Over the next year, the Tow Center will be putting a lot of time, effort, resources and thought into sensors and journalism. Journalists at the cutting edge of the news industry have started exploring drones – just one of the new ways to get sensors into the field. And most often, drones come equipped with cameras, just one of the types of sensors the industry can use. In this guest post the 2013 CBC News Vancouver Scholar and Masters of Journalism Graduate from the University of British Columbia, Alexandra Gibb, surveys drones in journalism and picks out some of the ethical and legal traps.

If you’re interested in this field, you’ll probably want to take a look at the Tow Center’s Workshop on Sensor Journalism. 

Flying machines with no human in the cockpit get called many names: drones, pilotless aircraft, remotely piloted aircraft, remotely piloted vehicles, robot planes, unmanned aerial vehicles, unmanned aerial systems, unmanned aircraft, and even zenanas—Arabic jargon for nagging wives. They’ve dominated headlines for their controversial role in hunting and killing suspected militants in the Middle East and northern Africa. But as worldwide spending on unmanned technology burgeons over the next decade, so will the appropriation and use of drones by civil society.

Journalists striving for an edge in the changing news industry are among the earliest civilian adopters of drone technology. “Once you’ve learned this drone technology … it opens up this whole other world,” said Aaron Brodie, a freelance photojournalist and producer for CNN who is pioneering the field of drone journalism. And to navigate that world, journalists need to learn the skills to run a wide range of new hardware and practices.

Drones come in a variety of shapes and sizes with vastly different capabilities. Civilian drones, in particular, range from do-it-yourself designs built by journalists in their backyards, basements, and garages to $300 smartphone- or tablet-controlled models available at airport gift shops, Costcos, and Radio Shacks across North America to costly professional systems equipped with the latest cameras and sensors. Already, drones are providing reporters with a powerful new means of obtaining aerial imagery and collecting data that will propel storytelling to new heights. But the panoptic nature of drone technology also raises new concerns over safety and ethics.

Thus far, one of the most common uses of news drones is to enhance breaking and daily news coverage. Activist and journalist Tim Pool used a modified Parrot AR.Drone to live-stream coverage of the 2011 Occupy Wall Street protests and police response in New York’s Zuccotti Park. CBS 60 Minutes used a drone to capture never-before-seen video inside the half-submerged Costa Concordia luxury cruise liner, which ran aground off the coast of Tuscany in January 2012. And FOX5-Vegas morning meteorologist Ted Pretty uses a drone he built from a DIY kit to supplement his daily weather forecasts. “It’s perfect for a local meteorologist to get video of a nice day, to get people going about their lives in the park … some shots of the local mountains,” said Pretty in an interview last November.

Drones have been used to collect aerial images and sensory data for investigative purposes. In January 2012, for example, an unidentified radio control hobbyist in Dallas, Texas was flying his camera-equipped drone over Trinity River when he discovered a local slaughterhouse was pumping pig blood into a nearby creek. The hobbyist reported his findings to the authorities and a lengthy investigation ensued. Additionally, drones—particularly those built from scratch—can be equipped with sensors that allow operators to gather vast amounts of information from the air: gas whiffers, electronic eavesdroppers, laser mapping systems, infrared and hyperspectral cameras, and so on. Most can be purchased online for a few dollars to a few hundred dollars, installed on microcontrollers, and programmed to gather and display the desired data.

Journalists have used drones to gain a drone’s-eye view of conflicts and disasters. As the former producer/news manager for multimedia and interactive storytelling at CNN.com, Brodie traveled to Tuscaloosa, Ala. in April 2011 to photograph the aftermath of a tornado that had recently devastated the region. Inspired by an aerial video he had seen on Rupert Murdoch’s now defunct tablet newspaper The Daily, Brodie borrowed a Parrot AR.Drone from a coworker, rigged a GoPro camera to it, and flew it over demolished homes, through tangled power lines and across debris-strewn roads. The captured video was edited and posted on CNN’s This Just In blog. “[Drones] allow us to get pictures we would never get without putting someone’s life in serious danger. Or that we might just never get, period. And that’s why we’re here—to tell those stories,” said Brodie in an interview last October. He has also used drones to film a New Jersey beach before and after Hurricane Irene, the aftermath of a Texas wildfire, and tornado damage in Forney, Texas.

And finally, journalists are using drones to take groundbreaking aerials. National Geographic photographer Michael Nichols used them to document wildlife in the Serengeti. In a letter to his editor, Nichols describes flying a drone more than 65 feet over migrating wildebeests. “This gives us an image that could be made no other way. Not by helicopter—too noisy and too costly. Not by balloon—too scary and balloons only go where the wind goes. Not by traditional airplane—too fast and too high and also too scary.” The drone, he adds, “is a big tsetse fly and so far the scariest thing for the wildebeests is the commotion we make preparing for the launch.” Others have used drones to capture panoramas of global landmarks, the streets of Prague at night, the granite spires of Pakistan’s Karakoram mountain range, and NASCAR races.

But while many journalists are enamoured with drone technology and what it could mean for their craft, it also raises concerns over public safety. Drones pose a collision risk because they still lack the technology to sense and avoid other aircraft. They are also challenging to fly and prone to crashing. For example, a $200 million Global Hawk surveillance drone owned and operated by the U.S. Navy crashed during a routine training operation near Salisbury, Maryland in June 2011 when its pilot lost control. Law enforcement officers in Montgomery County, Texas crashed a $300,000 ShadowHawk drone into their SWAT vehicle in March 2012 when the drone lost contact with the operator’s controls. And drones built by do-it-yourself enthusiasts often fall out of the sky due to pilot and technical errors or simply a gust of wind. “These things crash,” said Brodie. “These things are not commercial aircraft. Batteries can run out, motors can fail, propellers can break and, most importantly, you’re flying at a low level. I mean, I’ve crashed into trees.”

The panoptic nature of drone technology also challenges traditional journalism ethics. Rumours of paparazzi using drones to stalk celebrities and photograph Hollywood events have caused fear of abuse and privacy invasion. The ability of journalists to use drones to collect primary data may prompt authorities to subpoena that information for law enforcement purposes, jeopardizing journalistic independence. And just as drones have “gamified” war, so too may they gamify news by blurring entertainment and reality. Finally, the deployment of news drones—which are sometimes indistinguishable from tactical military drones—over conflicts and disasters may inflict further psychological harm among already traumatized publics.

To address these concerns, drone and data journalist Matthew Schroyer founded the Professional Society of Drone Journalists and published a “Drone Journalism Code of Ethics.” It builds on existing journalism ethics codes and suggests the following requirements be met—in order—before reporters deploy drones: 1) the information sought must be newsworthy and unobtainable by other means; 2) drones must be in suitable working condition and operated safely; 3) drone operations must follow regulations and cause minimal public disruption; 4) drone operators must respect individual privacy rights; and 5) drone operators must comply with traditional journalism ethics codes.

This is a great start. But with so much at stake, drone journalists must continue devising specific rules guiding behaviour. Fortunately, there are at least two more years until the U.S. Federal Aviation Administration legalizes commercial drone operations, giving the news media time to reflect on its values and develop appropriate ethical principles for drone journalists. Ultimately, however, the pioneers of drone journalism insist that safety and ethical concerns will be resolved and that the benefits of drone journalism will outweigh the risks: “When you turn on the news it’s either crime, fires, police chases, or it’s two politicians yelling at each other,” said Brodie. “I just don’t think that’s going to fly anymore for young audiences. … We’ve got to find a better way to give them something that actually interests them, engages them, or else the problems we have here in America will just keep getting worse because nobody will care. If we can use a drone occasionally to do that, I’m all for it.”

This blog post is a condensed version of Alexandra Gibb’s Drone Journalism Master’s Thesis (Downloadable PDF file)

Standard operating procedure, for the majority of journalists working with data, has been to analyze information generated by someone else; whether government agencies like the EPA, or researchers working in the university sector. However, as the dividing lines between technologists and journalists blur, sensing technology is becoming cheaper, and possibilities are emerging for journalists to be pro-active in taking recordings of our physical surroundings – of light, heat, sound, pressure, vibration, air quality, moisture, radioactivity, pollution – generating data that increases the depth and range of the kind of stories they can tell. Meanwhile, drones, (sometimes called small-unmanned aerial vehicles or sUAVs), equipped with cameras and a variety of sensors, are becoming more common and economical. With the help of a collaborative, public-minded hacker culture, these sensors can be programmed to record specific data—particles in the air, conversations on the ground, 3D topographical maps, or toxins indiscernible to human senses. With stealth, mobility and sensing capabilities, drones and sensor networks present a new range of possibilities for environmental journalism.

Mapping Disasters

“Sensors are like our senses—at scale,” says Matthew Waite, of the University of Nebraska-Lincoln’s Drone Journalism Lab, a pioneer in the field. Sensors can be deployed to give massive scale environmental stories an immediate, personal, or, as Waite puts it, “selfish” context. Sensors can answer questions of individual interest, such as: ‘Will the flood surge reach my lawn?’ ‘Is toxic run-off getting into my water supply?’ Or, as his lab enquired last summer: ‘Has my city’s main water source dried up?’ “Last year, Nebraska had 9.5 inches of rain in the entire year,” Waite says. “It was the worst drought since the Dust Bowl.” With the help of an unmanned aerial vehicle, his lab sought to investigate “a proof of concept drone journalism story”: whether the Platte River, a significant water source that runs the length of the state, had gone bone-dry over the summer. (It had.) Waite’s work collected arresting aerial images without the expense of flying a manned helicopter. He also developed a DIY sensor kit which detects how dry the soil in his backyard is, and uploads the results to be collated and graphed. While the drone-collected images visualized the story, the backyard soil sensor laid down what Waite suggests is an important precept for crowd-sourced, collaborative sensor journalism culture: “Harnessing selfish interest to look at communal topics.”

For those interested in filming extreme weather systems, a decidedly perilous activity, drones can serve as “data mules” controlled from a safe distance. In April 2011, Aaron Brodie, a freelance photojournalist and producer for CNN, who is also a radio-controlled aircraft hobbyist, used a drone to capture footage of an intensely violent EF4 tornado that hit Tuscaloosa, Alabama.

Sensor data can help track disasters and map out where a city’s weakest spots lie. With the so-called 100-year-disaster becoming an annual affair, the Open Infrared (OpenIR) project by the Brooklyn-based DuKode Studio, translates geo-located infrared satellite data into map layers, each of them conveying information on water depths, oil spread, fault lines, burn scars, elevation, vegetation, and built-up areas—all valuable information during times of crises. After Sandy, they generated a Flood Risk Index Map for New York, showing extracted environment features, an approach that they say “has proven effective in showing, for public access, the populated low-lying regions most vulnerable to flood damage.” The studio’s co-founder, Arlene Ducao, is currently working on a similar mapping project that’ll highlight areas of acute ecological vulnerability in Indonesia, which she hopes will help in emergency preparedness in a rapidly developing region besieged by disasters: floods, earthquakes, volcanic eruptions, and climate change.

Monitoring and Detecting Pollution

As flexible and cheap sensing machines, drones are ideal companions for enterprising investigative journalists, particularly those interested in interrogating official data—or catching polluters red-handed. In January 2012, a sUAV hobbyist flying his drone over the Trinity River in Texas saw it had captured images of a blood-red creek. An investigation later revealed that the culprit was the Oak Cliff slaughterhouse, which was funneling pig blood straight into a creek that flowed into the river.

Low-tech DIY tools, borne on kites and balloons, can be just as effective, as Liz Barry and her collaborators at the Public Laboratory for Open Technology and Science (PLOTS) demonstrate. They use a simple and ingenious combination of kites, balloons, and cheap point-and-shoot cameras to investigate everything from illegal logging in the Czech republic to the 2010 oil well explosion in the Gulf of Mexico. The balloon-borne cameras that Barry’s team sent sailing over the spill site generated 700,000 images and dozens of maps across 200 miles of shoreline. The images she captured as a “spontaneous response to the crisis”—to aid boats and crews on the clean up—are now making their way through the courts as evidence.

PLOTS have been focusing its efforts closer home of late. With the help of infrared and thermal imaging, they are conducting extensive aerial surveys of the Gowanus Canal in Brooklyn, a Superfund site chockfull of carcinogenic Polycyclic Aromatic Hydrocarbons (PAHs). The inquiry, according to the PLOTS website, “seeks to address the 300M gallons of untreated sewage that will continue entering the canal yearly even after the EPA finishes their Superfund cleanup of the toxic sediments at the bottom of the canal.”

Albert Chao is a collaborator on the Air Quality Egg, a low-cost, relatively imprecise sensor system that samples NO2 and CO2 concentrations—an index of air pollution—outside the homes of participants around the world. His project functions on the sensor journalism precept that Waite subsequently laid out—using crowd-sourced personal interest data to draw communal-interest conclusions. But Chao’s project also reveals an interesting limitation of user-generated data. The sensors we use are really cheap, and uncalibrated,” Chao concedes, “and we’ve made that clear many, many times. But hopefully after a while, we can get a critical mass, and we’ll start to see trends in the data.”

Cheap and accurate air quality sensors are a sought-after but currently unavailable technology.  Chao points out that the least expensive, most accurate air quality sensor on the market costs $120. The DIY sensor in the Air Quality Egg can be cobbled together at a price point of around $10.

Interrogating Government Data

A similar air sampling effort made headlines on the eve of the 2008 Beijing Olympics. As John Keefe, WNYC’s data news editor explained, the MIT Civic Data Design Lab collaborated with the Associated Press to question the Chinese government’s official data on particulate matter pollution and carbon monoxide levels. To pull this off, the reporters used handheld aerosol monitoring systems, which, as Keefe pointed out, the team had smuggled in pretending they were camera equipment. They used the sensors to chart, in real time, the heightened levels of particulate matter and carbon monoxide near Olympic venues around Beijing, contradicting the official line.

In the aftermath of the March 2011 earthquake, tsunami and the nuclear meltdown in Japan, the public was skeptical of the data the government was releasing on radiation leaks. At this point, a grassroots radiation sensor network, powered by DIY Geiger counters, provided journalists and the public with an alternative set of numbers—from as many as 7 million data points. The project was led by SafeCast, an intercontinental network of hackers, in collaboration with a computer communications professor at Keio University.

Sometimes, data is not willfully falsified, but just buried under layers of dirt and neglect. USA Today’s award-winning Ghost Factories project revisited more than 230 long-defunct lead smelting sites across the country, and used (rented out) $41,000 hand-held XRF analyzers to test 800 surface soil samples, generating entirely new data that showed alarmingly high levels of soil contamination in public parks and residential neighborhoods across 13 states. The team had the samples independently verified by a soil-sampling expert, Howard Mielke of Tulane University, who conducted field tests to measure lead levels, first absorbing it from the soil using nitric acid, and then analyzing samples through atomic absorption spectroscopy.

#Tracking Ecology and Wildlife

National Geographic photographer Michael Nickols uses a small arsenal of remote-controlled micro-copters, night-vision goggles, infrared cameras and sophisticated camera-traps to take thrillingly intimate pictures of lions and sprawling aerial images of wildebeest migrations.

Aaron Huslage, a North Carolina-based hacker and citizen scientist, once part of the Safecast team, has now turned his attentions to mapping the mangrove ecosystems outside Doha in Qatar, along with Qatar Foundation International and Conservation International. For environmentalists interested in climate change, mangroves serve two purposes; they mitigate the cause while indicating the severity. Mangroves absorb large amounts of CO2, converting it into oxygen as all plants do, but they’re also vulnerable to changes in their habitat. Huslage’s project deploys temperature and CO2 sensors into mangrove fields off Al Dhakira, north of Doha. The pods sense the health of the mangrove fields and send their data back to a central field hub, which uploads the data to the internet over cellular connections. Huslage says his intention is to “raise awareness about them through remote sensing data—and bring quantifiable data to the party.”

Cicada Tracker, another citizen science effort run by by WNYC and Radiolab, has no problem attracting people to the party. Once every 17 summers, cicadas emerge when the soil 8 inches below ground hits a steady 64°F.  About 300 participants have joined in the project; planting home-built temperature sensors in their yards, and through 1700 data points, they’re helping predict this year’s ‘Swarmaggedon’.

Ethical Caveats

Sensors may be able to generate enterprising, public-spirited journalism, but it is not without risks. The maker culture nurturing sensor journalism fosters a collaborative, innovate-as-you-go spirit, and encourages important partnerships between data scientists, technicians, statisticians, journalists, hackers, experts in specific research domains, and citizen science enthusiasts. But with new endeavors come new ethical questions.

Foremost among these is privacy. There are, at this point, no privacy or trespass laws that specifically relate to non-recreational domestic drone operations. In addition the increasing sophistication of data mining, correlation and triangulation techniques mean that when a journalist decides to create and publish new data about a person – even supposedly anonymized data, they run the risk of inadvertently compromising privacy. The organizers of the Cicada Tracker had to exclude the data generated by at least one person, who was under a protective order, for fear of tipping off a former acquaintance of their whereabouts.

Legal protection or redress is not a possibility just yet. Although these technologies are rapidly advancing, the law is slow to catch up. Kord Davis, author of ‘Ethics of Big Data: Balancing Risk and Innovation’, puts it succinctly: “It’s never possible to act on a law that doesn’t exist yet.” As Professor Joanne Irene Gabrynowicz  says,, laws are often written only in reaction to events, and even then, they have narrow, focused scopes. For instance, Judge Bork’s video rental history was leaked to the press, and despite its innocuousness, the press published. This led to the passing of the 1988 Video Privacy Protection Act. “The law is reactive in nature,” points out Joanne Irene Gabrynowicz a Research Professor of Law at the National center for Remote Sensing, Air and Space Law. “Courts don’t look for problems to solve.”

The second concern is that of accuracy. The field of journalism has had far more experience establishing truth in the social sciences than in the sciences of maths, statistics and physical measurements. “Data is not information,” says Gabrynowicz. “It gets sliced, diced and chopped up. Interpreting it is more than a science—it’s an art.” Despite the traditional appeal of the lone wolf journalist uncovering scandals a la Upton Sinclair, 21^st century sensor journalism cannot, in the interest of fairness and accuracy, avoid collaborations with data scientists and researchers or technicians in relevant fields.

Arlene Ducao and Ilias Koen are the Principals of The DuKode Studio. Ducao is also a Research Affiliate at MIT Media Lab.

In past years, we’ve been feeling ecological crises much more severely. And when you think about the people, Ducau said, it is usually those in developing and poor regions who are most vulnerable.

In the past, we’ve turned to radio and television to get connected to resources. Nowadays, many news organizations use street maps to show the public where to get help. But these maps are not perfect. “When it comes to regions that are not as information-dense as New York City, the web maps systems have varied data,” Ducau said.

One way that mappers and geo-spatial professionals have been dealing with this missing information is by using remote sensing. These professionals use images from an interagency satellite system administered by NASA, ESA and a Japanese space agency. This resource, combined with sensors that are closer to the ground–airplanes, cell towers, quadcopters, drones and more–can help fill in the gaps.

Seeing this data in standard maps is where OpenIR comes in. The group wants to “increase accessibility and increase the understandability of environmental sense imagery from public satellite data sources,” Ducau said. Data viewers can take bank combinations that highlight vegetation, urban impervious surfaces, soil, elevation, moisture and irrigation very clearly. The OpenIR team publishes sets of code and pushes them through Github, so others can build these sorts of data viewers for any given location on the planet.

Initial studies have been done in New York and Indonesia, in conjunction with partnerships with NGOs, government officials, and citizen journalism networks. The OpenIR team also sought to empower farmers themselves as citizen journalists.

“[We're] giving context to satellite images,” Koen said, describing the visualization aspect of the software. “You can click on a picture, get a spectrographic analysis and potentially get a way to identify what that material is.” The other function of the software, he explained, is visualizing rates of change–so for instance, you can compare two images of land, one from 1991 and another from 2004, to see what kind of ecological change has taken place after the Mamasa watershed in Indonesia.

“In the case of an environmental situation, if you can highlight the areas that are environmentally robust and could be good areas of refuge, and highlight areas that are more environmentally vulnerable, or areas that should be prioritized as rescue,” Ducau said, “that can help you give priority to different kinds of social inputs that you can layer on to.”

“Our project is a scientific look at what the arts might mean,” said Pierce Crosby. He stood beside Rachael Johnson onstage, presenting a project where they’d recently strapped motion sensors on the ankles of two dancers—one expert, the other amateur—while they executed ballet moves. Crosby and Johnson were journalism students in Mark Hansen’s class on Formats, Protocols and Algorithms in the past school year. Their goal was to encode “grace” in data, finding the figures behind the performance.

“Grace is basically a fluidity of movement that’s created by muscle contractions made to look effortless,” Johnson said. But that definition was too broad, so the pair decided to focus on sensing motion in three specific movements in ballet: the tendu, the sauté, and the tamble.

The researchers chose to use accelerometers instead of a motion capture system because they could get those over-the-counter, and they bought a Nintendo Wii Remote for $30. They connected the system to the computer via Bluetooth, and loaded up a program called OSCulator—which reads in data every hundredths of a millisecond, depending on where the Wii Remote is pointed toward. Then they programmed a simple Python program to crunch the numbers.

Visually, each move, encoded in data, looked like a kind of polygraph—for the tendu, there were three valleys, four peaks. The journalists had the dancers do each move five times in a row, and looked at the plots.

The professional dancer had jump patterns that were similar each time; the amateur ballerina had more erratic graphs. The journalists’ outcomes and findings, thus, was that precision and consistency in repetition is somehow indicative of grace in ballet.

“For future ideas, we think there’s a lot of great implications for ballerinas overall,” Crosby said. (It had just been revealed that he was the mystery amateur ballerina.) “What this could mean for medical compensation is an example—they exert a lot of force on their bodies.”

Bryan Nunez is the Technology Manager at WITNESS. Bryan joined WITNESS in 2002. He oversees technology for the organization as well as the development of projects like the Hub, a site for citizen human rights media, and the Secure Smart Cam, a camera-phone app for human rights activists. Prior to WITNESS, he was a technology strategist and consultant on a variety of projects ranging from online banking to interactive television. He is an alumnus of the Interactive Telecommunications Program at NYU and has a BA in anthropology from UC Berkeley.

Sensors are everywhere, and not just in places where they’ve been deployed by researchers. We text, email, and chat on them everyday. Our smartphones are wired with all kinds of sensing tools—GPS, audio, video, and more—and we carry them in our pockets wherever we go.

Bryan Nunez and his colleagues at Informacam recognized this, and came up with a smart plan to leverage it. They programmed an app that would tag any media it took—photos, videos, audio—with metadata, so that the validity of the recorded thing would be certified. “I guess we’re sort of taking a more traditional journalistic approach to this idea of sensor journalism,” he told the crowd during his lightning talk. “Most of the projects we’ve seen so far are thinking beyond traditional journalism with using sensors to actually generate the content for stories. We’re looking at sensors to verify and validate what traditional media has collected in the course of journalism.”

The endeavor is important, Nunez said, because there is a lot of media being produced in the world right now, much of it citizen generated.

In every piece of media, there is a type of device that took the video or image, someone who took it, and what’s actually in it. If you’re going to use it for evidence, Nunez asked, how do you determine the chain of custody from collection to submission, in some sort of legal setting?

Informacam is one approach. What happens is when you take a photo, Informacam creates a digital fingerprint that signs the media collected. Sensors are turned on—GPS sensors, wireless, etc.—and metadata is collected and embedded into the file itself. An extra layer of security is added, too—when the data is transmitted to the secure server, it is encrypted.

But Nunez said that ultimately, it is not important to their team that Informacam as an app succeeds. He’d rather see the development of a standard, like the one the team developed. They dubbed it J3M (for JSON Evidentiary Mobile Media Metadata Standard). The standard the group envisions will make the metadata captured with mobile phones and other devices interoperable and human-readable. It could also work on platforms like YouTube.

“[That] goes beyond this use of media as evidence, but provides future ways of telling interactive stories using metadata and geo-temporal data,” Nunez said. “You can use video for multiple viewpoints, etc.”

Ákos Lédeczi is an Associate Professor at the Electrical Engineering and Computer Science Department and a Senior Research Scientist at the Institute for Software Integrated Systems (ISIS) at Vanderbilt University.

From knowledge of the acoustic events of a typical rifle shot, researchers hacked together a system of sensors, comprised of four microphones and a mount, that could listen for a shooter and locate his position. Akos Ledeczi, an associate professor at Vanderbilt University, spoke about the project which started ten years ago with DARPA funding.

Ledeczi said you needed two things to measure to be able to localize a gunshot. First, there’s the muzzle blast, or the sound you hear coming from a shotgun, which propagates from the shooter to the sensor. Next, there’s the shockwave, generated by the bullet. You can measure both with microphones and time their arrival to locate a shooter.

Of course, Ledeczi explained, you need multiple microphones to do this; one isn’t enough. He showed the audience a microphone rig called the “Boomerang,” which was used in the Iraqi war. It had seven microphones sticking out of a bulky contraption. It was, Ledeczi said, a bit “big, but works quite well.”

The researchers were soon struck with an innovative idea: army soldiers always carried smartphones, and smartphones have microphones in them. They brainstormed technology that could leverage the smartphone’s microphone to find a shooter. Finally they accomplished this, designing an additional sensor that would boost the microphone’s capability.

However, five or six of these receiver-boosted smartphones would be needed to locate a single shooter. “DARPA didn’t like the fact that there would need to be six soldiers running around to be able to locate a shooter, so they gave us a few months to come up with a better solution,” Ledeczi said.

So he dug up some hardware from an old existing project, a sensor with four microphones that did a better job of picking up audio. Now only two of these would be needed to locate a shooter.

The researchers tested the tech out. The system worked from about one to ten degrees accuracy.

Aaron Huslage has led the effort, Mapping the Mangroves, since it launched in the spring of 2012 in partnership with Qatar Foundation International and Conservation International. Citizen involvement played a big role in the project. Huslage said the team was working with students, teachers and, hopefully in the future, some journalists. Students learned how to build solar-powered sensors, deployed them and aggregated field data to build a data map of the Qatar mangrove forests. The aim was to raise awareness around the decline of the forests, their role in the environment and the importance of conservation on the whole. As the forest trees dwindled, so did their capacity for carbon sequestration.

“They basically sequester 150% more carbon than any other plant in the world,” Huslage said. “They’re very efficient at this work.” According to Huslage, the planet has lost 50% of all the mangroves in the world in the last fifty years.

In Doha, the researchers worked on prototypes of different sensors. The first was a dual temperature sensor to put in mangrove forests. They also developed CO2 soil carbon sensors—which measures the CO2 coming out of the soil—to understand the health of living trees.

“We take all the data from these remote sensors, we aggregate it with a computer in the field, that data is transmitted back to the Internet, and we put it all to a map,” Huslage described. There a few places working with these sensors right now; there are maps for Brazil, Central America, and in the U.S., Florida, which is the only region of the States that has a mangrove population.

Now, the group expanding this work with the help of other companies: to seagrass beds, coral reefs, and other coastal communities. They’re also aiming to bring the hardware cost down and the accuracy up.

They’ll also figure out a way to allow users to plug in all kinds of new sensors—sensors to measure oxygen reduction potential (a measure of the quality of the water), soil salinity, water salinity, pH levels, CO2 and more.

“We’re trying to characterize in real-time the health of these coastal ecosystems, which has not really been done before,” said Huslage.

Emily Bell is the Director of The Tow Center, formerly the Director of Digital Content at The Guardian.

Kord Davis is the author of The Ethics Of Big Data, and has twenty years experience providing business strategy, analysis, and technical consulting to more than 100 organizations of all sizes including: Microsoft, Intel, Nike, Bonneville Power Administration (BPA), Bill & Melinda Gates Foundation, Western Digital, and Gardenburger. He holds a BA in Philosophy from Reed College and professional certifications in communication, systems modeling, and enterprise transformation.

The Public Laboratory for Open Technology and Science is a community that develops and applies open-source tools to environmental exploration and investigation. Liz Barry, their director of urban development, spoke about the organization and one of its data projects, balloon mapping, in this brief lightning talk.

The Public Laboratory has a growing community of researchers who use open source hardware and software tools and methods to generate knowledge and share data about health and the environment. Its mantras are accountability through evidence, and getting boots on the ground to be able to investigate and hold government and industries accountable when necessary.

The Public Lab has also launched a program to manufacture and sell affordable “civic science kits” which have become extremely popular; these include aerial photography “watchdog” kits and spectrometry kits for measuring contaminants. The balloon mapping kit sells is one of the most popular kits of the group. It allows civic-minded citizens to collect aerial photos from 1000 feet, and has already been used to fly over the Jamaican Bay, and in New York City-based projects among others, according to Barry.

Javaun Moradi is a product manager at NPR digital. He runs their APIs, which you can now learn by taking a Codecademy course. He’s one of the inventors of the Infinite Player, an experiment in continuous, personalized listening and is also very interested in what open hardware sensors mean for news and society. He’s a big fan of open source software and open data and loves the public media mission to create a more informed, culturally-enriched society.

Javaun Moradi “waded in” to sensor journalism territory after he started talking to the experts. He spoke to a geochemist at Columbia University. He met people in the open source community. Many conversations later, he realized that he–and all of us–live in a truly exciting time for tapping into the potential of data. “If data didn’t exist, we could make it,” Moradi said. The way he saw it, independent groups and the civic media were already solving problems that journalists would be interested in–and they were doing it without reporting.

Moradi outlined five of the most important points for sensor journalism.

Look outside of journalism
Innovations in sensing are happening all over: in open data, hacker communities and citizen science efforts. We need to broaden our perspectives.

More collaboration
This means engaging an audience; the most successful sensor projects tend to be the ones that have community baked into their plans from the beginning. But it also means teaming up with the government. “Sometimes we FOIA them, sometimes they subpoena us,” Moradi half-joked. “But they’ve got great data.”

Revisit privacy and ethics
Moradi says “we have ethics guidelines, but the data is coming from unexpected places and so quickly,” he said. Journalists need to create better rules to make sure the communities they are supposed to be helping are not inadvertently harmed.

Data Control and Access
Just because someone buys a sensor, he said, doesn’t mean that person owns the data. Moreover, we should ask: Is the data even valuable? Does it solve a meaningful problem? Is there a story?

A universal, stable platform
Finally, Moradi says we need to have a universal, stable platform that can crunch large data sets across each other. One city project working with another city project, for instance, could have very interesting potential.

Willie Shubert is the Senior Project Coordinator for Internews’ Earth Journalism Network. As a coordinator of a global network of environmental journalists, Willie helps make tools that enable people to connect with each other, find material support, and amplify their local stories to global audiences.

William Shubert acquainted the audience with the GeoJournalism mapping platform called InfoAmazonia, essentially a tool that journalists can pull into their stories to provide more context and evidence about an unraveling environmental issue. InfoAmazonia in particular is an interactive map that combines georeferenced environmental news articles with data on deforestation, fires, protected areas, and oil and gas concessions in the Amazon. The platform lets users upload and download news and data about the Amazon basin, expanding its database through audience engagement and participation.

InfoAmazonia is the brainchild and flagship GeoJournalism project of non-profit group Internews and O Eco, a Brazilian environmental news agency. The project was created by leveraging the vast resources of the 4,500-plus members in Internews’ Earth Journalism Network as well as using open access data from dozens of partner organizations.