In recent years, the rise of cross-border migration has captured media and public attention around the globe. At the same time, significant numbers of people are migrating within the borders of their home countries rather than beyond them. Although the drivers of this internal migration are multifaceted, it’s clear that the slow onset impacts of climate change—such as increased temperatures, drought, reduced agricultural productivity, rising sea levels, and storm surge—have spurred people’s movements from vulnerable to more viable areas of their countries in search of a better life, and will continue to do so for the foreseeable future.
According to The World Bank’s Groundswell report, “Climate change, an increasingly potent driver of migration, could force 216 million people across six world regions to move within their countries. Hotspots of internal climate migration could emerge as early as 2030 and continue to spread and intensify by 2050. By [that time], Sub-Saharan Africa could see as many as 86 million internal climate migrants; East Asia and the Pacific, 49 million; South Asia, 40 million; North Africa, 19 million; Latin America, 17 million; and Eastern Europe and Central Asia, 5 million.”
To help members of the international development and humanitarian assistance communities, policymakers, and others better understand how slow-onset climate change might affect internal migration in low- to middle-income countries in the coming decades, NASA’s Socioeconomic Data and Applications Center (SEDAC) has announced the release of the Groundswell Spatial Population and Migration Projections at One-Eighth Degree According to SSPs and RCPs, v1 (2010 – 2050) dataset.
The Groundswell dataset provides projections of future population distribution and migration at a resolution of one-eighth degree (or 7.5 arc-minutes, which are approximately 15-kilometer grid cells at the equator) for 112 countries. The data were produced by the Groundswell project of The World Bank, which uses a novel population gravity modeling approach based on gridded population data. The Groundswell methodology relies on spatial population projections with and without climate impacts using a combination of Shared Socioeconomic Pathways (SSPs) only, for the former, and SSPs combined with climate impacts according to different Representative Concentration Pathways (RCPs) for the latter. (Note: an in-depth explanation of the gravity modeling approach and the Groundswell methodology is available in the dataset documentation.)
The modeling work was led by Dr. Bryan Jones at the City University of New York (CUNY) Institute for Demographic Research (CIDR) and Dr. Alex de Sherbinin at Columbia University’s Center for International Earth Science Information Network (CIESIN), who also serves as SEDAC deputy manager. They are co-authors on both the original 2018 Groundswell report and the subsequent update, Groundswell II, published in 2021.
The dataset includes the following variables for each 7.5 arc-minute grid cell:
- Baseline 2010 population counts and density per square kilometer
- Projected population counts and densities for each time slice (2020, 2030, 2040, 2050) for the SSP-only “no climate” projections, individual model runs that include climate impacts, and ensembles for the four scenarios (SSP-RCP combinations)
- Projected climate migration for each time slice (2020, 2030, 2040, 2050) for each model run and scenario
- Hotspots of climate migration (2030, 2040, 2050), indicating agreement among the scenarios for the areas with greatest in- or out-migration stemming from climate impacts.
According to de Sherbinin, “Owing to the wide reach of the Groundswell reports, statistics from the reports have been widely cited in the media. Having well-documented data available to other researchers is important for transparency and open science since some may wish to interrogate the results.”
As de Sherbinin points out, anyone interested in scrutinizing the report’s results should note that anytime there are projections of future social dynamics in a complex socio-ecological system there are going to be big uncertainties. “Climate projections have well-documented uncertainties, which are carefully qualified in Intergovernmental Panel on Climate Change reports,” he said. “We take climate projections two steps further: we use sectoral impact models of crop production and water availability that use the climate projections as inputs. Then we calibrate the population gravity model by assessing the sensitivity of past shifts in population distribution to these sectoral impacts and project future shifts in population on the assumption that the population will maintain a similar sensitivity to climate impacts out to the year 2050. This is a big assumption.”
Such uncertainties can be addressed, at least in part, by running multiple scenarios, a common practice in the climate literature. Nevertheless, de Sherbinin advises that Groundswell data should be understood as plausible future scenarios with high levels of underlying uncertainty. An additional uncertainty is the fact that future climate adaptation measures are explicitly not taken into account.
“They might be best thought of as one input among many that a decision-maker might use to inform policies and programs [on] where to invest climate adaptation funding, how much infrastructure to build out in anticipation of population growth, [and forecasts of] where people may be moving towards areas of vulnerability, such as the coastal zone, etc.,” he said.
To that end, de Sherbinin said he could foresee researchers “using the data in conjunction with remotely sensed and other spatial data to understand potential future trends in climate mobility in given geographical zones, such as the low elevation coastal zone, mountainous areas, forest lands, and the like.”
He also suggested that national governments may want to use these data in conjunction with other data to better understand future climate impacts on population distribution.
SEDAC, which is operated by CIESIN, is the NASA Earth Observing System Data and Information System (EOSDIS) Distributed Active Archive Center (DAAC) responsible for archiving and distributing socioeconomic data within the EOSDIS collection. As such, it offers three additional population-focused datasets that, according to de Sherbinin, may serve as a complement to Groundswell data:
- Global One-Eighth Degree Population Base Year and Projection Grids Based on the SSPs, v1.01 (2000 – 2100)
Provides global urban, rural, and total population base year and projection grids based on SSP data at a resolution of one-eighth degree (unlike Groundswell, which uses two SSPs, these provide scenarios according to all five SSPs)
- Global 1-km Downscaled Population Base Year and Projection Grids Based on the SSPs, v1.01 (2000 – 2100)
Provides global urban, rural, and total population base year and projection grids based on the SSPs at a resolution of 1-kilometer (this is essentially a higher resolution version of the dataset above)
- Global Estimated Net Migration Grids By Decade, v1 (1970 – 2000)
Provides estimates of net-migration (in-migration minus out-migration) per one-kilometer grid cell on a decadal basis for the 1970s, 1980s, and 1990s
The Groundswell dataset’s projections of future population distributions with and without climate impacts, along with future estimates of climate migration, have already been used by The World Bank for country consultations related to development planning. Further, using the dataset to anticipate the potential scale of population shifts as a result of climate impacts can be useful to the humanitarian assistance community, which uses forecasts of displacement to locate aid and social services; urban planners, who endeavor to understand the magnitude of future flows from rural areas; and policymakers, who use “what if” scenarios based on alternative futures to anticipate the effects of proposed policies.
Of course, the Groundswell dataset’s projections of future population distributions are just that, projections; how internal climate migration actually plays out during the next half-century will, according to The World Bank, “depend on our collective action on climate change and development in the next few years.” Slowing that migration, the The World Bank says, will require urgent “action to reduce greenhouse gas emissions to reduce the climate pressures that drive internal climate migration.”
The World Bank acknowledges that not all migration can be prevented. However, if well-managed, such shifts in the distribution of populations can become part of an effective adaptation strategy, allowing people to rise out of poverty and build resilient livelihoods. Having the Groundswell dataset freely and openly available from NASA’s SEDAC can benefit the development and management of any internal climate migration plan.
How to Access the Data
The Groundswell Spatial Population and Migration Projections at One-Eighth Degree According to SSPs and RCPs, v1 (2010 – 2050) dataset is available through Earthdata Search as well as from the SEDAC website. At SEDAC, users also will find several maps of Groundswell Projections 1/8th degree SSPs and RCPs (2030, 2050) for Africa, South Asia, and East Asia.
- Groundswell data are available in GeoPackage (GPKG) and Esri File Geodatabase (GDB) formats for each of the 112 countries modeled.
- The countries are listed in Table 5 of the product documentation (see page 24) and each file includes the International Organization for Standards three-character alphabetic code (ISO3 code) listed in the table. The data dictionary includes field names and descriptors for 204 variables found in each file. This is packaged with each download as a separate Excel file.
- Each downloadable file is a compressed zip file containing the GPKG or GDB file, a Microsoft Excel file (XLSX) data dictionary, and PDF documentation.