John Goolgasian III is the Director of the National Geospatial-Intelligence Agency (NGA) Source Operations and Management Directorate, a position he has held since 2014. He has also served the NGA as Deputy Director of Source Operations and Management Directorate, Director of the Foundation GEOINT Group in Source, and the Director of the Office of Counterterrorism. Goolgasian began his government career in 1993 as a cartographer with the Defense Mapping Agency. Esri writer Jim Baumann interviewed Goolgasian about the rollout of the new IC GIS portal, open data, data collection and more.
Goolgasian: Today, our agency collects, analyzes, and uses data outside of our traditional authoritative data sources on a regular basis. This includes social media and other voluntary geospatial information, such as OpenStreetMap. This is all very important data to us because it contributes to a more comprehensive view of an incident or area.
However, the challenge for us is synthesizing this vast amount of voluntary data through our standard data production process so that it becomes verifiable and actionable intelligence and can be used in our standard GEOINT [geospatial intelligence] products. So we are making a fairly significant effort to determine how we can better leverage the exponential growth in data collection from the greater geospatial community.
Goolgasian: Well, I don’t think it’s just the Source Directorate; I think it’s all of NGA. For the first time in our history—or at least the 23 years I’ve been here—we have a one-stop shop for all geospatial information across the entire intelligence community.
This is a significant advantage for every agency, not just NGA—and specifically, not just the Source Directorate. We can now seamlessly share information between NGA, DIA [Defense Intelligence Agency], and CIA [Central Intelligence Agency], and bring all of that information to bear on the hardest intelligence problems and operational requirements. So this is a tremendous advancement in our ability to determine who has what information and how to use that information in solving our most challenging problems.
Goolgasian: In addition to my role as director of Source, I’m also the content portfolio manager. And within my portfolio, I have an initiative called Object-Based Production. This is a combination of the Foundation GEOINT mission, and what we’re doing from a data collection standpoint, and our Structure Observation mission, and how we are moving our analysts from developing purely written assessments to creating spatially enabled objects of their intelligence observations. And we’re creating those spatially enabled intelligence observations to the same standards that we build our Foundation GEOINT data.
So, for the first time, we’re actually merging the two domains into a single, object-based production process so that we can analyze intelligence observations in context with natural and man-made objects in real time. So as information is being collected and analysts are looking at imagery, they can collect that information in the same format and in the same space as our Foundation GEOINT data. We’re really bringing the two sides together from a technology standpoint, from a standards standpoint, and from a process standpoint.
Goolgasian: Do I think the organization will continue to expand in its worldwide response and support of humanitarian and other nonclassified efforts? Absolutely! It is part of our director’s initiative to operate more in the open—or operate with the open, as we are now calling it—not solely in an unclassified sense. But we are currently examining how we can best bring to bear the combination of unclassified information and classified information for the greatest impact. We are taking the unclassified information we have that is shareable and providing that for humanitarian disaster relief for worldwide response actions and for scientific studies like the Arctic Open Data application. So it’s an effort that we will continue to expand, and we are refining the process along the way.
The first time we did this, it took a number of weeks to get the necessary data needed to stand up an unclassified site and make it publicly available. In our latest effort, we were able to provide unclassified information to the public on our website within 24 hours. So it is a battle rhythm, if you will, that we’re going to continue to repeat as the need arises.
Goolgasian: This is one of the most exciting projects that we are currently working on, especially inside the Source Directorate. We’ve really taken a major effort to open up our data stores to the science community. We’ve done this in the past, but this is probably our greatest initiative. This partnership is with a number of organizations, both international and domestic. The International Hydrographic Office is one of our biggest partners on this project. [The office has] an Arctic regional board that’s made up of representatives from the United States, Canada, Denmark, Norway, and Russia. We’re sharing information with all these countries through the web portal to ensure that we all have a common understanding of the Arctic environment. In addition to the international partnerships, the domestic partners include the National Science Foundation; United States Geological Survey; and the Polar Geospatial Center, at the University of Minnesota. So it’s really exciting for us to be participating in this collaboration. Will we work on other, similar projects? I don’t see why we wouldn’t. This is really setting an important precedent in international cooperation, and we’ll continue to look for similar opportunities.
Goolgasian: Our next-gen tasking approach is based on predictive analysis and modeling. This is our determining policy when collecting, tasking, and leveraging GEOINT in the future. And this is for all data, including national technical sources, commercial imagery, and open-source information.
This will allow us to develop normalcy patterns so that we can identify when there’s abnormal activity—things that are out of scope/out of range from the norm. We can then use our unique assets to collect additional data to further evaluate these abnormal activities.
With the growth of commercial imagery, we’re going to have ubiquitous coverage of the earth on a daily basis in the not-too-distant future. So our ability to do predictive analysis and understand normalcy—and variants from that normalcy—is going to increase tremendously. We plan on leveraging that modeling process to identify those areas that we want to further examine with other sources of information—and not just from a collections standpoint but how [we can] use that to drive and to focus what our analysts do on a daily basis. This type of modeling will help us identify suspicious activities for further examination by our analysts.
Goolgasian: I’m going to link this right back to your question on next-gen tasking. Where we’re considering geodata collection in the future is not solely based on imagery collection or overhead surveillance. It is really looking at the collection of spatial information that can be used to focus our resources and guide our analytical activities. For example, how do we look at the collection of social media? How do we look at open content in combination with what we do with classified imagery? We’re really trying to define processes that we can use in modeling and predictive analysis to bring together literal information—like imagery and imagery-derived observations—with nonliteral information like social media that can be combined in a geospatial context to provide a clear understanding of an area on the ground in a normal state, so that we can better recognize abnormal activity in that area.
Goolgasian: We are currently in the process of evaluating our procedures and establishing our goals for the next five years from a technology, workforce, and mission standpoint. And a big piece of that is training.
We want to stay current on the latest COTS and open-source technologies, but it’s about tradecraft and methodology training as well.
As we shift to leveraging more unclassified voluntary information, our processes are changing—especially in the foundational sense—from traditional data collection, processing, and distribution to a more nimble, responsive process.
We are evaluating the skills the agency workforce will need in the future and how to best achieve the necessary skill set. We have determined that it’s a combination of training and hiring. For example, the need for classic cartographic skills will remain. The need for earth science skills in the geomatics arena will remain. We’re also looking at new skills and commercial technology that we will need to further expand into areas such as data science, and how that applies to our foundational mission and our intelligence missions as well.
Goolgasian: The Source organization is made up of three main components: our Foundation GEOINT Group, [which] does the mapping, charting, and geodesy; our media replication group, [which] does both hard- and soft-copy media replication of our products; and our collections group.
Let me talk a little bit about the Foundation GEOINT Group and how we see that in the future. Five years from now, I still see the need of foundation data and foundation products for our primary customers—the US military commands and services. That is, we’re still going to be creating maps and charts. However, the way we create and disseminate them is likely to change. We are already shifting from the traditional creation of paper products and magnetic media to the delivery of some of our information on smart devices. As previously mentioned, we are shifting the production back end of that as well, from solely collecting information from imagery and other sources to leveraging open spatial data and voluntary geographic information.
We’re looking at how we work with international partners as well. We have tremendous international agreements with countries all around the world. We need to reexamine how we cooperatively create and share geospatial information. So the big change for us is how we will be creating content and how we can disseminate that content in a more dynamic way to our mission partners.
I do believe that we will still be producing paper maps and magnetic media five years from now, but it will be less than what we are producing today because of our efforts to implement more of an online/on-demand approach to our spatial information. We’ve already made tremendous strides with moving all of our Foundation GEOINT content to the Map of the World, our online capability. And, again, we’re trying to get more data out to disconnected users through smart tablets and other electronic formats.
Finally, I believe that our collection procedures are going to change from the current process of looking at the requirements that come into the agency and the capabilities that we have and then allocating those resources to meet the requirements. Our future collection process will include more dynamic modeling and sensing of the areas on earth where we detect change—some abnormality from previous observations—[which] will allow us to perform more precise analysis to determine the necessary course of action.