Honduras 2020

Tropical cyclones Eta and Iota in 2020

In November 2020, amid the COVID-19 pandemic, Honduras was struck by two hurricanes within two weeks. 

  • Hurricane Eta made landfall on November 4, bringing torrential rain, flooding, and landslides - causing at least 74 deaths. While recovery was still underway, 
  • Hurricane Iota hit on November 17, impacting many of the same areas and causing 13 more deaths. Major rivers such as the Ulúa, Chamelecón, Choluteca, Aguán, and Leán overflowed, inundating large parts of the country.

Together, the hurricanes directly affected 437,000 people and indirectly 4.5 million, displacing 937,000 and leaving 2.8 million in need of assistance. Economic losses reached $1.9 billion, with national growth falling by 0.8%. In total, 95 lives were lost, and over 90,000 homes were damaged. Despite early warnings, evacuations were hindered by holidays, insecurity, and the pandemic. The experience has since strengthened Honduras’s commitment to preparedness planning and early action.

Prior to Eta and Iota, Honduras was already facing high levels of structural vulnerability driven by poverty, rapid urbanization in flood-prone valleys, environmental degradation, and limited disaster-resilient infrastructure. Years of recurrent floods, droughts, and storms had weakened livelihoods, especially in agriculture, while migration pressures and insecurity further strained coping capacities. The COVID-19 pandemic compounded these challenges by disrupting services, reducing income opportunities, and stretching emergency response systems. As a result, when the two hurricanes struck in quick succession, pre-existing socio-economic fragilities amplified their impact, exposing underlying risks.

Key impacts from the two hurricanes include:

  • Heavy rainfall, river flooding and landslides
  • Loss of life
  • Roads, bridges and other transportation infrastructure damaged
    - limited access reduced the speed and reach of early rescue and relief operations
  • Large scale displacement
  • Damage and interruption to health care facilities, residential, school and other community structures
  • Agricultural losses and food insecurity

Geographical context

The tropical cyclones affected communities across a wide area in Honduras. The most significantly impacted departments in general were Cortés, Santa Bárbara, and Lempira. The Atlantic coast and the region surrounding San Pedro Sula, especially areas along the Chamelecón and Ulúa rivers, sustained particularly heavy damage. In the wake of flooding and landslides, nearly 369,000 people were cut off from surrounding areas for weeks, with the Colón department being the most affected by the isolation. 

More specifically, Cortés, Atlántida, and Francisco Morazán, had the highest concentration of schools being damaged and some permanently destroyed. The Cortés department was particularly affected by impacts on health infrastructure, with 70% of its health facilities reporting damage. Some of the most widespread physical damage was in relation to agriculture, with losses up to 80% in certain areas, from the Sula Valley and northern Atlantic departments to southern, western, and northeastern regions like Gracias a Dios.

Map of Eta and Iota storm tracks and population affected across the region (Mapaction).

Historical context

Three important societal factors affected the way in which the events of hurricanes Eta and Iota unfolded. These included high levels of previous displacement and migration, violence, as well as pre-existing health crises due to dengue and COVID-19.

Pre-existing internal displacement and migrants returning from Mexico, meant there were large numbers of people living in low-cost neighbourhoods, often in informal settlements. This led to more people living in precarious conditions, lacking both economic resources and social networks to help them cope with shocks. In areas like Sula Valley, flooding from Eta and Iota impacted some of the poorest and most violent areas. The threat of violence, including a fear of being robbed or inability to return, affected the willingness of individuals to leave their homes in advance of the storms. Lastly, the health system was already under strain from historical funding challenges, COVID-19 and the dengue epidemic. The system was unprepared to deal with the damages to health facilities from Eta and Iota, as well as secondary health impacts.

COMPOUND HAZARd

The compounding nature of Hurricanes
Eta and Iota   

The Hurricanes Eta and Iota resulted from a convergence of atmospheric, oceanic, and environmental factors that unfolded across the Atlantic and Central American regions. Hurricane Eta originated as a tropical wave moving westward from the western coast of Africa, being pushed by a high-pressure system over the northern Caribbean. Once it entered the Caribbean Sea, the environment was perfect for it to strengthen: the winds above were calm, and the ocean was unusually warm, giving the storm plenty of energy. Eta moved very slowly as it approached Central America. This slow movement led to extremely heavy and prolonged rainfall across Honduras, worsening flooding and landslides.

Hurricane Iota developed in much the same way as Eta, beginning as a weather disturbance off the coast of West Africa and moving west across the Atlantic. Its path toward Central America was shaped by larger wind patterns that steered it into the Caribbean. At first, Iota strengthened slowly because the surrounding winds were disruptive. But as it entered the southwestern Caribbean, conditions became ideal for a powerful storm: calm upper-level winds, very warm ocean waters, and lots of moisture in the air. This allowed Iota to intensify rapidly, becoming one of the strongest storms ever recorded so late in the season. However, as it moved across waters previously cooled by Eta, the storm weakened from its peak strength.

These events interacted with pre-existing conditions in Honduras, many of which stemmed from longer-term climate patterns. A prolonged drought linked to both El Niño Southern Oscillation ENSO variability and human-caused climate change had already disrupted rainfall, increased temperatures, and stressed water systems across the country, alongside land use changes and altered water management practices. This left communities especially susceptible to flooding and landslides from extreme rainfall.

In the weeks before Eta, northern Honduras experienced heavy precipitation from an earlier cold front, saturating soils and filling up dam reservoirs in advance of the hurricane’s landfall. The 2020 Atlantic hurricane season had already delivered an unusually high number of storms to the region, compounding soil saturation. Water reservoirs, previously depleted by drought, had reached near-full capacity by the time Eta approached, reducing their ability to absorb additional rainfall. Furthermore, the ENSO was in a La Niña phase , which typically increases precipitation and tropical cyclone activity in Central America. When Eta and Iota led to intense rainfall, these conditions contributed to widespread landslides across multiple regions, magnifying the hurricane’ severe impacts.

Observed impacts

The compounding impacts of Hurricanes
Eta and Iota

The impacts of Hurricanes Eta and Iota unfolded over several weeks in November 2020, creating a compound emergency for Honduras during the height of the COVID-19 pandemic. Eta made landfall on November 4, bringing strong winds, heavy rainfall, and widespread flooding. As major rivers including the Ulúa, Chamelecón, Choluteca, Aguán, and Leán overflowed, large areas of both urban and rural Honduras experienced inundation. Landslides occurred in upland regions where soils were already saturated, damaging homes, farms, and local road networks. Eta resulted in at least 74 deaths and significant disruption to transportation, services, and livelihoods.

While recovery operations were still underway and many areas remained waterlogged, Hurricane Iota followed a similar path, making landfall on November 17. Though slightly weaker on arrival, it affected many of the same communities, causing additional flooding and 13 more deaths. The repeated rainfall further strained already weakened infrastructure. 

Across both storms, the scale of human impact was substantial. An estimated 437,000 people were directly affected, while another 4.5 million experienced indirect impacts such as disrupted services, damaged infrastructure, or loss of income. Approximately 937,000 people were displaced—some temporarily, others for extended periods due to the destruction of over 90,000 homes. In total, 95 lives were lost. The storms also left 2.8 million people in need of assistance, placing considerable pressure on local authorities, humanitarian organizations, and community networks.

Economic impacts were far-reaching. Losses reached an estimated $1.9 billion, driven by damages to housing, agriculture, transport infrastructure, and public services. The national economy contracted by 0.8%, reflecting setbacks in key sectors and the challenges of responding to the disaster amid ongoing pandemic restrictions. Agricultural losses in regions such as the Sula Valley reduced food production and affected livelihoods for months.

Evacuations and response efforts were complicated by the COVID-19 context, holiday periods, and localized insecurity, which limited the movement of both responders and affected households. However, the experience of managing the back-to-back storms strengthened national discussions on preparedness, early action, and risk reduction. In the years that followed, Honduras placed greater emphasis on improving forecast-based decision-making, community readiness, and coordination mechanisms to reduce the impacts of future hazards.

Population exposure to floods during Hurricane Eta and Iota 2020 – Deltares flood model 2025; Map – Climate Centre 2025

Impact summary

Impact Type of impact Magnitude
Lives lost Human 95
Missing persons Human 10 (mainly fishermen)
Homes affected Human/Infrastructure 90,000
People evacuated Human 437,000
Roads damaged Infrastructure 927
Bridges damaged Infrastructure 72 damaged, 62 destroyed
Health facilities affected Infrastructure 400 damaged, 120 inoperative
Schools damaged Infrastructure 3% of all buildings
Education disrupted Human / Infrastructure Schools used as shelters
Agricultural losses Economic / Livelihoods Up to 80% (coffee, banana, plantain, sugarcane)
Isolation of communities Human / Access Weeks of isolation
Increase in COVID19 cases Human / Infrastructure Documented increase due to overcrowded shelters
Economic losses Economic 1.9 billion USD

Causal map of impact pathways and vulnerability factors based on retrospective analysis of Eta and Iota impacts

Response

The response to Hurricanes Eta and Iota in Honduras was largely reactive and marked by significant challenges across all levels of the disaster risk management system.

In the immediate aftermath of Eta, authorities and humanitarian organizations concentrated on life-saving activities. Evacuations were conducted in areas facing severe flooding and landslides, and emergency assistance such as food, shelter, and water, sanitation, hygiene (WASH) support was rapidly mobilized. However, these efforts were significantly constrained by heavy infrastructure damage. Roads and bridges were destroyed across multiple departments, leaving more than 368,000 people cut off from assistance for weeks, especially in Colón department. These access limitations severely reduced the speed and reach of early rescue and relief operations. 

Early warning and forecasting capacities were limited, relying primarily on manual monitoring of river levels without integrating rainfall data. Early warning systems were fragmented with many communities depending on localized alerts that were poorly connected to national-level mechanisms. Institutional challenges such as inadequate staffing, funding shortages, and weak coordination across municipal, regional, and national authorities further weakened early action efforts. When forecasts for Iota emerged, responders were still overwhelmed by Eta’s impacts, leaving little opportunity to shift from emergency relief to proactive risk reduction.

Communication gaps further weakened preparedness. Many early warning messages lacked clarity, detail, and actionable instructions. Communities often misunderstood the risks, including the likelihood of secondary impacts such as landslides or renewed flooding. Younger populations, who had limited experience with major hurricanes, showed particularly low risk awareness. In northern Honduras, vague or inconsistent guidance led many residents to underestimate the threat or feel uncertain about protective actions.

Some improvements were observed before Iota’s landfall. Messages became more explicit, including guidance on evacuation routes and protective measures. A second round of evacuations was carried out, and authorities attempted to reduce COVID-19 transmission within shelters through measures such as distancing, hygiene practices, and mask use. Nonetheless, many viewed these precautions as insufficient.

A critical overarching gap was the absence of a multi-hazard early warning system. Hazard data was not integrated with information on vulnerabilities such as food insecurity, health conditions, housing instability, gender-based violence, or migration patterns. This limited responders’ ability to anticipate compounding impacts. Risk monitoring remained ad hoc, with high-risk areas recognized informally rather than through systematic analysis. Information gaps were especially severe for rural, Indigenous, and migrant populations, restricting effective planning and targeted support.

Key gaps included:

  • Lack of an integrated, multi-hazard early warning system 
  • Fragmented communication and coordination
  • Insufficient data on vulnerable and remote populations

Attribution analysis 

The attribution modelling framework is used to understand how much climate change may have influenced the flooding from the combined impacts of Hurricanes Eta and Iota. The approach uses a concept known as storylines to analyse how the events unfold under different climate scenarios. The two simulations are:

1.     Factual, representing the world as it is today, and

2.     Counterfactual, representing a world without long-term climate change, similar to pre-industrial conditions.

The flood event itself is defined using modelled flood maps that show the maximum extent and depth of floodwaters.

For the factual simulation, this used real-world data and reanalysis datasets. One dataset provides information such as rainfall, temperature, and evapotranspiration. This information is used to estimate river runoff. To analyse coastal water levels, wind and pressure data are combined with a wind model. These inputs are combined with hurricane track data to simulate how river flooding, rainfall, and coastal surges interact. These factors are brought together to produce a full picture of compound flooding under today’s climate.

For the counterfactual simulation, the same modelling process is used but with rainfall reduced by 9%. This reduction follows the Clausius-Clapeyron relationship, which says warmer air holds more moisture. Lower rainfall in the counterfactual world allows for a comparison of how flooding might have differed without human-driven warming.

Attribution results

The analysis compares what actually happened during Hurricanes Eta and Iota with what might have happened in a world without long-term climate change. The modelling showed complicated differences between the two scenarios. For instance, in some places, the counterfactual model showed flooding where the factual model did not. Interestingly, the model estimated that around 462,693 people were exposed to flooding, which is very close to the 437,000 people who were actually evacuated. This suggests the modelling may be more accurate than satellite flood maps. The counterfactual simulation produced only slightly less flooding, with 454,498 people exposed, a difference of just 1.8%. Because the difference is small and the modelling challenges are significant, these results should be interpreted carefully.

The climate-related modeling showed only small differences between factual and counterfactual flood impacts. This is likely due to modeling issues where the hydrology and topography of the region, including very narrow river valleys upstream of the main flooding area may have prevented the model from realistically propagating the changing rainfall intensity into changing flooded areas. The results highlight the need for flexible, well-validated modeling systems.

Comparison of flooding extent between factual and counterfactual simulations

For buildings, the factual model estimated 111,668 buildings were affected, again close to the official number of 90,000 reported. The counterfactual simulation showed only slightly fewer buildings flooded. The unexpectedly small difference between the factual and counterfactual is because of the limited change in flooding extent under the counterfactual simulation.

The analysis also explored how impacts would have changed if key non-climatic risk factors in Honduras, violence, displacement, and COVID-19, had not existed. These factors significantly shaped the way in which people and communities experienced the storms.

 

Violence and Insecurity

Honduras is exposed to widespread gang violence, especially in San Pedro Sula and surrounding areas. On an individual level, exposure to violence has a broad impact on vulnerability to disasters. For instance, violence drives displacement to flood and landslide-prone marginal areas, increases individual social and economic vulnerability and influences choices due to the risk of exposure to violence when faced with disasters such as the ones triggered by Eta and Iota. Many people did not evacuate during Eta and Iota because they feared robbery or gang retaliation. In some neighbourhoods, the dissemination of storm warnings was limited, and emergency teams could not enter without negotiating with local actors.

If there had been no violence:

  • Evacuation and warning systems would have been more easily implemented.
  • Humanitarian aid would have reached communities faster.
  • More people would have used official shelters without fear.
  • Businesses and farms could have restarted sooner.
  • Fewer people would have lived in high-risk flood zones.
  • Trust in authorities and information sharing would have improved.

COVID-19 Pandemic

Before the Eta and Iota, Honduras was already struggling with COVID-19. The health system was overwhelmed, the economy suffered, and many families lost income. Poverty, food insecurity, and migration increased. COVID-19 also deepened mistrust in government, partly due to alleged misappropriation of COVID19 medical funds.

In this context, where individuals and households were dealing with the health and economic consequences of COVID19, compounding impacts from Hurricane Eta and Iota further aggravated these vulnerabilities.

If COVID-19 had not existed:

  • Households would have had more purchasing power to support their economic recovery.
  • Food insecurity would have been lower.
  • Health centers would have had more capacity to treat storm-related illnesses.
  • Displacement and migration pressures would have been reduced.
  • Shelter conditions and risks such as gender-based violence, would have been less severe.
  • Trust in government and response systems would have been higher.

The findings show clearly that non-climate factors, including violence, displacement, and COVID-19, played a much larger role in shaping the impacts and on the ground experiences of those affected by Eta and Iota than the climate-based counterfactual differences

Causal map of impact pathways and vulnerability factors based on retrospective analysis of Eta and Iota impacts

Policy implications 

This case study highlights several structural and operational gaps that hinder the resilience of Honduras to tropical cyclones. Many of these insights were gathered through a stakeholder workshop including representatives from local government, the Honduran Red Cross and NGO. These gaps provide clear pathways for policy action.

Strengthening information systems. COPECO (the entity coordinating disaster relief in Honduras) has limited access to critical data, weak data-sharing practices across institutions, and insufficient independence in information management. This hinders effective risk assessments and decision making for responses. Policies that improve institutional coordination would substantially enhance national preparedness.

Improving early warning and monitoring capabilities is another important policy need. Low confidence in weather forecasts, inadequate early warning systems, and sparse weather-station coverage, particularly at the basin scale, are major obstacles to timely and accurate response. Investments in hydrometeorological infrastructure, basin-level data integration, and modernization of warning systems would support more proactive preparedness measures.

Adopting multi-hazard and systems-based approaches can help anticipate cascading risks and prioritize interventions more effectively. Stakeholders emphasized that tropical cyclone impacts in Honduras are multicausal, multifactorial, and often compounding, highlighting the need for policy frameworks that reflect this complexity. In addition, systematic documentation of past events is essential for institutional learning and improved planning.

Finally, the  policy relevance of prioritizing retrospective attribution studies is an important consideration. These studies can fill key knowledge gaps and strengthen the understanding of the drivers and intersecting factors which shape cyclone-related risks. Policies that support this analytical work, can result in more robust, forward-looking strategies.