As a non-profit, they have a goal of a sustainable planet, but we're going to learn how they've learned to measure what was once unmeasurable -- and then to share that data to promote change and improvement.
To learn how big data helps manage environmental impact, BriefingsDirect sat down with Eric Fegraus, Director of Information Systems at Conservation International. The discussion is moderated by me, Dana Gardner, Principal Analyst at Interarbor Solutions.
Here are some excerpts:
Gardner: First, tell us the relationship with technology. Conservation International recently announced HP Earth Insights. What is that all about?
Fegraus: HP Earth Insights is a partnership between Conservation International and HP and it's really about using technology to accelerate the work and impact of some of the programs within Conservation International. What we've been able to do is bring the analytics and a data-driven approach to build indices of wildlife communities in tropical forests and to be able to monitor them in near-real-time.
Fegraus: This is really a telling line. We really don’t know what’s happening in tropical forests. We know some general things. We can use satellite imagery and see how forests are increasing or decreasing from year to year and from time period to time period. But we really don't know the finer scale measurements. We don't know what's happening within the forest or what animal species are increasing or are decreasing.
There's some technology that we have out in the field that we call camera traps, which take images or photos of the animals as they pass by. There are also some temperature sensors in them. Through that technology and some of the data analytics, we're able to actually evaluate and monitor those species over time.
Gardner: One of the interesting concepts that we've seen is that for a certain quantity of data, let's say 10,000 data points, you can get magnitude of order more inference points. How does that work for you, Eric? Even though you're getting a lot of data, how does that translate into even larger insights?
Fegraus: We have some of the largest datasets in our field in terms of camera trapping data and wildlife communities. But within that, you also have to have a modeling approach to be able to utilize that data, use some of the best statistics, transform that into meaningful data products, and then have the IT infrastructure to be able to handle it and store it. Then, you need the data visualization tools to have those insights pop out at you.
Gardner: So, not only are you involved with HP in terms of the Earth Insights Project, but you're a consumer of HP technology. Tell us a little bit about Vertica and HP Haven, if that also is something you are involved with?
Fegraus: Yes. All of our servers are HP ProLiant servers. We've created an analytical space within our environment using the HP ProLiant servers, as well as HP Vertica. That's really the backbone of our analytical environment. We're also using R and we're now exploring with Distributed R within the Vertica context.
We’re using the HP Cloud for data storage and back up and we’re working on making the cloud a centerpiece for data exchange and analysis for wildlife monitoring. In terms of Haven, we're exploring other parts of Haven, in particular HP Autonomy, and a few other concepts, to help with unstructured data types.
Gardner: Eric, let’s talk a little bit about what you get when you do good data analytics and how it changes the game in a lot of industries, not just conservation. I'm thinking about being able to project into people’s understanding of change.
So for someone to absorb an understanding that things need to happen in order for things to improve, there is a sense of convincing. What is big data bringing to the table for you when you go to governments or companies and try to promulgate change in these environments?
Fegraus: From our perspective, what we want to do is get the best available data at the right spatial and temporal scales, the best science, and the right technology. Then, when we package all this together, we can present unbiased information to decision makers, which can lead to hopefully good sustainable development and conservation decisions.
These decision makers can be public officials setting conservation policies or making land use decisions. They can be private companies seeking to value natural capital or assess the impacts of sourcing operations in sensitive ecosystems.
Of course, you never have control over which way legislation and regulations can go, but our goal is to bring that kind of factual information to the people that need it.
Gardner: And one of the interesting things for me is how people are using different data sets from areas that you wouldn't think would have any relationship to one another, but then when you join and analyze those datasets, you can come up with astounding results. Is this the case with you? Are you not only gathering your own datasets but finding the means to jibe that with other data and therefore come up with other levels of empirical analysis?
Fegraus: We are. A lot of the analysis today has been focused on the data that we've collected within our network. Obviously, there are a lot of other kinds of big data sets out there, for example, provided by governments and weather services, that are very relevant to what we're doing. We're looking at trying to utilize those data sets as best we can.
Of course, you also have to be careful. One of the key things we want to do is look for patterns, but we want to make sure that the patterns we're seeing, and the correlations we detect, all make sense within our scientific domain. You don’t want to create false correlations and improbable correlations.
Gardner: And among those correlations that you have been able to determine so far, about 12 percent of species are declining in the tropical forest. This information is thanks to your Tropical Ecology Assessment and Monitoring (TEAM) and HP Earth Insights. And there are many cases not yet perceived as being endangered. So maybe you could just share some of the findings, some of the outcome from all this activity.
Fegraus: We've actually worked up a paper, and that’s one of the insights. It’s telling, because species are ranked by “whether they are considered endangered or not.” So species that are considered “least concerned” according to the International Union for the Conservation of Nature (IUCN), we assume that they are doing okay.
So you wouldn’t expect to find that those species are actually declining. That can really serve as an early warning, a wake-up call, to protected-area managers and government officials in charge of those areas. There are actually some unexpected things happening here. The things that we thought were safe are not that safe.
Gardner: And, for me, another telling indicator was that on an aggregate basis, some species are being measured and there isn’t any sense of danger or problem, but when you go localized, when you look at specific regions and ecosystems, you develop a different story. Was there an ability for your data gathering to give you more a tactical and insights that are specific?
Fegraus: That’s one of the really nice things about the TEAM Network, a partnership between Conservation International, the Wildlife Conservation Society and the Smithsonian Institution. In a lot of the work that TEAM does, we really work across the globe. Even though we're using the same methodologies, the same standards, whether we are in the Amazon or whether we're in a forest in Asia or Indonesia, we can have results that are important locally.
Then, as you aggregate them through sub-national level efforts, national-levels, or even continental levels, that's where we're trying to have the data flow up and down those spatial scales as needed.
For example, even though a particular species may be endangered worldwide we may find that locally, in a particular protected area, that species is stable. This provides important information to the protected area manager that the measures that are in place seem to be working for that species. It can really help in evaluating practices, measuring conservation goals and establishing smart policy.
Sense of confidence
Gardner: I've also spoken to some folks who express a sense of relief that they can go at whatever data they want and have a sense of confidence that they have systems and platforms that can handle the scale and the velocity of that data. It is sort of a freeing attitude that they don’t have to be concerned at the data level. They can go after the results and then determine the means to get the analysis that they need.
Is that something that you also share, that with your partnership with HP and with others, that this is about the determination of the analysis and the science, and you're not limited by some sort of speeds-and-feeds barrier?
Fegraus: This gets to a larger issue within the conservation community, the non-profits, and the environmental consulting firms. Traditionally, IT and technology has been all about keeping the lights on and making sure everyone has a laptop. There's a saying that people can share data, but the problem has really been bringing the technology, analytics, and tools to the programs that are mission critical, bringing all of this to business driven programs that are really doing the work.
One of the great outcomes of this is that we've pushed that technology to a program like TEAM and we're getting the cutting-edge technology that a program like TEAM needs into their hands, which has really changed the dynamic, compared to the status quo.
Gardner: So scale really isn't the issue any longer. It's now about your priorities and your requirements for the scientific activity?
Fegraus: Yes. It's making sure that technology meets the requirements in scientific and program objectives. And that's going to vary quite a bit depending on the program and the group that we were talking about, but ultimately it’s about enabling and accelerating the mission critical work of organizations like Conservation International.
Read a full transcript. Get the mobile app for iOS or Android.
You may also be interested in:
- How Globe Testing helps startups make the leap to cloud- and mobile-first development
- GoodData analytics developers on what they look for in a big data platform
- ITIL-ITSM tagteam boosts Mexican ISP INFOTEC's operations quality
- Novel consumer retail behavior analysis from InfoScout relies on HP Vertica big data chops
- IT Operations Modernization Helps Energy Powerhouse Exelon Acquire Businesses
- ECommerce portal Avito uses big data to master rapid fraud detection
- How a Hackathon Approach Juices Innovation on Big Data Applications for Thomson Reuters
- How Waste Management Builds a Powerful Services Contiunuum Across Operations, Infrastructure, Development, and IT Processes
- GSN Games hits top prize using big data to uncover deep insights into gamer preferences
- Hybrid cloud models demand more infrastructure standardization, says global service provider Steria
- Service providers gain new levels of actionable customer intelligence from big data analytics
- How UK data solutions developer Systems Mechanics uses HP Vertica for BI, streaming and data analysis