August 2

Contents

Editor's Introduction

Last week, I proposed the concept of public GIS centers. [Due to technical problems with our new mass e-mailing system, last week's issue went out late. If you have not yet read my proposal, I suggest that you click on the link above before reading further.] This week, I discussed related issues with two GIS professionals who endeavor to make the tools of their trade available to as many people as possible: Breece Robertson, National GIS Director for The Trust for Public Land, who helps communities develop and use GIS models to identify conservation priorities as a guide to planning and for the protection of parks and natural resources, and Gina Clemmer, the founder and president of New Urban Research, an organization that trains nearly 10,000 people a year to use ArcGIS to research and document community issues. Both are national projects that empower local governments and communities to gather, analyze, and display local data using GIS. In this issue, I also bring you one reaction to my proposal and 26 press releases.

— Matteo Luccio

Greenprinting: A conversation with Breece Robertson, National GIS Director, The Trust for Public Land

Fifteen years ago, after college and a couple of years of work as an exercise physiologist, Breece Robertson wanted to do something more related to conservation and sustainability. So, she earned a master's degree in geography and planning at Appalachian State University in North Carolina, then went to work nearby for the Yadkin-Pee Dee Lakes Project, which helped to create tourism as an economic development tool in a 15-county region. Next, she got a job at AllPoints GIS in Boulder, Colorado, teaching mostly to forest service and local government employees how to use ESRI software as well as CommunityViz — an application designed by the Orton Family Foundation to bring communities to consensus on land use planning. For a year, Robertson helped develop the training curriculum for CommunityViz and traveled around the country teaching it.

In 2001, Robertson accepted a position with The Trust for Public Land (TPL) in Santa Fe, New Mexico, as its first full-time GIS specialist. Her first assignment was to assist Los Angeles County bring together GIS data and layers using CommunityViz and develop a greenprint — which is an interactive, community-based process that uses GIS models to identify conservation priorities as a guide to planning and for the protection of parks and natural resources. Since then, the greenprint concept has grown within TPL, which is now building a suite of services around it.

A greenprint, Robertson explains, uses datasets that reflect a full range of benefits from land conservation. "With the greenprint," she says, "we are able to go into a community and identify all the things that are important to it for conservation. Often it is a lot more than just parks and open spaces for recreation use — it is looking at lands you can protect for water quality, wildlife habitat, or floodwater management."

Soon, word spread about TPL's work. The Department of Natural Resources for King County, Washington asked the organization to create a greenprint model to help its five programs achieve consensus as to which lands it should protect, based on their programmatic goals. "It really helped all of those programs agree on land acquisition priorities," Robertson recalls. "They found that they were competing for funding for similar pieces of land. The greenprint helped them identify land that all five programs wanted to acquire, based on their common goals." Five years later, she says, the DNR programs are still using greenprinting that way.

Since 2003, TPL has completed more than 45 greenprints for local governments across the nation, mostly for counties, but also a few for multi-county regions, including one in process now in Northern Ohio that covers 20 counties. Robertson now has eight full-time staff working on greenprints.

By developing greenprint models, TPL helps local governments and communities create a conservation vision. "It is much more than a mapping exercise," Robertson says. "It is really using the GIS as a decision-support tool." The first step in greenprinting is a stakeholder process, in which TPL brings together ten to thirty key people in the community to discuss local conservation priorities. "We help them identify which ones of those goals are mappable — that is, what we can plug into the greenprint model, which ones of those goals might take some sort of a regulatory approach from the local government — and then how you need to educate the community to reach those goals," Robertson explains. TPL then takes those goals and creates a greenprint model that reflects them. "The final product of the greenprint model is a priority map that drills down to the parcel level and shows the community or the local government which parcels rank the highest on those goals."

During the first stakeholder meeting, TPL also identifies a technical advisory team, consisting of people who understand the local data and scientific issues. They help TPL understand what has already been done in the community with regard to a given issue, identify available datasets, and decide how to use them in the models. The team also helps with quality control —making sure that the final maps reflect local knowledge.

During the final stakeholder meeting, participants review the maps and, using weighting tools straight out of the ESRI software, play around with the weights assigned to different priorities — such as wildlife habitat, water quality, or trail connectivity — and see immediately how this changes the maps. The goal is to come to consensus on one priority map that local government can use to guide its land acquisition. "We can create the model and then deliver it to the city or the county or a local nonprofit," says Robertson, "and they can have the model there live, so they can re-run it or change the values of variables."

"We created some custom tools that allows us to drill into the top priority parcels," Robertson explains. "It pulls out every single goal that went into the model and it associates either an acreage number or a percentage number for that parcel that reflects those goals. So you can start to see exactly why the parcel is important."

All of this, Robertson explains, ties into another important TPL focus: conservation finance. "That is where those weightings come into play. For the next two years you might have a million dollars for land acquisition for water quality. After that money runs out, your community might pass a bond measure for parks and open spaces, so you can shift those weightings, which shift the priorities, to better match those goals."

TPL uses ArcInfo 9.2 and switched from CommunityViz to Model Builder when ESRI released it as part of ArcGIS 9.0. "It is mainly because we do our analysis in raster," says Robertson, citing the saying that "Raster is faster, vector is correcter."

In their models, TPL's GIS staff often have to crunch together up to 50 datasets, which takes too much memory and processing power to do in vector format. "So we convert everything to raster and we run it through Spatial Analyst using Model Builder," she explains. On top of that, TPL built Visual Basic tools, such as Model Investigator. "It is like a little extension that our partners can load onto their computers."

To run the models, users need to understand ArcGIS software and Spatial Analyst, but they don't need to be proficient in ArcInfo, because they can run them on the much simpler ArcView. "We make sure that they are transferable at that level, because we don't want cities and counties to have to purchase additional software and send staff to additional training," Robertson says. TPL also provides user manuals, training, a support contract, and an ArcIMS Internet mapping site solution. "If a community does not have any GIS expertise, we can provide all the results from the greenprint through an ArcIMS site. It can be password-protected for the stakeholders or you can upload the maps to a public site and allow the users some type of functionality. It is much easier for them and they don't need any GIS skills."

Teaching GIS to Community Activists: A conversation with Gina Clemmer, President, New Urban Research

Gina Clemmer first studied GIS in the early 1990s at the University of Iowa, as part of a masters' degree in urban planning with an emphasis on low income housing development. "At that time, GIS was an emerging tool in urban planning," she recalls. "The technology was not very user-friendly and it was extremely expensive to get the hardware and software, the plotters to prints maps, and so on." However, a very forward-thinking professor introduced GIS into the planning program.

"Immediately," she says, "there was a rift between the qualitative researchers and the quantitative researchers, for whom technology started to be the sub-text. I fell in the second camp and really believed that technology was a great tool to empower social activists. That was what brought me to urban planning: to make the world a better place." She discovered that she could use GIS to visualize changes in communities and to locate pockets of poverty and dilapidated housing stocks. "I got really jazzed up about that and thought that GIS was just the best tool to do that sort of thing."

Clemmer then moved to Portland, Oregon, where she had a summer internship with the Portland Development Commission, worked briefly for a nonprofit called Network for Oregon Affordable Housing, then got a job with the city's Bureau of Housing and Community Development. "BHCD was my dream job," Clemmer says, "because it is so community-oriented and because Portland had a lot of money to make change happen. They needed a data analyst and someone to do research for them. I was hired to do quantitative research on neighborhoods and be their technology person."

Clemmer began to receive many requests to help people find data — especially from nonprofits that had to report to BHCD and to federal agencies the number of people they served, the demographics of that population, the extent to which their intervention had reduced poverty in an area, and how much money it had taken to effect that transition. When the Bush administration took over from the Clinton administration, Clemmer recalls, it cared less about community stories and innovative ideas and mostly about how many people an organization served. "That single statistic was really elusive to many nonprofits. They would ask for technical assistance to figure out how to get data and how to present it and I handled a lot of that."

After about three years at BHCD, Clemmer decided to continue her work on her own. "At that time I was thinking about pursuing a Ph.D. in social research, but then I decided that I would start a business instead." A few months ago it became an ESRI business partner, so that now it can sell software. "I had in mind a place where average folks who were working in communities of color, with poor people, with vulnerable communities, would be able to access tools of the dominant power structure — at that time, it was PowerPoint, data, percentages, and then eventually GIS — and be able to articulate their effectiveness to the people in charge. I wanted to help people do that and that was what spurred me on to start my company."

For three years now, Clemmer's company has been teaching a one-day GIS class. It is now taught in about 50 cities to nearly 10,000 people a year — mostly GIS beginners working for municipal agencies and nonprofit social service groups. About one third of them work on health-related projects — such as mapping out the locations of women who have cancer and of mammogram vans, in order to ensure that women, especially poor ones, have access to that service. Clemmer finds this very gratifying. "We give people tools to empower them to make the world a better place," she says. "GIS is one tool in that toolbox."

The company's GIS class is "very fast-paced," says Clemmer. "People must have pretty good computer skills, but they don't need to have any GIS skills." The class consists of seven exercises, all done using data for the city in which it is being taught:

1. Creating a basic reference map using shapefiles, changing their color, labeling them, and creating a basic reference map of the city.
2. Downloading Census data directly into ArcGIS.
3. Cleaning up data using Microsoft Excel files, which can be imported directly into ArcGIS 9.2.
4. Importing and joining data to create a thematic map.
5. Importing and geocoding addresses.
6. Doing location queries, attribute queries, and custom selection.
7. Critiquing the newly created maps and discussing what it means to make a good map.

At the end of the class, students get a workbook with all of the exercises, a copy of the slide presentation, some handouts, and a 30-day subscription to New Urban Research's website, which has about 20 downloadable shapefiles for each U.S. county. This allows them to access all the files used in the class and go through the exercises again. The company then orders a 60-day free trial version of ArcGIS for each student.

New Urban Research also teaches a separate class on Census data, which it is now restructuring to include other data resources. "In the GIS class," Clemmer says, "we give you just enough information [about Census data] to be dangerous: where to get it. We don't talk about data accuracy rates or methodologies." The Census class does not use GIS at all.

The ability to use Census information is vital for social researchers and activists, Clemmer explains, because all federal funding streams are tied to it. Census data on such variables as age, race, ethnicity, language, commuting patterns, per capita income, and who receives public assistance are a great starting point for any sort of community analysis. "People who take our class generally want to get basic demographics about a particular area, such as a neighborhood, a school district, or a crime zone," Clemmer says. GIS then helps them compare neighborhoods: "You need a visual model to be able to see spatial trends in that information, to see how a community stacks up to other ones. With data in a table or a chart it would take you a long time to arrive at the same conclusion you can draw from a thematic map at a glance."

GIS, Clemmer says, is now much more accessible than it used to be to the average person who wants to map out something. Her students range in age from 16 to 75 and an increasing number of them are women. "Part of GIS is data management, which is a whole separate set of technology skills: working with Microsoft Excel, in a database environment, or through Internet Explorer, learning how to unzip files, and so on. GIS analysts spend only half their time using GIS software."

According to Clemmer, the biggest barrier remaining to widespread use of GIS is cost. "Most agencies cannot afford to buy a GIS server to get all their users using it," she says. "So, typically, one or two people, who are usually the most technology-driven or have the biggest need, end up getting the software and trying to teach themselves. Then everyone in their organization begins to direct requests for maps to them."

Often, users are not aware of data that is available at no additional cost, Clemmer points out. She cites the example of the State of California, which has bought its data from TeleAtlas and makes it accessible to all its employees. "I still get California state employees who come to my class and don't know that they can get free shapefiles," she says.

Consistent with her effort to further democratize this technology, Clemmer is enthusiastic about the possibilities for Web-enabled GIS, which she sees as "the wave of the future." She envisions an online service, available to anyone for a $100 annual subscription and accessible through a Web browser, which would automatically create maps from user data. Instructions on the site would guide users in preparing their data (for example, an Excel table with poverty rates by neighborhood), uploading it, and selecting a map type from a series of templates. The system would then automatically geocode the data as needed, join it to the appropriate shapefiles, run spatial queries, and create maps, ready for users to download. "I think that would be just awesome," she says. Such a system, she hastens to add, could only handle the most common GIS tasks and would not include very detailed spatial analytical tools. Perhaps, she muses, such tools could be available for a higher subscription fee.

Clemmer readily acknowledges that a system like the one she envisions represents "the lowest common denominator" of GIS needs and capabilities. "It dumbs down the software, to a certain extent. It takes out all the bells and whistles and just focuses on the most primary tasks — which, for most of the people I work with, is thematic mapping, geocoding, and some spatial queries."

Does Clemmer worry that people with very little training in GIS could cause mischief by using geospatial tools improperly? "Even with four years of college training people get into a lot of trouble!" she points out. "There is a big potential for people to get it wrong. A common mistake I see is that people often map out raw numbers in thematic maps. Those numbers are relative reflections of population in those geographies. There you should be taking a percentage. But, my experience is that people love to show their maps off and when they display them to their peers, their peers will very quickly tell them if they don't make any sense."

Letter to the editor

A reader, who does not wish to be identified, had the following to say about my proposal for public GIS centers:

Public GIS centers? What a dumb idea! In case you hadn't noticed, 90 percent of the people in the world just don't get maps. Of the remaining 10 percent, 90 percent are unable, by nature or training, to apply analytical thinking to maps or spatial problems. Even people with knowledge of GIS — if they do not routinely apply their skills — will flail around on basic operations, much less perform a complex analysis efficiently. Joe Homeowner would be at a complete loss in the centers that you envision.

GIS is, and will remain, a field in which skilled practitioners (much like architects or engineers) provide services to individuals, agencies, companies, and the public. It takes constant practice and long experience to create usable maps and analyses of spatial issues, especially those that involve large segments of the public. It takes even more skill to facilitate stakeholder engagement based on or using such materials. We would be much better off upgrading the standards of and expectations for professional GIS practitioners, and preparing them for public interactions, than wasting resources on hardware, software, and facilities that will not be effective.

» Back to our GIS Monitor AUG 2007 Issue