Timothy Nyerges is a professor in the Department of Geography at the University of Washington. He has long been interested in the connection between GIS and sustainability, having worked on a number of projects, including a curriculum development project in 2001 on GIS for Sustainability Science. V1 Magazine editor Matt Ball spoke with professor Nyerges about the connection between GIS and sustainable development, as well as the ongoing need to develop better tools to tackle sustainability issues.
V1: You have been thinking about GIS and sustainability for some time, not only in how you teach but in how GIS tools are utilized. How do you define the connection between sustainability and GIS?
Nyerges: The social and economic issues and ecological conditions of the world are the characteristics that one has to pay attention to for land use, transportation and water resources, among other important activities. The practical reason why I don’t think that much of the world has made headway on sustainability is that a lot of the people talk about the social, economic and ecological aspects, but you have to tie those into a pervasive activity.
On the ground, what are people doing? They’re changing the landscape with land use, they have a need to move from place to place, and water resources are something that no community can do without. Those are pervasive activities in almost all communities. It turns out to be a matrix, a three-by-three matrix of characteristics that are important for sustainability (activity features by conditions).
That’s also the same input for GIS. I like to think about three conditions – social, economic, and ecological – for land use activity when designing a GIS for any project. How do we measure those conditions of the world? How do we capture those conditions in large GIS databases? How should software be organized to treat these kinds of conditions on various decision horizons for these pervasive activities?
We have many time and space decision horizons when making decisions about sustainability. I look at this issue of sustainability in terms of decision theory and GIS. We have to believe in the decisions that we made in the past and move forward to the future. If we’re always coming back to our decisions and redoing them in conflicting ways, then that certainly isn’t a sustainable trajectory for any community.
Unfortunately the inputs and outputs for these decisions can get very complicated, but you have to be able to track all of these characteristics. GIS fortunately is a platform where you can track all of these metrics.
V1: How do we move forward in harnessing the power of GIS for sustainable development?
Nyerges: When we look at sustainability projects that exist now, we focus on communities that like to plan — some do, some don’t. Communities moving toward sustainability are those that like to raise revenue and funding to put their plans into action. To put the plan into action, we focus on projects.
In most of the sustainability literature, people talk about sustainable planning and sustainable project development. Those are the two ends of the space and time decision horizons. A lot of people have missed the element in between, and that’s funding through programming.
No organization in the world can do without a revenue stream. You have to have volunteer effort or someone has to pay something at some point in time to get things done. Revenue funds need to be established in an organized manner. For the last 25 years I’ve dealt with transportation activity, and in the transportation world it’s called improvement programming.
The local/regional urban planning process is a “Balkanized” decision process. Balkanized in the sense that so many people are trying to make decisions that it overwhelms processes — process rich, and outcome poor.
Sustainable development links the planning horizon with the programming horizon, and with the project implementation horizon. Without the linkages between planning, programming and project implementation within GIS databases, over broad areas (focusing on small areas within broad areas) and over broad time (focusing on week to week, month to month, year to year), we won’t get a handle on sustainability.
I have a textbook on this topic under contract with Guilford Press, it’s called GIS and Decision Support. I’ve teamed with Piotr Jankowski again on this. The final draft copy should be done soon. All of what I’ve mentioned here is in the textbook, and this is what I teach and talk about in my Urban GIS class.
V1: Do you feel that the current geospatial toolset is adequate to fill this role for sustainability? One of the things that we’ve discussed in Vector1 Media is a missing temporal capability. Is that something that you can work around?
Nyerges: Software tools aren’t up to speed. One of the ways that I keep so busy is that Piotr and I are on our fourth National Science Foundation project where we build tools and evaluate software.
We had a National Science Foundation project that built a tool called GeoChoicePerspectives. It is an extension to ArcView 3, and runs under ArcView 3.3. I’m using that in a class this week for habitat site redevelopment and decision process. It works to combine maps with decision analysis and group-based software. The group software allows you to bring together different rankings of sites.
Piotr and I have been writing text, doing research, and teaching in order to try to push things forward on multiple fronts. No single front in service, teaching or research is going to be the answer.
I work locally mostly on the sustainability ideas. That’s the advantage here in the Pacific Northwest, because it’s a huge topic. A lot of government organizations are interested, and it’s easy to do service-related projects here.
V1: Certainly in the initial research that I’ve done on this topic, Portland and Seattle are among the cities that stand out in terms of sustainable development.
Nyerges: Yes, I’d group Portland, Santa Monica, Boulder, and Seattle as some of the few selected cities that are trying to put something together. Another overriding issue in each of these cities is the issue of affordable housing.
When we talk about the social component of the sustainability triad (social, economic, environment), I look at affordable housing as an overriding sustainability issue that must be addressed. If we can’t solve affordable housing, then the city as a whole won’t have a sustainable outlook from a balanced social justice perspective.
V1: Do you work with urban planners, architects and others on projects of a multidisciplinary nature?
Nyerges: The cross-disciplinary issue is often a problem with regard to scale. Because of differences in scale, it’s hard sometimes to connect architects and planners and other disciplines as well.
An architect that spends a great deal of time on the sustainability detail from a LEED perspective, working from building to building, doesn’t have a sense of this work on a city-wide scale. When you get into the built environment and infrastructure issues, there’s so much information associated with multiple buildings in small area planning, it’s a challenge to come to grips with all that information.
You have to get a handle on the information technology to be able to treat this much complex information in a straightforward manner. That’s one of the biggest dilemmas, and that’s why I consider this area of research to have tremendous opportunity.
The technology to treat these complex multi scalar process issues hasn’t really been developed yet. It’s kind of a new thing in the world, and a lot of people need to pay more attention to do more in this area. It’s been one of my goals for the past 15 years, and there’s a lot yet to be done.
V1: Do you have an interest in the interface between the different design tools (CAD, GIS and most recently Building Information Models)?
Nyerges: I have been a systems integrator for the majority of my career. One of the challenges here is the reference frame of an individual. How far can one really go with the nature of systems integration? In any one person’s waking day, or any one person’s given project, there are a large number of tools that can come into play.
The systems integration perspective that I take is general enough to look at any appropriate tool. If a tool has a capability associated with it, and the tools being developed do not, then pull in the one that does. Why redevelop a capability? Instead you should integrate it.
There’s something to be said for how people make use of information from any one interface. There are tightly integrated systems, meaning that an information user would not have to go to multiple interfaces.
Let’s take a typical GIS user, such as a resource manager in an urban/regional context. How many different interfaces would this individual face to invoke capabilities that don’t exist in a single tool, but exist across multiple tools?
With the Web Services idea, and federated databases, there’s progress toward bringing different tool capabilities together. Because the user interface is such a big part of systems these days, we wouldn’t want to belabor the information user with too many different ways to look at the same information.
There’s something to be said about multiple ways of looking at information, but not to increase the cognitive load of an individual by having them face a CAD interface vs. a GIS interface vs. BIM interface. Each tool has its own interface.
When you’re trying to integrate the tools, which of the interfaces does one really use? There are some big questions there, and user interface design and human/computer interaction is something we should pay more attention to.
The systems that we’ve been developing here have investigated five different technologies for decision support – database management, spatial analysis and geovisualization – the three standard technologies in GIS – but also communications technology and decision analysis technology.
V1: Do you feel that the interface of GIS, with its broader geographic view, will have the greatest input to sustainable development?
Nyerges: I think it’s clearly a leader because of the three core technologies in GIS. I’ve been promoting that for more than 20 years. Database technology in and of itself is a great contributor to problem solving. Spatial analysis, as a number of people have shown, can be a big contributor to solving problems. Visualization is a tremendous contributor to solving problems. When you bring the three together, you really do have something unique, and bringing them together is really the nature of the data integration engine in GIS.
When you look to the world, and the last 25 years of development in information technology, there aren’t any other systems that have done a better job in service to the world of bringing together technologies to address complex problems. That’s why GIS has such a huge growth opportunity, for the last two decades, and it will continue.
If you look at any particular industry player, and it doesn’t matter who it is, bringing together the multiple technologies and making a stable platform to solve problems is a tremendous opportunity.
For the first five years after my Ph.D. work, I was in private industry. I was part of four start-up companies and consulting organizations. The whole software development sphere, although it’s changed substantially in volume of growth, is the key to this GIS technology. It’s still about three or more fundamental technologies that need to be integrated as a solid platform for people to be able to address complex problems.
V1: Sustainability certainly is a complex problem.
Nyerges: That’s the theme throughout the last 15 years of my work. What is the nature of complex problems? How do you address complex problems? Sustainability at a macro scale is a complex problem.
When is the world truly going to come to grips with climate change? It has to become practical to come to grips with it. We are going to have to do some interesting GIS-related work to show the practice of being able to deal with climate change.
The Intergovernmental Program of Climate Change (IPCC) has done fabulous work on the science of climate change. To be able to make real changes on the ground, and truly understand the causes of climate change over the past 100 years, is going to be the real challenge for the world.
V1: I see that you’ve been working with NOAA on the impact of climate change on the coast. What type of work are you doing?
Nyerges: I’ve had a fascination with oceans and coastal systems since a young age, and finally now in the last three years, I’ve been devoting time and research to this topic. It poses fascinating problems to come to grips with.
Just yesterday, interestingly enough, we were discussing in the research group about the Intergovernmental Program of Climate Change and their vulnerability assessment framework. It’s much different than the risk analysis/assessment framework that’s been developed and used in environmental health, ecological health and toxicology for a number of years.
How could the climate people adopt terms that don’t match what all of the other risk people have been doing for years? Just in the term “vulnerability” for example, it’s very different. It’s an interesting challenge to bridge that. I’ve been doing risk analysis for years, but addressing climate change is a component that’s new to me.
I decided about four years ago to focus on the land/water interface, and then made a push into substantive areas that would allow me to do research and teaching. I have a course in coastal GIS that I’ve been teaching for two years.
Now all of a sudden, I see a number of researchers telling decision makers that population growth along the coast with climate change is probably one of the biggest problems in the world. It’s fortunate for me that I elected to focus on this, and can be in a place to contribute.
V1: The term vulnerability seems to connote that there may not really be a problem.
Nyerges: It’s an interesting distinction between risk and vulnerability. I published a paper in 1997 on use of GIS in environmental and ecological health. Basically it was a risk assessment paper that did a fairly detailed analysis to develop a framework that linked vulnerability models and risk models.
In vulnerability there’s a potential for harm. With risk there’s a probability for harm. The difference between the two is that insurance companies would never write a policy for potential for harm. They don’t know what the metric means. But they will write a policy based on the percentage of chance. They need to do the actuarial work and can economically finance the problem if there’s percentage of chance associated with it.
That’s the interesting thing about the climate change language that’s being published. With vulnerability, there’s an impact that’s understood to occur, but there are other external effects outside of the direct effect of the impact. The term vulnerability (in the IPCC literature) doesn’t seem to have much to do with probability. It has something to do with the breadth of the receptors that are being impacted by climate change. The language of vulnerability as that stands is not a very operable way to approach the language of climate change.
Hans-Martin Füssell from Stanford, who is on the IPCC, wrote an article titled, ” Climate Change Vulnerability Assessments: An Evolution of Conceptual Thinking,” that tries to deal with the problem, and the different ways that vulnerability and risk have been dealt with in the literature. He had to ultimately side with the climate group, because that’s the group he’s with, and then do his assessment framework along those lines. But without a bridge over to what other risk assessment/risk managers have been doing over the years, there will be some significant disconnects. There is still much to be clarified in the GIS, risk and climate change literature to make concepts operable.
People have been writing insurance policies for hundreds of years, and not necessarily paying attention to environmental or human health from the perspective of risk. It’s been a fairly short window that agencies have been dealing with the probabilities for human health in connection to the environment. The EPA is only 40 years old.
V1: It’s really fascinating to me that we’re just getting a handle on the human interaction with Earth systems. Do we have good enough metrics to be able to really understand our impact on the planet?
Nyerges: There are a number of really good studies that have been done, and at a large scale. The Millennium Assessment of Global Ecosystems that was done between 2001-2005 is significant. And Europe has been doing a good job with their interest in sustainability ever since the Brundtland Report was published in 1987.
Europe has been more interested in looking at the nature of indicators. What’s the nature of an indicator, and what does it do for decision makers, stakeholders and society in general? There are a number of good indicator inventories that are out there, and a number of good books have been published.
One of the big issues around indicators is that they’re undoubtedly a publicly understood data measurement, and they haven’t been taken for very long. We haven’t been monitoring various aspects of our environment for very long.
The EPA got started in 1968 and really got going in 1969 with the National Environmental Policy Act. The nature of environmental indicators is a newer thing to the nature of most societies across the world. We really didn’t believe that as a world population we were having these detrimental effects, and perhaps even more so these days, cumulative detrimental effects.
The thing about climate change is that it’s not so much about a single car going out on the highway and driving, but you put them all together and look at the growth rate, and realize that we should have been looking at this for quite some time. Here we are, and at the turn of the millennium essentially, we are now looking at what CO2 emissions are like en mass.
Again, the question is how much does the public really understand related to indicators? That’s why in all of my sustainability projects, I build public decision support systems. That’s where I think the greatest need is.
V1: Have you worked on some large-scale local projects with public participation as a component?
Nyerges: I learned after three declines from the National Science Foundation in the late 1990’s that the term sustainability didn’t resonate. The review panel tripped over the first paragraph. Although I defined the term sustainability, they really didn’t understand it.
I quit using the term sustainability, and went for a more practical approach. If I never used the term sustainability, but still did it, it would still have a meaning to society.
The most recent project that I’ve done is a transportation improvement project. A group of 250 users used a Web interface for one month in a decision experiment that had them selecting various transportation projects in the region. It’s out there on the Web by going to www.letsimprovetransportation.org.
We weren’t able to assimilate public databases that would support what the public wanted to do, although we scoured local government and regional transportation databases. A next logical step to me is to have the public design the database rather than professional planners.
V1: The complexity of a coastal system with tides, current and storm events is really mind boggling to me. How do you address that with a GIS? Does it involve input from sensors?
Nyerges: The reason why I started to teach a coastal GIS class was to get after that exact thing. I’m selfish in one way and forward-looking in another way. I wanted to write a set of assignments, with an instructional research assistant, that try to take into consideration flow of water. How does water, as it falls onto a coastal watershed, find its way to open water and what kind of mixing occurs with tide and currents.
With ArcGIS there is a raindrop tool that gets after some of the issue. But I wanted to test a conceptual hydrological process, tracking the water flow out into open water.
A report that was recently completed for the Washington Department of Ecology found that the largest contributor of pollution into Puget Sound is non-point source pollution from transportation. Vehicular deposition of oil, gas or other contaminants onto highways makes its way into the waterways and is the worst source of pollution.
It was fortunate that I’ve done Transportation GIS for a number of years, and have been teaching this Coastal GIS course from a frame of mind of tracking water circulation. I now have students doing research about non-point source pollution.
V1: What are your next steps in terms of tool design?
Nyerges: I’m trying to operationalize the second step in Carl Steinitz’s six phases of GIS modeling. Carl Steinitz has taught at the Harvard Graduate School of Design for thirty-five years. He came up with this framework in the early 90s, and I picked it up in the late 1990s as a basis for model development in public participatory GIS.
• The first phase is representation modeling (database design)
• The second phase is process modeling (pure process as in system and world processes, water flow, land use change, whatever temporality in geospatial processes)
• The third phase is scenario modeling (he calls it evaluation, but it is still selecting out from the process model the things you want to look at)
• The fourth phase is change modeling (before and after effects, differentiation of things that occur in the world)
• The fifth phase is impact modeling (risk assessment, impact from land use, transportation, water, etc.)
• Phase six is decision modeling
I’ve been designing software to operationalize all six phases for some time now as a participatory system. And essentially, that’s why I chose water process as a core issue to research, because I’ve wanted to address process modeling.
There are various modeling software packages such as GoldSim from GoldSim Technology Group and VenSim from Ventana Systems for simulating process, but they don’t adequately address the spatial component. There was an academic version of geospatial process modeling from Robert Costanza from when he was at the University of Maryland. He has since moved and unfortunately the development of that software seems to have left off. Another possibility is the Recursive Porous Agent Simulation Toolkit also referred to as Repast.
Right now I’m gearing up for a major grant writing effort to do the next phase of a geospatial process model that will likely require substantial funds. Either that or I’ll borrow from someone else and perform the systems integration we discussed earlier.