The mission of the National Geospatial-Intelligence Agency (NGA) is to provide timely, relevant, and accurate geospatial intelligence to support national security. NGA defines geospatial intelligence (GEOINT) as “the exploitation and analysis of imagery and geospatial information to describe, assess, and visually depict physical features and geographically referenced activities on the Earth.” NGA faces a crisis of need in the post-9/11 world. Without effective GEOINT, the nation and its armed services are vulnerable to security risks and threats. NGA must improve the speed, rigor, accuracy, fidelity and relevance of its geospatial analyses while the sources of data increase in number and type, and data volume grows. Because GEOINT is rapidly moving to ever finer temporal, spatial, radiometric and spectral resolutions, increased volumes and more complex data must be absorbed: that is, captured, stored, analyzed and reported. The time horizons of problems that the intelligence community seeks to understand have gone from months and days to hours and minutes. Other challenges include adopting and spearheading new methods and technologies while maintaining fully operational existing systems; integrating data from a host of old and new sources through rapid georectification and spatial analysis; improving uncertainty management, including dealing with denial and deception; dealing with data volume issues, especially the need to automate human interpretation tasks; ubiquity of access, including web-based systems and the effective reuse of existing data; and the ability to work effectively within a broadening pool of partners and allies while maintaining appropriate security control. The challenges can be summarized as the conversion of what is today data into distilled information and knowledge. Yet analysis methods have not evolved to rapidly integrate multiple sources of data to create actionable intelligence. Nor do today's means of information dissemination, indexing and preservation suit this new agenda or future needs. NGA will play a major role for the entire intelligence community in creating the next-generation National System for Geospatial-Intelligence and has set forth a consistent vision of what this next generation GEOINT should be. The vision is intended to see NGA through the transition into a new era. NGA also plays a leading role in supporting fundamental research for the next generation of GEOINT, termed GEOINT2 in this document. It is within this context that the National Academies was asked by the NGA to identify research priorities and strategic directions in geospatial science for the NGA’s Basic and Applied Research Program. The goal of the study was to examine both “hard problems” in geospatial science that must be addressed to improve geospatial intelligence, as well as promising methods and tools to pursue in geospatial science and related disciplines. The results of this study are intended to help NGA’s Chief Scientist to anticipate and prioritize geospatial science research directions and, by doing so, to enhance NGA’s mix of research as it addresses these priorities. NGA has defined their “Top 10 Challenges” for GEOINT. Using these as a base, along with knowledge of the current state of the art in geospatial information science, the hard research problems associated with each of the GEOINT challenges were identified, leading to a total of 12 recommendations. The hard problems are summarized in Table S-1 below. Several promising methods and techniques for approaching each of the hard problems are addressed in the body of this report.
While it is useful to associate the hard research problems with the GEOINT challenges, it is also instructive to look at them in the context of the GEOINT process. This study puts forth a framework that describes the GEOINT2 process information flow. The key stages in this geospatial information flow are to acquire, identify, integrate, analyze, disseminate, and preserve. Consequently the hard problems are linked to one or more steps in the process flow which they impact. Looking at the hard problems both in terms of an overall GEOINT challenge and in goals. The success of the research program in creating new technologies and techniques to address NGA’s GEOINT vision is dependent not only on the focus of the research, but also on the research process itself. NGA-led research is conducted through a wide variety of programs inside and outside of NGA, including academic research grants, Broad Area Announcements, contracts, and funding to various agencies and organizations. The effectiveness of the research process has become even more important since a considerable part of the research activity in geographic information science now has some roots in NGAfunded programs. Therefore, this study makes five recommendations to increase the effectiveness of the research process. To improve the coordination of the research program, the committee recommends increasing the use of peer review and better defining the roles, responsibilities and relationships of the various participants in NGA research. To increase the number of basic research projects that result in the development of new technologies and techniques that can be incorporated into the GEOINT process, the committee recommends an improved definition of the current and future information systems architectures, and a clear plan for migrating R&D projects into these architectures, including better integration with open systems architectures. To maximize the pool of research expertise available to NGA, the committee recommends working to involve the geospatial science and technology community from our coalition countries. Finally, the complete set of recommendations is given a priority of 1 to 3, with 1 being the highest priority. The priorities are summarized in Table 6.2 of the report, and are proposed for consideration by NGA as it works to develop a research agenda to support the evolution to GEOINT2, the geospatial intelligence infrastructure for the twenty-first century.