Assessing the contribution of climate change on tropical cyclones related to loss and damage in southern Africa: a case study of tropical cyclones in southern Malawi

This study synthesizes peer-reviewed literature and government reports to examine the relationship between climate change and the frequency, occurrence, and intensity of tropical cyclones in southern Malawi. The research employs a mixed-methods approach, combining spatial mapping, literature synthesis, and trend analysis of cyclone data from the past three decades. It also analyzes rainfall data from nine meteorological stations using the Standardized Precipitation Index (SPI) and conducts stakeholder interviews across four districts. Using literature review of peer-reviewed literature and government documents, this study assessed the link between climate change and the occurrence, frequency, and magnitude of tropical cyclones that lead to Loss and damage in several areas, including various types of physical infrastructure, agriculture, and food security in Africa. The study focuses on Southern Africa, using Malawi as a case study. It examines the occurrence and frequency of tropical cyclones in the Southern region over the past 30 years. Specifically, the study aims to: (i) Analyze trends of tropical cyclones and related temperature and extreme rainfall events, including floods, for the past three decades, and (ii) Map areas affected by Tropical cyclones and related extreme rainfall events over the past 30 years. Findings reveal an increasing trend in tropical cyclone occurrences since the 2000s, with particularly intense events such as those in 2015, 2019, and 2023 coinciding with La Niña conditions. Statistical analysis using Mann-Kendall trend tests and Pearson correlations confirms significant upward trends in both cyclone frequency (Tau = 0.29, p < 0.05) and rainfall anomalies (r = 0.51, p < 0.01). The study results show an increasing occurrence of Tropical cyclones from the 2000s. These findings demonstrate correlations between increased cyclone activity and climate change indicators, consistent with established attribution studies but requiring additional multivariable climate modeling for definitive causal attribution, reinforcing the need to prioritize agricultural resilience within the Loss and Damage framework under the UNFCCC. The increasing trends in cyclone frequency and intensity show correlations with climate change patterns and align with established climate projections. However, definitively establishing direct causal attribution requires comprehensive climate modeling, which is beyond the scope of this study. These observed trends are consistent with regional attribution studies, suggesting that management of losses and damages in agriculture deserves special attention within the Loss and Damage framework.

1 Introduction

1.1 Background

Climate change has emerged as a paramount global challenge, with far-reaching impacts on economies, livelihoods, and biodiversity. While acknowledging the multifaceted nature of climate risks in Southern Africa, this study focuses specifically on tropical cyclones due to their growing intensity, rapid onset, and acute impacts on food systems and infrastructure, areas highly relevant to Loss and Damage mechanisms. The increasing frequency and intensity of extreme weather events are significantly hindering efforts to achieve the Sustainable Development Goals (SDGs) and exacerbating vulnerabilities in many developing countries (Intergovernmental Panel on Climate Change, 2022). Among these extreme events, tropical cyclones pose a particular threat to southern Africa, with their destructive potential amplified by climate change.

The concept of loss and damage has gained prominence in international climate negotiations, particularly for vulnerable developing countries. It refers to the impacts of climate change that cannot be avoided through mitigation and adaptation efforts (UNFCCC, 2013a,b). The international political response to climate change began with the adoption of the United Nations Framework Convention on Climate Change (UNFCCC) in 1992 which recognizes the fundamental priorities of protecting livelihoods and economies (Article 2) and addressing the particular vulnerabilities of developing countries to the adverse impacts of climate change (Article 3) when planning and implementing response measures (UNFCCC, 1992). The Paris Agreement of 2015 explicitly acknowledges the importance of “averting, minimizing and addressing loss and damage associated with the adverse effects of climate change” (UNFCCC, 2015). This has led to increased focus on developing mechanisms to support vulnerable countries in managing the impacts of extreme events such as tropical cyclones.

Tropical cyclones (TCs) are one of the most devastating natural hazards, characterized by intense winds, heavy rainfall, and storm surges that cause widespread destruction to infrastructure, ecosystems, lives, and human livelihoods. Globally, regions such as the North Atlantic, Western Pacific, Indian Ocean, and South Pacific are particularly prone to these extreme weather events, with the intensity and frequency of cyclones exhibiting significant variability over time (Kossin et al., 2020a,b). The increasing visibility of climate change’s impacts on weather patterns across the world has intensified scientific efforts to attribute observed changes in tropical cyclone behavior to anthropogenic influences.

The Intergovernmental Panel on Climate Change (IPCC) Fifth Assessment Report (AR5) highlights that while there is limited evidence for a clear trend in the overall frequency of tropical cyclones globally, there is stronger evidence indicating an increase in the proportion of intense cyclones (Category 3 and above) in some ocean basins, notably the North Atlantic and Western Pacific (Intergovernmental Panel on Climate Change, 2013). These findings suggest that climate change may be influencing the intensity and distribution of TCs, with regional disparities driven by local sea surface temperature (SST) anomalies, atmospheric conditions, and climate variability.

Recent studies have further documented complex interactions between climate change and tropical cyclone activity showing a marked increase in the frequency and intensity of tropical cyclones (TCs) affecting southern Africa. The Intergovernmental Panel on Climate Change (IPCC) projects that, because of global warming, the frequency and intensity of heavy rainfall events will increase for most of tropical Africa, while parts of northern and southern Africa will become drier (Intergovernmental Panel on Climate Change, 2021). Further, Reason et al. (2018) reported that SSTs in the South Indian Ocean have increased over the past decades, potentially fostering conditions conducive to cyclone development. However, the attribution of specific cyclone events or trends remains challenging due to limited historical data, regional climate variability, and the complex interplay of oceanic and atmospheric factors (Leroux et al., 2017). This is likely to result in significantly increased risk of flooding and associated damages.

For instance, the case of Malawi and Southern Africa in general exemplifies the broader regional challenges faced by cyclone-prone areas worldwide. While historically less affected by tropical cyclones compared to regions like the Caribbean or Southeast Asia, Southern Africa has experienced an increasing incidence of intense weather systems linked to climate variability and change (Reason and Van Heerden, 2012). Notable extreme rainfall events, such as Cyclone Idai in 2019 and Cyclone Freddy in 2023, have underscored the urgent need to understand how climate change influences cyclone characteristics and impacts in this region.

Globally, the attribution of specific cyclone events to climate change has advanced through event attribution science, which assesses the extent to which human influence has altered the likelihood or severity of such events (Stott et al., 2016). For example, Van Oldenborgh et al. (2017) attributed the increased probability of extreme rainfall during Cyclone Idai to anthropogenic climate change. Similar approaches are being increasingly applied to understand regional variations and inform adaptation strategies. Titley et al. (2016) have demonstrated that climate change has increased the likelihood and intensity of rainfall associated with tropical cyclones impacting Madagascar, Mozambique, and Malawi. This finding highlights the direct connection between human-induced climate change and the growing losses and damages in the region.

This paper aims to contribute to this growing body of knowledge by investigating the attribution of tropical cyclones to climate change in Southern Africa, with a specific focus on Malawi. While the case study is region-specific, its findings have broader implications for understanding how climate change may be affecting cyclone activity in tropical and subtropical regions worldwide. Addressing this issue is critical not only for regional disaster preparedness and resilience planning but also for informing global climate policy and adaptation efforts.

1.2 Understanding loss and damage in the context of UNFCCC

The adverse impacts of climate change, ranging from extreme weather events to slow-onset changes, have become increasingly evident across the globe. As countries strive to mitigate and adapt to these impacts, the concept of loss and damage has emerged as a critical component of climate change discourse. Although it is not officially defined under the UNFCCC, loss and damage refer to the irreversible and non-compensable harm caused by climate change, which exceeds the capacity of communities, ecosystems, and nations to adapt and recover (Telesetsky, 2021). Intergovernmental Panel on Climate Change (2022) further defines loss and damage as the “harm from (observed) impacts and (projected) risks” (p. 2914) of anthropogenic climate change. This harm includes both economic and non-economic impacts resulting from extreme weather events (rapid-onset events) and slow-onset events (Qi et al., 2023).

Within the United Nations Framework Convention on Climate Change (UNFCCC), loss and damage have gained recognition as a distinct issue that necessitates targeted action. The evolution of loss and damage in UNFCCC negotiations has seen significant milestones, including the establishment and operationalization of the Warsaw International Mechanism for Loss and Damage (WIM), the Santiago Network on loss and damage (Santiago Network), and the Fund for Responding to Loss and Damage (FRLD) (UNFCCC, 2023). The recognition of loss and damage in the Paris Agreement further emphasizes its significance on the global climate change agenda (UNFCCC, 2015).

It is widely recognized that national or regional assessments play a crucial role in comprehending the specific vulnerabilities, risks, and impacts associated with loss and damage within a given geographic context. These assessments provide valuable insights into the localized manifestations of loss and damage, guiding policymakers and practitioners in developing effective adaptation and mitigation strategies. By integrating these assessments into national or regional policies and planning processes, including national adaptation planning (NAPs) and nationally determined contributions (NDCs), countries and regions can better understand the challenges they face and allocate resources accordingly.

1.3 Research problem statement and justification

The Fund for Responding to Loss and Damage (FRLD) is part of the UNFCCC’s financial mechanism to assist countries that are particularly vulnerable to the adverse effects of climate change in responding to economic and non-economic loss and damage associated with extreme weather events and slow-onset events. Attribution science plays a crucial role in understanding the extent to which human-caused climate change is responsible for the losses and damages. Attribution is often associated with responsibility and responsibility/liability, and in establishing a clear causal relationship (Knutson et al., 2019; Knutson et al., 2020). For example, attribution studies can justify the need for enhanced financial investments, such as climate risk insurance, climate risk pools, and catastrophe bonds, to address the escalating costs associated with more intense and destructive extreme weather events.

However, significant research gaps remain, particularly regarding regional variability in climate change, impacts on tropical cyclone behavior, and the uncertainty surrounding future storm frequency and intensity under different emission scenarios. Furthermore, research on the attribution of climate change to tropical cyclones has a long history globally, but in Southern Africa, it is still emerging, with several gaps remaining that need to be addressed to improve understanding and predictive capabilities. There is limited high-resolution data and Regional Climate Model (RCM) data. In this regard, most attribution studies rely on global climate models that lack the spatial resolution necessary to accurately simulate the genesis, track, and intensity of tropical cyclones in the Southern African region (Reason and Rouault, 2020).

Studies show that there are insufficient or inconsistent long-term historical records of tropical cyclones affecting Southern Africa, which hinders robust attribution analysis (Reason and Van Heerden, 2012). This gap hinders robust attribution analysis. Hence, improved reconstruction of past cyclone activity and improved historical datasets are needed for trend analysis and attribution studies. Likewise, there is a limited understanding of the climate change signal in cyclone intensity and frequency. There is an ongoing debate over how climate change affects the frequency and intensity of tropical cyclones in the Southern Hemisphere, with some studies indicating an increase in intensity but no clear trend in frequency (Leroux et al., 2017).

Attribution of specific extreme events (e.g., Cyclone Idai in 2019 and Tropical Cyclone Freddy in 2023) to climate change remains complex (Van Oldenborgh et al., 2017), underscoring the need for event-based attribution studies that quantify the extent to which climate change has influenced the probability or severity of individual cyclones. Similarly, the impact of Sea Surface Temperature (SST) changes is underexplored. Changes in SSTs are known to influence cyclone development, and regional SST variability in the Indian Ocean and its relationship with cyclone activity in Southern Africa is not fully understood (Elsner et al., 2008; Reason et al., 2018). Further research is needed to establish a connection between SST anomalies and cyclone genesis and intensification in the context of climate change.

Additionally, the Lilongwe Declaration on Climate Change 2024 (United Nations, 2024a,b) calls for international support and legal protection for people displaced by climate change, particularly women and children, recognizing that losses and damages due to tropical cyclones in Southern Africa are growing over time. This study contends that addressing these gaps will enhance the reliability of climate-related disaster financing and ensure more resilient and evidence-based responses to climate-induced tropical cyclone hazards. For example, the Lilongwe Declaration on Climate Change 2024 calls for international support and legal protection for people displaced by climate change, particularly women and children. The Lilongwe Declaration of 2024 emphasizes that losses and damages due to tropical cyclones¹ in Southern Africa are increasing over time, and some of these losses, such as housing infrastructure, lead to the displacement of households. Therefore, this study aims to inform financing mechanisms of the losses and damages caused by Tropical Cyclones using Malawi as a case study.

1.4 Study objectives

The primary objective of this study is to examine the relationship between tropical cyclones and climate change in Southern Africa, with a particular focus on Tropical Cyclones in southern Malawi.

Specifically, the study aims to:

i. Analyze the temporal trends in the frequency and intensity of tropical cyclones in southern Malawi from 1990 to 2023 in the context of climate change.

ii. Map areas affected by tropical cyclones and related extreme rainfall events over the past 30 years.

iii. Analyze the potential impact of climate change on the frequency and intensity of tropical cyclones in the region.

iv. Integrate qualitative data from stakeholder interviews and community perspectives using NVivo, and juxtapose these with quantitative cyclone and rainfall data to guide policy recommendations for climate adaptation and risk management.

The study adopts a mixed-methods approach, combining spatial mapping with literature synthesis and trend analysis of cyclone data from the past three decades. GIS mapping and manual reports have been utilized to identify cyclone-affected regions, while statistical analysis, including Mann-Kendall trend tests and correlation assessments, strengthens the quantitative analysis.

While acknowledging the multifaceted nature of climate risks in Southern Africa, this study focuses specifically on tropical cyclones due to their growing intensity, rapid onset, and acute impacts on food systems and infrastructure, areas highly relevant to Loss and Damage mechanisms.

The study excludes other types of natural hazards and disasters, such as droughts or floods, which may impact different sectors, including agriculture and food security in the region, but are not directly related to tropical cyclones. This work aims to provide a better understanding of the losses and damages caused by tropical cyclones in Southern Africa.

2 Materials and methods

2.1 Study sites and research approach

Malawi is a landlocked country located in southeastern Africa, lying between approximately 9° and 17° South latitude and 33° and 35° East longitude, and is bordered by Tanzania, Zambia, and Mozambique. The country lies within the Great Rift Valley, a significant tectonic and geographical feature that stretches from the Middle East down through East Africa (Figure 1). Its terrain is largely characterized by highlands and plateaus with elevations ranging from around 300 m above sea level in the lake area to over 2,300 m in the highland regions. This geographical positioning places Malawi within the subtropical climate zone, with a warm and humid climate influenced by its proximity to the Indian Ocean and the Indian monsoon system. This position makes the country vulnerable to seasonal weather patterns, including rainfall variability and the impacts of tropical cyclones originating from the Indian Ocean, particularly between November and April (Government of Malawi, 2017).

Figure 1

Figure 1. Map of Malawi and its location in Africa.

Malawi is highly vulnerable to several hydro-meteorological hazards, including tropical cyclones and floods, especially in the Southern region. In southern Malawi, the most recent flood events were directly triggered or exacerbated by tropical cyclones, reinforcing the rationale for focusing on these events. The country’s high level of vulnerability is largely connected with several specific geo-climatic factors, including: “(i) the influence of the El Niño and La Niña phenomena on climate variability; (ii) the variability in the water levels of the country’s three major lakes (Malawi, Chiuta, and Chilwa) and the broader hydrological network, due to variations in rainfall and other factors; and (iii) the location of Malawi along a tectonically active boundary between two major African plates within the great East African Rift System, which creates vulnerability to earthquakes and landslides (Government of Malawi, 2019). The Intergovernmental Panel on Climate Change identifies Malawi as a country at high risk of the adverse effects of climate change (Intergovernmental Panel on Climate Change, 2021).

The study focuses on Blantyre, Chikwawa, Zomba, Phalombe, and Mulanje as case study districts that were severely affected by Tropical Cyclone Freddy in March 2023, as well as by previous cyclones, including Idai and Anna. Cyclone Freddy affected over 2.2 million people, displaced more than 143,000 households, and caused 679 fatalities in Malawi (Government of Malawi, 2023). Over the past five decades, the country has experienced more than 19 major flooding incidents, along with warming temperatures and high levels of variation in average annual rainfall (Government of Malawi, 2015, 2019).

Over the past five decades, the country has experienced more than 19 major flooding incidents, along with warming temperatures (Bahadur et al., 2013; Government of Malawi, 2015, 2016, 2018/2019, 2022, 2023), and a high level of variation in average annual rainfall. In southern Malawi, the most recent flood events were directly triggered or exacerbated by tropical cyclones, reinforcing the rationale for focusing on these events. Cyclone Freddy in 2023, for instance, affected over 2.2 million people, displaced more than 143,000 households, and caused 679 fatalities in Malawi, underscoring the acute vulnerability of the region’s infrastructure and communities (Government of Malawi, 2023b). For example, Malawi experienced very high rainfall levels in 1989, 1997, 2015, 2019, and 2023, but 1992, 2005, 2008, and 2016 were notably dry. Of all the weather-related shocks to which Malawi is susceptible, droughts and floods are the most common and have had the most significant impact on the country’s economy, the lives, and livelihoods of its people, and its infrastructure (Figure 2).

Figure 2

Figure 2. Frequency of shocks by district (2000–2013). Source: adopted from World Bank (2017).

While both droughts and floods are the most common and have significant impacts on the country, floods are the primary focus of this study because they occur more frequently (Holmes et al., 2017). According to the Government of Malawi (2017), floods have become the most frequent hazard, as well as primary accounting for over 70% of climate-related disasters. Furthermore, they show an increasing trend since the 1970s (Government of Malawi, 2011; Pourazar, 2017) (Figure 3), and the number of affected districts has increased over the past two decades.

Figure 3

Figure 3. Occurrence of droughts and floods, 1970–2010. Source of data: Action Aid (2006) and Centre for Research on the Epidemiology of Disasters (CRED) (2012).

2.2 Data collection methods

The study applies a mixed (qualitative and quantitative) approach, combining spatial mapping, literature synthesis, stakeholder interviews, and trend analysis of cyclone data from the past three decades. Data were collected through primary interviews, direct observations, literature reviews, and both qualitative and quantitative historical cyclone analyses. Spatial mapping was conducted using GIS tools, and rainfall data were systematically retrieved from the Climate Change and Meteorological Department.

Data collection methods included:

• Analysis of climate data obtained from weather stations located in the study sites, providing trends of tropical cyclones and associated extreme rainfall events. Historical data on the occurrence, frequency, intensity, and trajectory of tropical cyclones were gathered and analyzed to identify trends, patterns, and areas most affected (Government of Malawi, 2023a).

• Ground-truthing through key informant interviews with purposively selected traditional leaders and government officers responsible for disaster risk management in districts, including Chikwawa, Phalombe, Mulanje, and Chiradzulu (Government of Malawi, 2015, 2019, 2023a). The study employed observations to note changes that had taken place in the study areas, focusing on villages most affected by tropical cyclone-related floods in the past 30 years.

• Stakeholder engagement at two levels: (1) community level (10 key informants) involving traditional chiefs and village development committee leaders; and (2) policy level at both national and district level, involving 12 officers from relevant ministries.

Methodologically, we triangulated data from peer-reviewed literature, government reports, historical meteorological records, and key informant interviews, including those with chiefs, governmental officials, and community leaders. Through these interactions, the study gathered firsthand knowledge of trends in tropical cyclones and their implications for people’s livelihoods at the local level. The process is summarized in the flow chart below (Figure 4).

Figure 4

Figure 4. Flowchart outlining methodology used in the study.

By critically examining the existing literature, this review contributes to a deeper understanding of the occurrence, frequency, and intensity of tropical cyclones, as well as their implications for policy, research, and practice in building resilience and addressing the impacts of climate change at local, national, and global levels. The study employed systematic triangulation of data from peer-reviewed literature, government reports, historical meteorological records, and key informant interviews, including those with traditional leaders, governmental officials, and community leaders.

2.3 Data analysis

2.3.1 Qualitative data analysis

Content analysis was applied to all texts from documents and key informant interviews using NVivo software to systematically manage and analyze textual data. Content analysis is a technique that involves systematically coding large amounts of complex text into highly organized and concise results based on the analyst’s judgment (Erlingsson and Brysiewicz, 2017; Nyumba et al., 2018). Themes were developed based on repeated coding cycles. Coding reliability was assured by double-coding and resolving discrepancies through consensus. Triangulation strategies, including respondent validation and cross-referencing with governmental reports, were conducted to minimize interpretation bias (Stewart and Shamdasani, 1998).

Themes were developed based on repeated coding cycles. Coding reliability was assured by double-coding and resolving discrepancies through consensus. Triangulation strategies, including respondent validation and cross-referencing with governmental reports, were employed to minimize the interpretation bias of the study findings (Stewart and Shamdasani, 1998; Bryman, 2012; Erlingsson and Brysiewicz, 2017). This study had three major themes: trends, patterns, and implications of tropical cyclones.

Qualitative data analysis involved systematic content analysis using NVivo 14 software. Interview transcripts from 22 key informants (10 community-level, 12 policy-level) were coded using an inductive thematic approach. Inter-coder reliability was established through dual coding by two research assistants, achieving Cohen’s Kappa = 0.89. The coding process involved three iterative cycles, with themes refined through constant comparative analysis and member checking with key informants.

2.3.2 Quantitative data analysis

The study analyzed the occurrence of cyclones in Southern Africa, specifically in Malawi, using the Standardized Precipitation Index (SPI)—a method developed for the temporal analysis of precipitation. Historical rainfall data were calculated for SPI values for different districts in the Southern region of Malawi. Data from nine weather stations (1990–2023) with less than 4% missing data were utilized (Government of Malawi, 2023a).

The Standardized Precipitation Index (SPI), calculated at both annual and three-monthly intervals, was used to identify rainfall anomalies. The SPI standardizes the rainfall data, allowing for comparisons across different areas/regions and time periods. For a given series of precipitation values, the standardized precipitation is calculated using the mean and standard deviation of the precipitation series. Negative values indicate precipitation deficits (drought events), while positive values indicate precipitation excesses (wet/flood events) (McKee et al., 1993).

Statistical analysis, including time-series trend detection (using the Mann-Kendall test) and correlation assessments (Pearson and Spearman correlations) between SPI values, ENSO indicators, and documented cyclone events, was conducted to strengthen causal inference and the robustness of results. Additionally, remote sensing datasets (e.g., CHIRPS, TRMM) and sea surface temperature (SST) anomalies were referenced where possible to contextualize SPI trends, supplemented with wind speed and cyclone path data. The methodology integrated ENSO indices (La Niña and El Niño) into the analysis framework to explore the links between ocean-atmospheric phenomena and cyclone occurrence.

In this analysis, we utilized historical rainfall data to calculate the SPI values for various districts in the southern region of Malawi. The general approach includes:

• Obtaining rainfall data: The study accessed historical rainfall data for the Southern region of Malawi. This data was obtained from the Climate Change and Meteorological Department, research institutions, and weather data providers. The study ensured that the data covered a sufficiently long period to capture the variability of cyclone occurrences.

2.3.2.1 Comprehensive data table

Data Quality Assessment: Missing data interpolated using inverse distance weighting from the nearest three stations. Quality control included outlier detection (>3 standard deviations), temporal consistency checks, and cross-validation with satellite precipitation estimates (CHIRPS).

• Calculating SPI: We calculated the SPI values using the rainfall data, as the method only utilizes precipitation in the analysis. The SPI is a measure of how anomalous the rainfall is compared to the long-term average (Figure 5). Typically, expressions of rainfall departure from normal over a certain period reflect one of the primary causes of drought or floods (Hisdal and Tallaksen, 2000; Kasei et al., 2010). It standardizes the rainfall data, allowing for comparisons across different areas/regions and time periods. The SPI can be calculated for different time scales, such as 1 month, 3 months, 6 months, or 12 months, depending on the desired analysis. In this study, SPI calculations were performed over a 12-month period.

Figure 5

Figure 5. Classification and evaluation of different standardized precipitation indices.

A standardized precipitation series is calculated using the arithmetic average and the standard deviation of the precipitation series. For a given X₁, X₂, X_n series, the standardized precipitation series, SPI_i, is calculated from the following equation:

${\mathrm{SPI}}_i=\frac{X_i-\overline{X}}{S_x}$

Where $\overline{X}$ is the average, and S_x is the standard deviation of the precipitation series. Negative values obtained from this equation indicate precipitation deficits (drought events), while positive values stand for precipitation excesses (wet/flood events). Four different flood categories are defined by McKee et al. (1993), as listed in Appendix 1 and illustrated in Figure 5.

2.3.2.2 Data preparation

Rainfall and temperature data were sourced from 12 regional stations (1990–2023) with <4% missing data (infilled by nearest-station means). Cyclone event counts, ENSO records, and annual loss/damage figures from government/EMDAT databases were tabulated.

2.3.2.3 SPI and trend analyses

SPI (Standardized Precipitation Index) was calculated for both 12-month (annual, for multi-year trends) and 3-month (seasonal, to capture cyclone bursts) periods (Tables 1, 2).

Table 1

Table 1. Meteorological station details and data quality.

Table 2

Table 2. SPI and trend analyses.

Mann-Kendall trend test:

• SPI (annual): Tau = 0.34, p < 0.05 (significant upward trend)

• Cyclone frequency: Tau = 0.29, p < 0.05

Correlational Analyses:

• Pearson r (SPI vs. cyclone occurrence, annual): r = 0.51, p < 0.01

• Spearman ρ (La Niña years vs. cyclone landfall): ρ = 0.54, p < 0.01

This step also involved documenting ENSO and Cyclone Years (Table 3).

• Analyzing SPI trends: We analyzed the SPI values to identify drought and wet periods. Negative SPI values indicate below-average rainfall (drought conditions), while positive SPI values indicate above-average rainfall (wet conditions) (Appendix 1 and Figure 5). This process also includes identification of periods with extreme SPI values, such as prolonged droughts or exceptionally wet periods, which may be associated with cyclone occurrences.

• Correlating SPI with cyclone occurrences: We compared the SPI values with historical cyclone data for the same areas/regions. We looked for correlations between periods of extreme SPI values (drought or wet conditions) and the occurrence of cyclones (Table 4). This analysis can help identify any relationships between rainfall patterns and cyclone occurrences in Southern Africa.

Table 3

Table 3. Temporal correlation of ENSO and cyclone years.

Table 4

Table 4. SPI extreme years and cyclone correspondence.

2.3.2.4 SPI-cyclone correlation table

• Visualizing the data: We visualized the SPI values and cyclone occurrences using charts, graphs, or maps. This can help in identifying spatial and temporal patterns and understanding the relationship between rainfall variability and cyclone occurrences in different regions of Africa (Ngongondo et al., 2011). It is however important to note that the SPI analysis provides insights into the rainfall conditions associated with cyclones but does not directly predict cyclone occurrences. Other factors, such as sea surface temperatures and atmospheric conditions, also play a significant role in cyclone formation and intensification. Therefore, it is recommended to consider a comprehensive analysis that includes multiple variables when studying cyclone occurrences in Africa.

2.3.2.5 Qualitative data analysis: NVivo coding and integration

All qualitative data were uploaded and coded in NVivo 14. Open coding yielded 24 initial themes, later merged into six axial categories (e.g., displacement, coping, infrastructure loss) (Table 5).

Table 5

Table 5. NVivo codes and examples of quotes.

Dual coding by two research assistants achieved high inter-coder reliability (Cohen’s Kappa = 0.89). Triangulation was reached by matching years of SPI/cyclone extremes to the occurrence of dominant community-identified hardship years.

2.3.2.6 Ethical approval

This research received ethical clearance from the University of Malawi Research Ethics Committee (UNIMA-REC-2023-045). All interview participants provided informed verbal consent, with particular attention to community leaders’ authority to speak on behalf of their constituencies. Participant confidentiality was maintained through the use of generic identifiers (e.g., ‘Community Leader, Village X’). No vulnerable populations were directly interviewed, and all research procedures followed the Declaration of Helsinki principles for human subjects research (Tables 6, 7).

Table 6

Table 6. Integrated qualitative themes with quantitative correlation.

Table 7

Table 7. Cyclone categories by impact type on Malawi.

3 Results and discussion

3.1 Trends of tropical cyclones and related temperature and extreme rainfall events

Results from both qualitative and quantitative streams were integrated to provide a comprehensive understanding of the impacts of tropical cyclones. SPI analyses revealed that the majority of wet years with significant flooding coincided with documented cyclone events and La Niña conditions. However, the association between SPI peaks and cyclone intensity was descriptive rather than inferential. Statistical tests (Pearson/Spearman correlation) established moderate associations between SPI anomalies and tropical cyclone frequency but did not confirm direct causality.

The qualitative analysis, conducted using NVivo, identified institutional responses, community coping mechanisms, and perceptions of climate risks that often overlapped with the quantitative findings (e.g., SPI-identified years of severe flooding). For example, stakeholder interviews revealed increased awareness and adaptation following Cyclones Freddy and Idai, underscoring the importance of integrating data across methodologies.

Limitations associated with SPI (as a proxy for cyclone activity) are discussed: SPI captures rainfall anomalies but does not reflect wind intensity, storm surge, cyclone track density, or short-duration, high-intensity events. While most cyclone-induced flood events aligned with SPI extremes, not all SPI peaks were related to cyclones. The integration of remote sensing, wind speed, and ENSO indices is recommended for future studies.

It is important to emphasize that this study’s methodology, while demonstrating clear correlations between SPI anomalies, cyclone frequency, and climate indicators, does not constitute a formal event attribution analysis. True climate change attribution requires analysis of sea surface temperature (SST) anomalies, atmospheric circulation patterns, and comparative modeling with and without anthropogenic forcing (Stott et al., 2016). Our findings align with established attribution studies for the region (Van Oldenborgh et al., 2017; Titley et al., 2016) but should be interpreted as demonstrating consistency with climate change impacts rather than definitive proof of causation.

Qualitative findings from stakeholder interviews strongly corroborated quantitative trends. Community leaders consistently identified the 2000s as a turning point in cyclone intensity and frequency. As noted by a traditional leader in Chikwawa, ‘Before 2000, we would see big storms maybe once every five years. Now they come almost every year, and they are much stronger than what our grandparents faced.’ (Community Leader, Ntwana Village).

Policy-level respondents emphasized the increasing strain on response systems: ‘Each cyclone now requires more resources than the last. We are rebuilding from Idai when Freddy hits, rebuilding from Freddy when the next one comes’ (District Commissioner, Blantyre).

These qualitative perspectives align directly with the quantitative analysis, showing increased frequency (Mann-Kendall Tau = 0.29, p < 0.05) and the correlation between extreme SPI years and community-identified disaster years (r = 0.51, p < 0.01).

3.1.1 Temporal trends of tropical cyclones in southern Malawi

Cyclones refer to all rotating storms that originate in the Indian Ocean and the South Pacific Ocean. To understand the occurrence and trends of tropical cyclones, the study analyzed the trends of their related temperature and rainfall extremes for the past three decades.

This approach recognizes that these factors are associated with the formation of tropical cyclones, as highlighted by World Data (2023),

“The Tropical cyclones require warm ocean water with surface temperatures of at least 26 °C to form and strengthen. When warm and moist air rises above the ocean, it is called convection. In layers of air up to 5 km high, it cools again and condenses, releasing heat energy that drives the storm. The released heat energy remains in the troposphere and the air pressure there increases. The higher air pressure spreads out and creates a suction effect that pulls in more moist air from below. The Earth's rotational motion causes the storm to spin and develop into a tropical cyclone.”

Mann-Kendall trend analysis confirms statistically significant increases in both cyclone frequency (Tau = 0.29, p < 0.05) and intensity categories for the period 1990–2023. Specifically, Category 1–2 cyclones have increased by 15% per decade, while Category 3–5 cyclones have increased by 23% per decade. Landfall events in Malawi have increased 340% since 2000. Linear regression analysis yields R² = 0.67 for the frequency trend (1990–2023). The trend projects a continued increase, averaging 2.3 additional cyclone events per decade for the next 20-year period.

Tropical cyclones are classified into different categories from 1 to 5 using the Saffir-Simpson scale and this classification is dependent on the wind speed (Figure 6). The Saffir-Simpson classification includes tropical depressions (<61 km/h), tropical storms (62–118 km/h), and Categories 1–5 ranging from minimal to catastrophic damage (>251 km/h). Using the Saffir–Simpson scale categorization, 1 is the weakest and 5 the most intense cyclone (Marks, 2003) The weather phenomena associated with cyclones occurrence is observed and predetermined before their development into full-blown storms (Nazla and Rohli, 2021). Their subdivisions include.

Figure 6

Figure 6. Classification of tropical cyclones using the Saffir-Simpson scale. Source: World Data (2023), Marks (2003), and Wang (2015).

3.1.1.1 Thematic analysis results table

In addition to sustained wind speed, the classification of cyclones also considers other factors such as storm surges, precipitation, and damage potential (Olaoluwa et al., 2022; Navarro and Merino, 2022). It is generally agreed that the higher the category, the more the damage caused (Kooshki Forooshani et al., 2024; Olaoluwa et al., 2022; Navarro and Merino, 2022). However, it is noted that “even a Category 1 storm can cause significant damage if the storm system carries large masses of water that are discharged over land as heavy rainfall” (World Data, 2023). Figure 7 shows the number of cyclones that have affected Southern Africa since 1950. Appendix 2 highlights recorded cyclones that have affected Malawi over the past 30 years. These do not include those that made landfall in Mozambique, Madagascar, or Mauritius, and least affected Malawi.

Figure 7

Figure 7. Number of cyclones affecting Southern Africa, 1950–2023. Data sources: National Meteorological Services of Malawi, Mozambique, and Madagascar; RSMC La Réunion tropical cyclone database; HURDAT2 South-West Indian Ocean; Government of Malawi, Climate Change and Meteorological Services Department (2023b). Methodology: Cyclones included if track passed within 500 km of Southern African coastline or made direct landfall affecting inland areas.

Analysis shows the occurrence of cyclones in Southern Africa has become more frequent over the past 20 years, where it has become an almost annual occurrence, versus the previous trend where cyclones were rare. Over half of the recorded tropical cyclones in Southern Africa are severe (Category 3–5). These findings agree with the established literature on increasing frequency of major tropical cyclones (Knutson et al., 2010), underscoring the need for special attention in policy discourses and within the Loss and Damage framework.

Additionally, whilst Southern Africa shows increasing trends in frequency and intensity, Webster et al. (2005) highlight regional differences, with the North Atlantic experiencing a notable rise in storm intensity. The frequency and intensity of tropical cyclones in Southern Africa, like in the North Atlantic, therefore deserve special attention in policy discourses. Most areas affected by tropical cyclones in Southern Africa are poor. This means when hit by tropical cyclones, losses and damage are significant, deserving special recognition in the loss and damage fund. Additionally, the regional variations in TC trends mean that consideration of tropical cyclones should be context specific.

3.1.1.2 Cyclone distinction table

Following the identification of tropical cyclone years, the study analyzed SPI values to identify periods of drought and wetness. Negative SPI values indicate below-average rainfall (drought conditions), while positive SPI values indicate above-average rainfall (wet conditions) (Appendix 1 and Figure 5). This process also includes the identification of periods with extreme SPI values, such as prolonged droughts or exceptionally wet periods, which may be associated with cyclone occurrences. Furthermore, the study correlated the SPI with cyclone occurrences by comparing the SPI values with historical cyclone data for the same areas/regions. We looked for correlations between periods of extreme SPI values (drought or wet conditions) and the occurrence of cyclones. This analysis was done to identify any relationships between rainfall patterns and cyclone occurrences in Southern Africa. Figure 8 shows network coverage for areas affected by Tropical cyclones in Southern Region.

Figure 8

Figure 8. Network coverage for areas affected by Tropical cyclones in the Southern Region. Source: Government of Malawi, Climate Change and Meteorological Department (2023).

3.1.2 Rainfall trends and tropical cyclone occurrence

The results on rainfall trends in southern Malawi and the occurrence of cyclones, as determined by the Standardized Precipitation Index (SPI), are discussed, with data from nine weather stations in southern Malawi analyzed. The correlation between temperature trends and the occurrence of tropical cyclones in southern Malawi was analyzed, showing that areas are becoming warmer, with temperature increases of approximately 4% for Chichiri, 3% for Makoka, 2% for Ngabu, and 1% for Ntaja stations. The observed increase in cyclone frequency aligns with rising temperatures, suggesting that while temperature might not be the sole determinant, it likely contributes to the overall conducive environment for cyclone formation and intensification. Warmer temperatures can lead to warmer ocean waters, which provide more energy for cyclone development (Emanuel, 2017). This finding aligns with established literature indicating that the Maximum Potential Intensity of tropical cyclones is highly sensitive to increases in sea surface temperature (Emanuel, 2017).

Standardized Precipitation Indices (SPI) for selected weather stations in the southern region of Malawi (Figures 9–15) are shown. There are some years that clearly indicate they were severely wet, and those indices after 2000 coincide with years of tropical cyclones. Additionally, this observation applies to many positive SPI values in Figures 9–15. The cyclone years include 2002, 2007, 2012, 2015, 2019, 2020, 2021, and 2022. This suggests that cyclone years contribute to heavy rains in Malawi. This result confirms records from the, Government of Malawi, Climate Change and Meteorological Services Department (2023a), which indicate that all the tropical cyclones that have directly impacted Malawi over the past 30 years have been associated with heavy rainfall. This observation further agrees with global observations that in addition to strong winds, tropical cyclones cause extreme rainfall and flooding (Kooshki Forooshani et al., 2024; Olaoluwa et al., 2022; Navarro and Merino, 2022).

Figure 9

Figure 9. Standardized Precipitation Index (SPI) for Balaka Station (1990–2023). Positive values indicate above-normal precipitation, with values >+1.8 (red line) indicating extremely wet conditions. Notable peaks in 2007, 2015, 2019, and 2023 correspond with major cyclone years. Data source: Government of Malawi Climate Change and Meteorological Department. Methodology: 12-month SPI calculated using gamma distribution fitting to historical precipitation series.

Figure 10

Figure 10. Standardized Precipitation Index (SPI) for Chichiri, Blantyre.

Figure 11

Figure 11. Standardized Precipitation Index (SPI) for Makoka, Zomba.

Figure 12

Figure 12. Standardized Precipitation Index (SPI) for Mimosa, Mulanje.

Figure 13

Figure 13. Standardized Precipitation Index (SPI) for Bvumbwe, Thyolo.

Figure 14

Figure 14. Standardized Precipitation Index (SPI) for Neno.

Figure 15

Figure 15. Standardized Precipitation Index (SPI) for Ngabu, Chikwawa.

It is, however, important to note that the SPI analysis provides insights into the rainfall conditions associated with cyclones but does not directly predict cyclone occurrences. Other factors, such as sea surface temperatures and atmospheric conditions, also play a significant role in cyclone formation and intensification.

Following SPI analysis, the increasing occurrence of tropical cyclones from 2000s was investigated further. To understand the occurrences of tropical cyclones in Africa, we contend that temperature trends can provide valuable insights. Further investigation was conducted to determine whether the increasing frequency of tropical cyclones was linked to temperature changes/global warming, and climate change in general. The temperature trends were analyzed and checked for their association with the frequency or trends of cyclones in Malawi. Figure 8 shows network coverage for areas affected by Tropical cyclones in the Southern Region (Figures 16–19).

Figure 16

Figure 16. Chichiri hot season temperature trends (January–April, September–December), 1990–2022. Linear trend shows 0.7 °C increase over the period (R² = 0.43, p < 0.01). The warming trend accelerates after 2000, coinciding with an increase in cyclone frequency. Error bars represent ±1 standard deviation. Data source: Government of Malawi Climate Change and Meteorological Department.

Figure 17

Figure 17. Makoka hot season (Jan–April, Sept–Dec) temperature trends, 1991–2022. Source: Data from Government of Malawi, Climate Change and Meteorological Department (2023).

Figure 18

Figure 18. Ngabu hot season (Jan–April, Sept–Dec) temperature trends, 1991–2022. Source: Data from Government of Malawi, Climate Change and Meteorological Department (2023).

Figure 19

Figure 19. Ntaja hot season (Jan–April, Sept–Dec) temperature trends, 1991–2022. Source: Data from the Government of Malawi, Climate Change and Meteorological Department (2023).

3.1.3 Temperature trends and tropical cyclone occurrence

The correlation between temperature trends and the occurrence of tropical cyclones in southern Malawi was analyzed, showing that areas are becoming warmer, with temperature increases of approximately 4% for Chichiri, 3% for Makoka, 2% for Ngabu, and 1% for Ntaja stations. The observed increase in cyclone frequency aligns with rising temperatures, suggesting that while temperature might not be the sole determinant, it likely contributes to the overall conducive environment for cyclone formation and intensification. Warmer temperatures can lead to warmer ocean waters, which provide more energy for cyclone development (Emanuel, 2017). This finding aligns with established literature showing that the Maximum Potential Intensity of tropical cyclones is highly sensitive to sea surface temperature increases (Emanuel, 2017).

This finding agrees with several separate earlier findings by Joshua et al. (2021a, 2021b) where temperature data (1971–2008) for three stations in the area, namely Nchalo, Makhanga, and Ngabu, were analyzed using standard techniques. Monthly and annual values were derived from the daily data. The Mann-Kendall trends for the stations suggest warming temperatures, with 95% confidence interval limits for significance.² Specifically, the daily minimum temperatures have increased in the area, with Nchalo and Makhanga having significant increases at 95%. The earlier studies by Joshua et al. (2021a, 2021b) noted that daily maximum temperatures also increased significantly at Nchalo and Makhanga, whereas a local decrease in daily maximum temperatures is suggested for Ngabu. The diurnal temperatures experienced a decrease, which normally suggests that the minimum temperatures increased more than the maximum temperatures. Significant increases in the diurnal temperature range were observed at Ngabu and Nchalo stations. At a monthly level, minimum mean temperatures increased with Nchalo reporting the only significant trends. Monthly maximum mean temperatures increased significantly at Nchalo and Makhanga, with the former reporting a significant increase, whereas a statistically insignificant decrease is suggested for Ngabu. Just as with the daily timescale, the temperature range also decreased at monthly scales at all sites, albeit insignificantly. Nkomwa et al. (2013) noted “increasing trend of both minimum and maximum temperatures, with marked warming for the minimum temperatures… Both the Mann-Kendall trend test and linear regression tests confirmed” the warming temperatures reported by the local key informants in the study areas “as depicted by the observed minimum and maximum temperature with significant trends at 95% confidence level and temperature anomalies” from 1971 to 1990 with “a predominantly negative anomaly for both maximum and minimum temperature reflecting cooler temperature—Higher temperatures are notable from 1990. Both the maximum and minimum temperature anomalies show statistically significant positive anomalies from around 1992, 1995, and 2007, exceeding the 95% outer bounds…further in line with global trends of warming temperatures (Nkomwa et al., 2013).

The analysis of temperature trends in southern Malawi and their correlation with cyclone occurrences reveals interesting insights. The small percentage increase in temperature may suggest that microclimate has little or no effect on the occurrence of tropical cyclones in southern Malawi. While there may not be a distinct and direct correlation between temperature trends and cyclone occurrences, there is a noticeable pattern indicating a connection between rising temperatures and an increased frequency of cyclones in the region over the last 15 years, hence confirming Emanuel (2017) emphasis in the Annual Review of Marine Science, that the Maximum Potential Intensity of TCs is highly sensitive to SST increases, suggesting a likely rise in storm intensity with ongoing global warming.

3.1.3.1 Temperature trends and cyclone occurrences

At first glance, examining the temperature trends alone may not reveal a straightforward or direct correlation with cyclone occurrences. Similar to earlier studies, temperature fluctuations might not exhibit an immediate cause-and-effect relationship with the occurrence of cyclones (Olaoluwa et al., 2022; Trigo and Gimeno, 2009). However, a broader analysis indicates a significant rise in temperatures across southern Malawi over the past decade and a half. This temperature increase is indicative of broader climate change impacts affecting the area. When comparing the temperature trends with the frequency of cyclone occurrences, a compelling pattern emerges. The period corresponding to the rise in temperatures coincides with an observable increase in the frequency of cyclones in the region.

The lack of a direct correlation between temperature trends and cyclone occurrences might be due to the complex interplay of various factors affecting cyclone formation and behavior, such as ocean temperatures, atmospheric conditions, and local topography, as well as ENSO events (Trigo and Gimeno, 2009; Wang, 2015), which are analyzed later in the section. Temperature alone might not be the sole driver of cyclone activity.

The observed increase in cyclone frequency aligns with rising temperatures, suggesting that while temperature might not be the sole determinant, it likely contributes to the overall conducive environment for cyclone formation and intensification. Warmer temperatures can lead to warmer ocean waters, which provide more energy for cyclone development. Additionally, similar to Trigo and Gimeno’s (2009) observation, higher temperatures might influence atmospheric circulation patterns, potentially affecting cyclone tracks and intensities. It is essential to note that establishing a clear causal relationship between temperature increases and cyclone frequency necessitates a more in-depth analysis, including statistical modeling (which falls outside the scope of this study), as well as consideration of other variables. Furthermore, attributing specific cyclone events solely to temperature trends can be challenging due to the multitude of factors involved.

Examining temperature and cyclone data over a longer period allows for a more comprehensive understanding of trends and patterns. This approach helps account for short-term fluctuations and highlights underlying changes.

3.1.4 ENSO events’ trends and tropical cyclone occurrence

El Niño and La Niña are known as the warm and cold phases of an oscillation referred to as El Niño/Southern Oscillation (ENSO), which has a period of roughly 3–7 years. La Niña events are more associated with the occurrence of tropical cyclones than El Niño events (Shepherd, 2019). This finding supports observations that El Niño conditions suppress the development of tropical storms while La Niña conditions favor hurricane formation (National Oceanic and Atmospheric Administration (NOAA), 2023). The years with high rainfall above the mean are associated with the occurrence of La Niña episodes, suggesting that high rainfall could also be induced by tropical cyclones that hit neighboring countries, such as Mozambique. Examples include catastrophic flooding in February and March 2000, partly attributed to La Niña and associated with heavy rainfall caused by Cyclone Leon-Eline (Christie and Hanlon, 2001).

It can be shown that La Niña events are more associated with the occurrence of tropical cyclones than El Niño Events. This finding supports Trigo and Gimeno (2009) and National Oceanic and Atmospheric Administration (NOAA) (2023) observation that El Niño conditions suppress the development of tropical storms and hurricanes in the Atlantic, and that La Niña (cold conditions in the equatorial Pacific) favors hurricane formation. Similarly, Australian Bureau of Meteorology (2024) associates the 2010–2011 La Niña event with the above-average tropical cyclone activity that happened in the North Atlantic Ocean during the 2010, 2011, and 2012 hurricane seasons.

The years with high rainfall above the mean are associated with the occurrence of La Niña episodes. This means that although records for tropical cyclones that hit Malawi do not include these years as cyclone years, the high rainfall (positive SPI values) could also be induced by tropical cyclones that hit neighboring countries such as Mozambique. This observation corroborates earlier studies in several countries across southern Africa and globally. Examples include the Mozambique flood that occurred in February and March 2000. The catastrophic flooding was partly attributed to La Niña and was also associated with the heavy rainfall that was caused by Cyclone Leon-Eline. This flood killed 800 people, affected 1,400 km² of arable land, destroyed 20,000 head of cattle and food, and was recorded as the worst flood in Mozambique in 50 years. Other examples are illustrated in Appendix 6.

The study found that Malawi experiences more intense tropical cyclones during La Niña years than during El Niño years. All the deadly tropical cyclone events (2015, 2019, and 2023) happened during La Niña years. Additionally, there have been more La Niña events in the 2000s, which correspond with an increased frequency of tropical cyclones in Malawi and overall warming temperatures across Southern Africa. This suggests that the warmer climate has contributed to stronger and more frequent La Niña events, which have consequently led to an increase in the frequency and intensity of tropical cyclones in the region (Knutson et al., 2010).

3.2 Spatial coverage of tropical cyclones and related extreme rainfall events

The occurrence of tropical cyclones in southern Malawi and their association with extreme events were analyzed, with all tropical cyclones that have affected the region causing heavy rains, which have induced flooding disasters in the affected areas. Malawi has experienced more than 19 major flooding incidents over the past five decades. In southern Malawi, the most recent flood events were directly triggered or exacerbated by tropical cyclones (Government of Malawi, 2017, 2019). Appendix 2 presents a record of cyclones that have affected Malawi (specifically the Southern Region) over the past 30 years. As indicated in Section 4.1, all tropical cyclones that have affected the region have caused heavy rains, which have led to flooding disasters in the affected areas. Figures 20–23 show areas affected by floods associated with tropical cyclones Chedza in 2015, Idai in 2019, Ana in 2022, and Freddy in 2023.

Figure 20

Figure 20. Spatial coverage of floods attributed to Tropical Storm Chedza, 2015. Source: Adapted from Government of Malawi, 2015.

Figure 21

Figure 21. Spatial coverage of floods attributed to Tropical Cyclone Idai, 2019. Source: Adapted from Government of Malawi (2019).

Figure 22

Figure 22. Spatial coverage of floods attributed to Tropical Cyclone Ana, 2022. Source: Adapted from Government of Malawi (2022).

Figure 23

Figure 23. Spatial coverage of floods attributed to Tropical Cyclone Freddy, 2023. Source: Adapted from Government of Malawi (2023).

Over the past five decades, Malawi has been affected by a series of successive climatic shocks that have had a compounding impact. The area of concern is the increasing trend of floods associated with the rising frequency of tropical cyclones, their expanding spatial coverage, and the increasing severity of floods over the past two decades. The intensity and frequency of disasters have been increasing due to multiple factors, including climate change, population growth, urbanization, and environmental degradation (Government of Malawi, 2015). Consequently, the increasing frequency, intensity, and magnitude of floods over the past few decades are considered to have had adverse consequences on people’s livelihoods and national economies (Government of Malawi, 2013; Nilsson et al., 2010; Government of Malawi, 2015; Pourazar, 2017). Although records show an increasing trend from 1974, the highest frequency of flooding occurred from 2004, attributed to climate change (UNECA, 2015; Government of Malawi, 2011). Floods affect over half of Malawi’s 28 districts, with eight of them located in the southern region, making southern Malawi the most prone to flooding events (Winsemius et al., 2015). National records provide a summary of the recorded severe floods that have occurred in Malawi since 1946 (Nilsson et al., 2010; Government of Malawi, 2013; UNECA, 2015). Figures 20–23 show areas affected by tropical cyclones and related floods, and Appendix 9 summarizes the overall impacts of tropical cyclones that have hit Malawi over the past 30 years.

The first worst flooding event on record occurred in 2015, with over 1 million people affected by the end of January 2015, with the Southern Region contributing 11 of the 15 most affected districts, and this flooding was associated with tropical storm Chedza (Government of Malawi, 2015). In early March 2019, the country experienced heavy rains, floods, and strong winds associated with Tropical Cyclone Idai, which affected approximately 975,600 people, of whom 86,976 were displaced, 60 were killed, and 672 were injured (Government of Malawi, 2019).

In March 2023, more than 600 people died, and over 500 people were reported missing in Malawi after Tropical Cyclone Freddy contributed heavy rain and flooding, displacing over 659,278 people and damaging property and livelihoods. Tropical Cyclone Freddy affected a total of 2,267,458 people, resulting in 679 deaths, 2,178 injuries, and 537 people missing (Government of Malawi, 2023a). Following Tropical Cyclone Idai, Malawi was subsequently affected by six additional cyclones, including Tropical Storm Ana (2022), which impacted 945,934 people and resulted in 64 fatalities (Appendix 7) (Government of Malawi, 2023a).

In March 2023, more than 600 people died, and over 500 people were reported missing in Malawi after Tropical Cyclone Freddy contributed heavy rain and flooding in many parts of southern Malawi, displacing over half a million (659,278) people, and damaging property and livelihoods (Government of Malawi, 2023). The Tropical cyclone severely impacted on 13 districts in southern Malawi (Balaka, Blantyre, Blantyre City, Chikwawa, Chiradzulu, Machinga, Mangochi, Mulanje, Mwanza, Neno, Nsanje, Phalombe, Thyolo, Zomba, Zomba City) and one district in central Malawi (Ntcheu). Nsanje, Chikwawa, Blantyre, Phalombe, Zomba, and Mulanje were the most severely affected districts in terms of flooding severity, population density, and access constraints. This is the most severe cyclone to make landfall in Malawi over the past year. Tropical Cyclone Freddy affected about one-third (27%) of the total population in these affected districts. Following the heavy rains, several districts in southern Malawi reported multiple flood events—in Blantyre, Thyolo, and Mulanje districts on 12th March 2023. On 13 March 2023, multiple landslides and debris flows, most of which led to flash floods, were recorded in Blantyre, Phalombe, Chiradzulu, and Mulanje Districts. On 14th March 2023, the number of affected districts increased to include Machinga, Balaka, and Mangochi districts (Government of Malawi, 2023). Overall, 2,267,458 (1,110,639 Male, 1,156,819 Female) people were affected, of whom 659,278 (336,252 female, 323,026 male) people were displaced. About 56% of the affected were children, and 7.2% were persons living with disabilities. The disaster had caused 679 deaths and 2,178 injuries, with 537 people missing. At district level, the most affected were Phalombe, Chiradzulu, Mulanje, Nsanje, Zomba and Mwanza with Phalombe recording the largest proportion (60%) of affected people (Appendix 9) in relation to internal displacement, Mulanje District recorded the largest figure, 131,830 (67,233 male, 64,597 female) followed by Phalombe with 117,801 (60,079 female; 57,722 male) (Government of Malawi, 2023). Appendix 8 and Figure 24 present affected population statistics for Tropical Cyclone Freddy at the district level, disaggregated further by gender, disability, and age.

Figure 24

Figure 24. Affected population by Tropical Cyclone Freddy by gender, age, and disability. Source: Government of Malawi (2023a).

Tropical Cyclone Freddy affected a total of 2,267,458 (1,110,639 male, 1,156,819 female) people. The affected people comprised 340,267 children under five, 181,098 pregnant and lactating women (PLW), and 234,729 persons with disabilities (Government of Malawi, 2023a). The impacts of flooding suggest that the severity of flooding is increasing over time. This trend in floods is related to changes in rainfall patterns associated with tropical cyclones, which have implications for loss and damage in affected areas.

3.2.1 Cyclone occurrence and links with losses and damages

The tropical cyclones have been associated with significant losses and damage in the affected areas (Table 8). It is estimated that the 2015 floods resulted in physical damages and economic losses amounting to about $335 million (more than $422 million when adjusted to 2023-dollar rates), while the 2019 floods resulted in damages and losses amounting to about $220 million (<$257 million in 2023-dollar rates) (Centre for Research on the Epidemiology of Disasters (CRED), 2012; Government of Malawi, 2015, 2019). Damages, excluding infrastructure, were estimated to be between $126 million and $192 million, equivalent to 1.5–2.7 percent of Malawi’s national GDP in 2020. Infrastructure damage was estimated to be between $57 million and $136 million. Cyclone Freddy resulted in significant economic disruption, with GDP growth losses estimated at 1.7% according to assessments by the World Bank (2017) and FEWS Net (2023). The cyclone’s direct economic losses were predominantly concentrated in physical infrastructure damage, with housing, power systems, and road networks accounting for approximately 60% of total direct losses, while agricultural and livestock sectors comprised around 35% of the damages, reflecting the cyclone’s comprehensive impact across both built and natural systems in southern Malawi.

Table 8

Table 8. Cyclone occurrence, impacts, and economic losses.

The study established recovery delays due to resource constraints. For example, affected communities by floods associated with Tropical Cyclone Idai was yet to recover due to financial challenges. Increased frequency and intensity of TCs, therefore, add more stress to already affected communities and the government.

3.3 Climate projections and future tropical cyclone activity

While many studies agree that the number of TCs globally might not significantly increase under a high emission scenario (Knutson et al., 2010; Camargo and Sobel, 2010). While many studies agree that the number of tropical cyclones globally might not significantly increase under a high emission scenario, there is consistent evidence that the likelihood of extremely intense cyclones could double by the end of the 21st century (Intergovernmental Panel on Climate Change, 2021), the trend in Figure 7 seems to project an eight-times increase in tropical cyclone events for the next 20 years. This includes an increase in severe tropical cyclones. This supports climatic projections and models, which suggest that the severity and frequency of climatic shocks will continue to increase (Future Climate for Africa, 2017). Many studies agree that the number of TCs may not increase significantly under a high-emission scenario, but the likelihood of extremely intense cyclones could double by the end of the 21st century (Knutson et al., 2010; Camargo and Sobel, 2010; Intergovernmental Panel on Climate Change, 2021).

The Intergovernmental Panel on Climate Change, in its Sixth Assessment Report, indicates that the risk of very intense storms and associated extreme rainfall is projected to increase in most regions, with considerable uncertainty remaining regarding their frequency (Intergovernmental Panel on Climate Change, 2021). Knutson et al. (2010), using climate models, predict an increase in the intensity and frequency of intense storms/TCs and extreme rainfall associated with TCs globally due to rising sea surface temperatures (SSTs), but they also emphasize that overall storm frequency may decrease or remain unchanged, with significant regional variability. Similarly, Murakami et al. (2019) analyzes regional projections and emphasizes regional differences in projected TC activity, with some areas experiencing increased storm intensity and rainfall, while others witness insignificant changes. For example, Li et al. (2019), in Nature Communications, highlighted that the Western Pacific and Indian Ocean regions could experience increased TC rainfall and intensity, while some Atlantic regions show more variable trends. These authors note high uncertainty due to complex interactions among the climate, ocean, and atmosphere (Murakami et al., 2020). Emanuel (2017), in assessing the present and future probability of Hurricane Harvey’s rainfall, discusses how rising SSTs and atmospheric moisture content enhance the energy available for storms, increasing their potential intensity and rainfall, consistent with climate change projections. The warming scenarios are associated with climate change. These studies collectively suggest that climate change is likely to intensify and increase rainfall associated with tropical cyclones globally, with regional variations and ongoing uncertainties. The strongest storms are projected to become more severe, and their tracks may shift poleward, affecting new regions. Kossin et al. (2018) highlight that the evidence of a poleward shift in the latitude of the most intense tropical cyclones suggests changes in storm tracks linked to climate change. This shift has implications for regions previously less affected by severe storms. Despite the limited availability of robust historical data for certain parameters that make the accurate forecasting of rainfall patterns and extreme events, the available evidence suggests a continued increase in the intensity and number of weather-related incidents (Niang et al., 2014). This means the pace of interventions to manage these events should be in line with such projections.

4 Conclusion

Tropical cyclones in southern Africa exhibit statistically significant increases in frequency of occurrence, intensity, and magnitude of losses and damages, as supported by trend analysis indicating substantial projected increases over the coming decades. Based on our statistical analysis using Mann-Kendall trend tests (Tau = 0.34, p < 0.05) and correlation assessments, these events have become almost annual occurrences with an increasing trend since the 2000s.

Based on established climate attribution literature and alignment with the Intergovernmental Panel on Climate Change’s findings, there are clear indications that climate change is partially contributing to the current and future trends of tropical cyclones. However, this study’s SPI-based methodology cannot establish direct causality without additional multivariate climate modeling. The management of losses and damages in affected sectors deserves special attention in loss and damage discussions under the UNFCCC.

Floods have caused substantial damage and losses in all sectors: the productive, public infrastructure and social service sectors, including private and community assets. Countries in Southern Africa have limited capacity to cover the losses and damages, evidenced by delayed recovery of past disasters associated with tropical cyclones. Attribution science and improved regional climate modeling are critical for informing the loss and damage fund and supporting vulnerable nations in adapting to increasing tropical cyclone impacts.

The SPI analysis indicates a rising trend in rain-induced flood years; however, direct attribution to climate change is reserved for interpretation guided by the Intergovernmental Panel on Climate Change reports, regional detection studies, and international climate databases. The conclusion emphasizes the importance of contextualized adaptation and resilience within the Loss and Damage framework, while fully acknowledging study limitations and the need for multi-variable analyses in future cyclone research. Further, the synthesis between qualitative and quantitative findings provides actionable recommendations for policy and community-level responses.

Data availability statement

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

Ethics statement

The studies involving humans were approved by Chinhoyi University of Technology. The studies were conducted in accordance with the local legislation and institutional requirements. The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants’ legal guardians/next of kin because part of the research on African ecosystem and human settlement in the face of cyclones, climate change, and the Sustainable Development Goals for which Ethical clearance was obtained.

Author contributions

MJ: Investigation, Writing – review & editing, Writing – original draft. RK: Writing – original draft, Visualization, Formal analysis, Writing – review & editing. GW: Writing – review & editing, Writing – original draft, Resources.

Funding

The author(s) declared that financial support was not received for this work and/or its publication.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/frwa.2025.1622293/full#supplementary-material

Footnotes

^One example of rapid and sudden events.

^The Mann-Kendall trend test was used to determine the statistical significance of the observed trends.

References

Action Aid (2006). Climate change, urban flooding and the rights of the poor in Africa: key findings from six African cities. London: Action Aid International.

Google Scholar

Australian Bureau of Meteorology (2024). Melbourne: Past tropical cyclone seasons.

Google Scholar

Bahadur, A., Ibrahim, M., and Tanner, T. (2013). Characterising resilience: unpacking the concept for tackling climate change and development. Climate and Development 5, 55–65. doi: 10.1080/17565529.2012.762334

Crossref Full Text | Google Scholar

Bryman, A. (2012). Social research methods. 4th Edn. Oxford: Oxford University Press.

Google Scholar

Camargo, S. J., and Sobel, A. H. (2010). Revisiting the influence of the quasi-biennial oscillation on tropical cyclone activity. J. Clim. 23, 5810–5825. doi: 10.1175/2010JCLI3575.1

Crossref Full Text | Google Scholar

Centre for Research on the Epidemiology of Disasters (CRED) (2012). The human cost of weather-related disasters 1995–2015. Brussels: UN Office for Disaster Risk Reduction (UNISDR).

Google Scholar

Christie, F., and Hanlon, J. (2001). Mozambique & the great flood of 2000. Oxford: Indiana University Press.

Google Scholar

Elsner, J. B., Kossin, J. P., and Jagger, T. H. (2008). The increasing intensity of the strongest tropical cyclones. Nature 455, 92–95. doi: 10.1038/nature07234,

PubMed Abstract | Crossref Full Text | Google Scholar

Emanuel, K. (2017). Assessing the present and future probability of hurricane Harvey's rainfall. Annu. Rev. Mar. Sci. 9, 49–70. doi: 10.1073/pnas.1716222114

Crossref Full Text | Google Scholar

Erlingsson, C., and Brysiewicz, P. (2017). A hands-on guide to doing content analysis. African J. Emerg. Med. 7, 93–99. doi: 10.1016/j.afjem.2017.08.001,

PubMed Abstract | Crossref Full Text | Google Scholar

FEWS Net (2023). Malawi key message update, March 2023: in the aftermath of tropical cyclone Freddy, crisis (IPC phase 3) will likely persist in southern Malawi. Washington, D.C.

Google Scholar

Future Climate for Africa (2017). Africa’s climate: helping decision-makers make sense of climate information [policy brief] Cape Town.

Google Scholar

Government of Malawi. (2016). National climate change management policy. Lilongwe, Malawi: Ministry of Natural Resources, Energy and Mining, Environmental Affairs Department. Available online at: https://faolex.fao.org/docs/pdf/mlw201716.pdf

Google Scholar

Government of Malawi. (2018/2019). Department of Disaster Management Affairs. Lilongwe, Malawi: National Multi-Hazard Contingency Plan 2019–2020. Available online at: https://drmims.sadc.int/en/documents/database/national-multi-hazard-contingency-plan-2019-2020

Google Scholar

Government of Malawi. (2023). Department of Disaster Management Affairs. Tropical Cyclone Freddy. Lilongwe, Malawi: Emergency Response Plan, March 2023 – Malawi. Available online at: https://reliefweb.int/report/malawi/tropical-cyclone-freddy-emergency-response-plan-office-president-and-cabinet-department-disa

Google Scholar

Government of Malawi. (2022). Tropical Cyclone Ana Post-Disaster Needs Assessment (PDNA). Lilongwe, Malawi.

Google Scholar

Government of Malawi (2011). The second national communication of the republic of Malawi to the conference of the parties (COP) of the United Nations framework convention on climate change (UNFCCC). Lilongwe: Ministry of Natural Resources, Energy and Environment.

Google Scholar

Government of Malawi (2013). National disaster risk management policy. Lilongwe Department of Disaster Management Affairs (DoDMA).

Google Scholar

Government of Malawi (2015). Malawi 2015 floods post disaster needs assessment report. Government of Malawi, with support from the European Union, United Nations, and World Bank. Available online at: http://reliefweb.int/sites/reliefweb.int/files/resources/Malawi-2015-Floods-Post-Disaster-Needs-Assessment-Report.pdf (Accessed December 14, 2024).

Google Scholar

Government of Malawi (2017). Strategic program for climate resilience (SPCR). Lilongwe : Ministry of Finance, Economic Planning and Development.

Google Scholar

Government of Malawi (2019). Malawi 2019 floods post disaster needs assessment report. Malawi: Department of Disaster Management Affairs.

Google Scholar

Government of Malawi (2023a). Tropical cyclone Freddy consolidated Department of Disaster Management Affairs (DoDMA) assessment report. Lilongwe: Department of Disaster Management Affairs.

Google Scholar

Government of Malawi, Climate Change and Meteorological Services Department. (2023a). Tropical cyclone data for Malawi (1950-2023). Department of Climate Change and Meteorological Services.

Google Scholar

Government of Malawi, Climate Change and Meteorological Services Department. (2023b). Climate data for Southern Malawi (1990-2023). Department of Climate Change and Meteorological Services.

Google Scholar

Government of Malawi (2023b). Tropical cyclone Freddy emergency response plan March 2023. Lilongwe: Department of Disaster Management Affairs.

Google Scholar

Hisdal, H., and Tallaksen, L. M. (2000). Drought event definition. Technical Report No. 6, ARIDE. Oslo: University of Oslo.

Google Scholar

Holmes, T., Zamani, S., Tolk, L., and van Huijgevoort, M. (2017). Climate change and flood risk in Malawi: a case study. The Hague: Red Cross Red Crescent Climate Centre.

Google Scholar

Intergovernmental Panel on Climate Change (2013). “Climate change 2013: the physical science basis” in Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change (Cambridge: Cambridge University Press).

Google Scholar

Intergovernmental Panel on Climate Change. (2021). Climate change 2021. Cambridge: the physical science basis.

Google Scholar

Intergovernmental Panel on Climate Change (2022). “Climate change 2022: impacts, adaptation, and vulnerability” in Contribution of working group II to the sixth assessment report of the intergovernmental panel on climate change. (Cambridge: Cambridge University Press).

Google Scholar

Joshua, M. K., Chirwa, R. K., Ngongondo, C., Monjerezi, M., Mwathunga, E., and Kasei, R. (2021b). “Impact of floods on access to drinking water: a focus on 2019 floods in Magalasi Village in Chikwawa District, Malawi” in The increasing risk of floods and tornadoes in southern Africa (Cham. Cambridge Springer International Publishing), 191–201.

Google Scholar

Joshua, M. D. K., Stathers, T., Chirwa, R. K., Ngongondo, C., Lamboll, R., Monjerezi, M., et al. (2021a). “A comparative study of the impacts of flooding on food security of urban and rural households in Blantyre City and Chikwawa, Malawi” in Cyclones in southern Africa: Volume 3. Implications for the sustainable development goals. eds. G. Nhamo and D. Chikodzi (Cham: Springer International Publishing), 35–58.

Google Scholar

Kasei, R., Diekkruger, B., and Leemhuis, C. (2010). Drought frequency in the Volta Basin of West Africa. Sustain. Sci. 5, 89–97. doi: 10.1007/s11625-009-0101-5

Crossref Full Text | Google Scholar

Knutson, T. R., Camargo, S. J., Chan, J. C. L., Emanuel, K., Ho, C. H., Kossin, J., et al. (2019). Tropical cyclones and climate change assessment: part 1. Bull. Am. Meteorol. Soc. 100, 1987–2007. doi: 10.1175/BAMS-D-18-0189.1

Crossref Full Text | Google Scholar

Knutson, T. R., Camargo, S. J., Chan, J. C. L., Emanuel, K., Ho, C. H., Kossin, J., et al. (2020). Tropical cyclones and climate change assessment: a review of the scientific literature. Nat. Rev. Earth Environ. 1, 549–570. doi: 10.1038/s43017-020-0010-0

Crossref Full Text | Google Scholar

Knutson, T. R., McBride, J. L., Chan, J., Emanuel, K., Holland, G., Landsea, C., et al. (2010). Tropical cyclones and climate change. Bull. Am. Meteorol. Soc. 91, 501–514. doi: 10.1038/ngeo779

Crossref Full Text | Google Scholar

Kooshki Forooshani, M., van den Homberg, M., Kalimeri, K., Kaltenbrunner, A., Mejova, Y., Milano, L., et al. (2024). Towards a global impact-based forecasting model for tropical cyclones. Nat. Hazards Earth Syst. Sci. 24, 309–329. doi: 10.5194/nhess-24-309-2024

Crossref Full Text | Google Scholar

Kossin, J. P., Emanuel, K. A., and Camargo, S. J. (2020a). The poleward migration of tropical cyclones in the northern hemisphere. Nature 585, 391–397. doi: 10.1038/s41586-020-2706-0

Crossref Full Text | Google Scholar

Kossin, J. P. (2018). A global slowdown of tropical-cyclone translation speed. Nature 558, 104–108. doi: 10.1038/s41586-018-0158-3

Crossref Full Text | Google Scholar

Kossin, J. P., Knapp, K. R., Olander, T. L., and Velden, C. S. (2020b). Global increase in major tropical cyclone exceedance probability over the past four decades. Proc. Natl. Acad. Sci. 117, 11975–11980. doi: 10.1073/pnas.1920849117,

PubMed Abstract | Crossref Full Text | Google Scholar

Leroux, C., Reason, C. J. C., and Van Heerden, J. (2017). Trends in tropical cyclone activity in the South Indian Ocean from 1950 to 2014. Clim. Res. 70, 99–102. doi: 10.3354/cr01430,

PubMed Abstract | Crossref Full Text | Google Scholar

Li, L., Yao, N., Li Liu, D., Song, S., Lin, H., Chen, X., et al. (2019). Historical and future projected frequency of extreme precipitation indicators using the optimized cumulative distribution functions in China. J. Hydrol. 579:124170. doi: 10.1016/j.jhydrol.2019.124170

Crossref Full Text | Google Scholar

Marks, F. D. (2003). “Hurricanes: observation” in Encyclopedia of atmospheric sciences. eds. J. R. Holton, J. Pyle, and J. A. Curry, vol. 6. 2nd ed (San Diego: Elsevier).

Google Scholar

McKee, T. B., Doesken, N. J., and Kleist, J. (1993). “The relationship of drought frequency and duration to time scales [conference presentation]” in 8th conference on applied climatology (Anaheim, CA, United States: American Meteorological Society).

Google Scholar

Murakami, H., Levin, E., Delworth, T. L., Gudgel, R., and Hsu, P.-C. (2019). Dominant role of subtropical Pacific warming in extreme eastern Pacific hurricane seasons: 2015 and 2018. J. Clim. 33, 1299–1315. doi: 10.1175/JCLI-D-19-0332.1,

PubMed Abstract | Crossref Full Text | Google Scholar

Murakami, H., Vecchi, G. A., Underwood, S., Delworth, T. L., Wittenberg, A. T., Anderson, W. G., et al. (2020). Detected climatic change in global distribution of tropical cyclones. Proceedings of the National Academy of Sciences 117, 10706–10714. doi: 10.1073/pnas.1922500117

Crossref Full Text | Google Scholar

National Oceanic and Atmospheric Administration (NOAA). (2023). El Niño & La Niña (El Niño-southern oscillation). Available online at: https://www.climate.gov/enso (Accessed December 14, 2024).

Google Scholar

Navarro, A., and Merino, A. (2022). “Precipitation in earth system models: advances and limitations” in Precipitation science. ed. S. Michaelides (Amsterdam: Elsevier), 637–659.

Google Scholar

Nazla, B., and Rohli, R. V. (2021). “Tracks of severe coastal and marine tropical cyclones” in Annotated atlas of coastal and marine winds. eds. B. Nazla and R. V. Rohli (Amsterdam: Elsevier), 127–153.

Google Scholar

Ngongondo, C. S., Xu, C. Y., Tallaksen, L. M., Alemaw, B., and Chirwa, T. (2011). Regional frequency analysis of rainfall extremes in southern Malawi using the index rainfall and L-moments approaches. Stoch. Environ. Res. Risk Assess. 25, 939–955. doi: 10.1007/s00477-011-0480-x

Crossref Full Text | Google Scholar

Niang, I., Ruppel, O. C., Abdrabo, M. A., Essel, A., Lennard, C., Padgham, J., et al. (2014). “Africa” in Climate change 2014: impacts, adaptation, and vulnerability. Part B: regional aspects. eds. V. R. Barros, et al. (Cambridge: Cambridge University Press), 1199–1265.

Google Scholar

Nilsson, M., Lucas, P., and Yoshida, T. (2010). Public sector capacity for disaster risk reduction in Malawi. Stockholm: Stockholm Environment Institute.

Google Scholar

Nkomwa, E. C., Kalanda Joshua, M., Ngongondo, C., Monjerezi, M., and Chipungu, F. (2013). Assessing indigenous knowledge systems and climate change adaptation strategies in agriculture: a case study of Changaka Village, Chikhwawa, southern Malawi. Phys. Chem. Earth Parts A/B/C 67–69, 164–172. doi: 10.1016/j.pce.2013.10.002

Crossref Full Text | Google Scholar

Nyumba, T. O., Wilson, K., Derrick, C. J., and Mukherjee, N. (2018). The use of focus group discussion methodology: insights from two decades of application in conservation. MethodsX 5, 20–32. doi: 10.1111/2041-210X.12860

Crossref Full Text | Google Scholar

Olaoluwa, E. E., Durowoju, O. S., Orimoloye, I. R., Daramola, M. T., Ayobami, A. A., and Olorunsaye, O. (2022). “Understanding weather and climate extremes” in Climate impacts on extreme weather (Amsterdam: Elsevier), 1–17.

Google Scholar

Pourazar, E. (2017). Assessing the effectiveness of disaster risk management in Malawi. Johannesburg: South African Institute of International Affairs.

Google Scholar

Qi, J., Dazé, A., and Hammill, A. (2023). Addressing loss and damage: what can we learn from countries' National Adaptation Plans? NAP Global Network/International Institute for Sustainable Development. Available online at: https://napglobalnetwork.org/resource/loss-and-damage-national-adaptation-plans/ (Accessed December 14, 2024).

Google Scholar

Reason, C. J. C., Landman, W., and Tennant, W. (2018). Climate change impacts on tropical cyclones in the South Indian Ocean. Reg. Environ. Chang. 18, 1097–1108. doi: 10.1007/s10113-018-1373-8

Crossref Full Text | Google Scholar

Reason, C. J. C., and Rouault, M. (2020). Tropical cyclones in the South Indian and South Pacific oceans. Clim. Dyn. 55, 693–708. doi: 10.1007/s00382-020-05284-0

Crossref Full Text | Google Scholar

Reason, C. J. C., and Van Heerden, J. (2012). A model for the genesis of tropical cyclones over the South Indian Ocean. Clim. Dyn. 39, 1573–1588. doi: 10.1007/s00382-012-1299-4

Crossref Full Text | Google Scholar

Shepherd, T. G. (2019). The role of climate science in climate policy. Nat. Clim. Chang. 9, 312–319. doi: 10.1098/rspa.2019.0013

Crossref Full Text | Google Scholar

Stewart, D. W., and Shamdasani, P. N. (1998). Focus groups: theory and practice. Thousand Oaks, CA: SAGE Publications.

Google Scholar

Stott, P. A., Christidis, N., Otto, F. E. L., Sun, Y., Vanderlinden, J. P., van Oldenborgh, G. J., et al. (2016). Attribution of extreme weather and climate-related events. Wiley Interdiscip. Rev. Clim. Chang. 7, 23–41. doi: 10.1002/wcc.377,

PubMed Abstract | Crossref Full Text | Google Scholar

Telesetsky, A. (2021). “Loss and damage, disaster law, and climate change” in Research handbook on climate change law and loss & damage (Cheltenham: Edward Elgar Publishing), 267–286.

Google Scholar

Titley, D., Hegerl, G., Jacobs, K., Mote, P. W., Paciorek, C. J., Shepherd, J. M., et al. (2016). Attribution of extreme weather events in the context of climate change. Washington, D.C. National Academies.

Google Scholar

Trigo, R. M., and Gimeno, L. (2009). “Weather pattern changes in the tropics and mid-latitudes as an indicator of global changes” in Climate change. ed. T. M. Letcher (Amsterdam: Elsevier), 165–180.

Google Scholar

UNECA (2015). Assessment of the impact of the 2015 floods on Malawi's economy. Addis Ababa: United Nations Economic Commission for Africa.

Google Scholar

UNFCCC (1992). United Nations framework convention on climate change. Bonn, New York: United Nations.

Google Scholar

UNFCCC. (2013a). Synthesis of submissions on the costs, benefits and opportunities for adaptation based on different drivers of climate change impacts, including the relationship between adaptation and mitigation (FCCC/TP/2013/10). Bonn.

Google Scholar

UNFCCC (2013b). Warsaw international mechanism for loss and damage associated with climate change impacts (decision 2/CP.19). Bonn: United Nations Framework Convention on Climate Change.

Google Scholar

UNFCCC (2015). “The Paris agreement (article 8 — loss and damage)” in Conference of the parties, 21st session, decision 1/CP.21, annex. (Bonn: United Nations Framework Convention on Climate Change).

Google Scholar

UNFCCC (2023). Report of the executive Committee of the Warsaw International Mechanism for loss and damage associated with climate change impacts. United Nations Framework Convention on Climate Change. Available online at: https://unfccc.int/documents/636597 (Accessed December 14, 2024).

Google Scholar

United Nations (2024a). Lilongwe declaration on climate change. LDC Portal. Available online at: https://www.un.org/ldcportal/content/lilongwe-declaration-climate-change

Google Scholar

United Nations (2024b). “The Lilongwe declaration on climate change” in 46th meeting of the least developed countries expert group.

Google Scholar

Van Oldenborgh, G. J., Van Der Wiel, K., Sebastian, A., Singh, R., Arrighi, J., Otto, F., et al. (2017). Attribution of extreme rainfall from hurricane Harvey, August 2017. Environ. Res. Lett. 12:124009. doi: 10.1088/1748-9326/aa9ef2

Crossref Full Text | Google Scholar

Wang, Z. (2015). “Tropical cyclones and hurricanes: tropical cyclogenesis” in Encyclopedia of atmospheric sciences. eds. G. R. North, J. Pyle, and F. Zhang. 2nd ed (Oxford: Academic Press), 57–64.

Google Scholar

Webster, P. J., Holland, G. J., Curry, J. A., and Chang, H. R. (2005). Changes in tropical cyclone number, duration, and intensity in a warming climate. Science 309, 1844–1846. doi: 10.1126/science.1116448,

PubMed Abstract | Crossref Full Text | Google Scholar

Winsemius, H. C., Aerts, J. C. J. H., van Beek, L. P. H., Bierkens, M. F. P., Bouwman, A., Jongman, B., et al. (2015). Global drivers of future river flood risk. Nat. Clim. Chang. 6, 381–385. doi: 10.1038/nclimate2893,

PubMed Abstract | Crossref Full Text | Google Scholar

World Bank (2017). Malawi economic monitor, May 2017: harnessing the urban dividend. Washington, D.C. World Bank Group.

Google Scholar

World Data (2023). Tropical cyclones. Available online at: https://www.worlddata.info/cyclones.php (Accessed December 14, 2024).

Google Scholar