final autoregressive model showing standardized coefficients; (B) analytical results for overall significance of final spatial model..................................................................................................121 Table 10. Results of stepwise multiple regression (ordinary least squares) model for mean household sustainability demand index (SDI) as a function of Republic of Moldova landscape form and population characteristics. (A) Final regression model showing standardized coefficients; (B) analysis of variance (ANOVA) for overall significance model................................................................................................................................127
xii
Table 11. Results of conditional autoregression (CAR) multiple regression model for mean household sustainability demand index (SDI) as a function of Republic of Moldova landscape form and population characteristics. (A) Final autoregressive model showing standardized coefficients; (B) analytical results for overall significance of final spatial model..................................................................................................132
xiii
ACKNOWLEDGEMENTS This dissertation has been supported by: The University of Wisconsin-‐ Milwaukee Graduate School, The University of Wisconsin-‐Milwaukee Geography Department, and the United States Department of State through a J. William Fulbright Fellowship.
I would like to thank my advisory committee: Dr. Christopher A. De Sousa, Dr. Timothy J. Ehlinger, Dr. Changshan Wu, Dr. Glen G. Fredlund, and Dr. Woonsup Choi for their continued support. Furthermore, I would like to thank Dr. Ghennadie Sirodoev and Dr. Igor Sirodoev from the Institute of Ecology and Geography, Academy of Sciences of Moldova for their patronage and expertise during my time in the Republic of Moldova. This international research would not have been made possible without the additional resources provided by the Institute of Ecology and Geography, Academy of Sciences of Moldova. The findings and conclusions within this dissertation research are those of the author and not to reflect on the supporting agencies. It is without question that this dissertation could not have been made possible without the continued encouragement and support of all those who have touched my life. With all of you, I share my greatest achievement.
xiv
1
1. Introduction We live on a planet with finite resources. Its ability to support a thriving diversity of species, including humans, is large but fundamentally limited (WWF 2008). As we embark though the beginning stages of the 21st century, it is imperative that we come to terms with the effects that the expansionist worldview is having on the Earth. As the dominant social paradigm, it views humans as cure-‐all capable (Rees 1995). Ultimately, this worldview concludes that new technologies and human ingenuity will improve human life and planet conditions (Simon 1995). However, due to the size and complexity of the Earth, the spatial and temporal effects of the expansionist worldview are seldom seen in a human lifetime. Because of this mechanism, it makes it possible for humanity to “tune out long-‐term trends over which (we) have no control” (White 1994) and let our preferences guide our decisions, rather than facts (Jones 1996). Despite the instinctual strengths of the expansionist worldview, an alternative ecological worldview has emerged. This perspective suggests that there are limits to the ability of the planet to support humanity; albeit, that human activity must be tempered to their long-‐term effects on natural resources and related services (Costanza 1989; Rees 1995; Wackernagel and Rees 1996). That if we are to avoid self-‐destruction, human behavior and use of our planet needs to be changed immediately (Ehrlich and Ehrlich 1992). The ecological worldview recognizes that unconstrained consumption of limited resources will ultimately lead to Garrett Hardin’s (1968) “tragedy of the commons” and social chaos (Ruckelshaus 1989).
2
The current integrity of the planet is being stressed beyond its biological capacity. The Living Planet Index (LPI) of global biodiversity has declined by roughly 30 percent since the 1970s, showing a loss of total vertebrate species throughout the world (WWF 2008). In 1998 the global population exceeded the Earth’s carrying capacity. Carrying capacity is the largest number of any given species (in this case, humankind) that a habitat (in this case, earth) can support
indefinitely (Keiner 2004). Humanity’s demands, measured by Ecological Footprint (EF), now exceed the planet’s natural regenerative capacity by roughly a 30 percent overshoot; furthermore, it has been projected that humanity will require the biocapacity equal to two planet Earths by the 2030s (WWF 2008). Anthropogenic stressors are projected to continue to increase as global population increases. It has been estimated that the global population will range between 9 – 13 billion by 2050 (United Nations 2005). We are now reaching a landmark in human history. Until recently, more people have lived in rural areas than urban areas (Crane and Kinzig 2005; UNEP 2005). In 1900, the population of cities worldwide was only 224 million people; by 1999, urban population had increased to 2.9 billion (UNPD 1999). Scientists have stated that urban populations are going to continue to increase (Figure 1). A study by the United Nations showed that global urban population rose from 29% in 1950 to 49% in 2000 (United Nations 2005). It has been projected by 2030 that 81% of Europeans and 85% of North Americans will live in urban settings (UNDP 2001). Overall, it has been projected that developed countries urban population will total roughly 84% by 2030 (UNDP 1999). It has been projected that by 2030 developing
3
countries urban population will increase by 20%; albeit, suggesting that 80% of global growth of urban population will take place in poorer countries from 2000 to 2030 (UNPD 2005). According to a United Nations Population Division report (2001), by the year 2015 there will be 58 cities with more than five million in population, that number up from 39 in 2000, and foresee 27 so-‐called ‘Maga-‐Cities’
with more than 10 million inhabitants. The amount of urban occupied land area on Earth is projected to increase from 0.3% in 2000 to 0.9% by 2030 (UNPD 1999).
Fig. 1. Global urban and rural population trend, 1950-‐2030 (From United Nations 2005).
As the world becomes more and more urbanized, it is imperative that there becomes better understanding of the systems underway. The structure, function, and dynamics of contemporary ecosystems are profoundly influenced by human
4
activities (Alberti 2005), and understanding the mechanisms responsible for environmental changes requires the integration of both natural and human processes. Pervasive social, economic, and ecological changes have occurred as a result of human activities (Alberti 2005). A change in land cover through the appropriation of natural landscapes to provide for human needs is one of the most pervasive alterations resulting from human activity (Vitousek 1994). Specifically,
the transition from the natural/native landscapes to urban landscapes is having the greatest impact on earth. Ecologically, urbanization alters environmental integrity through a range of processes including: fragmenting landscapes, isolating habitat patches, simplifying biodiversity, degrading natural habitats, modifying landforms and drainage networks, introducing exotic species, controlling and modifying disturbances (e.g., floods, forest fires), and disrupting energy flow and nutrient cycling (Picket et al. 2000; Alberti et al. 2003; Alberti 2005). The EPA (2001) concluded in Our Built and Natural Environments that the urban form directly affects habitat, ecosystems, endangered species, and water quality through land consumption, habitat fragmentation, and replacement of natural cover with impervious surfaces. The synergy effects of future urbanization are not completely understood; however, it has been stated that urbanization will continue to metabolizing landscapes surrounding existing cities (Daniels 1999; Theobald 2002; Crump 2003). Paralleling global urban expansion, there is a necessity for a sustainable transition toward a stable human population with an increase in living standards and the establishment of long-‐term balance between human development needs and
5
the planet’s environmental limits (NRC 1999; Kates et al. 2001; Clark and Dickson 2003; Parris and Kates 2003). Besides the environmental ramifications of urbanization, a major challenge worldwide is to understand how changes in social
organization and dynamics will impact the interactions between nature and society (NRC 2001; Kates et al. 2001; Clark and Dickson 2003). Close to a billion people currently live in “extreme economic poverty” (less than 1 US dollar a day), and lack access to essential natural resources to meet basic needs (World Bank 2008). The present worldwide trend toward urbanization is intimately related to economic development and to profound changes in social organization, land use, and patterns of human behavior (UNMP 2004; Crane and Kinzig 2005). It has been stated that economic growth is now increasing the world’s environmental burdens much faster than population growth (Hughes and Johnston 2005). To combat the problems associated with human population growth, and its affects on global evolution, a paradigm of international awareness was started in the 1940s. The inequalities of the world were exposed and became internationally known during World War II. After the end of World War II, President Harry S. Truman (1949) laid the foundations of international development with goals of alleviating poverty, reducing disparity, and improving the global standard of living. Internationally, the United Nations Educational, Scientific, and Cultural Organization (UNESCO) started the ecologically sustainable development (SD) discussion at the 1968 International Conference for Rational Use and Conservation of the Biosphere. In 1972, at the United Nations Conference on the Human Environment (UNCHE or Stockholm Conference), the United Nations Environmental Program (UNEP) stated
6
its mission: “To provide leadership and encourage partnership in caring for the
environment by inspiring information, and enabling nations and people to improve their quality of life without compromising that of the future generations”. In 1983, the World Commission on Environment and Development (WCED) or (Brundtland Commission) was created by UN-‐mandate to develop a long-‐term environmental strategy for achieving SD for the years 2000 and beyond. In 1987, the WCED responded to an emerging global awareness of social, economic, and environmental inequalities with the report Our Common Future (WCED 1987). In this report, the WCED expressed the importance of development that “extends to all the opportunity to fulfill their aspirations for a better life.” They went on to emphasize that the development must be “within the bounds of the world’s ecological means”. Ultimately, the Commission called for SD “that meets the needs of the present without compromising the ability of future generations to meet their own needs” (WCED 1987). In 1992, an action plan, called Agenda 21, for SD was initiated at the Rio De Janeiro Earth Summit (World Summit on Environment and Development) to balance the needs and aspirations of people with the health of ecosystems and overall integrity of the Earth (United Nations 1992). It was at the Rio Earth Summit, that SD was embraced as an important goal worldwide. The UN Millennium Development Goals were adapted (2000), and reiterated at the SD summit in Johannesburg (2002), for all countries to integrate the principles of SD into national policies and programs (UNDSD 2002). Although SD has become a priority for many countries throughout the world, there remains a need for better clarification and examples of practical applications.
7
In The Future of Sustainability, Marco Keiner (2006) states that the challenges for
improving the world now lie in operationalizing SD. Evidence suggests that SD has a lack of applied usefulness because of its all-‐encompassing theoretical vagueness (Voss 1997). There is a paradox within the term-‐ sustainable growth implies increase endlessly-‐ which is not possible on a planet with a finite amount of resources (Bartlett 2006). Additionally, growth has been used synonymously with development in context to urbanization and economic prosperity (Keiner 2006). To support the need for better clarification, Dobson (1996) found roughly 300 documented definitions for sustainability and SD. Some believe that achieving a common understanding of SD is more remote than ever (Jickling 2000). According to Grober (2007) “It has, some critics say, the smell and flexibility of plastics and feels like something thoroughly artificial.” It has been suggested that SD is too subjective and ultimately unreasonable for humans to achieve (Kemp and Martens 2007). In order to better understand the usefulness and applicability of SD, this research is organized into three main stages. The first stage of this research is to critically review information on composite indices of SD, and then to develop two composite indices of SD at the household scale for local spatial assessment. It is imperative to have a concise understanding of the usefulness, needs, and applicability of current SD indicators if they are to be implemented and used properly. The second stage of this research is to give examples of how SD can be operationalized for an applied practical purpose (in this case, regional planning). This step in the research required the collection and analysis of landscape form and
8
urban design factors, household community information, and property composition parameters for spatial analysis at the landscape level. Without a sound example of how to use SD for a practical purpose, people will continue to be confused on its usefulness and reluctant to move theory into practice. The third and final stage of this research will investigate the achievability of SD. This concluding phase of the research explores the relationships between-‐ sustainable urbanization and landscape form-‐ and human behavior, and the next steps that need to be taken to continue this research paradigm. 2. Background 2.1. Sustainability and sustainable development During the 1987 World Commission on Environment and Development (WCED), the Brundtland Commission’s report, Our Common Future, stated that global population had surpassed its ecological carrying capacity. To bring attention to this, and other issues, the UN General Assembly recognized that environmental problems were global in nature, had economic and social relationships, and should be taken into account when establishing policies (WCED 1987). Within the Brundtland Commission’s report, the most popularly used definition of ‘sustainable development’ was coined. The Brundtland Commission’s definition of sustainable development (SD) reads as follows: “Sustainable development is development that meets the needs of the present without compromising the ability of future generations to meet their own needs” (WCED 1987). The new buzz word: ‘sustainable development’ was later brought onto the global stage at the 1992 ‘Earth Summit’ in Rio de Janeiro when trying to establish a balance between the use and
9
preservation of nature’s potentials and resources (United Nations 1992). However, before the Brundtland Commission report and the ‘Earth Summit’, the conceptual roots of SD were long established. The roots of SD can be traced back to the era of early ‘European Enlightenment’ (Grober 2007). In the 16th century, many of the established nations were using their woodlands to build ships to support their imperialist and colonialist needs. Soon it was noticed that woodlands were diminishing and the future of their nations were at stake. To combat the diminishing lumber supply, and to assure England’s economic prosperity, John Evelyn dedicated his time to creating a forestry manual called: Sylva (1664). Within this manual, the foundations of sylvaculture were established. Sylvaculture is the practice of tree farming for cultivation purposes. From Evelyn (1664) work, the German nobleman Hanns Carl von Carlowitz developed the concept of ‘nachhaltigkeit’ or ‘sustainable yield’ in his work call: Sylvicultura Oeconomica (1713); making its debut into print more than 250 years before the Brundtland Commission’s report, Our Common Future. By the middle of the 18th century, the concept and usage of ‘sustainable yield’ was being applied in other context. In Moser (1757) work, “eine nachhaltige wirtshcaft” the concept was used to explain a sustainable economy in context to mining and the timber industry. Alexander von Humboldt made an early definition of ‘sustainable yield’ popular in 1792. von Humboldt (1792), considered ‘sustainable yield’: “A steady and safe husbandry aiming at the balance between offspring and annual consumption”.
10
As Europe was applying the conceptual roots of SD to almost all natural
resources, the United States resources were being consumed at an astounding rate. In Pinchot’s (1998) historical work, timber and other resources in the United States were considered a virtue and were consumed without any management until the early 1900s. According to Pinchot (1998) “Until 1905, not a single acre of timberland was under systematic forest management.” In 1901, Franklin D. Roosevelt sent a letter to Congress strongly commending the plan for a national forest reserve in the Southern Application region (New York Times 1901). Due to this and other forest management success, Roosevelt published a book entitled: The Use of National Forest Reserves in 1905. In this book, he stated that: “The prime objective of the forest reserves is wise use.” While the forest and its dependent interests must be made permanent and safe by preventing overcutting or injury to young growth, every reasonable effort will be made to satisfy legitimate demands” (Roosevelt 1905). By the 1930s Franklin D. Roosevelt was using the conceptual framework of ‘sustainable yield’ to support his ‘New Deal’ plan; combat the so called ‘Dustbowl’ and employ millions through the Civilian Conservation Corp (CCC) to restore nature and stimulate local economy. By the 1960s and 1970s, there were environmentalist, such as Rachel Carson and Aldo Leopold, using the conceptual roots and practicality of ‘sustainable yield’ in their work. “We abuse land because we regard it as a commodity belonging to us. When we see the land as a community to which we belong, we may begin to use it with love and respect” (Leopold 1970). Based on communal efforts, one of the largest historical sustainability hallmarks for
11
the United States came with the conception of the ‘National Environmental Policy Act of 1969’ (USEPA 1970).
Internationally, after the end of World War II, Truman (1949) laid the
foundations of international development with goals of alleviating poverty, reducing inequality, and improving the global standard of living. By the late 1960s, the concepts of ‘sustainability’ and ‘sustainable development’ were being proposed and suggested in international conventions. In 1968, at the International Conference for Rational Use and Conservation of the Biosphere, the United Nations Educational, Scientific, and Cultural Organization (UNESCO) started the ecological sustainable discussion. In 1972, at the United Nations Conference on the Human Environment (UNCHE), the United Nations Environmental Program (UNEP) proclaimed “to provide leadership and encourage partnership in caring for the environment by inspiring, informing, and enabling nations and people to improve their quality of life without compromising that of the future generations”. In 1983, at the World Commission on Environment and Development, the Brundtland Commission set a goal to: develop a long-‐term environmental strategy for achieving SD by the year 2000 and beyond. In 1984, during the International Conference on Environment and Economics, it was concluded that environment and economics should be mutually reinforced. In 1987, the Brundtland Commission’s report, Our Common Future, define the ‘sustainable development’ and re-‐popularized its conceptual roots. In 1992, at the United Nation’s Conference on Environment and Development, SD was established as a common goal for the 160+ countries that attended. In 1994, at the International Conference on Population and Development,
12
it was stated that: extreme poverty and environmental resource shortages can exacerbate ethnic and political divisions globally. In 2000, the eight Millennium
Development Goals were established for reducing poverty and improving live at the Millennium Summit (UNMS 2000). In 2002, at the World Summit on Sustainable Development, the Johannesburg Plan of implementation of Agenda 21 though public/private partnerships for SD was created. Even though the term ‘sustainable development’ has very applied roots, it has lost a lot of its applicability due to discrepancy of its meaning and overuse of the term. In 1996, there were over 300 documented definitions for sustainability and sustainable development (Dobson 1996). Voss (1997) said that there is a vague substance of the term sustainability itself, which leaves much room for interpretation. One such problem is with development being used synonymously with urbanization and economic prosperity (Keiner 2006). Another common discrepancy is when SD is confused with sustainable growth. Sustainable growth implies to increase endlessly, which is not possible on a planet with a finite size of resources (Bartlett 2006). Some people believe that achieving a common understanding of SD is more remote than ever (Jickling 2000). Within ‘sustainable development’ there is a debate between two opposing groups: the ‘ecological’ and ‘expansionist’ worldviews. The ‘ecological worldview’ is to avoid self-‐destruction; that human behavior and use of our planet needs to be changed immediately (Ehrlich and Ehrlich 1992). The ‘expansionist worldview’ believes that new technologies will improve human life and planet conditions exponentially through time (Simon 1995). Glasby (2002) argues that only a massive
13
decrease in world population and resource use would permit long-‐term occupancy of the earth. Further, it has been stated by Hughes and Johnston (2005) “Economic growth is now increasing the world’s environmental burdens much faster that population growth.” With world environmental issues being magnified, and globalization affecting economies and societies at astounding rates, implementation of SD is needed now more then ever. However, according to Hales and Prescott (2002) “making progress toward sustainability is like going to a destination we have never visited before, equipped with a sense of geography and the principles of navigation, but without a map or compass.” It has been noted that the challenges of SD now lie in its operationalization (Keiner 2006). Albeit, to have “implementation of initiatives that do not merely pay lip-‐service to the words, but actively do justice to the original concept” (Campbell 2000). 2.2. Measuring sustainable development “A concerted effort to enhance habitability of our planet is unlikely to succeed unless we know ‘where we are’ and ‘where we want to go’” (Thomas 1972). Since the industrial revolution of the 18th and 19th centuries, due to increasing anthropogenic stressors, there has been a greater push to monitor the environment in which we live (Harris and Browning 2005). By the early 1970s, environmental indicators were starting to gain popularity. In the United States, through the formation of the President’s Council on Environmental Quality, indicators were in demand to measure progress towards environmental goals and pollution control targets (Rogers et al. 1997). The foundations of indicator development can be
14
linked back to Herbert Inhaber’s (1976) work: Environmental Indices and Wayne Ott’s (1978) work Environmental Indices: Theory and Practice. After a lull in indicator research through the early 1980s, a renaissance came after their
applicability became apparent at the UNCED meeting and Agenda 21 (Rogers et al. 1997). An indicator is a single value from a single measure of quantity; whereas, an index is the combination or aggregation of more than one single indicator or single value (Ott 1978; WRI 1992). According to Rogers et al. (2008) the World Bank describes an indicator as “a performance measure that aggregates information into a usable form.” Indicators are formed by observed or estimated data (OECD 2002). Indicators are quantitative measures and only generic definitions of quality are used, making it very difficult to make accurate decisions at the margins (Rogers et al. 2008). The environment in which we live is multidimensional-‐ influenced by many different economic, social, and environmental phenomenon (Pezzoli 1997; Cabezas et al. 2003; Mayer et al. 2004). By accepting the sustainability challenge, countries have accepted the need for indicators as a measuring tool for SD (Moran et al. 2008). Sustainability and SD indices have been developed specifically to help policy-‐makers make their decisions (Mayer 2008). According to the United Nations (1992) “Indicators of sustainable development need to be developed to provide solid bases for decision-‐making at all levels and to contribute to a self-‐regulating sustainability of integrated environment and development systems.” Sets of sustainability indicators, and manipulation of these indicators into indices, are increasingly used
15
to make policy decisions (Oras 2005; Hezri and Dovers 2006), and it is critical to
understand strengths, weaknesses, scale-‐dependence, etc. when using them (Parris and Kates 2003; Morse and Fraser 2005; Ness et al. 2007). Hundreds of different indices have been suggested and more are under development by a growing number of institutions (see Tschirley 1997). The International Institute of Sustainable Development (IISDnet) compiled a list of SD indicators in 1995. However, it has been stated, that the results are “voluminous” and “not very well focused” (Rogers et al. 2008). The UN Department of Economic and Social Affairs-‐ Division for Sustainable Development-‐ finalized a list of 96 indicators, including a subset of 50 core indicators, in their report: Indicators of Sustainable Development: Guidelines and Methodologies (2007) to be used as a reference for countries. It has been said, that over 140 different indicators have been proposed for the OECD countries (Moffatt 2008). In the large and growing SD literature there are two views on measurement: weak measures and strong measures (Moffatt 2008). With strong sustainability, measures assume that some ecological functions and resources cannot be substituted with technological or other man-‐made replacement. The strong measures are based on the assumption that maintaining the planet’s ecology is vital, and economic and social activities have to remain well within the ecological means (Moffatt 2008). With weak sustainability, measures assume that there can be universal substitution. The weak measures are based from the long-‐standing tradition of neoclassical economics (see Atkinson and Pearce 1993; Pearce and Barbier 2000). Unfortunately, many of the sustainability indices were created with
16
similar methods and from similar data sources (e.g., UN, WB,). According to Mayer (2008) “The degree to which these indices differ in their results using the same data is due to their assumptions, biases, and methodological disparities, creating confusion for sustainability efforts.” It has been stated that SD indicators lie heavy on environmental needs while skimping on social and economic needs (Moldan et al. 2004; Moffatt 2008). Some of the more common composite sustainability indices are: Gross Domestic Product (GDP), wellbeing (WB), happiness (H), human development index (HDI), dashboard of sustainability (SDDSI), and Ecological footprint (EF) (OECD 2004; Wilson et al. 2007; Mayer 2008; Moffatt 2008; Rogers et at. 2008). 2.3. Landscape ecology, planning, and sustainability A paradigm started by the German biogeographer Carl Troll (1939), Landscape Ecology combines both the spatial approach of the geographer and the functional approach of the ecologist (Naveh and Lieberman 1984; Forman and Godron 1986). The principles of Landscape Ecology are relevant when addressing the dynamic makeup of the ever-‐changing landscape that humans inhabit. Landscape Ecology, as defined by Richard T. T. Forman, (1983) incorporates: (1) the spatial relationship among landscape elements, or ecosystems, (2) the flow of energy, minerals, nutrients, and species among the elements, and (3) the ecological dynamics of the landscape mosaic through time. Today, Landscape Ecology is considered to be an interdisciplinary science drawing from a variety of different disciplines (i.e., anthropology, architecture, biology, ecology, economics, geography, and forestry). A key component of Landscape Ecology addresses anthropogenic
17
effects on both natural and built landscapes; furthermore, understanding that human activity is a central factor for shaping the environment (Forman 1995; Turner 2001).
In Landscape Ecology, the patch is the fundamental element of the landscape. The patch, defined by Richard Forman (1995), is an area of specific type (e.g., agricultural land, woodlot, waterbody) that is different than its surrounding types in a landscape. The size and shape of the patch, its proximity to other patches, and its edges are particularly important patch characteristics that have significant ecological and environmental impacts (Forman 1995; Turner et al. 2001; Alberti 2005). The patch is the primary component in Landscape Ecology used for developing the analytical metrics in a land cover or land use analysis (McGarigal and Marks 1995; McGarigal et al. 2002). Urban to rural gradients provides ecologists with an opportunity to examine the urbanization process as a transformation of landscape patterns and functions (Huang 1998; McGarigal and Cushman 2005). One approach is to characterize the relationships between various arrangements of patch structure and ecosystem functions (Godron and Forman 1982; Turner 1989; Forman 1995; Collinge 1996). The use of landscape metrics for analyzing spatial patterns has become quite popular, and some effort has been made to examine the behavior and limitations of landscape metrics for better understanding urbanization (Turner 1989; Wu 1998; Jennerette and Wu 2001). Significant advances in computing, mathematical theory, and system analysis has occurred over the last two decades; furthermore, making progress in theory and tool development for addressing the level of complexity associated with
18
sustainability (Leitão et al. 2006). One of the most significant advances in Landscape Ecology came through the development of FRAGSTATS in 1995. FRAGSTATS is software developed for computing landscape metrics (McGarigal and Marks 1995; McGarigal et al. 2002) and is the tool of choice in basic and applied ecology literature (see Diaz 1996; Zorn and Upton 1997; Hargis et al. 1998; Tinker et al. 1998; Tischendorf 2001). FRAGSTATS is utilized for several reasons: (1) it contains the most relevant landscape metrics, (2) it supports distribution statistics such as median, mean, range, standard deviation, etc., (3) it includes a complete user’s guide, (4) inputs/outputs are compatible with a wide range of GIS software including ArcGIS (ESRI), and (5) it is available online as freeware through the University of Massachusetts (Leitão et al. 2006). The outcomes of the 1992 Rio Earth Summit had a profound influence on the development of Landscape Ecology (Potschin and Haines-‐Young 2006). The development of new theory and practice for sustainable landscape planning are now important focal points for landscape ecologist (Vos and Meekes 1999; Tait and Morris 2000; Saunders and Briggs 2002; Antrop 2006; Blaschke 2006; Leitão et al. 2006; Potschin and Haines-‐Young 2006). Despite numerous planning, management, conservation, and restoration projects taken on by landscape ecologists, to date Landscape Ecology is having limited impact on sustainable land use decision making (Naveh 2007). Despite the current limited impacts, some think Landscape Ecology is the perfect science for addressing the issues of sustainability and SD (Wu 2006; Naveh 2007). A heuristic model of the interaction between Sustainability Science and Landscape Ecology can be seen in (Figure 2). Landscape
19
Ecology and Sustainability Science both focus on the dynamic interaction between nature and society (Clark and Dickson 2003; Reitan 2005). According to Reitan (2005) they both involve “the cultivation, integration, and application of knowledge about Earth systems gained especially from holistic and historical sciences (such as geology, ecology, climatology, oceanography) coordinated with knowledge about human interrelationships gained from the social sciences and humanities.” There now lies a call for action from the science of Landscape Ecology. If Landscape Ecology is to move forward as a discipline that is relevant to contemporary issues then it must be aware of the needs of SD and respond accordingly (Bastian 2001; Opdam et al. 2002; Wu and Hobbs 2002). Fig. 2. Hierarchical and pluralistic view of landscape ecology and its relationship to sustainability (from Wu 2006).
20
Planners and planning academics have been exploring the role of sustainability in planning theory and practice (Healey and Shaw 1994; Beatley
1995; Spain 1995; Campbell 1996; McDonald 1996; Haughton 1999; Jepson 2001; Berke 2002; Godschalk 2004; Jepson 2004) and proposing and testing tools and techniques to plan and build sustainable places (Innes and Booher 2000; Kakee 2002; Jepson 2003; Portney 2003). Despite a plethora of definitions and debates, and uncertainty about implementation techniques, the field of planning increasingly is acknowledging SD as an influential concept (Godschalk 2004; Jepson 2004). Planning needs to anticipate future conditions-‐ where we want to go and we can go (Hardi and Zdan 1997). Spatial planning is the core discipline that steers the development of our present and future living space through social, economic, and environmental structures (Keiner 2006). To be successful at planning, at any scale, appropriate methods, procedures, and instruments are required (Keiner 2006); furthermore, the proper choice of indicators is essential for monitoring progress towards sustainable spatial development (Presscott-‐Allen 1997; Bossel 1999). In terms of ‘sustainable development,’ the planning community sees a need for implementation, but struggles at putting the concepts into action. “Along with the questions ‘should we?’ or ‘can we implement sustainable development?’ more and more the question ‘how can we apply this concept?’ dominates the literature” (Chifos 2007). To date, there exist no ‘ideal’ planning instruments for achieving sustainability neither on the regional nor the local level (Keiner 2006). In final, policy makers are now encouraging scientists to improve models, and develop new
21
techniques, for the integration of quantitative and qualitative analysis for regional sustainable development planning (Grosskurth 2007). 2.4. Human behavior and sustainability The human-‐environmental relationship is constantly evolving. As science and technology continue to advance, we improve our abilities of reaching sustainability and sustainable development. However, on the verge of making sustainable development technically operational, we need to revisit if human behavior allows humanity to do so. Jepson (2003) states that, while communities are engaging in the enactment of policies and techniques that are consistent with sustainable development, few show evidence of successfully integrating all three needs (environment, economics, social) of sustainable development. If humanity is to avoid the “tragedy of the commons,” driven by traditional neoclassical economic decision-‐making, there will need to be public/communal regulation (Meadows et al. 1992). Revisiting the expansionist and ecological worldviews, it is not too difficult to understand why sustainability and sustainably development is not forthcoming. Each of us carries some parts of both the expansionist and ecological perspectives. Evidence for the dominant expansionist view can be seen all around us through consumption levels, economic benefits, individual happiness, and waste production (Barbier 1987; Daly 1989; Despotakis et al. 1992). On the other hand, we have seen an increase of sustainability “wants” since Rachel Carson’s (1962) Silent Spring and the formation of Greenpeace (1972).
22
To better understand the interaction of sustainability and human behavior there will need to be collaboration between different disciplines (e.g.,
environmental psychology, education, geography, architecture, social psychology) (Stagl 2007; Gifford 2007). It is imperative to understand how individuals think and act, if we are move closer to our dream of operationalizing sustainability (Clark and Dickson 2003; Gifford 2007). Marina Alberti et al. (2003) states that questions now should address: “how do humans interacting with their biophysical environment generate emergent collective behaviors (of humans, other species, and the systems themselves) in landscapes?” It is suggested that by understanding human cognition and behavioral geography, sustainable development may be learned (Golledge and Stimson 1997; Golledge 2002; Stagl 2007). Jickling (2000) believes that sustainability is a stepping-‐stone in the evolution of human thinking. Finally, by looking at how humans behave within the landscape, we may be able to incorporate those wants and needs into sustainable development planning and policy-‐making. 3. Problem statement Despite it being over a decade since Agenda 21 first called for SD indicators, there is no consensus regarding the best approach for the design and use of SD indicator models (Wilson et al. 2007). As part of the ongoing process of moving SD into practice many research organizations, governments, and international agencies have created numerous sets of indicators. In 1974, the Organization for Economic Cooperation and Development (OECD) pioneered a “Core Set of Indicators” for SD at the international level. Through time, other agencies (i.e., UNEP, UNDP, WRI, WB, EEA) created many national and global level indicators for measuring sustainability
23
and SD. Many of the indicators to date fall under the physical/environmental sphere of sustainability and still do not address the holistic goal of combining social, economic and environment into a single index. Rogers et al. (2008) explains this by saying: we are good at measuring physical and chemical phenomena, but once we try to rationally assess the biological, social, and cultural aspects of our world, we have serious problems. With that said, I believe that an applied geographic review of composite indices would better inventory our current situation on SD indicators. By evaluating where we are-‐ we can move closer to developing and improving composite indices that combine economic, social, and environmental phenomenon. Specifically, a geographic analysis could help explain scale limitations of indices, and other geographical phenomenon, that the OECD (2002), Mayer (2008), and Ness et al. (2007) suggested affect the utility of indicators for decision-‐making. Although some SD indicators can bridge across scale, most are still being implemented at the national and international scales (see WWF 2008; WWI 2008; World Bank 2008; UNEP 2008). It has been stated that sustainability indicators have been embraced by researchers and policy-‐makers at local, national, and international scale (UNCED 1992; UNCCD 1994; Bell and Morse 2004); however, they are seldom used because of a lack of accessibility, understanding, and usefulness by people who manage at the local landscape level (Innes and Booher 2000; Carruthers and Tinning 2003; Reed et al. 2008). There is a growing body of literature suggesting that a combination of local knowledge and scientific knowledge may empower local communities to monitor and manage their local sustainability needs (e.g., Folke et al. 2002; Thomas and Twyman 2004; Reed et al.
24
2008). Although local planning agencies have access to a plethora of local data, they typically lack those data and/or analytical skills specific to creating a composite indicator of SD. Assuming that local data for creating a composite index of sustainability can be obtained, a remaining issue remains to its usefulness. Chen at al. (2006) work showed the effectiveness of a composite indicator of SD for a planning purpose at the province level of China. Successful examples of SD indicators used at the local and regional planning level should provide the necessary framework for operationalizing SD at the landscape scale. After providing a localized example (landscape scale) of how an indicator of SD can be utilized, the issue of if SD can truly be obtained still remains. By providing a local method for scoring SD, a means to plan spatially and temporally for SD can be created. However, the issue remains that even if you provide a strategic plan for regional sustainability, will humanity behave in a sustainable fashion? Additionally, if you can design a sustainable landscape in which humans are to function, will they interact within that landscape sustainably? It has been clarified that we should revisit learning processes, understand it further, and try and incorporate it into sustainable management practices and planning institutions (Stagel 2007). In summary, this dissertation research, through a two article approach, provides: 1) a better understanding of SD indicators through a survey of composite SD indices, and the creation of two multimetric SD indices for local spatial application; 2) local (landscape scale) methods and models for operationalizing SD for regional planning purposes; 3) a better understanding on spatial models and geographic information science (GIScience) techniques for SD; 4) research on
25
landscape design and human behavior for investigating the feasibility of
sustainability; 5) ultimately, a work that links SD theory with an applied application. 4. Assessing sustainable development using additive household community and property composition indices 4.1. Abstract To date, there exist no ‘ideal’ instrument for achieving sustainability neither on the regional nor the local scale. This paper approaches this problem by developing an assessment system for analyzing and evaluating sustainable development (SD) for use at the local scales of urban planning and regional development. Because humans are the major driving factor of global change, and their behaviors are modified by the landscape in which they exist, a holistic approach is applied at the household scale using multi-‐metric household community and property composition indices and additive construction. The purpose of this paper is to express the need for local (e.g., point level or disaggregated scale) measures of sustainability that capture the “triple bottom line” while remaining understandable, analytically accurate, and operationally uncomplicated. Two composite indices (Ecological Demand Index (EDI) and Sustainability Demand Index (SDI)) are developed for SD spatial assessment in the Republic of Moldova using a local demographic and health survey. Total sample size and spatial distribution was 11,066 households associated to 399 geographic locations throughout the Republic of Moldova, respectively. To investigate the analytical robustness of EDI and SDI, central tendency distribution of index values and Pearson product-‐moment correlation analysis was applied. Preliminary investigation suggests that this type of methodology can provide the flexibility needed for producing straightforward
26
composite indicators of SD that can account for specific local conditions and priorities with a country. 4.2. Introduction
Despite its complexities, the adaption of sustainable development (SD) by the United Nations World Conference on the Environment and Development (UNCED) in Rio de Janeiro (1992) marked a new era in global awareness. SD, defined by the Brundtland Report, is the equitable use of Earth’s resources that meets humanities present needs without compromising the ability of future generations to meet their own needs (WCED 1987). Nearly all societies throughout the world have now committed themselves to SD by integrating some form of environmental quality, social equity, and economic welfare into their day-‐to-‐day activities. As such, there is strong political desire at all spatial scales for the comprehensive assessment of changes in environmental, social, and economic conditions for evaluating current status, measuring progress, and setting future SD goals. Numerous studies have shown that humanities current practices exceed the natural limits of the earth (WCED 1987; UNEP 2005; WWF 2008). Besides the environmental ramifications of human behaviors, major global challenges remain between social groups and the interaction between societies and nature (Kates et al. 2001; Clark and Dickson 2003). In example of this, close to a billion people live in extreme economic poverty (e.g., less than 1 US dollar a day), and lack access to essential natural resources to meet basic needs (Word Bank 2008). Unfortunately, due to its all-‐encompassing goals and theoretical vagueness, SD has been found to be very difficult to measure. Further, a paradox is found within the term SD; albeit,
27
to continue growth indefinitely on a planet with finite resources is impossible (Bartlett 2006). With over 300 working definitions for sustainability and SD
(Dobson 1996), and some definitions contradicting each other (Goodland and Daly 1996), some feel that achieving SD is more remote than ever (Jickling 2000). SD has even been thought of as being too subjective and ultimately unreasonable for humans to achieve (Kemp and Martens 2007). Despite its flaws, a sustainability concept is still a seemingly rational guide to create a long-‐term, positive relationship between humanity and the planet; however, murky and conflicting goals hamper our ability to determine whether this relationship has been or will be achieved (Mayer et al. 2004). The current challenges of sustainability development now lie in its operationalization (Keiner 2006). Efforts must be made for the implementation of initiatives that do not merely pay lip-‐service to the words but actively do justice to its original roots (e.g., sustainable yield) (Campbell 2001). As part of this process, numerous researchers and governmental organizations have developed a plethora of indicators of SD. Indicators and composite indicators are increasingly recognized as useful tools for policy making, because they convey information on a country’s performance towards their specific sustainability goals. In Chapter 40 of Agenda 21 the need for SD indicators was articulated: “indicators of sustainable development need to be developed to provide solid bases for decision making at all levels and to contribute to a self-‐regulatory sustainability of integrated environment and development systems” (UN 1992). The main feature of indicators is their ability to summarize complex information of our dynamic world into manageable amount of
28
meaningful information. Despite the fact it has been almost two decades since
Agenda 21 first called for SD indicators, there remains no consensus regarding the best approach to their design and use. Hundreds of different indicators and indices for measuring SD have been suggested (Tschirley 1997), and over 140 different indicators been proposed for the Organization for Economic Co-‐operation and Development (OECD) countries (Moffatt 2008). Unfortunately, many of these indicators have quantitative flaws in their construction or lack the ability for implementation at the operational scale of local governments. Many indices differ significantly in their results although they are created using the same input datasets, because of their assumptions, biases, and methodological disparities, creating confusion for sustainability efforts (Mayer 2008). Furthermore, current SD indicators have been found to lie heavy on environmental needs while skimping on social and economic needs (Molden et al. 2004; Moffatt 2008). Overall, the current status of sustainability indicators has been said to be “voluminous” and “not well focused” (Rogers et al. 2008). To date, there exist no ‘ideal’ planning instruments for achieving sustainability neither on the regional nor the local level (Keiner 2006). Recently, policy makers have started to encourage scientists to improve models, and develop new techniques for integration of quantitative and qualitative analysis for local and regional SD planning (Grosskurth 2007). To be successful at planning at any scale, appropriate methods, procedures, and instructions are required (Keiner 2006). Specifically, the proper choice of indicators is essential for monitoring progress towards sustainable spatial development (Bossel 1999; Presscott-‐Allen 2001);
29
however, understanding their strengths, weaknesses, scale-‐dependencies, data needs, etc. when employing them is even more important (Parris and Kates 2003; Morse and Fraser 2005; Ness et al. 2007).
This paper describes a procedure for monitoring SD using data collected at the household scale. My contention is that by carefully monitoring the household community and their corresponding property composition, one can rapidly assess the sustainability status for local geographic areas. In short, carefully planned monitoring and assessment can rapidly and relatively inexpensively serve as an exploratory assessment of local SD quality that could be used in a variety of planning objectives in line with a country’s specific local conditions and priorities. Where impaired locations of sustainability are suggested by household monitoring, a more detailed assessment program can be implemented in search of causative agent(s) for remediation. 4.3. Sustainable development assessment and monitoring Due to growing environmental awareness, by the early 1970s environmental measuring tools were starting to gain popularity. According to Thomas (1972) “A concerted effort to enhance habitability of our planet is unlikely to succeed unless we know where we are and where we want to go.” In response to measuring progress towards environmental goals and pollution control targets, the foundations of indicator development can be linked back to Herbert Inhaber’s (1976) Environmental Indices and Wayne Ott’s (1978) Environmental Indices: Theory and Practice. Although a lull in indicator research occurred through the 1980s, a
30
renaissance came after their applicability became apparent for measuring progress towards SD at the (UNCED) in Rio de Janeiro (1992) (Rogers et al. 1997). An indicator is a single value from a single measure of quantity; whereas an index is the combination or aggregation of more than one single indicator or single value (Ott 1978). The World Bank (2008) describes an indicator as “a performance measure that aggregates information into a usable form”. Formed by observed or estimated data, indicators are quantitative measures that only use generic definitions of quality making it very difficult to make accurate decisions at the margins (Rogers et al. 2008). Sustainability indicators are increasingly recognized as useful tools for policy making and public communication of otherwise complex and complicated information (Singh et al. 2009). Although there remains debate on its theoretical design, we now see sustainability moving from an abstract concept to a measurable state of dynamic human-‐environmental systems (Mayer 2008). For practical purposes, the variety of SD indicators poses a huge problem for policy makers and scientists alike. Although there are a number of initiatives for compiling and understanding SD measures, there remains no subset of key variables for measuring sustainability at neither global nor local scales (Keiner 2006). At the global scale, studies have demonstrated that sustainability indicators do not rank countries consistently (Mayer 2008), and there is a lack of clear direction in how to best approach measuring SD (Wilson 2007). At the local scale there remain even fewer active SD indicator initiatives. A main reason for the lack of local SD measures is due to the established, or accepted, sustainability indicator data requirements; albeit, many of the popular SD indicators can only be calculated at the country scale
31
and are not transformable to local scales. Until there is clear analytical understanding of how data are collected, aggregated, classified, and weighted, SD
indicators risk being ineffective or even counterproductive. It is an underlying goal of the author, through this paper, not to create more indicators but to develop a flexible and transferable method for measuring SD at the local scale that can fit in-‐ line with a country’s specific local conditions, needs, and priorities. 4.3.1. Survey of common sustainability indices Since Agenda 21, a plethora of different measures of development have emerged to calculate the sustainability of locations. Of that suite, ten common indices widely used for assessing SD are briefly summarized in this section. 4.3.1.1. Gross domestic product (GDP) Gross Domestic Product (GDP) is one of the most well known indicators for measuring sustainability at the global scale (Shaker 2010a). GDP represents a metric where economic growth is considered the ultimate driver of SD (Beckermann 1992; CEC 2001; OECD 2001). The neo-‐liberal economic stance implies that strong economic growth is the best development strategy to improve environmental health (Economist 2000). GDP is calculated from the addition of private consumption, gross investment, governmental spending, and exports minus imports. Due to the scale of input data aggregation and how imports and exports are measured, GDP is seldom operationalized at scales other than the national level. 4.3.1.2. Human development index (HDI) The Human Development Index (HDI) is currently the most popular scientific based indicator for measuring SD (Shaker 2010a). HDI, the product of a noble prize
32
associated with SD, has been reported annually as part of the Human Development Report of the United Nations Development Programme (UNDP 2005). HDI consists of three equally weighted sub-‐indices (e.g., life expectancy index, education index, and a GNP index) that are aggregated by an arithmetic mean (Böhringer and Jochem 2007). Because HDI used a surrogate of GDP in its calculation, this index is seldom operationalized at scales other than the national level. 4.3.1.3. Wellbeing (WB)
The Wellbeing Assessment (WB), developed by Prescott-‐Allen (2001), is
based on the assumption that a healthy environment is necessary for healthy humans. The WB index is the arithmetic mean of the Human Well-‐Being Index (HWI) and an Ecosystem Wellbeing (EWI). The HWI is based on 36 indicators and the EWI is based on 51 indicators (Böhringer and Jochem 2007). The aggregation of the indicators from the HWI and EWI is conducted by a weighted arithmetic mean Presscott-‐Allen (2001), although the derivation of the weights is not explained in detail (Böhringer and Jochem 2007). With 87 sub-‐metrics employed, WB could be at risk of having multiple metrics statistically redundant of each other. Albeit, this many data requirements make WB difficult to operationalize at any scale for SD assessment. 4.3.1.4. Ecological footprint (EF) The Ecological Footprint (EF) was originally conceived as a simple and elegant method for comparing the sustainability of resource use among different populations (Rees 1992). The most basic rendition of EF is based on the quantification of land and water required to keep a national living standard into
33
infinity thereby assuming certain technological improvements (Wackernagel and
Rees 1997). As humans continue to metabolize the available biomass on the planet, less is available for use in providing for other goods and services (Harberl et al. 2004). EF captures the total amount of energy used or consumed by a system, and presents it as the area (in hectares) required for photosynthesizing organisms to fix that amount of solar energy, and to absorb the amount of waste produced by the system (Rees 2002). EF calculates total consumption by summing indicators for imports and domestic production, then subtracting exports (Mayer 2008). These indicators measure the number of hectares used in six major categories: croplands; grazing land; forestland for wood and non-‐wood products; marine fisheries; housing and infrastructure; and forestland necessary for absorbing domestic CO2 emissions (Wackernagel et al 2002). With over 60 input data requirements, although it can be presumed that this has been accounted for, EF could be at risk of having metric redundancy. Albeit, this many data requirements make EF difficult to operationalize at any scale for SD assessment. 4.3.1.5. Environmental sustainability index (ESI) The Environmental Sustainability Index (ESI), a 2005 pilot project, quantifies the likelihood that a country will be able to preserve valuable environmental resources effectively over the period of several decades (Esty et al. 2005). The ESI focuses on the environmental dimension of sustainability, and when calculated in 2005 it covered five major components through the use of 21 indicators; the 21 indicators are derived from 76 input variables (Böhringer and Jochem 2007). To date, ESI has only assessed SD at the national scale. Due to the scale of contributing
34
data aggregation, it would be hard to operationalize ESI at scales other than the national level. 4.3.1.6. Environmental performance index (EPI) Complementing the ESI, the Esty Research Group developed another pilot
project with the creation of the Environmental Performance Index (EPI). According to Esty et al. (2006) “The EPI is designed to address the need for gauging policy performance in reducing environmental stresses on human health and promoting ecosystem vitality and sound natural resource management. The EPI focuses on current on-‐the-‐ground outcomes across a core set of environmental issues tracked through six policy categories for which all governments are being held accountable.” The EPI is based on a proximity-‐to-‐target approach which measures country performance against an absolute target established by international agreements, national principles, or scientific agreement (Esty et al. 2006). The EPI scores range from zero to 100, with 100 being linked to the target and the minimum value of zero characterizing the worst competitor in the field (Böhringer and Jochem 2007). To date, EPI has only assessed SD at the national scale. Due to the scale of input data aggregation, it would be hard to operationalize EPI at scales other than the national level. 4.3.1.7. Living planet index (LPI) The World Wildlife Fund (WWF) (1998) developed the Living Planet Index (LPI) to measure the changing state of the world’s global biodiversity over time (Loh et al. 2005). LPI was first designed to measure trends in over 2000 populations of more than 1100 species of vertebrates in terrestrial, freshwater and seawater
35
ecosystems. The dataset now contains roughly 3000 population time series for over 1000 species (Loh et al. 2005). The LPI provides a sub-‐index for each of the three aforementioned ecospheres: for every species within an ecosphere, the ratio between its populations in pairs of consecutive years is calculated. The initial aim was to make the LPI as holistic and representative as possible with respect to vertebrate class, geography, and biome (Loh et al. 2005). The geometric mean of these measurements of different species multiplied with the index value of the former year then delivers the biodiversity index for the respected sphere with 1970 serving as the base year (Böhringer and Jochem 2007). The geometric mean of these indices is the LPI. Due to the scale of input data aggregation, and its design, it would be hard to operationalize LPI at scales other than large geographic areas (e.g., oceans, regions, biomes). 4.3.1.8. Green net national product (EDP) and SEEA The Green Net National Product, or more commonly known as the Environmentally Adjusted Net Domestic Product (EDP) has been developed within the scope of the System of Integrated Environmental and Economic Accounting (SEEA)-‐ UNEP 2000 and UN et al. 2003). Following inter alia Hanley (2000) three different versions of the EDP can be distinguished: (1) the EDPI which deducts the depreciations of natural resources caused by their extraction from the net national income (NNI); (2) the EDPII, which deducts from the NNI the costs necessary to reach the same state of the environment at the end of the period as existed at the beginning of the period; and (3) the EDPIII, which deducts the costs of environmental pressure and destruction (calculated by willingness-‐to-‐pay
36
methods). In these methods, aggregation takes place by simply adding up the monetarized vales (Singh et al. 2009). Although the number of countries that are implementing SEEA to calculate EDP (Böhringer and Jochem 2007), this SD index requires many input variables and would remain hard to operationalize at scales other than the national level. 4.3.1.9. City development index (CDI) The City Development Index (CDI) suggested by the United Nations Centre
for Human Settlements (HABITAT) consists of five sub-‐indices: (1) an infrastructure index, which builds on four (equally weighted) indicators as percentages of households which are connected to clean water, canalization, electricity and a phone network (without mobiles); (2) a twofold (equally weighted) waste index, which is composed of the percentage of untreated sewage in total wastewater and the percentage of disposal of solid waste in total solid waste; (3) a twofold diversely weighted health index, which considers the life expectancy and the infant mortality rate; (4) a twofold (equally weighted) education index which is calculated by adding the percentages of literacy and combined enrolment; and (5) a city product index, which is based on the logarithmic value of the city’s GDP (Böhringer and Jochem 2007). Although the CDI has been applied to cities, regions, and country scales, for SD assessment (Böhringer and Jochem 2007) the input data requirements needed, and its computational complexities, make it difficult to operationalize. 4.3.1.10. Environmental vulnerability index (EVI) The Environmental Vulnerability Index (EVI) comprises of 32 indicators of hazards, 8 indicators of resistance, and 10 indicators tat measure damage (SOPAC
37
2005). The EVI scale for normalization ranges between a value of 1(indicating high resilience/low vulnerability) and 7 (indicating low resilience/high vulnerability) (Singh et al. 2009). The 50 indicators in total are given equal weights and then aggregated by an arithmetic mean (EVI 2005). Although the EVI has been applied to 235 countries (SOPAC 2005), this index would be hard to operationalize at scales other than the national level due to large data requirements. 4.4. Why evaluate sustainability at the household scale? Household communities reflect SD condition since they are sensitive to changes in a wide array of environmental, social, and economic factors. Many phenomenon related to sustainability have been used or proposed as indicators of SD, but no single aspect of the human-‐environmental system has emerged as the favorite for policy-‐makers or sustainability scientists. Indeed, in the best circumstances, a SD monitoring program should be based on an integrated approach involving the use of several well-‐documented SD indicators at multiple spatial and temporal scales. However, limited funds and time for assessment, and planetary conditions argue for a more aggressive approach. Global scale analysis contrasting and comparing different countries progress towards SD has most frequently been found throughout the sustainability literature, implying that this scale is ideal for SD monitoring programs. Although most input data employed at the national scale directly or indirectly related to humans at a more local level, households are rarely used in comprehensive sustainability monitoring. Many efforts to use households in monitoring programs have been directed with the “bottom-‐up” approach to addressing the needs of sustainability.
38
Although these programs (e.g., environmental education) are highly valued and
needed, they lack the quantitative scoring mechanism needed to reach benchmarks or specific goals. Other household scale initiatives, such as carbon cycling analysis, are quantifiable, but remain complex and skimping on the social and economic needs of sustainability. Comparative studies of countries at the regional and global scales have been widely used in monitoring SD because of the availability of their theoretical and quantitative substructure that allows for a holistic integration of environmental, social, and economic measures. However, the national scale has major deficiencies for operationalizing SD. For example, at this spatial scale the measures typically require mass quantities of input data from multiple data sources; they entail specialized quantitative expertise; they are difficult and time-‐consuming to sample, sort, and calculate; background metadata is often lacking for sub-‐metrics recreation or deduction; the results are not transferable for local SD initiatives; and the findings are difficult to translate into values meaningful to the general public.
Households, on the other hand, have numerous advantages as the scale for
SD monitoring programs. These advantages include: 1. Many countries throughout the world have household data collection methods (e.g., U.S. Census Bureau) in place for local policy and inventory needs. 2. Household communities generally include a range of inhabitants that represent a variety of societal levels (e.g., children, adults, elderly); all of which require a different amount of resources that affect progress towards SD. 3. Humans, with their position at the top of the trophic system, make household evaluation an integrated view of their direct human-‐ environmental interaction.
39
4. Household related information is easy to identify. Technicians require relatively little training, and most samples question can be evaluated in the field without compromising the integrity of the households under study.
5. Environmental, social, and economic effects can be equally be evaluated through household community, their associated property information, and relative location. 6. The general public can easily relate to statements about conditions of household communities. 7. Households are typically present, even in the most remote and inhospitable places in the world.
8. Finally, the results of studies using households can be directly related to the Millennium Development Goals and Agenda 21 local action plan put forth by the United Nations. A number of disadvantages of monitoring households can also be cited. These include the selective nature of sampling, human mobility, objectivity of respondents, infrastructure cost, and human power needed for field sampling. But these disadvantages are associated with all human based surveys and methods have been developed to account for their shortcomings. The objective here is not to imply that humans are easy to sample or identify their corresponding information. Rather, the goal is to emphasize that, on a comparative basis, training periods for household composition and property surveys are likely to be shorter and the technology required is less sophisticated than other scales for assessing and monitoring SD. Obviously, all monitoring programs are expensive and time consuming, but it is hoped that the methodology presented here can even be flexible enough for implementation using local household surveys already in existence. My purpose here is to suggest that regular use of household surveys will improve the status of local SD monitoring and assessment programs.
40
4.5. The assessment system
Over the last several years, I have been working in Eastern Europe with the objective of developing a SD monitoring system that uses humans as the response variable for addressing local and regional SD needs. The purpose of this paper is to outline that system as well as provide a few samples of its use. At this point, I urge caution in transferring this method in wholesale fashion without further testing and addressing local needs. With that said, I hope that this manuscript will stimulate other sustainability scientists to react to the local call for action in similar fashion, perhaps even trying to adapt the method presented here in other geographic areas, thereby aiding in its improvement. I initially set out to develop a system with discrete classes (e.g., excellent, good, fair, poor) to evaluate a household communities demand for sustainability. However, as I worked with the assessment system for a time, I found it necessary to do away with discrete classes and rank the measures of sustainability from low to high using an interval level of measurement. Some of the terms used in the forthcoming descriptions are qualitative at best. Effective implementation of local sustainable development monitoring programs is in it infancy; albeit, there remains much in the lines of establishing standardized quantitative methods for local SD assessment initiatives. 4.6. The assessment criteria Prior efforts to monitor and evaluate the “triple bottom line” of sustainability typically involved use of many different criteria, often combined into an index. Although SD is a globally accepted concept, and many indexes have been developed
41
(e.g., Human Development Index, Ecological Footprint) to measure progress towards sustainability, there still exist issues for operationalization, such as development of tools and the appropriate geographic scale for helping to assess, achieve, and monitor sustainability (Chan and Haung 2004). As sustainability interests move away from the global scale towards empirical policy and planning initiatives at the local scales, it is necessary to understand the complexity of
interactions in human and natural systems (Nijkamp and Vreeker 2000). Because the principles of SD call for an integration of information related to environmental quality, social equity, and economic welfare in decision making (Kelly 1998), this research assesses sustainability through a systems approach of additive household community and property composition parameters. The parameters have been chosen based on their local relevance to environmental quality, social equity, and economic welfare, and can be grouped broadly into two sets: (1) Household Community Information and (2) Property Composition Factors. The choice of household community information and property composition factors as primary criteria is crucial for measuring the degree of local sustainability. Specific choice in metrics needs to be tailored to those measures that are significant to local sustainability goals and incorporate geographical relevance into their methodology. For example, the travel cost of a domestic product found within a country like Romania is not what you would find in the United States; albeit, many of the domestic goods found in the United States must travel farther than other countries because population density is geographically sparse. Because most national datasets track within their corresponding political borders, a degree of
42
environmental, social, and economic geographical relevance is maintained regionally. As a country increases in geographic complexity (e.g., area, elevation,
location) it may be important to assign scores to metrics based on a more similar or localized geographic region within a country. With that said, location of individual households is necessary for accounting for the human-‐environmental relationship needed for properly evaluating SD. 4.6.1. Household community information The household community (e.g., family members living in a home) provides pertinent information for measuring local progress towards sustainability. Although this research does not attempt to control the set of household community indicators for index creation, number of household members, income, education, and public health can serve as goal categories pertinent to SD and related to indicators commonly found in local datasets. Number of household members, or household richness, is related to all three aspects of sustainability (environmental quality, social equity, and economic welfare). In general, as the number of household members increases the amount of resources consumed also increases; however, the efficiency of more people sharing resources ultimately decreases ecological demand per person. In example, if you have two equal sized houses in the same location, one with five people and the other with three people, the house with five people splits heating costs five ways apposed to three making it more efficient. Social equity increases as the number of household members increases. In example, the more people providing food for the household increases the likelihood of overall enhanced household fitness through
43
burden sharing. Number of members within a household increases overall
economic welfare due to cost sharing. Although there is some complexity here with number of household members working or not working, in general the more people within a household will provide more working aged members for burden sharing. In combination with number of household members, a measure of income for neoclassical economically driven societies is potentially the most important category in measuring SD. In regards to environmental quality, income allows the means for the further consumption of resources; the greater the amount of income a household community has, the greater the amount of resources that that household community will consume through purchase. Because greater income provides more purchasing power, a greater level of household needs can be met with higher levels of income. With greater amounts of household income comes a greater amount of local economic ability and stimulation. Poverty and material deprivation are quite often accompanied by an incapability of full participation in social life, due to inadequate access to employment, education, health resources, etc. (Rodríguez-‐Pose and Fratesi 2004). Education at the household community scale is highly related to measuring SD. Environmentally, a higher degree of education would provide a higher degree of environmental education and thus an overall more logical use of natural resources. With a higher degree of education we would also find a greater amount of social equity, and economic welfare, respectively. Ultimately, education is the fundamental building block in which social and economic opportunities are built.
44
Public health, as a category goal for measuring local sustainability needs, can provide a strong avenue for measuring the often-‐overlooked social components of SD. Public health indicators can give a great indication of the local status of the environment. In example, if there is higher rates of “blue-‐baby” syndrome being recorded in an area, it would be environmental crucial to investigate nitrate levels in groundwater. Likewise, social equity is improved when lower levels of public health related illnesses are recorded. Economically, lower levels of public health related illnesses would decrease the amount of medical infrastructure and maintenance needed. 4.6.2. Property composition factors Along with household community information, property composition factors (e.g., household building material) provide pertinent information for measuring local progress towards sustainability. Although this research does not attempt to control the set of property composition indicators for index creation, household shelter composition, amount of land occupied, transportation related property (e.g., automobile), and major appliances and other significant household material goods can serve as goal categories pertinent to SD and related to indicators commonly found in local datasets. Household shelter composition (e.g., wall materials, number of rooms) is related to all three aspects of sustainability (environmental quality, social equity, and economic welfare). In general, the stronger the materials used to build a household dwelling the better it will be in terms of meeting the needs of sustainability. In example, a condominium complex made out of steel and cement
45
would be more sustainable overall than a freestanding house made out of adobe and wood. Although there could be some debate about this indicator, an environmental quality example is that the condominium allows for households to share walls making heating and cooling of the internal space of the individual household shelter more energy efficient than a freestanding house. Socially, the stronger the household structure is, as in the case of the condominium complex, the more unfailing the shelter is for long-‐term inhabitation. Economically, although the initial cost of the condominium complex is significantly more expensive, it is a better long-‐ term economic investment than the freestanding house made from adobe and wood. Amount of land occupied by a household community is relevant to measuring SD. Specifically to the “triple bottom line” of SD, amount of land occupied is most directly important to measuring the environmental and economic needs of sustainability. Environmentally, amount of land occupied by a household has a direct influence on waste assimilation and food production. In both cases, the more land available for a household community to utilize the more positive influence on SD. Amount of land owned in neoclassical economic driven societies is applicable to financial equity, which is directly related to economic welfare; albeit, the more land a household community owns the more economic and social stability they also have. Local production base from local land resources are linked with global economies and the consequence for local goods production is a decrease in imports and increase in exports creating multiplier effects in the local economy (Wong 2002). Transportation interconnected property is related to all three aspects of sustainability (environmental quality, social equity, and economic welfare). Within
46
this category goal of SD, transportation property (e.g., number of automobiles
owned) allows for quantifying household community mobility and the external cost associated with it. Environmentally, the more transportation property owned by a household community, such as a car or truck, increases demand on ecosystem services negatively affecting SD. In example, if a household community owns multiple automobiles the more likely they will drive themselves and not select mass transit for their mobility needs increasing their use on nonrenewable resources. Socially, transportation related property has a positive influence on SD. Household communities that own transportation property provide improved means for and flexibility with mobility, thus improving their social equity component of SD. Economically, in most neoclassical economic societies, transportation related property is also applicable to financial equity, which is directly related to economic welfare. Major appliances and other significant household material goods (e.g., computer, television) can provide further indication of all three aspects of sustainability (environmental quality, social equity, and economic welfare). Environmentally, major appliances and other significant household material goods put a considerable demand on nonrenewable resources thus decreasing objectives related to SD. Socially and economically, major appliances and other significant household material goods provide positive indication of SD. In example, computer technology provides a means for self-‐education, communication, information storage and analysis, thus improving the household communities social equity and economic welfare components of SD.
47
4.6.3. The geographic factor
Because landscapes have been viewed and taught as if their sole purpose is for building materials, instead of living or life-‐supporting agents, landscape resources have been progressively degraded by the actions of humans. Success in halting and reversing this degradation requires understanding of social and human-‐ environmental cohesion through new approaches to evaluating and analyzing SD at local scales. The first law of geography states that things that are near are more similar (spatially autocorrelated) than things that are farther apart (Tobler 1970; Fortin and Dale 2005). When investigating local SD conditions it is essential to understand that many different processes influence natural and social systems over space. Although progress has been made to incorporate space into surveying techniques much remains (e.g, disaggregating data) for improving SD spatial analysis. Aggregated data to political boundaries may be irrelevant or misleading for sustainability research (Mayer 2008). The flow of people, commodities, and pollution across national boundaries can be substantial (Panayotou 2000), and indices calculated at the national scale can underestimate the “leakage” phenomenon. (Mayer 2008). It should be noted, political boundaries do affect sustainability to the degree to which domestic laws affect movement of people, goods, services, etc. Geographic location of household community residence is highly interrelated to all three aspects of sustainability (environmental quality, social equity, and economic welfare). In all three aspects of SD, the closer a household is to a significant resource centers and large population (e.g., cities) the greater the
48
improvement of SD. Environmentally, a decrease in distance to a resource center would decrease direct and indirect travel cost associated with dependency on nonrenewable fossil fuels. Socially, a decrease in distance to a resource center
would decrease risk associated with travel cost. In example, in a medical emergency chance of survival drastically improves with a decrease in wait time for emergency response personnel. Economically, a decrease in distance to a resource center would decrease added cost associated with receiving production materials and delivering goods to markets and consumers. 4.7. The multimetric assessment process A variety of qualitative and quantitative indices have been used in measuring progress towards sustainable development, but it has been multimetric (composite) indicators that have been most widely applied and shows the most promise for incorporating all three standard dimensions of sustainability into a single decomposable indicator of local SD. Although there are many different analytical techniques available (see Böhringer and Jochem 2007; Mayer 2008; Singh et al. 2009), an additive, or multimetric, approach was developed here specifically due to its ease of assessment and management of sustainability quality. Operationally defining the indices employed by data availability, components of sustainability were evaluated based on their potential stress against an ideal reference condition found within the dataset. Because this method is designed to be flexible and transferable to different datasets, no ideal subset of metrics has been chosen at this time. However, the combination of metrics chosen is designed to reflect environmental quality, social equity, and economic welfare insights from
49
individual, population, household community, property composition, and geographic perspectives.
If we are serious in developing a single composite index as a measure of SD, then it is clear that environmental, social, and economic indicators have to be combined (Moffatt 2008). A composite index is an aggregation of indicators that have no common unit of measurement and no obvious way to assign weights to them. Every composite index should be considered a model, created for a specific purpose, with its construction following a series of useful and generally accepted steps (Kondyli 2010). These steps include the following (Jocobs et al. 2004; OECD 2005; Kondyli 2010): 1.
2.
Formulation of a theoretical framework: provide a sound basis for selecting and combining single indicators into a meaningful SD composite index. The theoretical framework must accurately define the SD phenomena targeted to be measured. Data selection: the SD indicators used should be selected based on their analytical soundness, measurability, and their relationship to each other. Use of proxy variables should be considered when data are scarce.
3.
Multivariate analysis: this encompasses a wide variety of methods, which can be distinguished into two main categories: exploratory and confirmatory analyses. In exploratory analyses, the overall structure of the indicators is examined. In confirmatory analyses, the purpose is not to describe but rather to examine specific assumptions based upon already developed theoretical frameworks.
4.
Accounting for missing data: three methods are available for cases with missing data: a) omission of cases with missing data, b) replacement with mean, median, regression, or other single imputation, c) advanced multiple imputation algorithm (e.g., Monte Carlo method).
5.
Data normalization: indicators should be normalized for comparability. A variety of techniques for normalization are available related to index creation (see OECD 2005).
50
6.
Weighting: weighting indicators greatly influence the overall output of the composite indicator. When constructing the SD composite index, indicators should be weighted either according to an underlying theoretical framework or based on empirical analysis from expert knowledge and/or stakeholder input. In general, there are three ways to assign weights to indicators within an index: to use statistical models, b) to adopt participatory methods, or c) to assign equal weights.
7.
Aggregation: the aggregation of the SD indicators can be linear, geometric, or based on multi-‐criteria analysis. In both linear and geometric aggregation, weights express trade-‐offs between indicators; multi-‐criteria analysis assures non-‐compensability by finding a compromise between two more legitimate goals.
8.
Robustness analysis: The results rendered from the SD composite index are due to the culmination of the methods (e.g., normalization, weighting) used in its creation. Therefore, it is important to determine whether the values of each SD composite index are affected by the uncertainty that might characterize the data and/or weighting scheme.
9.
Analysis of SD composite index structure: Because a SD composite indices are summary indicators, a decomposition to their individual parts may result in a better understanding of performance and therefore helping to address a locations specific SD conditions and goals.
10. Presentation and dissemination of results: A composite SD index must be able to provide accurate information to any interested party (e.g., planners, policy makers) in the sustainability process. A variety of techniques (e.g., maps, interactive websites) have developed for the dissemination and presentation of the SD composite index results.
4.8. Examples from the Republic of Moldova In this section, based on my personal expert knowledge, ‘systems thinking’ is applied to evaluate the complex concept of SD in the Republic of Moldova. 4.8.1. Study area and historical context The Republic of Moldova is an ideal location for conduction an operationalized SD study. The geographic area once known as Moldova was significantly larger than the area comprising the present day Republic of Moldova. The old principality of “Moldova” reached from the Carpathian Mountains and
51
Bocovina in northern Romania, south to Dobrogea, Romania, and included an area between the Dniester/Nistru and Danube rivers that is now part of Ukraine. Most of the area known as Bessarabia of the greater “Moldova”, the Republic of Moldova is now landlocked between the Ukraine and Romania. Specifically, the Republic of Moldova is located between the Prut and Dniester/Nistru Rivers to the west and east, respectively, and Ukraine to the north and south. The Bessarabia region of “Moldova” was first annexed to Russia in 1812. Following the Crimean War in 1856, Russia lost the southern area of Bessarabia to “Moldova,” only to gain it back from Romania in 1878 at the Congress of Berlin. With the collapse of the Russian Empire in 1918 the area of the present Republic of Moldova and some of the present-‐day Ukraine declared its independence and united with Romania. However, the formed Union of Soviet Socialist Republics (USSR) refused to recognize this unification. In 1924, the USSR created the “Moldovan Autonomous Soviet Socialist Republic (MASSR),” east of the Dniester/Nistru River in present day Transnistria. In June 1940, according to the Molotov-‐Ribbentrop pact, the remainder of the present day “Moldova” was annexed by the USSR to form the “Moldovan Soviet Socialist Republic (MSSR)”. During World War II, the MSSR area fell back to Romania and then was annexed for the final time to the Soviet Union. After the 1944 annexation of Moldova to the USSR, the progression of creating a uniformed, patriotic, and Soviet-‐cultured Empire began. This process officially introduced the Cyrillic alphabet and Russian language into everyday life-‐ replacing the traditional Latin alphabet and Romanian language. During this time, cultural re-‐education (e.g., removal of religion and Romanian holidays) was set into
52
action; albeit, manipulating the aboriginal people of “Moldova” to change their
native ways. Severe economic recession, transplanting of native Russians, and slow nation building shortly followed. Soviet rule promoted urbanization and industrialization, even though “Moldova” was traditionally agricultural. Soviet governmental methods soon changed with the introduction of “glasnost” (openness) by Soviet President Mikhail Gorbachev. This new policy set the stage for Moldova’s future independence. On 27 August 1991, after a failed coup in Moscow resulted in the ousting of Gorbachev from power and the political collapse of the USSR, the Moldovan Parliament and the Republic’s General Assembly declared independence. After declaring its independence, the Republic of Moldova was marked by conflict and turmoil. Early interests in reunifying greater Romania and the presence of a large ex-‐Soviet (Russian) military force provoked a short civil war in 1992-‐ 1993. This civil war led to the separation of Transnistria from the rest of the Republic. This area, east of the Dniester/Nistru River, still has sensitive ethnicity and language issues. The demographic breakdown of Transnistria is about 65 percent Romanian, 14 percent Ukrainians, 13 percent Russian, and the rest Gagauz, Jewish, and Bulgarian. Tiraspol is the capital of Transnistria, which is not recognized by the U.S. government. Significant social (e.g., human trafficking), economic (e.g., weapons manufacturing), and environmental (e.g., ground water nitrate poisoning) issues have been linked to Transnistria and the Republic of Moldova. Much of the issues in the Republic of Moldova are linked to lack of jobs and money. Moldova is the poorest country in Europe and has suffered a nearly 65 percent decline in income
53
since its independence. Since its independence and prior to 2000, the Republic of
Moldova recorded only one year of positive GDP growth. Sporadic and ineffective law enforcement, combined with economic and political uncertainty, and outstanding disputes with international investors, continues to discourage direct foreign investment. To aid in local SD goals, developed counties have tried to improve SD conditions in the Republic of Moldova; in 2005 the U.S. provided roughly $27 million in assistance to the Republic of Moldova to alleviate issues associated to post-‐land-‐privatization, promote democratic institutions and civil societies, improve law enforcement and boarder control, and combat human trafficking. To date, many of these issues remain unresolved in the Republic of Moldova and unacceptable globally. Currently, the Republic of Moldova occupies 33,846 Km2 of land with 4,317,483 people (2010 census); albeit, 42% of which live in an urban environment. 4.8.2. Official demographic and health survey data Data provided from the Moldova Demographic and Health Survey (2005) were used to describe the three aspects of sustainability (environmental quality, social equity, and economic welfare) through SD multimetric index development at the household scale. Information in this subsection comes from the Moldova Demographic and Health Survey (MDHS 2005). Over a two-‐month period from June 13th to August 18th, Moldova’s first nationally representative demographic and health sample survey of over 11,000 households was conducted to monitor the population and health situation in Moldova. The Moldova Demographic and Health Survey (MDHS) sample survey of over 11,000 households were selected from 400
54
sample points (clusters) throughout the Republic of Moldova (excluding the Transnistria region). The survey includes detailed information on fertility levels,
marriage, sexual activity, nutritional status of women and young children, childhood mortality, and household property and composition. Additional features of the MDHS (2005) include the collection of information on international emigration, language preference for reading, and domestic violence. The MDHS (2005) is based on a representative probability sample of over 11,000 households. This sample was designed to allow separate urban and rural estimates for key population and health indicators (e.g., fertility, contraceptive prevalence, infant mortality). Transnistria, the semiautonomous region in the eastern part of the country holding 15 percent of Moldova’s population was not included in the sample. The MDHS (2005) utilized a two-‐stage sample design. The first sate involved selecting a sample of cluster sectors pertinent to the 2004 Moldova Population and Housing Census. A total of 400 clusters in Moldova were selected from the master sampling framework. Clusters for urban and rural domains (233 urban and 167 rural) were selected using systematic sampling with probabilities proportional to the 2004 census distribution, and consequently neither is the final household distribution. A complete household listing operation was carried out from April to late May 2005 in all 400 clusters in order to provide a sampling framework for the second stage selection of households. The second stage selection involved the systematic selection of households from a complete listing of all households in each of the 400 clusters. The sample “take” in both urban and rural clusters was 30 households, which would make a total sample size of 12,000
55
households throughout the Republic of Moldova. Due to a geographic coordinate error with one cluster location and a number of household surveys missing data, 399 cluster locations corresponding with 11,066 households were utilized in this study (Figure 3). The Household Questionnaire was used to list all the usual members and visitors in the selected households and to identify women and men who were eligible for the individual interview. Basic information was collected on the characteristics of each person listed, including their age, sex, education, and relationship to the head of the household. In addition, a separate listing and basic information on former household members who had emigrated abroad was
collected. The Household Questionnaire was also designed to collect information on characteristics of the household’s dwelling unit, such as the source of water, type of toilet facilities, materials used for the floor and roof of the house, ownership of various durable goods, etc (MDHS 2005). A copy of the detailed household questionnaire used in the Moldova Demographic and Health Survey (MDHS 2005) can be found in Appendix A.
56
Fig. 3. Map of the demographic and health survey geographic distribution within the Republic of Moldova (47°24’N, 28°22’E).
57
4.8.3. Method Prior to index development, this approach requires: (1) identification and characterization of local SD conditions and priorities; (2) description of the
reference household community for each regional application; and (3) selection of appropriate SD attributes of the household community and property composition metrics that will be used to quantify the difference between observed and reference households. Using the aforementioned 399 cluster locations and 11,066 households from the Moldova Demographic and Health Survey (2005), two composite indicators (Ecological Demand Index, Sustainability Demand Index) for operationalizing SD were developed and computed. Seeking a more holistic, integrative and ecological approach, 10 metrics and 15 metrics were selected for creation of the Ecological Demand Index (EDI) and Sustainability Demand Index (SDI), respectively. The composite SD index EDI has an environmental centric designed, while the composite index SDI is designed to take into account all three aspects of sustainability (environmental quality, social equity, and economic welfare). By doing so, empirical questions related to SD index design and response to goals can be compared and contrasted.
Procedures for operationalizing SD assessment at the local scale should be
kept simple and understandable, and procedural and analytic complexity should be kept to a minimum. The EDI and SDI are designed to include a range of attributes of household community and property composition. The EDI’s 10 measures, or metrics, fall into six broad categories: Household Abundance, Structural Composition, Goods Production, Transportation, Services, and Income (Table 1).
58
The SDI’s 15 measures, or metrics, fall into ten broad categories: Household Abundance, Structural Composition, Goods Production, Transportation, Services, Income, Technology, Public Health, Psychology, and Education (Table 2). Remaining flexible in both broad category and specific indicators, EDI and SDI metrics were selected or developed from the available information found in the
MDHS (2005). Direction of metric response can vary due to overall SD index goals. For both EDI and SDI, the index is the sum of several standardized component metric scores; furthermore, each metric score is on an ordinal scale of 1 to 5 based on the strength of deviation from an excellent scenario at a reference site. Because the sample size of households from the MDHS is 11,066, it is assumed that at least one household meets the assumption of an excellent SD scenario at a reference location. Although specific degrees of threshold are not directly assessed through this methodology, as the overall EDI and SDI values increase there is a decrease in degree that SD is being achieved. The detailed framework and data for each metric used in EDI and SDI can be found forthcoming in Table 5 and Table 9 of Article Two, respectively. Table 1. Republic of Moldova household ecological demand index (EDI) metrics and relative scoring criteria. ,+)*'#$#$%&'($)*+,-%.*/.0' 893%09%3/)+:$2;$&;'+$4+3'&
