Quantifying environmental migration is challenging given the multiple drivers of such movement, related methodological challenges and the lack of data collection standards. Some quantitative data exist on population displacement within a country, and to a lesser degree across borders, due to natural hazards. However, for migration due to slow-onset environmental processes, such as drought or sea-level rise, most existing data are qualitative and based on case studies, with few comparative studies. While data gaps persist, research methodologies are constantly being improved.
Three key terms are important in the context of migration and environmental and climatic changes:
- Environmental migrants are defined as “persons or groups of persons who, predominantly for reasons of sudden or progressive changes in the environment that adversely affect their lives or living conditions, are obliged to leave their habitual homes, or choose to do so, either temporarily or permanently, and who move within their country or abroad.” (IOM, 2011: 33 in IOM, 2014:13).
- Environmentally displaced person refers to “persons who are displaced within their country of habitual residence or who have crossed an international border and for whom environmental degradation, deterioration or destruction is a major cause of their displacement, although not necessarily the sole one” (IOM, 2011:34 in IOM, 2014:13). The term disaster displacement “refers to situations, where people are forced or obliged to leave their homes or places of habitual residence, in particular as a result of or in order to avoid the effects of disasters triggered by natural hazards. Such displacement may take the form of spontaneous flight or an evacuation ordered or enforced by authorities. Such displacement can occur within a country, or across international borders. ” (The Nansen Protection Agenda, 2015)
- Planned relocation refers to persons whose livelihoods have been re-built in another place (IOM, 2014a). Others have defined planned relocation as referring solely to the collective movement of a community, the “permanent (or long-term) movement of a community (or a significant part of it) from one location to another, in which important characteristics of the original community, including its social structures, legal and political systems, cultural characteristics and worldviews are retained: the community stays together at the destination in a social form that is similar to the community of origin” (Campbell, 2010:58–59).
Although the term “climate refugees” is often used in relation to forced migration in the context of climate and environmental change, this is not a legally valid term as the 1951 Refugee Convention does not recognize environmental factors as criteria to define a refugee.
At the end of 2019, around 5.1 million people in 95 countries and territories were living in displacement as a result of disasters that happened not only in 2019, but also in previous years. (IDMC, 2020). The countries with the highest number of internally displaced persons were Afghanistan (1.2 million); India (590,000); Ethiopia (390,000), Philippines (364,000) and Sudan (272,000) (ibid.).
In 2019, nearly 2,000 disasters triggered 24.9 million new internal displacements across 140 countries and territories; this is the highest figure recorded since 2012 and three times the number of displacements caused by conflict and violence (ibid.) Most of the disaster displacements were the result of tropical storms and monsoon rains in South Asia and East Asia and Pacific; four countries accounted for more than 17 million new internal displacements due to disaster: India (5 million), the Philippines (4.1 million), Bangladesh (4.1 million), and China (4 million) (ibid.)
While the majority of mobility in the context of environmental and climate change more generally, including disaster displacement, occurs within the borders of countries, some people are forced to move abroad. Global data on cross-border movement in the context of disasters are, however, limited, with only a few notable cases being examined so far (Nansen Initiative, 2015; Ionesco, Mokhnacheva and Gemenne, 2017). In some cases, official sources on humanitarian visas by countries such as the United States (US), Brazil and Argentina for Haitians can be used.
Slow-onset processes such as droughts or sea level rise also increasingly affect people’s mobility worldwide. Though specific data are not available, case studies are highlighted by existing research, for example: Foresight, 2011; Piguet and Laczko, 2014; Ionesco, Mokhnacheva and Gemenne, 2017.
The relocation of communities in the context of environmental and climate change is also increasingly implemented by governments (for a summary of recent relocation programmes see Ionesco, Mokhnacheva and Gemenne, 2016; Benton, 2017 and Georgetown University, UNHCR and IOM, 2017). For instance, tens of thousands of people have been relocated in Haiti (Pierre, 2015) and in Viet Nam (UN Viet Nam, 2014; Chun 2014; Entzinger and Scholten, 2015); hundreds of thousands in Ethiopia (Foresight, 2011: 177); about a million in the Philippines (Ranque and Quetulio-Navarra, 2015; Thomas, 2015; Brookings and UNHCR, 2015: 3-4) and several millions in China (Foresight, 2011: 177).
Comprehensive datasets on environmental migration or planned relocation do not yet exist at the global level, but several initiatives have started to collect information across several countries. The following list provides an overview of the existing available information, including more qualitative research.
Primary data collection:
National authorities collect information on displacement and evacuations linked to disasters, in particular fast onset ones. Local-level disaster displacement data are available from (international and national) humanitarian agencies (NGOs, UN agencies) engaged in relief operations, which collect data in order to respond to the needs of affected populations.
Administrative data sources, such as the numbers of humanitarian visas (such as in the US, Brazil, Ecuador or Mexico) or residence permits granted (for instance, by Argentina) linked to disasters, can provide information on cross-border displacement and movements in the context of environmental events more generally.
IOM’s Displacement Tracking Matrix (DTM) is a system used to track and monitor disaster displacement and population mobility. Data are regularly captured, processed and disseminated to provide a better understanding of the movements and evolving needs of displaced populations and migrants, whether in situ or en route, before, during and in the aftermath of disasters. The data are presented in the DTM Data Portal.
Innovative data sources include mobile phone-based sources such as call detail records (CDRs). Big data generated by mobile phone users before and after disasters, such as the 2010 earthquake in Haiti (Bengtsson et al., 2011) and several typhoons in the Philippines and Bangladesh (Lu et al., 2016), can indicate where displaced persons moved to and help deliver prompt and targeted humanitarian assistance or to understand internal movements (Laczko and Rango, 2014; GMG, 2017). This can be a means to collect complementary quantitative data on movements at small-scale and on seasonal patterns linked to adaptation to environmental change and disasters that are difficult to account for in traditional household survey tools (Lu et al., 2016). Other projects aim at using big data sources, such as satellite images or social media data, to identify early the environmental stressors that could lead to displacement (see for instance Isaacman et al., 2017).
Several research projects have and are collecting new data on the links between the environment and human mobility, but few with a comparative approach. There are two notable exceptions. First, the Migration, Environment and Climate Change: Evidence for Policy (MECLEP) project, implemented by IOM and six research partners in 2014-2017, and funded by the EU, conducted a comparative quantitative and qualitative study of six countries (Dominican Republic, Haiti, Kenya, Papua New Guinea, Mauritius and Viet Nam). The methodology developed for the project could easily be applied to other countries.
Second, the Pacific Climate Change and Migration (PCCM) project by ILO, UNESCAP and UNDP focused on Tuvalu, Nauru and Kiribati. The United Nations University Institute for Environment and Human Security (UNU-EHS) released findings detailing how climatic changes are impacting these Pacific island states.
Secondary data sources and research:
The Internal Displacement Monitoring Centre (IDMC) has compiled data on internal displacement in the context of disasters since 2008 globally (data are generated by event, not by country) through its online Global Internal Displacement Database (GIDD). The estimates are based on information by national authorities, UN agencies such as IOM, the International Federation of the Red Cross (IFRC) and the UN Office for the Coordination of Humanitarian Affairs (OCHA), non-governmental organizations and in particular media reports. Figures are published in the annual Global Report on Displacement (GRID), which also covers internal displacement due to conflict and violence. IDMC is developing methodologies to map and assess future disaster displacement risks and is starting to gather data on cross-border displacement.
The HELIX project (High-End Climate Impacts and Extremes) provided research on climate impacts and adaptation in relation to varying global warming scenarios (2, 4 and 6 degrees Celsius), using predictive analytics. Human migration was included in the impact studies. The recent Groundswell: Preparing for internal climate migration report (Rigaud et al., 2018) developed a model for future population distribution in 2050 in three regions (sub-Saharan Africa, South Asia and Latin America) if no action is taken.
The CLIMIG database of studies on environmental migration, both of qualitative and quantitative nature, was developed by the University of Neuchatel (Switzerland).
The Environmental Migration Portal by IOM, features a searchable research database, initially based on the People on the Move in a Changing Climate: A Bibliography, published by IOM in collaboration with University of Neuchatel. The database also includes migration and environment country assessments published by IOM.
The thematic working group on “Environmental change and migration” of the Global Knowledge Partnership on Migration and Development (KNOMAD) produced an annotated bibliography on Environmental Migration and developed a toolkit on planned relocation with many case study examples (Georgetown University, UNHCR and IOM, 2017).
The first “Atlas on Environmental Migration” was produced by IOM and Sciences Po, Paris (published with Routledge in 2017). The publication brings together, for the first time, existing knowledge on the links between migration and environmental change, presented through comprehensive maps, diagrams and case studies.
The Hugo Observatory at the University of Liége (Belgium) focuses on research on environmental changes and Migration.Back to top
Data strengths & limitations
Over the past decade, important advances on methodologies and data collection have been made. Academic researchers and specialized agencies are working on improved methodologies for comparative cross-country or cross-region studies, agent-based models and multi-factor simulators designed to predict future trends (such as drought-induced displacement modelling, Ginnetti and Franck, 2014, or IDMC’s Global Displacement Risk Model focused on sudden-onset disasters based on housing destructions), and hotspot identification triangulating environmental and social data, all of which can contribute greatly to improving current evidence and future projections of environmental migration trends so as to better inform policies and action.
Innovative data sources: Big data can provide opportunities that can further be strengthened in trying to estimate the extent of movements in contexts of disasters and degrading environments. These new methods can fill gaps in time series data, indicate where people have moved from and to and enhance the timeliness of this information. In some cases, these new methods could be used to inform life-saving early warnings. At the same time, privacy safeguards and ethical considerations need to be adhered to.
Nonetheless, difficulties remain.
- It is challenging to differentiate when the environment is the main factor triggering migration, rather than or in combination with other factors: In most cases, environmental factors are closely linked to socioeconomic, political, demographic, cultural and personal factors that play a role in leading to or preventing mobility (Laczko and Aghazarm, 2009; Foresight, 2011), which makes data collection beyond fast onset disasters leading to evacuations difficult. Information on people moving due to more gradual, so-called slow-onset processes like sea level rise or salinization, is scarce for methodological reasons.
- The most comprehensive data available only track people newly displaced internally that year: Thanks to IDMC’s work, data on internal displacement due to natural hazards are available for almost all countries. However, differing definitions used by data providers and a lack of reporting by countries remains a challenge, leading to media reports being an important source of events covered in the estimates. IDMC’s data aggregates focus exclusively on people newly displaced during the year of interest. This figure reflects the flows of people during one year (or stocks by the end of the year) and does not capture the duration of people’s displacement, their return home or relocation elsewhere, those not sheltered in camps or people caught in long-term displacement, so-called protracted situations, from year to year. Data collection on cross-border movements after disasters is only starting and limited to localized case studies (IDMC, 2018 (a)). Further research on cross-border disaster displacement is being supported as part of the work of the Data and Knowledge Working Group of the State-led Platform on Disaster Displacement.
- Underreporting: The quality and the availability of data on displacement vary between countries and from event to event: small-scale events or disasters that occur in isolated and marginalized areas are under-reported and thus not included in the available aggregate estimates (IDMC, 2017: 98; IDMC, 2018 (a)).
- Little information on links between conflict and disaster displacement: In cases where conflict is linked to disasters information on movements is lacking, in particular on displacement histories that could inform future predictions (IDMC data for instance are only available since 2008 but since 2017 include drought figures).
- Comprehensive datasets on environmental migration or planned relocation are needed: Data on environmental migration and planned relocation have improved in recent years, as an increasing number of studies have been conducted in affected areas. The research databases listed above are important tools providing an overview of the existing available information. However, comparable quantitative, longitudinal, disaggregated and georeferenced data are needed to assess how different forms of mobility can be a beneficial adaptation strategy and what potential risks need to be minimized. The majority of existing surveys focus mainly on the links between migration and the environment as a driver, and are mostly qualitative in nature. More information is needed on the impacts of those movements on adaptation to environmental and climate change.
- Few data on trapped populations: Some populations affected by environmental degradation and disasters may not be able to move due to a lack of financial resources or social networks. They are highly vulnerable populations, but data to inform action and protection are scarce.
- Better predictive analytics are needed: When it comes to predicting future trends, the disconnection between the environmental sciences and social sciences communities constitutes an additional challenge, in a context where environmental migration research would greatly benefit from multidisciplinary research and better integration of climate and population data.
Back to top