Improving GEOINT Access for Health and Humanitarian Work in the Global South

Case studies on resource inequity with respect to GEOINT in the Global South

global-south-3

The “Global North” and “Global South” are generally distinguished by their respectively higher and lower economic and development profiles. With respect to geospatial intelligence (GEOINT), they also exist as parallel yet distinctly different worlds. The marked dominance of the U.S. within the GEOINT sphere diminishes our appreciation for operational challenges in the Global South, where critical authoritative data and geospatial infrastructure are lacking. Humanitarian activities, including disaster mitigation, service delivery to refugees and internally displaced people, and multinational efforts such as the Global Health Security Agenda’s mission to secure the world from “global health threats,” are constrained by that region’s variable geospatial capacity. Spatial data, also known as geospatial data, is information about a physical object that can be represented by numerical values in a geographic coordinate system. The increasing availability of geographically referenced base layer data, geo-referenced imagery sources, improved processing, and crowdsourced data enable rigorous and complex analyses with more granular outputs that allow analysts to target specific locations and populations. However, owing to a dearth of geospatial expertise, core data layers, and technical and financial resources, GEOINT capabilities remain out of reach to many countries. Such resource inequity presents a significant challenge that is further amplified in conflict areas, where current, precise, and, wherever possible, verified ground-reference data are mission-critical.

In the Global North, discussions on the “state of the art” reflect the ubiquity of fundamental GEOINT capacities including automated feature extraction and change detection; big data analytics and geospatial presentation; access to topical, relevant, and quality geospatial data; the tools and knowledge required to execute fundamental geospatial processes; and to a lesser but increasing degree, machine learning (ML) and artificial intelligence (AI). While there are notable exceptions, outside of capitals and major cities, a significant part of the Global South is bereft of basic information and communications technology (ICT) prerequisites—such as consistent electricity and internet access—needed to routinely and accurately conduct geospatial work.

National and local government support for health and humanitarian efforts vary, and the onus of procuring quality geospatial data may be left to aid and health agencies, few of which have the capacity to meet this immense need. Additionally, GEOINT fundamentals, such as current census data or authoritative base layers, are often outdated or non-existent, sometimes at the country-level, and especially below second- or third-order administrative-level boundaries. Further complicating access to authoritative data, governmental and other institutions may restrict data for a variety of reasons (which may run counter to their missions to improve the well-being of their constituencies). Restriction of these authoritative datasets may arise from political sensitivities, protection of funding streams through data dominance, or deflection of questions concerning data quality.

In the absence of open-source, authoritative data, crowdsourcing platforms such as OpenStreetMap, HealthMap, Wikimapia, and CrisisMappers fill important gaps by providing egalitarian scaffolding that supports data aggregation, curation, and management. However, it is important to recognize the intrinsic limitations of user-generated and “found” data. Free and open-source geospatial platforms such as Google Earth and QGIS have had a similarly democratizing impact on geospatial software utilization within the minority of Global South countries with dependable internet access. However, while data and tools are necessary, they are not sufficient to enable true access. Geospatial expertise is the third leg of the access “stool” required to maximize data utilization. In the Global South, the preponderance of technical capability and data dwells among national-level government, multilateral, and academic institutions rather than among implementing staff or non-governmental organizations (NGOs) that operate at the subnational level.

Given this divide in geospatial resources, improved collaboration is critical among private, bilateral, and multilateral stakeholders that have access to data, expertise, and imagery. Recent examples from the global health arena illustrate how GEOINT practitioners have contributed to effectively target service delivery through a combination of imagery analysis and inexpensive, creative, low-tech ground-referenced datasets. Further field and sky coordination of GEOINT capabilities in conjunction with activity-based analytics hold significant potential to strengthen disaster mitigation response as well as civil and military humanitarian actions. We discuss the role of data access and recent achievements targeting infectious disease and humanitarian responses in remote and conflict-ridden areas as examples of successful collaborative and multidisciplinary approaches to GEOINT of benefit to the Global South.

  • This article is part of USGIF’s 2018 State & Future of GEOINT Report. Download the PDF to view the report in its entirety and to read this article with citations. 

Case Studies 

The resource inequity with respect to GEOINT that the Global South faces necessitates a continuous stream of outside financial, technical, and human resources to establish and maintain parity with the Global North. To some degree, this may account for the prominent rise of crowdsourced labor and online data sharing platforms (e.g., Ushahidi, Swift River, OpenStreetMap, Tomnod) for near real-time reporting. There have been several recent public-private efforts, however, to create sustainable solutions by investing in GEOINT infrastructure and expertise that illustrate the long-term value proposition to both donors and countries. The following case studies illustrate how providing access to authoritative base layers as well as specialized knowledge and resources such as imagery classification tools and automated feature extraction can solve problems, leverage further investment, and highlight new opportunities to bridge the North-South GEOINT divide.

1. The Global Polio Eradication Initiative (GPEI), a public-private partnership with the goal to eradicate polio worldwide, exemplifies the application of geospatial data and analysis to solve a humanitarian problem while building technical capacity to create sustainable geospatial infrastructure. The use of GIS has significantly changed the trajectory of GPEI since 2007, when Google Earth was first used to develop “the river strategy”—a tactic devised to interrupt transmission of poliovirus along the Congo River in the Democratic Republic of Congo.This effort used imagery to identify settlements along the river, visualize potential trade routes and related population movement patterns, and facilitate vaccine distribution logistics by examining navigation patterns. Analytics were subsequently used to evaluate the geographic coverage of house-to-house vaccination teams; assess team performance and campaign coverage; collect location coordinates for all suspected polio cases; track post-campaign coverage surveys; and collect microcensus data to support imagery-based population estimates. The granular geospatial reference data collected in Nigeria for polio eradication also resulted in the Vaccination Tracking System (VTS) platform, which is arguably the most complete synthesis of population and health program data in Sub-Saharan Africa. In 2015, this system was adapted and successfully repurposed to avert the spread of Ebola within Nigeria.

The VTS has also provided the foundation for a major breakthrough in the field of demography, resulting from a collaboration among the Geographic Information Science and Technology (GIST) Group at Oak Ridge National Laboratory, the Bill & Melinda Gates Foundation, and Sweden-based Flowminder Foundation. This group developed population estimates for gender and standard 0-12-month and five-year age groupings at a resolution of 90 meters, based on settlement feature extraction and microcensus data.

The creation of this extensive GIS infrastructure in Nigeria led to additional base-mapping efforts in the other Lake Chad Basin nations of Cameroon, Chad, and Niger, as well as the Democratic Republic of Congo, Mozambique, and Somalia. These activities revealed significant data gaps such as the identification of hundreds (Mozambique, Somalia) and sometimes thousands (Nigeria) of previously unrecorded place names and error rates in authoritative data that have been known to exceed 50 percent.

This work also spurred the formation of two informal, virtual stakeholder GIS working groups with representation from the U.S. government, UN, and private organizations and NGOs for East and West Africa. These working groups afford an important opportunity to exchange information on planned and completed regional activities and a professionally curated library through which geospatial data, tools, and analyses can be shared among partners. This activity has facilitated the exchange of base layer data among local humanitarian efforts with regional and supraregional organizations in remote environments in Cameroon, Mozambique, and Somalia.

2. Civil and military conflicts also pose an obvious barrier to humanitarian and disease control efforts. In the face of limited authoritative geospatial and census data, a number of multilateral, humanitarian, and academic groups have designed innovative, multisourced solutions to conduct needs assessments, deliver services, and monitor human rights violations in the region. For example, Boko Haram insurgents have occupied and destroyed villages throughout Northern Nigeria since 2008. Airstrikes and military raids have wrought further destruction, leaving humanitarian agencies reliant upon imagery and analysis from donors and commercial entities to maintain situational awareness in non-permissive areas. Even with myriad resources used to identify locations and estimate populations, remotely sensed data have limitations and a network of reliable human informants is required to validate information gleaned in these high-threat areas. By fusing imagery analysis and fresh key-informant data, villages can potentially be described as sustaining complete structural damage or partially/fully intact—and potentially whether inhabited—informing how the flow of internally displaced populations and refugees is monitored within the region.

3. A combination of geospatial data, imagery, and activity-based analysis has also been used to investigate and respond to outbreaks of guinea worm disease in humans, dogs, and baboons. Individuals are infected through the consumption of water contaminated with the parasite’s larvae. Breaking the transmission cycle requires the treatment of water sources to kill the larvae in the intermediate host, identification of other cases in the area, and preventive efforts through education and water filtration. Mounting a comprehensive response thus requires identification of all stagnant water features proximal to areas inhabited by infected humans, dogs, and baboons. In remote areas of Ethiopia, where the disease was detected in a baboon troop, authoritative geospatial data are sparse, and, while maps displaying water features may be available, seasonality plays a major role in water level, flows, etc. Thus, seasonally accurate, high-resolution imagery granular enough to reveal large game trails and walking paths used by baboons and humans to reach water sources was critical to formulate a response plan. Two-dimensional printed paper maps, rather than tablet- or computer-displayed imagery, also played a key role in communicating with local guides unfamiliar with digitally displayed data. In this case, the provision of technical assistance in addition to geospatial assets has not simply supported guinea worm eradication efforts in southwestern Ethiopia, it also increased GEOINT capacity where there was little and introduced new ways to approach a complex logistical problem.

Capacitating Access and Utilization

In addition to these examples, potential use cases with benefits that extend beyond the Global South are plentiful. For example, service delivery to refugee and internally displaced persons (IDPs) could benefit significantly from improved data fusion. While the United Nations High Commissioner for Refugees (UNHCR) and NGOs strive to maintain current maps of IDP and refugee camps, these data are not always geo-referenced and thus opportunities to integrate multiple data types may be missed. The creation of “neighborhood” level maps that enumerate households would facilitate linkage of specific populations with appropriate services and follow-up. In the absence of such granular data, it may be incumbent upon residents to seek social, health, and protection services, which may be difficult for the infirm, aged, unaccompanied children, or women without freedom of movement. Geo-referenced, neighborhood-level, multilayer IDP and refugee camp data could also be used to evaluate the equitable distribution of services and, in conjunction with human activity patterns, monitor security incidents within the camp, while simultaneously assisting with situational awareness throughout the host area. Finally, this type of data affords an extension of services for returnees and protection monitoring as people transition from the care of agencies, such as UNHCR and implementing partners, back to their areas/countries of origin.

The increasing role of GEOINT as a form of social, political, programmatic, and technical currency is a countervailing influence on multilateral efforts to build sustainable technical and human capacity. In the absence of an incentivized sharing culture, a unified effort by the global development community can be successful in breaking this data-sharing impasse. One such effort is the Geospatial Reference Information Database (GRID) project, which aims to create open-source geospatial reference layers in priority developing countries selected by donor-partners, along with building local capacity to use, manage, and sustain the datasets at the country level. Co-funded by the United Kingdom’s Department for International Development, GRID will engage the United Nations Population Fund to support geo-referenced national censuses in all countries, which represent the “gold standard” in reference data. A key requirement of GRID is countries must be willing to expose their base reference data layers to include settlement names and locations, key points of interest, validated administrative boundaries, and GIS-modeled population estimates to a global, public platform. Such significant and freely available GIS infrastructure can directly improve digital democracy and potentially attract further investment, which could bolster the labor market for geographers, GIS specialists, and related expertise.

Additionally, several for-profit firms such as DigitalGlobe, Planet, Google, and Esri facilitate access to the imagery, data, tools, and expertise required for humanitarian and other activities consistent with their missions. In the non-profit sector, organizations such as Radiant Earth offer free access to open-source satellite, aerial, and drone imagery archives from across the globe, alongside the analytic tools that enable greater access for organizations with less technical and financial resources. Other opportunities to link sky and field lie in the integration of geospatial data with complementary computational capabilities such as AI and ML—as Palantir and the Carter Center have effectively demonstrated through the Syria Conflict Mapping Project. Through these efforts, the playing field is slowly being leveled to create parity between those with the greatest capacity and those whose access is currently dependent upon educational institutions, donors, or fee-for-service expertise. Access to free imagery, geospatial data, and analytic capacity alone is a social good in that it moves forward academic research on modeling and methods for validation. However, significant unmet needs awaiting creative and synergistic solutions remain, so continued support for robust, open-source GEOINT tools and expertise is essential to provide effective and sustainable support to the Global South.

Headline Image: A child is vaccinated against polio in the Central African Republic, November 2017. Photo Credit: UNICEF CAR

Do We Need A “Civilian ARPA” for AI?

The case for leveraging artificial intelligence to improve public service

Making Sense of Compression

Understanding the three common types of raster image compression

The Necessity of GPUs

Resolutions have surpassed Moore's Law: What does that mean for data analysis?