The Berkeley Science Review: Articles

Location, Location, Location
WebGIS puts science on the map (view PDF)
by Tim De Chant

Maggi Kelly, Associate Professor of Environmental Science, Policy, and Management (ESPM) at UC Berkeley, was sitting at her desk staring at the coast live oak just outside her window. Reports had been pouring in from Marin County of a mysterious new disease that was attacking these icons of the California countryside. At first one by one, but then by the hillside, this disease had been draining the coast live oaks of their deep and luscious greens, their leaves rapidly giving way to yellows and browns as the plants died. The disease, called Sudden Oak Death (SOD), was threatening to turn California’s oak woodlands into relics of a distant era. Kelly knew she had to get the word out about this potential environmental catastrophe. More importantly, she knew she had to get the word in.

Getting the word into webGIS
Kelly’s brainstorm resulted in an early and innovative application of internet-enabled geographic information systems (GIS), also known as webGIS, which are being increasingly utilized by researchers across the UC Berkeley campus and elsewhere. With SOD moving rapidly, Kelly needed a way to quickly disseminate new, spatially explicit scientific information about the disease to get the word out, as well as to gather information from the public about the locations of new infections—to get the word in. Her solution was a webGIS called OakMapper (oakmapper.org). California’s oak woodlands occupy 10 percent of the state, making traditional scientific monitoring difficult at best. OakMapper allows concerned members of both the scientific community and the public at large to help monitor the spread of SOD by logging on and inputting the locations of diseased trees. Like the first telephone call across the Atlantic, OakMapper has helped to bridge the communication divide between environmental scientists and the public, vastly improving the speed and efficacy of those relations.

A type of software widely used in the environmental sciences, geographic information systems allow geographically referenced digital data to be stored, displayed, and analyzed; what Ken-ichi Ueda describes as “computerized maps, but better!” The “web” in webGIS simply means the “GIS” is going to be hosted on the Internet. Ueda, a graduate student in the School of Information and veteran GIS and webGIS practitioner, explains: “If you had a [digital] map of the Bay Area with layers of water features, streams of the bay, and topographic information, GIS would allow you to generate hydrological models and say, ‘If we get X amount of rain this year, we’ll see these kinds of runoff effects in different parts of the Bay.’ That’s the kind of interesting quantitative analysis that you can’t do with traditional maps... Or you could, but it would be a lot slower.”

The great thing about webGIS, Kelly says, “is that the client, the person using the computer, doesn’t have to have special, expensive software on their computer—they just have to have an internet browser.” The benefits of this distributed framework are three-fold: Most webGIS applications are free, relatively easy to use, and broadly accessible. Since its inception in the 1980s, GIS has been an expert-driven establishment. By simplifying the user interface for the web, webGIS designers have opened up the field of GIS to the untrained layperson. As Kelly puts it, “WebGIS says, ‘Hey, we need information from you.’” And that’s exactly the model Kelly followed when she and ESPM graduate student Karin Tuxen developed OakMapper.

OakMapper is a prime example of the how webGIS can facilitate extensive and distributed data collection. “The audience is really diverse,” says Kelly. “Everyone from scientists to government officials to the public at large has used OakMapper in some capacity. I think it’s woven into the management community for the disease. That’s really exciting,” she adds. But the real impact of public monitoring has been to guide researchers to some areas where SOD was not previously known to exist. A goal of the OakMapper project, according to Tuxen, is “to harness the manpower of the general public to be more eyes and ears for the researchers.” Those eyes and ears have aided scientists by scouting locations for further testing, fostering a better understanding of Sudden Oak Death.

Viva las Google Maps
Since 2000, when OakMapper was released, webGIS has positively exploded, thanks in no small part to the successes of Google Maps and Google Earth. Google’s pervasiveness in the field has led some to oversimplify the definition of webGIS as “Google Maps.” More product than definition, Google Maps is also more web mapping than it is true webGIS. True webGIS (which is not a brand of software but simply a genre of applications) enables users to ask open-ended questions, like Ueda’s runoff example. That is not, however, meant to slight either of Google’s offerings. “I can’t overestimate how much Google Earth and Google Maps have raised the visibility of mapping and webGIS in particular,” says Kelly. Google Earth, according to Tuxen, has “enough of the cool factor and the utility factor so that millions of people around the world have downloaded it.”

Part of the appeal of Google’s mapping applications is the vast amount of data they make both readily available and interpretable. “There’s all sorts of interesting geographic information floating around on the Internet, in people’s databases, in government storage facilities,” says Jess Lee, product manager for Google Maps. “What we’re trying to do at Google is make it so that all that content is universally accessible to anyone and organized so it can be searched easily.”

Google Maps has done far more than simply raise the visibility of the entire field—it has enabled developers to create even more sophisticated webGIS applications. From his home base on a 320-acre farm in the scenic Oregon countryside, John Deck, programmer analyst and webGIS guru for the UC Berkeley Museum of Invertebrate Zoology, administers and develops the BerkeleyMapper application (berkeleymapper.berkeley.edu). Hosted by the Berkeley Natural History Museum, BerkeleyMapper was originally created to map the locations of origin of natural history specimens. “It started simply,” Deck says. “I think the first version of BerkeleyMapper was a tool to highlight the county a specimen was located in.”

Now the program integrates multiple sources of data, allowing users to ask more complex and spatially explicit questions. For example, a user can search the Natural History Museum’s records for bald eagle specimens collected in California, and output those results to a digitized map; this map can then be panned, queried, and even overlaid with ecological boundaries. This expansion of BerkeleyMapper’s capabilities is due in part to Google Maps. “When Google Maps came out,” Deck recalls, “I immediately saw that it was basically an unlimited resource for geographic information.” Google Maps handled the basic satellite imagery, road maps, and geopolitical boundaries—those same layers that Deck had previously worked to supply himself—so well that he has been able to use his time to expand the scope of BerkeleyMapper, allowing for more complex analyses. The extensibility of BerkeleyMapper has inspired 11 museums and organizations to use it as the foundation for exhibiting their specimens in a geographic context.

iNaturalist and the future of museums
Spatially-aware, web savvy museum collections are an obvious first step toward bringing natural history to the masses. The visitors to such online museums would no longer be passive observers, but instead become active contributors and de facto researchers. However, museum collections do have their limits. Imagine the chaos that would ensue if visitors to a museum brought along a lizard from their backyard or a list of birds they saw over the weekend. Now imagine if they brought 50 such specimens. Or one hundred. The museum would cease to be a museum and instead be transformed into an anarchic hullabaloo of mammoth proportions. Nevertheless, while museums must constrain their collections to fit within four walls, these accumulated observations of nature could be extremely useful for researchers if collected in a webGIS.

“Bird watchers have a legendary reputation for being huge nerds,” jokes Ueda. (A self-described nature geek, he believes it takes one to know one.) His pet project, iNaturalist.org, “will be a place where people interested in nature can record their data online and share it with others,” he explains. He knows that many bird watchers keep lists of every bird they have ever witnessed, with some lists stretching into the thousands. “The life list,” says Ueda, “that’s great; that’s really cool. But that data could be really useful if it was in the hands of everybody else.”

There are hosts of scientists out there “who may want to tap that huge population of monitors who are out there making observations at a spatial and temporal density that no research project could ever achieve,” says Ueda. iNaturalist’s users could provide ornithologists with incredibly detailed records of bird migrations, or herpetologists with the spatial distribution and temporal habits of rare and endangered reptiles.

Coyote Bytes
With the success of OakMapper under their belts, Kelly and Tuxen, along with new collaborators Robert Timm of the UC Hopland Research and Extension Center, Craig Coolahan of the U.S. Fish and Wildlife Service, and Ray Smith of the Los Angeles County Department of Agriculture, are targeting another high-profile environmental concern: human-coyote encounters. Hot on the heels of the recent coyote attacks in Golden Gate Park in San Francisco, their new webGIS, dubbed Coyote Bytes (coyotebytes.org), could not be more timely. “It’s just been released in beta form.” Kelly explains. “It’s a pilot site for Southern California for participants to tell us where they’re either seeing coyotes or have experienced a coyote attack on one of their pets.” With the continual expansion of urban areas into natural landscapes, such encounters are increasingly inevitable. “That example in Golden Gate Park is what we’re trying to capture,” says Kelly. “Not in an anecdotal way, but in a very organized and systematic way so that we can understand what’s actually happening in these wildland-urban interface zones.”

Mapping of the masses, by the masses, and for the masses is not without its issues. One major problem is colloquially known as “garbage in, garbage out”. “Letting the masses map the world presents this problem of having that many people submitting to it,” Tuxen points out. “You might get a lot of garbage.” The payoff, however, is so great that most people are willing to risk that. “Whenever you get information from the public,” she says, “you may get some bad data, but hopefully you’re going to have a lot more good data.”

Mapping the future, minding the detours
With such a broad audience available to any webGIS application, it is also easy to forget that the global internet-using population fails to match the world’s population as a whole. “Part of it is the digital divide,” says Kelly of webGIS’s other challenge, referencing the gap between the haves and have-nots in the game of technology. The consequences of digital divide with respect to webGIS are particularly important: Imagine that a large corporation wishes to site a metal smelting plant in a particular location. Those with access to webGIS technologies could easily study the effects of this plant on air quality in their neighborhood, allowing them to adeptly argue against its placement. Those without access to webGIS may not have ready access to such information and analyses, hindering their arguments against the plant’s construction in their neighborhood. Furthermore, the larger webGIS community also suffers from a paucity of data in regions lacking internet access. Frog diversity, for example, is likely to be well-documented in Central Park, but rather lacking in Chiapas, Mexico. These hurdles are indeed difficult, but not intractable.

As technology becomes more pervasive, so too will webGIS. The key, then, is to eliminate the remaining boundaries. “One of the things we try to do in our work,” Kelly explains, “is reach communities in all their diversity.”

WebGIS will continue to shape the way we understand our world, regardless of the challenges it faces. Issues of privacy have begun to crop up, especially with the advent of Google’s new “Street View” that not only shows storefronts and front yards, but also people exiting the shops and mowing their lawns. While the near future of webGIS may be overshadowed by privacy-related lawsuits, the pervasiveness and abundance of spatial data all but ensure its long term survival. WebGIS-related technologies have already expanded their reach beyond the desktop and onto millions of phones and “smart” devices like the iPhone. Future advances may even merge sunglasses with feature recognition software to create an “augmented reality,” providing hands-free, real-time information about your geographic location. Whatever the case, it is sure to greatly impact how we view and interact with the world. Such leaps in information access have occurred twice in the past century, when the radio brought us a world of sound and the television brought us a world in motion. In the next century, webGIS will bring us a world in context.

Tim De Chant is a graduate student in environmental science, policy, and management, and a member of the Kelly lab.

Comments on this article? Drop us a line at with 'letter to the editor' in the subject!