« Justice Kagan for Emperor | Main | Oulde Fields, New Corn-cerns »

Saturday, July 06, 2013

Environmental Intermediaries

It seemed like a great idea. A refinery applies for a land use permit. The towns nearby are still reeling from an accident that should have led to immediate shutdown but was allowed to continue for days. The county conditions permit approval on what sounds like science fiction: "infrared or other state-of-the-art remote sensing technology." Nevertheless, an MOU is signed. It resolves all kinds of issues: What kinds of sensors? FTIR? TDLS? UV? Where should they be set up around the facility? What compounds can they detect? What detection limits should be used? Who is going to maintain the system? How to address rain and fog? What to do with data speed, transfer, and security? This was a storied moment in the annals of environmental monitoring. As expected, there were problems with implementation. But the goal was clear, and bold: reflect beams of light along the sides of a giant facility, detect dozens of compounds as they crossed the fenceline, and gather these data in the interest of right-to-know and emergency response. But here's the catch: the data were "view-only." A single person could watch as they streamed across his computer screen. These raw spectral data, gathered by one company in California, cleaned and converted by another in Texas, and then, weeks later, after they were dumped into dispersion models to estimate pollution concentrations downwind, sent to a homeowner near the refinery. I took an interest in the technology years after another resident, a camera specialist, taught himself environmental monitoring so that he could help negotiate the MOU. I drove a couple of hours to meet him, and found this treasure trove of monthly data, which was at the time some of the most advanced local air quality data available. It was stacked neatly on CDs in a barn in Oregon.

That was years ago. There is now a vast range of environmental monitors on the market, and the choices among them are not as painful, and not nearly as dichotomous - expensive, fixed equipment versus unreliable, time-limited handhelds - as they once were. We've witnessed the global diffusion of grab samplers in chemical corridors and the rise of the nanosensor. EPA employees can sit in an office in New Jersey and watch real-time air quality data stream from wireless networks at sites around the world. Citizen monitors can attach small devices to cars, or their clothing, and silently map the VOCs or sulfur compounds that invade their daily routines. We are, or soon will be, awash in these kinds of data, although relevant, actionable data will still be very much up for grabs. This will pose serious challenges, which I discuss in "The Architecture of Ignorance." For a future article, I hope to zero in on a related issue, one that is grounded in that pile of CDs that bear the logos of Terra Air Services and Petris Technology. To do so, let me introduce an organization that has yet to receive much attention in law reviews, with the exception of a piece by its founder in Environmental Law Reporter. The issue is the role of "environmental intermediaries." The organization is SkyTruth.

As attorneys, we grow accustomed to certain modes of argument: doctrinal, prudential, and the like. We make propositions and set out to show their truth. We spot logical fallacies and evaluate claims. Sometimes, the forms of argument conflict, and we bring out another set of interpretive tools to reconcile them or appeal to the primacy of one over the other. We carry this out across walls of words. But there are other forms of argument and interpretation. The remote sensing and related technologies that upend our image of life on Earth and now inform debates over climate change, natural resource management, and disaster response are brimming with them. What I have in mind are military and intelligence technologies that are repurposed for commercial use, such as digital mapping, the dual civilian and military use of GPS satellites that started in 1991, the spread of Geographic Information Systems to desktop use, and privatized surveillance satellites. These spatial tools, whether viewed as a system or in isolation, hide layers of implicit biases. Peter Galison would say that they leave us a bit untethered. Your location on a map is no longer fixed, but relative to an invisible grid held together by satellites that spin around the earth at 12,500 miles per hour. The maps that we use online no longer reveal an objective space - they represent and interpret that space. Laura Kurgan, in her excellent Close Up at a Distance (2013), describes Google Earth not as an application and a database but a "patchwork of archived aerial and satellite images of varying origins, sources, motivations, and resolutions...[the consumers of which] will never know who has tasked a satellite to take a picture." Classified satellite images were once gathered on film and ejected into the Pacific Ocean for retrieval. Now, private satellites gather digital images that are used by the government and Google alike, athough their operating licenses require a lower resolution for some customers. Since digital images do not exist until they are sorted and rendered, we are at the mercy of interpretive decisions that accumulate along the way. And when they are subjected to further statistical operations, layerings, corrections, and visualizations, with GIS and other tools, the interpretations mount, and are more difficult to retrace.

These layers of interpretation silently, and regularly, inform environmental policy. "Open data" efforts can paradoxically add layers of data practices and veiled choices that are over time taken for granted. Enter the environmental intermediary. For now, I'll use a very broad definition: entities (public, private, non-profit) that gather, process, merge, and present geospatial data to inform public debates over environmental quality. You find them everywhere these days. Take unconventional energy as an example. A town in West Texas runs out of water near a heavily fracked shale play. The Texas Railroad Commission has a limited sense of how much water is used to operate drilling wells in the area. So newspaper interns merge data from a private, industry-funded website and a groundwater conservation district to find out for themselves, and have their work checked by a nonprofit research institute. The proposed Keystone XL pipeline is intended for diluted bitumen, the impacts of which EPA claimed were not fully understood. Yet even more basic data must be gathered and processed from disparate public sources, including the pipeline's precise 1,700-mile route (e.g., its mile post markers and water crossings). In response, the Keystone Mapping Project uses state data and different mapping tools to piece the route together. The result is an interactive public square, "allowing communities, stakeholders, journalists, and non-profits to more easily evaluate the pipeline and its impacts." Other impacts simply cannot be appreciated in a spreadsheet or EIS appendix. They are engaged by other non-profits, whose data visualizations simulate the appearance of five-acre well pads in Colorado's Thompson Divide or lead a Google Earth tour across a Manhattan covered in 65 billion gallons of tainted water used in fracking operations.  

One of these non-profits is SkyTruth. Its founder, John Amos, is a geologist who started something exciting and essential: a civic-minded environmental intermediary. SkyTruth is a company "based in West Virginia that uses satellite and aerial images, digital mapping, data analysis and digital graphics to investigate and illustrate environmental issues and incidents." To get a sense of his team in action, read the organization's "Sky Truth Alerts." Some of them take incident reports from the National Response Center and add further analysis, such as latitude/longitude, sheen size, volume, and minimum estimates of hazardous materials released. The incidents range from large spills to slow-motion crises of the kind that Thomas Beamish theorized after years of fieldwork in the Guadalupe Dunes. One of them involves a Taylor Energy oil platform damaged by Hurricane Ivan. The rig once commanded 28 oil and gas wells in the Gulf of Mexico. Now its mangled remains sit hundreds of feet below the surface. Efforts to plug the wells were abandoned, and an international well control company created a proprietary plan to manage the oil flow instead. Some of the Sky Truth Alerts characterize this chronic spill, showing slicks that extend for miles from the site. SkyTruth teams up with the Gulf Monitoring Consortium to estimate the extent of the spill with the help of NOAA satellite data. This does not sit well with the Unified Command set up by the Coast Guard to coordinate response efforts. It publicly criticized SkyTruth's work, and the value of NOAA's own satellite images.

We often read about Internet, financial, healthcare, and other data intermediaries in law reviews. A lot has been done to chart everything from their aggregation and use of personal data to how they contribute to systemic risk. There is comparatively little research on environmental intermediaries such as SkyTruth, and the layers and styles of interpretation that represent digital, geospatial data for use in policymaking, monitoring, and enforcement. This is true even as the fruits of their labor, from Scorecard in the early days of TRI to the Louisiana Bucket Brigade's "Oil Spill Crisis Map," are given due attention. Other disciplines share this sense of urgency. A recent article in the American Journal of Public Health notes that we know very little about the research capacity of community-based organizations, which are key partners in the work of environmental intermediaries. I hope to build on the work of scholars such as Holly Doremus, who recognized the role of intermediaries in adaptive management, and Oren Perez, who argues that we should open up the technical choices that influence how raw data are shared. John Amos gave his own thoughts on the subject in recent testimony before Congress. His target was FracFocus. This is the online, voluntary registry that discloses chemical use in hydraulic fracturing. It is now a mandatory disclosure tool in some states. The Bureau of Land Management may follow suit for drilling on public lands. Amos sketches the many valuable uses of its data, the data aggregation and analysis that intermediaries would need to carry out, and how FracFocus undermines those tasks. His testimony and work offer brilliant examples of the opacity of disclosure and its hindrance of environmental protection.  

Posted by Gregg P. Macey on July 6, 2013 at 08:59 PM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference Environmental Intermediaries:


I will tell you that whatever tools we have it just is not working. Fracking is coming to my state of Pennsylvania and the people really do not want it but the politicians and businesses surely do. Wonder who will win that fight. This has been a problem for decades and it continues to be kicked down the road. I do not believe anything will change as money and commerce comes before peoples safety and well being almost all the time.

Posted by: STEVEN J. FROMM | Jul 7, 2013 10:46:09 AM

@Stephen J Fromm - What you mean fracking is coming to Pennsylvaniait's been here for years and is one of the largest fracking States in the country.

What bothers me is that couple months back there was a university report that Found that only 20% of the fracking wells in Pennsylvania have been inspected by state regulators so for all we know, there have been instances of groundwater aquifers or private wells being contaminated.

Thank you, Gov. Corbett.

Posted by: Ken Glick (EEI) | Jul 10, 2013 3:09:56 PM

Post a comment