Mozambique 2019
Tropical Cyclones Idai and KennethOn 14 March 2019, Tropical Cyclone (TC) Idai made landfall close to Beira City in Mozambique. Idai was a category 3 TC when making landfall, causing heavy rainfall, strong winds and a significant storm surge. Beira and the surrounding areas received up to 700 mm of rainfall in 48 hours, leading to flooding events in different parts of the country. The storm caused major damages and ultimately affected approximately 3 million people. The event is considered to be one of the Southern Hemisphere's most devastating storms on record.
Just six weeks later, another TC hit the region. This time, TC Kenneth made landfall in northern Mozambique. Kenneth was the strongest TC to ever make landfall in the African continent, and it was the first time in recorded history that Mozambique was affected by two strong TCs (at or above category 2) in the same season. The accumulated rainfall reached values up to 400 mm during Kenneth's occurrence. While less extensive than Idai, the heavy rainfall still caused flooding in the affected northern regions of Mozambique.
Impact summary
Key impact data of Idai
1303 fatalities
3 million people affected, two-thirds in Mozambique
240,000 homes damaged or destroyed
Source: OCHA, PreventionWeb
Key impact data of Kenneth
45 fatalities
374,000 people in need
45,000 homes damages or destroyed
Source: OCHA
Key impact data of Idai
715k ha of crops flooded, just before the harvest in area already affected by poverty and drought
Source: OCHA, Mozambique after the cyclone by UN Humanitarian
Key impact data of Kenneth
USD 300 million in damage
Key impact data of Idai
USD 3 billion in damages
Source: Nhundu, K., Sibanda, M., & Chaminuka, P. (2021). Economic Losses from Cyclones Idai and Kenneth and Floods in Southern Africa: Implications on Sustainable Development Goals. In G. Nhamo & D. Chikodzi (Eds.), Cyclones in Southern Africa: Volume 3: Implications for the Sustainable Development Goals (pp. 289–303). Springer International Publishing. https://doi.org/10.1007/978-3-030-74303-1_19
Key impact data of Kenneth
30% of the national road network was damaged
20 bridges destroyed
Source: PDNA
Key impact data of Idai
Idai damaged 90% of the the infrastructure of the city Beira, including the country’s main port. Transportation, electricity and communication infrastructure got severely damaged.
Lost tourism days due to the flooding in Somerset
Geographical context
TC Idai had a major impact in three countries: Mozambique, Malawi and Zimbabwe. TC Idai made landfall very close to Beira, a low-lying city located in the Sofala province at the convergence of the Indian ocean and two major rivers, the Buzi and the Pungwe rivers. This makes the city, and its surrounding low-lying lands, prone to both coastal and riverine flooding. Deforestation upstream of the estuary impacts runoff characteristics, which can further exacerbate flooding from extreme events. TC Kenneth caused significant damage in the Comoros islands, northern Mozambique, southern Tanzania and Madagascar. In Mozambique, TC Kenneth made landfall north of the city of Pemba, located in the Cabo Delgado province, unusually far North. Multiple big rivers cross the Cabo Delgado province, with the Lúrio River in south, the Messalo River in the middle, and the Rovuma River in the north of the province, bordering Tanzania. The combination of the low-lying coastal areas and big rivers make this region prone to flooding.
While climate change may have led to additional flood impacts, TC Idai struck Mozambique in a context of high poverty levels, food insecurity, weak infrastructure, political fragility, and social inequality. The 2019 TC season underscored the urgent need for socially inclusive climate resilience and disaster preparedness. Recent urbanization and population growth have increased the exposure in Beira, possibly amplifying impacts of TC Idai. The North of Mozambique, where TC Kenneth hit, has been faced with civil unrest since 2017, making this region particularly vulnerable to extreme events. As Mozambique’s northernmost coastal province, Cabo Delgado had seen floods but not cyclones, especially not of Kenneth’s calibre, leaving its society unprepared.
COMPOUND HAZARd
Tropical Cyclones Idai and Kenneth
Tropical Cyclones Idai and Kenneth caused severe flooding and damage in Mozambique, not only because they were powerful storms, but because several compounding and cascading dynamics shaped their impacts. Each cyclone brought heavy rainfall, strong winds, and high sea levels at the same time. Winds pushed seawater onto the coast, while intense rain caused rivers to overflow inland, and together these effects amplified flooding. Idai and Kenneth were also the first cyclones to strike Mozambique consecutively in the same season. Kenneth arrived only a month after Idai, hitting communities still responding to the first disaster and placing additional strain on people, infrastructure, and emergency services.
The broader weather patterns of 2019 played an important role. Although forecasts did not indicate an unusually active cyclone season, it became one of the busiest on record in the South Indian Ocean. Warm ocean waters and calm upper-level winds, conditions that favour cyclone development, created a compounding set of factors that helped both Idai and Kenneth intensify.
Idai made landfall near Beira in mid-March as a Category 2 cyclone and moved slowly across the region. Its slow movement allowed it to drop extremely high amounts of rain, up to 700 mm in places, resulting in widespread inland flooding. Strong winds and low pressure generated a storm surge of nearly 4 metres combined with the tide along the coast. Although Idai arrived during low (neap) tides, reducing what could have been even more severe coastal impacts, its unusual looping path kept the TC in the region longer and created cascading effects by prolonging rainfall and flooding.
Kenneth struck northern Mozambique only a month later, near Pemba. It was the strongest cyclone ever recorded to make landfall in the country and occurred unusually far north and late in the season. Abnormally warm ocean waters and fast movement allowed Kenneth to intensify rapidly before landfall. Kenneth produced up to 400 mm of rain and caused flooding in northern areas. Kenneth was primarily driven by rainfall, and its occurrence within 6 weeks of Idai, the event added significant pressure at a time when national response systems were already overstretched.
For both events, our use case analysis focuses on the mainland of Mozambique. We focus on the compound flooding and resulting direct damages. As such, we do not consider direct damage from wind, despite evidence that it also contributed to the societal impacts in both cases.
Observed impacts
The combined impact of TCs Idai and Kenneth on Mozambique was profound.
There were severe humanitarian impacts, including hundreds of casualties and millions of displaced people. For many, this was secondary displacement due to other reasons such as conflict. There was also significant damage to the national infrastructure, with extensive damage to the road network, electricity, water, hospitals, schools, and telecommunication infrastructure. The floods had a severe impact on the agricultural sector and exacerbated food insecurity and the outbreak of diseases such as cholera and malaria.
Underlying drivers of land use changes such as charcoal production, driving deforestation and informal urban expansion, which in turn enhance flooding are important to consider. It emphasizes that impacts of TC Idai were not only the result of intense climatic hazards but were amplified by complex, deeply rooted socio-economic processes, that create reinforcing cycles of vulnerability and exposure.
Vulnerability and exposure strongly influenced impacts. Ongoing conflict and displacement in Cabo Delgado increased risks, while conditions of poverty, food insecurity, weak infrastructure, and social and political fragility intensified the effects of Idai. Resulting from key informant interviews, vulnerabilities also differed between urban and rural areas, and land-use drivers such as deforestation, charcoal production, and informal urban expansion increased exposure to flooding.
A causal loop diagram (CLD), which maps the causal relationships of some of the main drivers and feedback loop of flood impacts in rural areas. It, for example, highlights that charcoal production contribute to increased deforestation and flooding, which in turn can further exacerbate socio-economic challenges faced by the communities that rely on charcoal production for their livelihoods. Source: (Webb, P., 2025).
Response
The combined impact of TCs Idai and Kenneth on Mozambique was profound. There were severe humanitarian impacts, including hundreds of casualties and millions of displaced persons. For many this was secondary displacement due to other reasons such as conflict. Mozambique has limited capacity for early warning and disaster response. Moreover, the difficulties for financing was a major constraint in recovery efforts and there were significant challenges in restoring livelihoods. There was also significant damage to the national infrastructure, with extensive damages to the road network, electricity, water, hospitals, schools and telecommunication infrastructure. The floods had a severe impact on the agricultural sector and also exacerbated food insecurity and the outbreak of diseases such as cholera and malaria.
Attribution analysis
The attribution analysis for TC Idai and Kenneth uses a storyline approach that compares a factual simulation of the flood event, representing present-day climate conditions, with a counterfactual simulation representing a world without long-term climate change.
The factual scenario includes widespread flooding for Idai across the Sofala province, especially around Beira and the Buzi and Pungwe floodplains, with about 3,600 km² flooded, 84,000 buildings damaged, and roughly 288,000 people exposed. Kenneth produced flooding mainly around Pemba Bay, with about 1,000 km² flooded, 13,000 buildings damaged, and around 53,000 people exposed.
The attribution analysis compares these factual simulations with counterfactual ones representing a world without long-term climate change. For Idai, the counterfactual scenario is based on the best estimates from literature and assumes the following: rainfall reduction according to Clausius-Clapeyron (7%/°C), corresponding to 8% for 1.1 degree of warming; 0.14 m of sea-level rise removal, and maximum wind speeds reduction of 10%. For Kenneth, only rainfall is reduced, as precipitation was the main driver of flooding. Other factors, such as tropical cyclone track and size are kept constant.
Attribution results are derived by comparing flood extent, flood depth between the factual and counterfactual scenarios, producing an event-specific assessment of climate change influence. We also attribute the flood impacts by also calculating and comparing the affected population and economic damages, based on building damages as a function of flood depth.
Attribution results
According to model simulations, climate change exacerbated flooding for both Idai and Kenneth. For Idai, the flood extent decreases to 3507 km2 in a scenario without climate change. This corresponds to an increase of 2% in flood extent that can be attributed to climate change. The mean increase in flood depth is 0.14 m, corresponding to a relative increase of 14%. The areas that see the largest increases in flood depth (>0.3 m) include the banks of the Buzi River and the coastal areas east and west of Beira. The total flood volume increases by 9% due to climate change.
When comparing the factual against the counterfactual impact, the total damages increase with $98 million USD, which corresponds to a 29% increase. The total number of buildings that experienced damage increases by 9,000, corresponding to a 11% increase. This indicates that most of the additional damage from climate change is caused by larger flood depth, instead of flood extent, exposing an additional 34,000 people to severe flooding; an 18% increase attributable to climate change.
Change in flood depth for TC Idai (upper panel) and TC Kenneth (lower panel) when comparing the factual against the counterfactual
For Kenneth, the pattern is similar, although the overall scale of the event is smaller. The total flooded area increase of 3% can be attributed to climate change.
As with Idai, flood depths increase by about 14 centimeters due to climate change, with the largest increases of more than 30 centimeters occurring upstream of Pemba Bay and in northern areas affected by river flooding.
Climate change resulted in an increase of damages by $7 million USD, which corresponds to a 14% increase. The total number of buildings that experienced damage increases by 800, corresponding to a 6% increase. There is a 10% increase in the number of people affected by flood depths above 20 cm, corresponding to about 3,500 additional people affected due to climate change.
When synthesizing the results for the two TCs, we find that the impact of climate change on flood damage is more profound than on flood hazard. It shows that the societal consequences of TCs can be amplified, even for relatively small increases in individual flood drivers, highlighting the role of a region’s exposure and vulnerability.
Climate attribution of the flood impacts of TC Idai (top) and TC Kenneth (bottom) for total damage (left) and total exposed population (right).
Across both cyclones, the results show a consistent pattern: climate change increases flood hazards, but its impact on flood damage is even larger. While flood extents grow by only 2-3%, flood depths grow by around 14%, which greatly amplifies economic losses and the number of people exposed. Even small changes in rainfall or sea level can significantly increase real-world impacts when communities are already vulnerable.
These findings are broadly consistent with past research on Idai, which has also shown increases in displacement and impacts linked to climate change. Kenneth, however, has received much less scientific attention, making this analysis one of the first of its kind.
Finally, it is important to recognize that attributing changes in TC-related flooding in data-scarce regions is challenging due to lack of good-quality input and validation data. Tropical cyclones are rare, complex events, and climate datasets do not always capture them perfectly. Flood observations from satellites vary between products, adding uncertainty. Even so, the results provide strong evidence that climate change increased the severity of flooding and its impacts for both Idai and Kenneth.
Policy implications
The active tropical cyclone season in 2019 highlighted the vulnerability of Mozambique and the larger eastern Africa region to the devastating impacts of tropical cyclones, and the aggravating influence of climate change. Our results indicate that climate change may increase the severity of TCs, putting additional pressure on the governments to invest in early warning systems, resilient infrastructure, and better protect vulnerable communities. When synthesizing the results for the two TCs, we find that the impact of climate change on flood impact is more profound than on flood hazard. It shows that the societal consequences of TCs can be amplified, even for relatively small increases in individual flood drivers, which highlights the added value of flood impact modelling in the field of climate attribution.