Skip to content

Climate Change and Migration: Definitions and Mechanisms

Climate Change and Migration: A Concept Map | MSHA-540 Module 2
— Concept Map / MSHA-540 / Module 2 — Vol. 01 — Iss. 01

Climate Change
and Migration

Four manifestations, five drivers, two-axis scenarios — and the political infrastructure actively dismantling them. A graphic organizer of the field, and of what the assigned literature cannot account for.

— research artifact —

What the Readings Cannot Know

A compact timeline of publication dates against ground-truth events. The three assigned readings predate every event below the axis. The framework they assume operates inside an institutional landscape that has since been dismantled.

2018 2019 2020 2021 2023 Jan 25 Jul 25 2026 ↑ PUBLICATION ↓ GROUND TRUTH Mechler et al. Oct 2018 Loss & Damage Ferris Dec 2020 “field in formation” Groundswell II Sept 2021 3.5 year gap → Jan 2025 EO foreign aid freeze Feb 2025 AfD 20.8% May 2025 Italy Remig. Summit July 2025 USAID SHUT Nov 2025 DHS: “Remigration” Dec 2025 WH remig. post
Sources: Mechler et al. (2018); Ferris (2020); Clement et al. (2021); KFF USAID Timeline (2026); GPAHE (2026); CSO Hate Studies (2026); Time (2026).

The cluster of yellow and orange dots in the final twelve months of the axis marks a categorical shift — not in the empirical drivers of climate-related mobility, but in the political infrastructure assumed by every recommendation in the literature. Each of the three readings is internally coherent. None can address the actor-collapse and frame-reversal that occurred after publication.

Master Concept Map

Six conceptual layers. Each adds a modulator on the one above. The final layer is what the assigned literature cannot see.

01 — CLIMATE HAZARDS sudden-onset (cyclones, floods, wildfires) · slow-onset (SLR, drought, salinization) [IPCC AR6 WGII] 02 — DRIVER CATEGORIES (FORESIGHT PENTAGON) environmental + economic + social + political + demographic climate change does not act alone — it modifies all five [Black et al. 2011; Foresight 2011] 03 — MANIFESTATIONS ▸ ADAPTIVE household choice ▸ INVOLUNTARY displacement ▸ PLANNED organized relocation ▸ IMMOBILITY “trapped populations” [IPCC AR6 WGII Ch. 7, Cross-Chapter Box MIGRATE] 04 — PROJECTED OUTCOMES (with likelihood bands) internal climate migrants by 2050: 44M216M · cross-border: uncertain · trapped: rising [Clement et al. 2021; IPCC AR6 high confidence on direction, contested numbers] 05 — MODULATED BY RESPONSE SCENARIOS emissions trajectory × development pathway → range 44M ↔ 216M (Δ80%) presupposes: functioning multilateralism, donor states, host-state capacity [Groundswell II, 2021] 06 — ACTIVE COUNTER-MECHANISMS / “THE UNDOING” USAID dismantled · “remigration” policy export · 3rd-country deportation pipeline · asylum gutting refugee admissions collapse · Paris withdrawal · L&D mechanism stalled · protection regime defection [Trump admin 2025–26 · AfD / FPÖ / PVV / FvD / Lega / Brothers of Italy / Reform UK · empirical, ongoing]
Layers 01–05 reproduce the structure implied by the assigned literature. Layer 06 is the political infrastructure that determines whether any of the Layer 05 scenarios can materialize. It is not optional context — it is the modulator with the largest current variance.
— · — · — · —
Panel 01

Manifestations

What climate-related human movement looks like on the ground. The IPCC AR6 WGII identifies four canonical categories that overlap across two axes: agency (voluntary ↔ forced) and motion (moving ↔ staying).

↑ MOVING ↓ STAYING VOLUNTARY ← → FORCED ADAPTIVE MIGRATION household choice INVOLUNTARY DISPLACEMENT no options remain PLANNED RELOCATION state- / community-organized — crosses voluntary/forced IMMOBILITY (“TRAPPED”) too poor / sick / old to move VOLUNTARY IMMOBILITY (rooted / attached)
The four IPCC categories arranged on two axes. Planned relocation crosses the voluntary/forced line because state- and community-organized resettlement can be either. “Voluntary immobility” (rooted populations) is implicit in the literature but rarely a stand-alone category.

Why this matters

Each category contains the others in latent form. A household’s adaptive seasonal migration becomes involuntary when conditions cross a threshold. Planned relocation produces immobility for those who refuse to move. People classified as trapped often have adaptive strategies that aid agencies fail to see.

Regional predominance

RegionPredominant form
Asia-PacificInvoluntary displacement (cyclones, monsoons)
Pacific atollsPlanned relocation
SahelAdaptive (seasonal labor)
BangladeshAdaptive + involuntary (rural→urban)
Dryland AfricaImmobility (trapped)
C. AmericaCross-border adaptive turning involuntary
US Gulf / LAPlanned (Isle de Jean Charles)
Mekong DeltaPlanned + adaptive
Panel 02

Mechanisms

The causal pathways that translate climate change into mobility decisions. The Foresight (2011) framework resolves the older “environmental refugee” determinism: climate change rarely acts alone — it modifies five interacting driver categories.

ENVIRON- MENTAL ECONOMIC livelihood SOCIAL networks POLITICAL conflict DEMO- GRAPHIC CLIMATE CHANGE modifies all 5 → MIGRATION DECISION
The Foresight “pentagon” framework (Black et al. 2011). Climate change is not a stand-alone driver: it modifies all five driver categories that shape whether, how, and where people move. Migration is the multi-causal output, not a single-cause response.

Direct mechanisms

  • Sudden-onset — cyclones, floods, wildfires, storm surges (IDMC: 225M+ internal disaster displacements in Asia-Pacific 2010–21)
  • Slow-onset — SLR inundation; chronic drought; saltwater intrusion rendering land unusable

Indirect mechanisms

  • Livelihood erosion — yield decline, fisheries collapse, water scarcity reduce income, prompting migration as economic adaptation
  • Food insecurity + price shocks — harvest failure cascades into urban food prices
  • Conflict mediation — resource scarcity intersects with weak governance (Syria, Sahel debates remain contested)
  • Health system stress — vector-borne disease range shifts, heat morbidity
  • Maladaptation — dams, sea walls, “green grabbing” displace populations even pre-impact
key inversion
Climate stress can also reduce migration by impoverishing households below the threshold needed to move at all — producing involuntary immobility (trapped populations).

Regional dominant pathways

Central America
drought → livelihood erosion → out-migration to US
Bangladesh
SLR + cyclones (direct) → rural to Dhaka
Sahel
drought → conflict mediation → mixed displacement
Pacific atolls
SLR (direct) → planned relocation
Panel 03

Likelihood

How confident the literature is in different claims, and how wide projection ranges are. The field has moved away from headline “X million by Y” figures toward probabilistic, scenario-conditional projections — because outcomes depend heavily on policy choices.

Confidence matrix (IPCC AR6 WGII)

Claim IPCC confidence
Climate hazards are a growing driver of involuntary migration and displacementHigh
Causal link between extreme weather and mortality, morbidity, food/waterborne diseaseVery high
SIDS disproportionately affected relative to population sizeVery high
Climate-induced displacement is generating and perpetuating vulnerabilityHigh
Direct climate → conflict causal pathway (e.g. Syria/drought)Contested / mixed
Specific quantitative cross-border projectionsLow
Tipping-point migration (AMOC slowdown, Amazon dieback)Low confidence, high consequence
Reducing future risks possible through cooperative international effortsMedium (preconditions now contested — see Panel 05)

Projection range: internal climate migrants by 2050

0 50M 100M 150M 200M 250M 500M 1B number of climate migrants (note: scale compressed after 250M) Groundswell II (2021) 44M climate-friendly 216M pessimistic Statista (2023): ~170M Myers (2002): 200M [methodologically contested] Stern (2006): up to 1B [legacy estimate] PROJECTION RANGES — INTERNAL CLIMATE MIGRANTS BY 2050 scenario-conditional. green = best case · red = worst case · earlier point estimates shown for contrast
Groundswell II projections are explicitly conservative — they cover only slow-onset drivers (water scarcity, crop productivity, SLR + storm surge), internal movement, and exclude high-income countries, Middle East, and SIDS. Cross-border numbers are not modeled. Worst case is 4.9× the best case.
216M
pessimistic reference
high emissions + unequal development
~130M
inclusive development
high emissions + better development
44M
climate-friendly
low emissions + inclusive development
75%
stay internal
most climate-mobile populations move within their own country
Panel 04

Scenarios

Outcomes are not fixed. They depend on the interaction of emissions trajectory and development pathway. Groundswell II uses three illustrative scenarios that map onto SSP/RCP-style framings — the same two-axis logic the IPCC uses for climate futures.

→ EMISSIONS ↑ DEV. EQUITY low / Paris-aligned high / SSP3–5 inclusive unequal [low-emit unequal — unmodeled in GW] PESSIMISTIC REFERENCE INCLUSIVE DEVELOPMENT CLIMATE- FRIENDLY 216M migrants ~130M migrants 44M migrants ↘ Δ80% GROUNDSWELL II SCENARIOS (2050) two-axis: emissions × development
Three illustrative scenarios from Clement et al. (2021). The diagonal Δ80% reduction is the World Bank’s headline finding: combined mitigation + inclusive development can reduce internal climate migration by up to 80% by 2050. This entire chart presupposes the actors named in Panel 05 are functioning.

Response inputs that modulate the outcome

  • Mitigation — cutting emissions to limit hazard exposure (Paris-aligned targets)
  • Adaptation — reducing vulnerability, increasing local options (livelihood diversification, social protection)
  • Development — inclusive growth, governance, gender equity
  • Anticipatory action — forecast-based financing, planned relocation, ex-ante humanitarian

Scenario flow

POLICY INPUTS mitigation + adaptation + development MODULATE MECHANISMS livelihood resilience, hazard exposure, governance SHAPE MANIFESTATIONS adaptive ↑, involuntary ↓, planned ↑, trapped ↓ OUTCOME RANGE 44M ↔ 216M migrants by 2050
The Groundswell scenarios are constructed within a Northern-donor, state-stability paradigm. They project numbers to be managed, not rights to be honored. The same projection figure functions very differently when invoked by Frontex, USAID, or a Pacific climate-justice coalition. — framing extending Duffield (2007), de Waal (2018), Lancet CHH Commission (2024)
Panel 05 — Critical Extension

The Undoing

What the four assigned dimensions cannot register: a distinct fifth layer of active counter-mechanisms operating on every node above. The assigned readings model a system being asked to adapt. The system has, since publication, been re-tasked to do something else.

Geographic atlas — three undoings, one map

ACTIVE COUNTER-MECHANISMS — 2025/2026 deportation destinations · remigration adopters · USAID collapse zones
SCALE OF UNDOING
● 20 deportation destinations
of 27 signed; ~51 approached
◆ 10 remigration-platform states
parties in government or opposition coalitions
▨ 7 USAID collapse zones (Oxfam, 2025)
Deportation destinations: third-country removal agreements signed or operational (MPI, 2025; State Dept & DHS sources, 2025–26). Remigration adopters: parties that have formally embraced the term in platform or government statements (GPAHE, 2026; CSO Hate Studies, 2026). USAID collapse zones: regions with severe humanitarian-response shortfalls following USAID dissolution (Oxfam, 2025; KFF, 2026). Map is stylized for relational reading, not cartographic measurement.

The three inversions

Three distinct dismantlements, operating concurrently, that the assigned literature cannot register because they post-date publication and contradict the literature’s core operating assumptions.

DIMENSION USAID COLLAPSE REMIGRATION FRAME 3RD-COUNTRY DEPORTATION
Key date Jan 20 → Jul 1, 2025 Feb 2025 — Dec 2025 Through 2025; ongoing
Mechanism EO foreign-aid freeze; 82–90% of programs terminated; agency dissolved; replaced by $11B “America First Global Health Strategy” Mainstreaming of ethnonationalist removal as policy vocabulary across 10+ Northern states Negotiated transfer of asylum-seekers to states with no prior tie to the deportee; ~27 agreements
Effect on framework Removes the largest single implementing actor for “adaptive migration” interventions Inverts the donor frame: from “manage flows” to “expel populations” Inverts the direction of forced movement: toward, not away from, exposure
Reading silent All three (predate dissolution) All three (predate mainstreaming) All three (predate Lyons / Las Americas litigation)

Three structural reasons the readings cannot reach the present

01 — Unit of analysis. The literature frames climate-induced movement as a humanitarian problem requiring development response. The dominant frame in 2025–26 is ethnonationalist threat requiring removal. Same population, opposite policy verb. Ferris (2020) anticipates contested terminology; she does not anticipate that “climate migrant” would become a category targeted for active expulsion by the states most responsible for the emissions producing it.

02 — Implementing actor. Mechler et al. (2018), the Foresight inheritance, and Groundswell II all presuppose a working development-aid architecture — bilateral programs, multilateral funds, in-country implementing partners. USAID’s dissolution (July 1, 2025) removed a $51.6B-a-year line of that architecture. The “comprehensive risk management” toolkit at the center of Mechler et al. now has nobody to execute it at scale in the most exposed regions.

03 — Direction of movement. Groundswell II projects 44M–216M internal migrants by 2050, moving largely away from climate-exposed origins toward viable destinations. Third-country deportation networks route people the opposite way: from the donor states (where livability is constructed) to El Salvador’s CECOT, South Sudan, Eswatini, Uzbekistan, Libya. Anti-adaptive migration, produced by policy. Groundswell’s projection math holds; its political economy assumption does not.

The pattern lines up with what the Transborder Protection & Resilience Network concept note (MSHA-520, 2026) identifies as the central analytical move required: treating the deportation pipeline as a slow-onset, policy-driven disaster system. Once that move is made, the four assigned dimensions read coherently — Manifestations include forced involuntary outflows from donor states; Mechanisms include policy as a hazard accelerator; Likelihood folds in legal-status precarity; Scenarios admit a sixth quadrant the original 2×2 cannot draw. — extending the literature, not refusing it
Thesis

What the four panels add up to

The climate-migration nexus, read through the assigned literature, is a system of manifestations, mechanisms, likelihoods, and scenarios. Read against ground truth in 2026, it is also a system being actively contained: its predicted flows are now political liabilities that donor states are reorganizing themselves to refuse. The dominant Northern-donor humanitarian model, as Duffield (2007), de Waal (2018), Fassin (2011), and the Lancet CHH Commission (2024) have argued in different registers, was never purely a response to suffering. The 2025–26 inflection makes that legible at the level of policy vocabulary: the same projection figure invoked at Frontex headquarters and at a Pacific climate-justice coalition is doing entirely different political work. A graphic organizer that maps the four assigned dimensions without naming the fifth — the containment apparatus operating on every node above — describes the field at a moment that no longer exists.

This artifact is a research deliverable.

References

APA 7th edition. Three required readings listed first, in italics; supporting and ground-truth sources follow alphabetically.

Clement, V., Rigaud, K. K., de Sherbinin, A., Jones, B., Adamo, S., Schewe, J., Sadiq, N., & Shabahat, E. (2021). Groundswell Part 2: Acting on internal climate migration. World Bank. https://openknowledge.worldbank.org/handle/10986/36248

Ferris, E. (2020). Research on climate change and migration: Where are we and where are we going? Migration Studies, 8(4), 612–625. https://doi.org/10.1093/migration/mnaa028

Mechler, R., Bouwer, L. M., Schinko, T., Surminski, S., & Linnerooth-Bayer, J. (Eds.). (2018). Loss and damage from climate change: Concepts, methods and policy options. Springer. https://doi.org/10.1007/978-3-319-72026-5

— — —

de Waal, A. (2018). Mass starvation: The history and future of famine. Polity Press.

Duffield, M. (2007). Development, security and unending war: Governing the world of peoples. Polity Press.

European Council on Refugees and Exiles. (2026). Asylum policy tracker. https://ecre.org/

Fassin, D. (2011). Humanitarian reason: A moral history of the present (R. Gomme, Trans.). University of California Press.

Foresight. (2011). Migration and global environmental change: Final project report. UK Government Office for Science.

Global Project Against Hate and Extremism. (2026). Remigration tracker. https://globalextremism.org/

IPCC. (2022). Climate change 2022: Impacts, adaptation and vulnerability [H.-O. Pörtner, D. C. Roberts, M. Tignor, E. S. Poloczanska, K. Mintenbeck, A. Alegría, M. Craig, S. Langsdorf, S. Löschke, V. Möller, A. Okem, & B. Rama, Eds.]. Cambridge University Press. https://doi.org/10.1017/9781009325844

KFF. (2026). USAID timeline: A chronology of dissolution. https://www.kff.org/

Martin, A., & Prysner, M. (Directors). (2025). Earth’s greatest enemy [Film]. Empire Files.

Migration Policy Institute. (2025). Third-country deportation agreements: 2025 review. https://www.migrationpolicy.org/

Oxfam International. (2025). Humanitarian collapse zones following USAID dissolution. https://www.oxfam.org/

Romanello, M., Walawender, M., Hsu, S.-C., Moskeland, A., Palmeiro-Silva, Y., Scamman, D., Ali, Z., Ameli, N., Angelova, D., Ayeb-Karlsson, S., Basart, S., Beagley, J., Beggs, P. J., Blanco-Villafuerte, L., Cai, W., Callaghan, M., Campbell-Lendrum, D., Chambers, J., Chicmana-Zapata, V., … Costello, A. (2024). The 2024 report of the Lancet Countdown on health and climate change. The Lancet, 404(10465), 1847–1896. https://doi.org/10.1016/S0140-6736(24)01822-1

U.S. Department of Homeland Security. (2025, November 28). Press release on third-country removal procedures. https://www.dhs.gov/