visualization of sustainability indicators: a conceptual...

Visualization of Sustainability Indicators: A Conceptual Framework

Ray Quay Khanin Hutanuwatr

Arizona State University

January 9th 2007 Abstract Though there are numerous models for defining and organizing Sustainability Indicators, some published work defining methods of weighting and aggregation, and some published work on developing visualization methods and tools there has been little published on a theoretical framework for organization and visualization of Sustainability Indicators. This article reviews the current state of the art for sustainability indicator organization and visualization. Then a theoretical framework for sustainability indicator organization and visualization is proposed and described. Introduction Indicators are an effective tool to provide an assessment of sustainability at various spatial, functional, and temporal scales. However, there has been little research on indicator visualization techniques that can be used to convey sustainability information that matches the range of interests and needs of public decision makers. This article explores the current sustainability indicator visualization research and proposes a theoretical visualization framework that provides a high degree of flexibility for conveying sustainability indicator data at the various topical, spatial, and temporal scales that decision makers may need to make decisions on a wide range of urban and regional issues. Sustainability can be a complicated concept that spans such a broad range of topical areas that few (if any) individuals can be experts in all aspects of sustainability. The common response among sustainability researchers and advocates is to focus on a particular area of sustainability, either scale or subject. Such an example would be those that focus on buildings and the LEEDS criteria or those that focus on national energy policy, fuel efficiency in cars, power generation, or oil dependency in general. This approach is useful when the focus of decision making is sustainability itself, however this represents a very small portion of decisions made at the local, state, or national level. If sustainability is to be successfully inserted into public decision it will need to be one one of many factors that may be used for a wide range of decisions (EUROCITIES, 2004) and sustainability information will have to be relevant to this wide range of economic, social or rights issues on which decision makers are focused (Clark, 2003; Nyerges, 2002). Given the complexity of these issues, decisions makers will need sustainability information simplified to a manageable level so that they can be considered along with the other factors of importance to the decision at hand(Forester, 1989.; Lindblom, 1995).

Visualization of Sustainability Indicators: A Conceptual Framework 1/9/07 1

Indicators are one tool that can be used to help simplify the understanding of sustainability issues(Gudmundsson, 2003). For example, though the issue of water use in a community can be very complicated, there are measures of water use, such as gallons of water consumption per capita per day (GPCD), which can be used to assess the water efficiency of different urban policies. Across most topics of sustainability, there are a variety of measures that can be used as indicators of a relative level of sustainability and a variety of models for organizing these measures and indicators have been proposed. Though there is a wide variation in these sustainability indicator models (See Table 1) they all have two things in common. First they are based on the collection of basic environmental, social, and economic measurements that are used to calculate some new metric that is used as the basis for an indicator. Second they all have some approach to reporting or visualizing these indicators that can vary from simply a table to a complex visual map.

In general most sustainability indicator models report or summarize information across three basic scales: topical, spatial, and temporal. As an example, the GPCD measure cited above can be reported topically based on various customer attributes (single family or multi family, renter or owner,). It can be reported spatially at a national, state, city, neighborhood, or even building scale. Finally it can be reported temporally as a snap shot in time or summarized as a delta value over various time periods. The interests and needs of decision makers can vary across these scales based on the nature of the various issues they are evaluating. (Prescott-Allen, 2001) City Council members may be only interested in the indicators at their City scale such as the City as whole, individual neighborhoods, or perhaps a singe building, such as a fire station. They may also only be interested in specific topical area, such as heat island, water and energy efficiency, air quality, and resource utilization. For some decision makers, their interest may even

Table 1: Sustainability Indicator Models o Canada Sustainability Report (SRP, 2004) o Central Texas Sustainability Indicators Project (Central Texas

Sustainability Indicators Project, 2004) o Ecological Foot Print (Venetoulis, Chazan, & Gaudet, 2004; Wackernagel

et al., 1997) o Interagency Working Group on Sustainable Development Indicators

(Sustainable Measures, 2002) o Neighborhood Plans (Crossroads Resource Center, 1999; Manglani &

Pijawka, 2003) o Oregon Benchmarks (Sustainable Measures, 2002) o Sustainability Counts (SDU, 1998) o Sustainable Seattle (Best, Dusen, & Conlin, 1998; Sustainable Measures,

2002) o The Montreal Process (MPO, 1995) o The Natural Step Process (Natural Step, 1997) o The State of the Nation’s Ecosystems (Heinz Center, 2002) o UN Indicators of Sustainable Development (Sustainable Measures, 2002) o UNCHS (Habitat) indicators program (Auclair, 1997) o Yale Environmental Performance Index (Esty et al., 2006; Heinz Center,

2002)


change over the course of the issue discussion. For example, the general level of sustainability for various General Plan options could be initially evaluated using an ecological footprint (EF), but decision makers may then want to understand why an EF is high for an option that has other important community values, such as a new retail center. In this case, the Council may want to shift their focused to the specific indicators of retail projects to try to improve the sustainability or understand the reason so it can be weighed against other factors, such as revenue generation. Effectively delivering sustainability information to decision makers will require visualization techniques that allow decision makers to easily explore simple sustainability indicators at these various topical, spatial and temporal scales (Clark, 2003; Gallopín, 2004; Nyerges, 2001). The two key factors to providing this level of flexibility is the method used to organize the indicators and the techniques used to visualize the indicators Current Organizational Concepts Most models for sustainability indicators use two basic concepts for organization. First environmental, social, and / or economic measures are used directly or in some normalized fashion to create indicators or a trend analysis. For example different types of crimes for an area may be reported, and then combined to create a crime index, such as total crimes per capita. Second these indicators are organized into categories, usually hierarchical, based on some topical or systems classification scheme. For example a Crime index may be combined with Health index under the category of Human Welfare. In some cases, these indices for each of the topics may be combined to create an index for the category as a whole. Table 2 shows the organization structure for Prescott-Allen's The Well Being of Nations Sustainability Indicators.(Prescott-Allen, 2001) These indicators are based on a range of measures of various environmental and social-economic attributes. These measures are used to calculate specific indicator values for a wide range of specific conditions. These specific indicators are aggregated to various higher level indicators that represent a more general topic. For example, in Table 1, the highest level indicator is over all well being. This can be broken down to Human Wellbeing and Eco Wellbeing and so on. Table 2 shows how the hierarchy for two lower level indicators, crime and air quality can be aggregated up to several higher level indicators. Generally, there is little consistency among these indicator models in how indicators are developed or aggregated functionally, spatially, or temporally. Some indicators system just focus on natural environmental indicators(Esty et al., 2006; Heinz Center, 2002; Venetoulis et al., 2004), while most others include some human condition indicators. Generally at the highest topical hierarchical level these indicators models have two main indicators, Human and Ecological. Below this level what data is used and how it is


Table 2: Sample Hierarchical structure from The Well Being of Nations Sustainability Indicators (Prescott-Allen, 2001) Total Level 1 Level 2 Level 3 Level 4 Measures Wellbeing Human

Wellbeing

Heath and Population

Wealth Community Freedom &

Governance

Political rights Civil liberties Press freedom Corruption

perception

Peace & Order Armed conflicts % defense

expenditure of GDP

Peace Crime Homicides

(Number and Rate) Rapes

(Number and Rate) Robberies

(Number and Rate) Assaults

(Number and Rate) Other violence

(Number and Rate) Knowledge Equity Eco Wellbeing Land Water Air CO2 Per Person CO2 CO2/ha CO2/person Ozone Depletion Ozone depleting

substance metric ton

ODS/ha (g) ODS/person (g) Global Air Quality Local Air Quality Species &

Genes


aggregated to different levels varies widely. Each model creates its own standard classification, weighting, and aggregation standards. But even in this area, rules for such aggregation or weighting are seldom presented. For example, in The Wellbeing of Nations Prescott-Allen does provide rules for topical aggregation of the indicators but not rules for spatial aggregation. In the presentation of the data, tables of country indicators indices are grouped by region, however, no method is provided to create an aggregated index for a region. There has been some work on the methods that could be used for weighting and aggregation of data and indicators to create composite indicators for different hierarchical indicator levels (Nardo, Saisana, Saltelli, & Tarantola, 2005) Current Visualization Concepts Though there is a rich set of systems for defining sustainability indicators, there has been less work on general methods of organization and visualization of these indicators (Williams, 2004). Each model is typically presented with one method of reporting its indicators. This may vary from a simple table such as used in the LEED system to maps that show index values for varies geographic units such as those used in The Wellbeing of Nations. However few of the models utilize advanced methods for displaying hierarchical data. Though few of the models cited in this article provide a robust method for exploring and visualizing indicators, there are some decision support systems being developed that utilize indicators from one or more of these models. The most prominent sustainability visualization tool is The Dashboard of Sustainability developed by International Institute for Sustainable Development (Consultative Group on Sustainable Development Indicators, 2006) (http://www.iisd.org/cgsdi/dashboard.asp). Though this tool uses the motif of a vehicle dash board to display sustainability status information as various gauges, it interface is much more in depth and interactive. The tool is a windows based stand alone application that allows the user to interact with a hierarchical structured database of indicator data. Dashboard is able to display to indicator data for various topical, spatial, and temporal scale data in a variety of ways including the dashboard guage, maps, scatter plots and value plots by spatial unit. The tool does allow the user to view data summarized at different levels of the topical hierarchy and does aggregate the data internally for each grouping of indicators. Three levels of hierarchy are supported, lower data level, group data indicator level, and then a single aggregated indicator. Indicators Figure 1 shows the typical Dashboard display of the United States with the indicator gauges. Each colored part of the gauge corresponds to a data indicator, with each center being the group aggregated indicator value. Different gauges can be selected. For viewing. Figure 1: Sustainability Dashboard for United States 2004


Figure 2 Shows a similar data display but in map view. In this view the map is interactive and data from any country can be displayed in the dashboard by clicking on the map and data values for any country can be displayed by clicking on the dash board gauge. Gauges can also be displayed side by side with other countries (up to eight). Other views include a scatter plot of all countries (figure 3) based on values for two indicators, and a nodal view (figure 4) of the hierarchy itself for one country.


Figure 2: United States Map View;

Figure 3: Scatter Plot View Figure 4: Hierarchy Node View


A Proposed Theoretical Framework for Visualization of Sustainability Indicators Though there are numerous models for defining and organizing Sustainability Indicators (see Table 1), some published work defining methods of weighting and aggregation (Nardo et al., 2005), and some published work on developing visualization methods and tools (Consultative Group on Sustainable Development Indicators, 2006; Williams, 2004) there has been little published on a theoretical framework for organization and visualization of Sustainability Indicators. This is a critical need. Sustainability is of increasing importance to the public, businesses, and government. At all these levels the public, managers, and politician's are eager to incorporate sustainable practices into our daily lives, business, and governments. In today's information society the lack of information will not be a key barrier to this effort, rather it will be the understanding of an over whelming amount of information that well be key. Thus sustainability systems that provide easily accessible and understood information are needed. To meet these anticipated information needs, a theoretical framework of data organization and visualization for sustainability indicators is here proposed. This concept has four components. 1) A data organizational system that defines indicator data based on hierarchies of topical, spatial, and temporal attributes; 2) methods for aggregation of indicator indices at various levels of topical, spatial, and temporal hierarchies; 3) Visual techniques to display indicator data at in various views that either are topically, spatially, or temporally focused or defined; and 4) Visual Anlaytic techniques to analytically assess indicator data, and 5) A simple and intuitive visual framework that would allow the decision maker to easily explore sustainability indicators and analysis through various views and scales. Organization of Indicators in a Multi-Dimension Hierarchical Data Structure The calculation of an indicator is typically based on some rule which defines what data is used, what the calculation is (if any) is used to create a metric, and what normalization is done (if any) to create a numeric index or indicator. In addition to this numeric, these indicators have attributes of topic, space, and time. For example, an indicator value created from CO2 levels is an indicator that describes some aspect of air quality. It is collected from a specific place such as an individual collector located at Broadway and 40th St. It is also collected as a specific time, such as June 21st 2005, 6:00 am. Thus in addition to its value, an indicator can be described by topic, space, and time. These other attributes are not static, but can describe as part of a hierarchical structure. Space can be defined as a single parcel of land, or aggregated into a single block, or to a square mile, or to a city, state, nation, hemisphere and planet. Topic can be described as a single measurement, such as CO2 levels, combined with others to describe green house gases, combined with other measures (Particulates etc) to describe air quality, or as part of a summary of environmental health or general global well being. Time can be described as now, or the day, or the month, year, decade, century or as the difference between two points in time.


Though these hierarchy schemes will be used to organize data, they can be described independent of the data itself. Such hierarchies are typically defined within a nodal network. Each node may be linked to one or more other nodes. For each attribute scheme, this structure could be described in XML as follows: <HIERARCHY NAME=”TOPICAL”>

<GLOBAL NAME=”Global Wellbeing Index”> <HUMAN NAME=”Human Well Being Index”>

<WEALTH NAME=”Wealth Index”> </WEALTH> <FREEDOM NAME=”Fredom Index”>

<VOTE NAME=”Voting Rights Index></VOTE> </FREEDOM>

</HUMAN> <ECOSYSTEM NAME=”Ecosystem Wellbeing Index”>

<AIR NAME=”Air Quality Index> <CO2 NAME=”CO2 Levels”></CO2> <OZ NAME=”Ozone Levels”></OZ> </AIR>

</ECOSYSTEM </GLOBAL>

<HIERARCHY> These attributes can be used to organize indicator data. But this is more than just a hierarchical classification scheme. In this case, each level of the hierarchy represents a summary of all the values below this level. This data summarization of indicators from one level to another is also based on rules of how the indicators or data from lower levels is aggregated at the higher level. These rules will be different for different hierarchical schemes. For example, the aggregation of the various air quality indicators to a single air quality index may simply be done by averaging the values. However, in a spatial hierarchy, an air quality index of several countries may be aggregated to a region by weighting each countries air quality index by the size of the country. Or a human welfare indicator may be weighted by the population of each country. Thus within the hierarchical indicator scheme, rules must be developed for how each indicator is aggregated These rules can either be defined as part of the hierarchy, or simply calculated and included in the data and associated with the aggregated node. For example, the Air Quality Node in the above scheme can simply have a value assigned to it. This value can be based on CO2 and Ozone, but may be based on some complicated formula that may even rely on other data not included in the indicator data set. But these values cold also be based on a rule, and the aggregated value is calculated at runtime. Averaging and weighted averaging would be the most common rules. Such simple methods could be reflected in XML data as follows.

<HUMAN NAME=”Human Well Being Index” AGMETHOD=”AVERAGE”>


Spatial and temporal hierarchies can also be defined in a similar fashion. <HIERARCHY NAME=”SPATIAL”>

<GLOBAL NAME=”World Wide”> <NHEMISPHERE NAME=”Northern Hemisphere”>

<NAMERICA NAME=”North America”> <US NAME=”United States”> <TEXAS NAME=”State of Texas”> <BEXAR NAME=”Bexar County”> <SA NAME=”City of San Antonio”></SA> </BEXAR> < /TEXAS> </US> </NAMERICA>

</NHEMISPHERE> </GLOBAL>

<HIERARCHY> <HIERARCHY NAME=”TEMPORAL”>

<1900CENT NAME=”20th Centrury”> <1990DEC NAME=”Decade 1990-1999”>

<1990 NAME=”1990”></1990> <1991 NAME=”1991”></1991> ….. <1999 NAME=”1999”></1999>

</1990DEC> </1900CENT>

<HIERARCHY> Given this concept of hierarchical attributes, indicator data can be described in a multi dimensional matrix. Each data point represents a value for a specific set of spatial, topical, and temporal attributes. This code be coded in XML as follows: <DATAVALUES>

<DATA SPATIAL=”TEXAS” TOPIC=”HUMAN” TEMPORAL=”1990”>56.4</DATA> <DATA SPATIAL=”TEXAS” TOPIC=”ECOSYSTEM” TEMPORAL=”1990”>65.4</DATA> <DATA SPATIAL=”ARIZONA” TOPIC=”HUMAN” TEMPORAL=”1990”>70.4</DATA> <DATA SPATIAL=”ARIZONA” TOPIC=”ECOSYSTEM” TEMPORAL=”2000”>55.4</DATA> <DATA SPATIAL=”TEXAS” TOPIC=”HUMAN” TEMPORAL=”1999”>56.4</DATA>


Chart 1: Topical View of Well Being of Nations Indicators For North America

Visualization of Sustainability Indicators Under this concept of SI organization, people interested in exploring the SI database could view the indicator data in many ways based on the organization of the data and the scales of topical, spatial and temporal interest. Each of these views would present indicator data in a hierarchical fashion based on the topical or spatial level of interest. Chart 1 shows this basic concept for a topical view using indicators from the Well Being of Nations. Initially a user could see the indicators organized by topic. In this example this starts with a display of Total Wellbeing with a breakdown of the two basic indicators that are used to calculate the total well being indicator, Human Wellbeing and Ecosystems Wellbeing. The user could look for more specific topical information by selecting a level in this case Human Well Being and expanding it. As each level is expanded, all of the indicators for this level and their values would be displayed. Each view has a particular temporal and spatial reference. Chart 1 is North America and Chart two shows Canada. This concept can also be applied to a spatial view but in this case the spatial data is displayed in a hierarchal fashion, either list or map. A spatial view would be based on geographical organization of data such as city, nation, continent, and world (in the case of a LEEDS system, this may be a building plan). A list view could start with the world summary indicator for a particular topic, with the various country indicators that make up the world indicator value. Spatially the user could explore indicators at lower levels of geography by expanding the levels of spatial definition, in this case cities. This could also be done with a map, where world map of continents with each continents indicator reflected by a graduate color. A continent could be selected and the map would zoom to


Chart 2: Topical View of Well Being of Nations Indicators For Canada

a continent map of countries, or zoom to cities, and so forth to the bottom level of geography. In the context of this visualization concept, temporal information is not considered a view, but rather an attribute of both the spatial and topical views. Within each of these views the indicators must be displayed for some temporal value. In our example, The Well Being of Nations, this temporal value is the year 2000. The temporal scale of sustainability indicators does not appear to fall below a year in time and it is not anticipated that anytime in the near future sufficient data will be available for all indicators at a smaller time frame (month or day). However, some of the older sustainability indicator efforts do now have historical data by year and it is anticipated that at some point sustainability indicators will be connected to forecasting models which could provide estimates of future indicator values. Thus selection of the time frame of the indicator data will be important. This will be true not only for snap-shot-in-time data, but also changes in the indicators over time. Table 3 provides a more detailed description of these two views in terms of organization and data display. Using these requirements, a web based prototype of this visualization concept was created using a sub set of the indicators from The Well Being of Nations. It can be found at http://www.public.asu.edu/~mcquay/sivp/


Table 3: Sustainability Indicator Views Topical

Organization Concept Hierarchical Category Trees- Each category would have an indicator

value for that category that would represent some type of summation of all indicators underneath.

Scales – Data would be reported for one spatial and temporal scale Different Schemes

System – Example: Prescott-Allen's The Well Being of Nations Principles – Example: Natural Step http://www.naturalstep.org/, United

Nations Rio Declaration, American Planning Association’s Policy Guide Planning for Sustainability, World Summit on Sustainable Development..

Display Concepts Data Display – Data could be viewed in a list based or map based format.

List based would list indicator topics in an expandable list format, each topic level could be expanded to show the indicators in the next level. Map based would be based on a Tree Map where indicators for each level would be displayed in a tree map with each indicator map size based on their % contribution to the indicator of the higher level.

Navigation - Navigation would allow the user to select different topical schemes and then allow navigation through hierarchy by list based or mapped based hierarchical views. In this view values for indicators would be displayed for one temporal scale and one spatial scale which could be selected.

Spatial Organizational Concepts

Predefined Geographic Units – This would be a set of predefined geographical units where in the lower level geographical units, such as cities, the indicators can be aggregated up to the higher level units, such as countries, which can be aggregated to a high level, such as continents.

Scales – Data would be reported for one topical and temporal scale

Different Schemes GeoPolitical – Values summarized for a specific geopolitical spatial

definition (city County etc) Socio-Economic – Values summarized for a socio economic geography,

census based (census tract block group), service area based (police beats)

Ecosystem – Values summarized by spatial ecosystem units such Biomes, continent, climate

Display Concepts


Data Display – Data could be viewed in a list based or map based format. List based would list indicators based on a verbal description of the geography, such as City name or country name topics, in an expandable list format, each spatial level could be expanded to show the indicators in the next level. Map based would be based on an actual map, which would display indicators for one level of the geography.

Navigation - Navigation would allow the user to select different spatial levels of the hierarchy by either expanding the list the list of spatial units or selecting a geographic feature In this view values for indicators would be displayed for one temporal scale and one spatial scale which could be selected.

Temporal (Options for Spatial and Topical Views)

Organizational Concepts This may be a dependent on availability of historical data which may vary

by spatial and functional scales. Different Schemes

Snap Shot – Current, Future, Specific Time (By Year Month ?) Delta – Change Over Time

Display Concepts Data Display – This would be an option for data reporting within the

Spatial or Topical views, but not a view by itself. For snap shot data, this would simply be a reporting of the indicator for the date selected (past, present or future). For change over time (delta) an actual numerical value of change could be displayed (magnitude of positive or negative change), multiple temporal values (Five different year snapshots), or it could be done through an animation that displayed the values for each of the time intervals of desired history or future.

Navigation – This would be a selection within the Spatial or Topical view

Within this framework of topical, spatial and temporal hierarchical data organization, an interface that allows the user to browse the data based on each hierarchy is suggested. Such an interface would provide a main display frame, in which the user would browse data based on one of the hierarchies, based on single or multiple attributes selected from the other hierarchies. Figure 5 provides an overview of such and interface. Also conceptual frameworks and prototypes are being developed on this basis and are located at www.public.asu.edu/~mcquay/sivp


Figure 5: Conceptual User Interface

Indicator Data Browser Display Various Views of Heirarchy Lists Maps Trees Graphs

Indicator Attribute Selection Select what attributes are displayed by the indicator browser Topic, Spatial, Time

Select an Indicator Browser: Topical,Spatial or Time. Each allows theuser to explore the indicator databased on a Topical, Spatial, or Timelineclassification heirarchy.

Attribute selectors can displaythe attributes based on different'views of the heirarchy: List, Map,Tree


Future Research Though this article provides a logical visualization framework for decision makers to explore sustainability indicators and the measures behind them, however most of the current sustainability indicator systems do not have sufficient organization and or data to utilize the full capabilities of such a framework. Six key areas of research are needed.

1) Not all sustainability schemes include a topical hierarchical organization for the indicators. A lack of such a hierarchy even for SI schemes with only a few indicators will be difficult for decision makers to use when comparing several different alternative actions. Methods of organizing indicators in hierarchical structures needs to be researched. (Gallopín, 2004; Nyerges, 2002)

2) Most indicators are not normalized, that is converted to some common scale that allows comparison of indicators created from different units of measure, as well as across topics, such as air quality and bio diversity. (Allard, Cherqui, Wurtz, & Mora, 2004; Nyerges, 2001) Though there is substantial literature on the normalization of indexes in general, there is little literature on application to sustainability indicators.

3) Most indicator system define their indicators only at one or two geographic scales. This is likely due to a lack of data to report all indicators at all geographic scales. However, this does not mean that for some indicators, data is not available at different geographic scales. (N.L.Leake, Adamowicz, & Boxall, 2002) In these cases this framework would allow exploration of indicators at various geographic scales where indicator are available at that level.

4) Most indicators do not provide rules for summation of indicators to higher topical and spatial levels. This is a critical aspect to this proposed framework. Such summation allows a decision maker to reduce complexity in areas of minor or no interest and expand complexity in areas of high interest. (Nyerges, 2001)

5) There is general lack of historical indicator data. This is most likely because such indicators have only become widely collected and reported for the last few years. Over time this will change but it will be important to maintain datasets of these indicators and maintain some consistency in how they are defined from one year to the next. However, there currently appears to be little research into the use of sustainability indicators with futures analysis. This is due in part because forecasting of a wide array of measures used in the sustainability indicators by one model is currently not practical. However, as various environmental and urban systems model are becoming capable of using standard inputs (land use, climate, economic growth) to forecast various future conditions. As these data sets are combined into common scenarios, linking these model outputs to standard sustainability indicator systems will simplify the visualization of futures analysis results.

6) Though there is much research on visualization of hierarchical data (tree maps, nodal mapping, etc) there is little research on the application of these techniques to the visualization of hierarchical sustainability data.


Conclusion Decision makers usually have to deal with variety of issues for which sustainability is only one of the factors that will be used in the decision making process. In addition, their focus on what aspects of sustainability may shift over the course of discussion of an issue. Thus there is a need for tools with high degree of flexibility providing both summarized and details in a variety of levels of place and time. The proposed organizational and visualization framework presented here tries to provide flexibilities reflecting the three major scales: topic, space, and time. This method provides a basis for a data standard that would allow a single user interface to display a wide range of indicator systems. More research and development of sustainability indicators will be needed to fully exploit this flexibility. References Allard, F., Cherqui, F., Wurtz, E., & Mora, L. (2004). A methodology to assess the

sustainability of rehabilitation projects in urban buildings. Retrieved 9/15/05, 2005, from http://www.costc16.org/downloads/WORKING%20GROUP%203/AllardF%20Sustainable%20Rehabilitation%20.pdf

Auclair, C. (1997). The unchs (habitat) indicators program. In R. Moldan & e. al (Eds.), Sustainable indicators: A report on the project on indicators of sustainable development.New York: John Willey& Sons.

Best, A., Dusen, H. V., & Conlin, R. (1998). Indicators of sustainable community:Sustainable Seattle.

Central Texas Sustainability Indicators Project. (2004). Fith annual report.Austin, Texas: Central Texas Sustainability Indicators Project.

Clark, W. C. (2003). Sustainability science: Challenges for the new millennium.University of East Anglia, Norwich UK: An address at the official opening of the Zuckerman Institute for Connective Environmental Research, September 4, 2003.

Consultative Group on Sustainable Development Indicators. (2006). The dashboard of sustainability. Retrieved 3/17/2006, 2006, from http://www.iisd.org/cgsdi/dashboard.asp

Crossroads Resource Center. (1999). Neighborhood sustainability indicators guidebook.Minneapolis, Minnesota: Urban Ecology Coalition (Minneapolis).

Esty, D. C., Srebotnjak, T., Kim, C. H., Levy, M. A., Sherbinin, A. d., & Anderson, B. (2006). Pilot 2006 environmental performance index.New Haven: Yale Center for Environmental Law & Policy.

EUROCITIES. (2004). Towards a strategy on the urban environment: Eurocities statement. In E. T. N. o. M. E. Cities (Ed.):European Union.

Forester, J. (1989.). Planning in the face of power.: University of California Press. Gallopín, G. (2004). Sustainable development: Epistemological challenges to science and

technology, Workshop on Sustainable Development: Epistemological Challenges to Science

and Technology, ECLAC.Santiago de Chile: ECLAC.


Gudmundsson, H. (2003). Making concepts matter: Sustainable mobility and indicator systems in transport policy. International Social Science Journal, 55(2), 199-217.

Heinz Center, T. (2002). The state of the nation’s ecosystems.Washington, DC: The H. John Heinz III Center For Science, Economics, and the Environment.

Lindblom, C. E. (1995). Science of muddling through. In D. C. McCool (Ed.), Public policy theories, models, and concepts (pp. 142-157). Elnglewood Cliffs, New Jersey 07632: Prentice Hall.

Manglani, P., & Pijawka, K. D. (2003, November 9 - 12, 2003). Measuring environmental impacts of sustainable neighhood plans. Paper presented at the PLEA 2003 - The 20th Conference on Passive and Low Energy Architecture, Chile.

MPO, T. M. P. O. (1995). Criteria and indicators for the conservation and sustainable management of temperate and boreal forests.

N.L.Leake, Adamowicz, W. L., & Boxall, P. C. (2002). An examination of economic sustainability indicators in forest dependent communities in canada:Sustainable Forest Management Network.

Nardo, M., Saisana, M., Saltelli, A., & Tarantola, S. (2005). Tools for composite indicators building.Ispra, Italy: European Commission, Directorate General, Joint Research Center.

Natural Step, T. (1997). The natural step. Nyerges, T. (2002). Visualizations in sustainability modeling: An approach using

participatory gis for decision support.Los Angeles: Association of American Geographers.

Nyerges, T. L. (2001, October 1-2, 2001). Research needs for participatory, geospatial decision support: Linked representations for sustainability modeling. Paper presented at the Intersections of Geospatial Information and Information Technology Workshop.

Prescott-Allen, R. (2001). The wellbeing of nations: A country-by-country index of quality of life and the environment.Washington: Island Press.

SDU, S. D. U. (1998). Sustainability counts. In F. a. R. A. Department for Environment (Ed.):United Kingdom.

SRP, S. R. P. (2004). Canada sustainability report:Institute for Research and Innovation in Sustainability.

Sustainable Measures, I. (2002, 6/25/02). Comparative summary of existing indicator sets - part 1. Retrieved 9/15/05, 2005, from http://www.sustainabilityindicators.org/workgroups/NIC/PocanticoComparativeSummaryPart1.pdf

Sustainable Measures, I. (2002, 6/25/02). Comparative summary of existing indicator sets - part 2. Retrieved 9/15/05, 2005, from http://www.sustainabilityindicators.org/workgroups/NIC/PocanticoComparativeSummaryPart2.pdf

Venetoulis, D. J., Chazan, D., & Gaudet, C. (2004). Ecological footprint of nations:Redefining Progress.

Wackernagel, M., Onisto, L., Linares, A. C., Falfán, I. S. L., García, J. M., Guerrero, A. I. S., et al. (1997, 9/15/2005). Ecological footprints of nations. Retrieved 9/15/05, 2005


Williams, M. (2004). Questvis and mdsteer: The visualization of high-dimensional environmental sustainability data. The University of British Columbia.

visualization of sustainability indicators: a conceptual...

Documents