Sensors and Systems
Breaking News
Swift Navigation utilizes Sitetracker to manage the global expansion of its network of reference stations
Rating12345– Swift Navigation is utilizing Sitetracker to track its...
Esri, GeoComm, RapidSOS Partner to Bring Advanced Situational Awareness Solution to First Responders
Rating12345REDLANDS, Calif.- Esri, the global leader in location intelligence,...
Citadel Releases AI Software that Automates CUAS Protection for Critical Infrastructure
Rating12345SAN DIEGO- Citadel Defense has released new AI-powered software...

April 15th, 2012
Collaborative Visualization to Advance Landscape Planning

  • Rating12345

sheppard steven thumbRecently the University of British Columbia unveiled a decision theatre, an interactive and immersive computer visualization lab for collaborative advancement of landscape planning. Special correspondent Matteo Luccio spoke with Stephen R.J. Sheppard, the project lead and professor in the Department of Forest Resources Management at the University of British Columbia, Canada. Sheppard is also the author of Visualizing Climate Change: A Guide to Visual Communication of Climate Change and Developing Local Solutions, which was just published by Routledge at the end of March. In this interview, Luccio delves into the details about the Decision Theatre, including both the technology and the approach.

sheppard stevenRecently the University of British Columbia unveiled The Decision Theatre, an interactive and immersive computer visualization lab for collaborative advancement of landscape planning. Special correspondent Matteo Luccio spoke with Stephen R.J. Sheppard, the project lead and professor in the Department of Forest Resources Management at the University of British Columbia, Canada. Sheppard is also the author of Visualizing Climate Change: A Guide to Visual Communication of Climate Change and Developing Local Solutions, which was just published by Routledge at the end of March. In this interview, Luccio delves into the details about the Decision Theatre, including both the technology and the approach.

ML: When did you start working on this project? How long did it take you to set it up? When did it go live?

There’s two aspects, really: there’s the actual visualization work and then there’s the decision theater as a place to present that and engage people in it. We’ve been working on these kinds of engagement approaches, using geospatial and visualization tools, for more than a decade. We’ve been applying it to sustainability and particularly climate change for at least the last seven years. The official name of the decision theater is the BC Hydro Theater at the University of British Columbia. We officially launched it in November 2011.

ML: When you went live, did you have a party?

Well, it was quite a big deal. We had a launch. The whole building was being opened up, so it was a combination of the theater and the larger building. It is a four-story building, it is the greenest building in North America, at least at the moment. It lives largely within its own footprint in terms of energy and water and carbon. When we did the theater opening, we had a number of different demos running in the theater — we had big screen demos, we had touch-screen demos, and we were running UBC’s campus 3D model and showing people how that worked.

ML: On what variables, questions, and issues does your visualization focus?

The work we do uses 3D visualization, by which I mean perspective three-dimensional views of, usually, real landscapes. Sometimes those are very realistic pictures that are fully modeled in 3D, sometimes they are more simple. We use things like Google Earth, Sketchup, sometimes even Photoshop, as well as 3D modeling programs like Visual Nature Studio. What we are showing, the content, is usually how communities will change in the future, what their different development paths might be. So, we usually start off with existing conditions.

The questions and issues that we are looking at really have to do with how a community will plan its own future. We find that most people don’t think very systematically about their own future. They have never seen a picture of where they live in the future and that’s one of the things we do. So, we call the decision theater “a window on the future.” Users could be planners, researchers, members of the public, city councilors. We want them to be able to immediately see the outcomes of different policy choices and what their community would look like. We might have up on the big wall four different views of what their community could be in the future. This is usually a bit of an eye opener, because, they haven’t seen that before and we’re confronting them with the kinds of choices they actually have to make. What would happen if they kept going on their current path? What if they went very green? What if they brought in renewable energy or electric vehicles? What would happen not only to the landscape but also other variables that may not be as visible? Things like carbon footprints, the number of jobs, the population, or ecological health. We try to use the visualizations and mapping as well as standard media, like charts and graphs and the outputs of various models to explain the tradeoffs.

ML: What are your data types and sources?

Most of our data is spatial. We start with a GIS database, usually, and we most commonly work at a community scale. We get a city’s database, if we can, or assemble information from various sources — land use mapping, topography obviously, vegetation cover. Increasingly, we have LiDAR data, as cities acquire it, and we have very precise mapping and visualization. Also, whole series of socio-economic data that we might be able to get on things like population or census tract. Then we use various models — usually spatial models that are already out there, or that we can develop simply in GIS. So, this is not a single system, it is not a single set of software, it’s not a single model. It’s a hybrid of whatever information is relevant and available. We pull it together in GIS, usually.

ML: What geographic extents does it cover?

Sheppard: Sometimes it’s regional-scale and occasionally even larger, because, if we are using Google Earth or GIS, we can cross over the various scales. We work all the way down to the neighborhood or the block, so that people get a sense of what it’s like to live there. We’re really interested in the cognitive information — the numbers, the facts, the science — but we’re also interested in the qualitative information — what it looks and feels like. We try to make this experiential. What’s it like to walk down the street? What’s it like to live opposite this new land use development or this new district energy plant? That’s the kind of thing we want to get at. We want to be able to combine the numbers and the feeling of being there.

ML: What were the key challenges in setting up the system?

It is really more the process. We call it local visioning or climate change visioning, depending on what we are focusing on. If we are visualizing, let’s say, a small town in the future under climate change, then we have to assemble a lot of information that’s never been assembled before. One of the key challenges for that, particularly with climate change, is getting hold of down-scaled climate change projections. One of the things that most communities don’t currently know is what is going to happen to their climate. Are we going to get more rain? Are we going to get floods? Are we going to get more heat waves? What’s going to start happening out there? That kind of information is only now starting to become available.

Very often, we have to go beyond the data. We’ll get whatever data we can from the climate scientists — regionally or sometimes even nationally — and try to get them around the table with us and with the local stakeholders and the planners, so that they can describe for us, sometimes verbally, what they think might happen or different assumptions that we could play out. So, one of the key challenges is getting local climate change projections. They are starting to be more available on the regional scale. But they always require interpretation and they never go as far as you want them to go, because they don’t tell you everything you need to know about forest fires or heat from buildings or transportation patterns.

Usually, there are other models that people have, locally, or at universities, for example, that we can tap into to fill in some of those gaps or at least suggest the way some of these things work. It is messy, but we’ve developed some guidelines for communities that want to do this and we try and follow them when we are developing a project around an area, a region, or a city.

ML: Of course, the more local your focus is, the greater the variability and the unknowns.

In a way, that’s true, because there’s lots of data at the national level, but that’s not data that an engineer, a planner, or a homeowner can use. The projections may tell us that global temperatures could go up by 3 degrees over the next 50 years and we can make these assumptions with a climate change model, but that doesn’t really help you locally.

So, for example, we’ve done a lot of work in North Vancouver and the North Shore on snow pack. You may have seen some of the visualizations showing how the snow pack in the spring time will shrink because of warmer temperatures and a lot more precipitation coming as rain instead of snow. In that case, we had pretty good data and we could map on the mountains the average elevation of the snow pack on April 1. There’s a little animation of the time sequences — “time travel” we call it — that allows you to see the snow line going up the mountain, always on April 1, but in 2020, 2050, and 2100. That’s very powerful, because it forces you to confront what before you only imagined.

Now you can actually see it on a place you know and love. You can see it in relation to Grouse Mountain or to the valleys where the reservoirs are. That snow pack is the main source of the water supply for the whole of the lower mainland for the entire summer. We’ve done research to see what the impact of those images is. They have been used countless times. They do seem to have an impact on people’s awareness, their understanding, their sense of urgency about what are we going to do about climate change locally.

ML: I assume you have some disclaimers about the uncertainty levels.

Sheppard: Yes, that’s very important and that’s one of the things that does often get missed. When the media use it, we are always worried. We usually try to package the visuals with some information that describe the scenario. What year is it? What set of assumptions went into it? If those stay with the image, then it’s OK, but if not, it can be open to misinterpretation by people. It’s pretty important to show alternative scenarios, because these are not predictions of the future. We can’t say which of these things will happen or what else will happen. Our job is to try and suggest several different possibilities, deliberate choices, or outcomes as to the future conditions that a city or community might have to deal with. How will you prepare for these things? What can you do to make this better? It’s up to them to make those kinds of choices. It is really important to always describe the uncertainty in the models and in the assumptions that go with the models. This is certainly partly science — we try to base it on the best science we can — but you inevitably have to go beyond the science to make plausible stories, really. These are really stories in 3D of what could happen if a certain set of assumptions kick into play.

ML: Going forward, how do you plan to develop and adapt the system?

Sheppard: What we would like to do is make this whole system more interactive, more real-time, and give people more access to twiddling the knobs themselves and seeing what happens. We’re trying to enable this so that they feel like they are walking down the street while they are doing that. We are not all the way there yet. Through Ron Kellett’s work at UBC, we have some of these live tools for moving buildings around, for example, or changing different land uses within a city. We’d like to expand that to larger scale work, where we could play out the impacts of things like forest management or bioenergy harvesting in a forest.

We’ve got the base models, we’ve got data. We’ve actually done this for a couple of communities in British Columbia, for example, looking not just at the city but at the land uses around it, because they are all interdependent. So, things like food, or energy, or water. We’d like to be able to say that if you develop in that part of the watershed, you are going to reduce water supply or increase flooding risk. Or, if you get a pest outbreak in the forest, how that is going to affect the flooding risk downstream in the city. Those kinds of issues.

ML: Have you experimented with inserting some economic variables, such as the cost of environmental mitigation?

Sheppard: We’re starting some of that work already. Our research group, the Collaborative for Advanced Landscape Planning (CALP), is really an interdisciplinary group. We combine expertise in visualization, GIS, and spatial analysis, but also landscape planning, land use, urban design, and a range of the natural sciences. We package that with some environmental psychology approaches to how people perceive and how they interact with visual interfaces. Then we pull in other researchers, to help to generate numbers or bring their models into the theater, so that we can connect them to some of the other work that is going on. So, we have a bit of a hub function.

We might bring in an economist, we work a lot with engineers, various kinds of modelers, hydrologists, foresters, architects, people like that, depending on what the key issue is. For example, we are doing right now a basic assessment of flood impacts and adaptation strategies in South Delta, which is in the Fraser River flood plain and very vulnerable to sea level rise. We’ve been playing out four different scenarios — options like raising the dikes vs. just raising the buildings and letting the sea come in occasionally vs. managed retreat away from more vulnerable communities and farmland. Then we use GIS to generate basic facts and figures on things like the length of infrastructure affected or the land values of the farmland or properties.

In most cases, we developed these scenarios not as experts — it’s not a top-down process — but jointly with the affected communities and our stakeholders. That’s true also of the visualizations: we don’t just drop them on people, we have a participatory process, to guide and recommend where and what to visualize and then to evaluate the visualizations before we take them to a wider public.

ML: Have you worked with people in Hollywood or at Dreamworks in terms of the high end of visualization and imaging?

Sheppard: We have been playing a little bit with video games and using things like the Unity game engine, to try to turn the South Delta data set into a game that local people can play themselves to explore things like different levels of flood risk or adaptation to flooding. At the same time, they could use it to look at some of the energy implications for their communities: what would their carbon footprint be and how could they reduce it? It could also be used in the schools. We are very keen to collaborate with the higher end video game developers and movie makers — Vancouver is a big center for that. We are really very interested in personalizing this scientific data and making it fun for people but also keeping it instructive at the same time and maybe trying to engage people who aren’t typically involved in things like planning.

ML: Who are the users of your system?

They can be quite varied. In a typical community visioning project — where we are trying to get a community to think about climate change implications and to start planning for it — the users would certainly include city planners and engineers. Also, the city council, in terms of presenting some of the outputs and some of the key conclusions or getting them involved in some of the development workshops along the way. Then stakeholders, members of the public. We work with farmers, landowners, developers, local businesses, recreation providers, forest and park users, representatives of local housing associations. We’re trying to get a mixture of users, so that they can pool their knowledge, as well as carry the information back to their stakeholder groups.

ML: What interests participants the most? Do local government staff and local residents typically have different concerns?

Sheppard: There is usually a lot of overlap. If it’s around climate change, the kinds of people we’ve got so far have had a significant amount of concern. The average Canadian has quite a bit of concern about climate change, a little less so in the United States, on average. Many people are concerned about it, but they don’t know how it will impact them.

I think the biggest thing that we’ve seen is that people are actually very interested and very curious about what is going to happen in their area. What are the possibilities? Are we at risk of flood? How soon will sea level rise impact our neighborhood? Does forest fire risk increase? Can we get solar energy in our area?

There are some pretty good applications now for answering such questions. Sometimes we are telling them about things they haven’t thought of. A lot of people don’t think very explicitly about their carbon footprints or recognize that there is this massive flow of carbon dioxide through their community every day, all the time, all totally invisible. We are trying to use the visualization capabilities to sometimes make the invisible visible. So, things like energy, heat loss from buildings or carbon dioxide flowing through the community, the pipes underground, the emissions coming out of the chimneys. We try to make that much more visible to people. They tend to be concerned about the impacts of climate change, stuff they’ve heard about. They tend to think less about energy sources or energy prices.

The staff and the engineers usually have much more precise questions and it can be quite frustrating for them because they want to know, well, how high should we plan for sea level rises? Is it a meter? Is it 1.2 meters? Is it 5 meters? While there are no definitive answers to those questions, only ranges of possibilities, guidelines are beginning to emerge. Developers like to know where the areas of risk are, so that they do not, for example, build high-end condos in places that might have liabilities.

ML: Do you survey the users pre- and post-visualization to see how it affects their knowledge and opinions?

Sheppard: That’s one of the methods we use. If we are doing a full test of people’s responses, then we will ask them to come in for a two-hour workshop. We’ll do a pre-visualization survey and then we’ll take them through a workshop, sometimes it is a PowerPoint presentation, sometimes it is live Google Earth, and walk them through some of these stories and scenarios and take their questions. Afterwards, we ask them to fill out a survey again.

Some of the questions in that survey will be identical to the questions in the first survey. That’s one way to track what they’ve learned, whether they’ve changed their attitude on anything, and also just how well they thought the visualizations conveyed the facts or the issues. Quite often we get statistically significant increases in awareness, sometimes in motivation and support for policies. That’s the kind of stuff that local politicians really need to know, that people are willing to make changes in some areas. Without doing these kinds of surveys you’d never be really sure.

If readers want to learn more about how local communities can deal with climate change using these and other visualization approaches, the techniques are summarized in my new book Visualizing Climate Change: A Guide to Visual Communication of Climate Change and Developing Local Solutions, published by Earthscan; see

Leave a Reply

Your email address will not be published. Required fields are marked *