September 16, 2013

BY KEVIN MERRILL

A national land-use study chaired by SNRE Professor Dan Brown today is calling for a new generation of modeling techniques and even more interdisciplinary and data-driven approaches.

The National Research Council carried out the study to improve the science around land use modeling, which examines the vegetation and buildings covering the land and humans’ impact on it. The committee set out to survey current land-change models, suggest how to use them better and more collaboratively, and describe ways to improve their integration. Brown, an international expert on land-use and modeling, and a professor in SNRE’s Environmental Informatics field of study, was selected to chair the work.

“The time is ripe to envision, plan for, and invest in the next generation of land-change models for an increasingly interdisciplinary scientific enterprise that takes advantage of the best available knowledge, data and computing resources,” the report states. “If appropriately planned and executed, the next generation of models can be increasingly process based, link processes in social and natural systems from the parcel scale to regional and global scales, and make use of better methods for process validation, in order to enhance both their predictive skill and their utility for policy analyses.”

According to the report, changes in how land is used by people (land use) and in the vegetation, rock, buildings, and other physical material that cover the Earth’s surface (land cover) can be described and future land change can be projected using land-change models (LCMs). LCMs are a key means for understanding how humans are reshaping the Earth’s surface, forecasting future landscape conditions, and developing policies to manage resources at a range of scales, from individual parcels of land in a city to vast expanses of forests around the world.

The report is titled “Advancing Land Change Modeling: Opportunities and Research Requirements.” The U.S. Geological Survey and National Aeronautics and Space Administration commissioned the study.

U.S. Landsat satellites have provided a 40-year record of global land cover change, providing input to LCMs, yielding new scientific insights, and informing policy on issues from agriculture to regional planning and disaster relief. But recent growth in the number and types of new land observations and monitoring data, modeling approaches, and computational infrastructure has ushered in a new generation of LCMs. These new modeling approaches come with different strengths, weaknesses, and applications. As a result, greater need exists for evaluating and forecasting these impacts as well as the data available.

The study, begun in 2011, provided guidance on the strategies, data, and research requirements needed to enhance the next generation of models. In particular, it assessed the analytical capabilities and science and policy applications of existing and emerging modeling approaches; described the theoretical and empirical basis and the major technical, research, and data development challenges associated with each modeling approach; and described opportunities for improved integration of observation strategies (including ground-based survey, satellite, and remote sensing data) with land change modeling to improve land change model outputs to better fulfill scientific and decision making requirements.

Brown spoke about the committee and its report in an interview:

Q: What groups are behind this report and why did they create your committee? What was the issue they wanted you to address?

Brown: The funding agencies are the U.S. Geological Survey and the National Aeronautics and Space Administration, both of which have been engaged in modeling activities, some of them related to carbon cycle. USGS has a big carbon program and NASA’s interested in combining the data that they generate with models.  And they do that across all the different fields they work in: atmospheric science and geology, for example.  And so one of the fields they work in is land use and land cover.

There are a lot of models getting used in the scientific literature and in applications, and there doesn’t seem to be a very good understanding about the models and which ones are proliferating and why.  And people don’t really quite understand them all.  And so the idea was to get a handle on what types of modeling approaches are out there, what they’re good for, how they interact with different kinds of data, and what kinds of research needs there are in order to advance modeling as a tool in understanding land systems.

Q: Even though it is funded by two U.S. agencies, these are global questions, not U.S. questions?  Or are you looking at it more from a U.S. perspective.

Brown: Well, it’s a method, so it doesn’t matter. We actually have a member of the committee from the Netherlands.  So there’s experience on this from all over the world.

Q: So is it more than just an inventory of modeling forms?

Brown: Right.  It includes future directions for research, but also guidance on best practices, creation of an inventory and descriptions of different modeling approaches in the context of application: what they’re good for.

Q: What did you learn through this process that you did not know already?

Brown: I think there’s going to be some really useful distillation and that the report will be a really valuable resource.  NRC reports can have a big impact.  My hope is that it will have an impact; researchers will read it and will hopefully refocus the science of modeling on the questions addressed in this report.

My hope is that the report places modeling in the context of what it’s good for, and help researchers in the community think about that as they are working on modeling and guide them in developing models toward specific kinds of applications in a more deliberate way. I also hope that it captures the attention of some of the funding agencies.

Q: And they may begin requesting those in RFP?

Brown: Exactly. And in fact I know that’s part of why NASA was interested in this study because they do fund research in this area, and they’ll be looking to it for ideas as to what they should be doing. 

Q: When the average citizen thinks of land-use modeling, should they think of it as a mathematical formula? A software application? A philosophical approach?

Brown: It’s all of those things. But when it comes down to actually doing it, you could do it mathematically, and that’s the way it sort of had been done.  But when we start talking about complex environmental systems, it’s essentially computer code: a software application that you write or you download and you use it to simulate things in the world.  So they’re basically simulation platforms, not just some sort of broad conceptual models or even simple and mathematical models.

Q: And right now we have kind of a Baskin Robbins approach to modeling?

Brown: In this field, there’s a bunch of different approaches and they sort of cluster around a few sort of broad approaches. And everybody is sort of kniddling away on making one sort of minor advance in this one approach, which really doesn’t do much necessarily to advance the whole field. 

But I think the bigger problem is that people tend to learn an approach and use that for everything.  And so we have different sorts of intellectual traditions coming together.  You have the computer scientists with maybe a cellular atomic automata approach, and the economists coming an econometric approach. They are starting to converge on this same question, because it is an interdisciplinary problem.  So you have people coming at it, but because of the different approaches, they can’t speak to each other.

Q: Is there a Rosetta Stone somewhere in your report?

Brown: I hope that it lays out a sort of roadmap to the terrain: this interdisciplinary terrain, that will help people talk to each other.