How to Reduce Africa’s Undue Exposure to Climate Risks

Climate and weather-related disasters, including tropical cyclones, storm surges, floods and droughts, are on the rise. Over the past 50 years, rates have increased fivefold globally, and the damages associated with them have swelled by 70 times1,2. This will only get worse as climate change increases the frequency and intensity of extreme weather. And some places are feeling the brunt much more than others — notably Africa.

In the Horn of Africa, since 2020, the most prolonged drought in the region’s recent history is continuing to cripple communities in Ethiopia, Somalia and Kenya, with more than 20 million people experiencing malnutrition. In 2017, flood-driven landslides in Sierra Leone claimed the lives of more than 1,000 people and left thousands homeless, many to this day. Torrential rains that hit Somalia in 1997 caused more than 2,000 deaths and displaced 250,000 people. The Ethiopian drought of 1983–84, which caused at least 300,000 casualties and wrecked millions of lives, is considered one of the worst hydrometeorological disasters of the past 50 years.

In such experiences, Africa stands out from other regions. During 2000–22, it had more floods and flood-related deaths than did North America and Europe combined (see ‘Africa’s disproportionate toll’). Only South and southeast Asia experienced more, owing in part to dense populations in flood-prone countries such as low-lying Bangladesh. With Africa hosting many low- and middle-income nations, it faces its own set of challenges, notably a lack of disaster preparedness and response.

To see why this matters, compare the effects of two similar storms: Intense Tropical Cyclone Idai, which wreaked havoc across Madagascar, Malawi, Mozambique and Zimbabwe in 2019; and Hurricane Ida, which swept across the eastern United States in 2021. Both were category 4 events with wind speeds of more than 200 kilometres per hour — in fact, Ida was slightly more intense, with higher maximum winds. Whereas US residents were alerted to evacuate before Hurricane Ida made landfall, Cyclone Idai caught African populations by surprise. The death toll in Africa was more than 1,000, compared with just under 100 in the United States.

Preparedness and response capabilities are fundamental to managing hydrological and meteorological (or hydromet) risks. They rely on multi-layered hydromet systems — consisting of weather monitoring, forecasting and nowcasting (local forecasts up to 6 hours in advance) and early warnings. Crucially, each layer cannot function without the preceding one. Whereas hydromet systems have been taken for granted by wealthy countries in the global north for decades, these layers are missing, outmoded or malfunctioning across Africa — more so than in any other global region.

The radar database run by the World Meteorological Organization (WMO) depicts a stark contrast in monitoring capacity, for example (see go.nature.com/3qufhdj). In Europe and the United States, there are 636 radar stations for a population of 1.1 billion and a landmass of 20 million square kilometres. In Africa, there are just 37 stations for a comparable population of 1.2 billion and a greater landmass of 30 million km2 (see ‘Weather station deficit’). Of the weather stations that do exist across Africa, more than 50% do not provide data that are accurate enough to enable forecasting and nowcasting.

Until recently, this deficiency has not been prioritized by intergovernmental agencies. The Intergovernmental Panel on Climate Change, for example, failed to recognize these gaps in its Sixth Assessment Report. A glimmer of hope came from the Early Warnings for All initiative (go.nature.com/3trrsvt), led by the WMO and the United Nations, which in November 2022 pledged US$3.1 billion for hydromet systems in 30 countries, including 13 African nations.

This is a good start, but it remains lacklustre. The United States, for example, has in the past 30 years installed, maintained and modernized 122 next-generation weather radar stations at a cost of $3.1 billion — equivalent to the whole Early Warnings for All pledge. Africa is more than three times the area of the United States, so from a purely geographical perspective, the pledge should be tripled. And the infrastructure deployment and training priorities need to be much better defined.

Globally, disaster-risk reduction, too, is vastly underfunded. In 2020, total international aid amounted to $194 billion, with $95 billion targeted at Africa. For every $100 of such aid spent between 2010–18, just $0.47 was allocated to disaster-risk reduction (see go.nature.com/3dtavvh). Importantly, gaps in hydromet systems often render other development investments redundant: there is no point investing in smallholder farms, for example, if floods are simply going to wash them away.

This underfunding is unacceptable, given the potential savings that could come from investment. Climate change is expected to cost the continent more than $50 billion annually by 2050. Meanwhile, hydromet systems are expected to save African countries $13 billion in asset losses and $22 billion in livelihood losses per year (see go.nature.com/3fw6bdp). The Early Warnings for All pledge is too little, too late. African countries deserve a more nuanced and comprehensive road map. Here we provide five priorities for such work.

Identify the most at-risk

Risk is best thought of as a function of the magnitude and location of a hazard, a given population’s exposure to that hazard and its intrinsic vulnerability (its degree of defencelessness) — such as those caused by socioeconomic factors. This approach rightly underpins the Australian Government’s framework for climate and disaster-risk reduction, for example.

Some Western narratives tend to portray Africa as a single homogeneous hydromet unit. It is not. There are notable differences in risk across the continent, and these are expected to change as time goes on. For example, in terms of future climate hazards, East Africa is expected to experience warmer temperatures and more-protracted droughts; the Niger River Delta in western Africa is expected to experience delays in the onset of the wet season; and both the frequency and intensity of cyclones is expected to increase south of Madagascar.

In terms of vulnerability, at present, the Sahel region (the semiarid part of western and north-central Africa) hosts many of the continent’s most vulnerable citizens, with populations already distressed and displaced by disease, conflicts and climate change. In future, vulnerability is expected to increase in the south and on the west coast.

Maps of hyperlocal risk, evaluated as a function of hazard, exposure and vulnerability, should be used to prioritize action. Risk maps have been produced for Africa, but these are not nearly detailed enough. Methodical examinations to identify the highest-risk populations are currently lacking in the literature3. The research community needs to help assess exposure and vulnerabilities to risks at greater granularity and in various climate-change scenarios. For example, researchers could identify farming communities that rely on rain-fed agriculture and might face greater exposure to hydromet hazards and rainfall variability.

Research funding should address these gaps in data collection and analysis, and data must be publicly available, easily accessible and shared with African governments and intergovernmental organizations. One example is the Global Flood Partnership, an interdisciplinary team of scientists and risk managers established in 2014 to develop flood monitoring, forecasting and impact-assessment tools, including hazard maps (www.globalfloodpartnership.org). It has helped to strengthen preparedness and response; more work such as this can and should be done across other kinds of hazard.

Invest in weather monitoring and forecasting

There is a crucial need to boost the number of weather stations in Africa, in particular those with weather-surveillance radars that have dual-polarization capabilities. These can detect storm clouds, evaluate storm structures and help to forecast severe weather. More will be needed. These radars can ‘see’ only about 1,800–3,000 metres above ground, and cannot detect lower convective systems, including tornadoes, which are strongest below about 1,500 metres. Radar can also be disrupted by mountainous terrain, and these systems are expensive: a single weather radar station can cost more than $1 million.

To fill gaps in radar coverage, ground observations can be used. Software applications such as the Meteorological Phenomena Identification near the Ground (mPING) project run by the US National Oceanic and Atmospheric Administration could be extended to Africa to crowdsource weather reports from trained weather spotters and untrained local citizens. Yet, such observations provide only partial coverage and are limited during weather-related disasters, when people on the ground are at risk and telecommunications are disrupted.

Satellite-based weather monitoring and forecasting is also crucial and must be improved. Indeed, unlocking the potential of existing satellite coverage could enable Africa to leapfrog capital-intensive ‘bricks and mortar’ infrastructure approaches such as radar stations. Most national hydromet services in Africa already have access to satellite imagery through collaboration with the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT). Some of these data are less granular for Africa than for the United States and Europe. But the more important problem is that there aren’t enough skilled meteorologists in African hydromet services to interpret this satellite imagery for forecasting.

In addition, global ground-based lightning detection networks (such as GLD360) already offer good data coverage of Africa, and can and should be used to detect markers of patches of severe weather.

This all requires significant investment in local infrastructure and upskilling. Policymakers will have to seek out financing solutions and funds, such as the WMO’s Systematic Observations Financing Facility and the Africa Climate Change Fund.

Promote computational nowcasting

Monitoring and forecasting are not enough on their own. A sophisticated form of forecasting called numerical weather prediction, which involves real-time weather simulation and computer modelling, is needed to get good information with a lead time of 24 hours or less. Computational nowcasting is necessary to pick up extreme weather with local detailed forecasts of just minutes to a few hours, and is crucial when it comes to tracking unfolding disasters.

For example, despite increasing weather monitoring on Africa’s Lake Victoria, without advanced nowcasting, thousands of people still die each year from thunderstorms that capsize fishing vessels4.

Climate and weather-related disasters, including tropical cyclones, storm surges, floods and droughts, are on the rise. Over the past 50 years, rates have increased fivefold globally, and the damages associated with them have swelled by 70 times1,2. This will only get worse as climate change increases the frequency and intensity of extreme weather. And some places are feeling the brunt much more than others — notably Africa.

In the Horn of Africa, since 2020, the most prolonged drought in the region’s recent history is continuing to cripple communities in Ethiopia, Somalia and Kenya, with more than 20 million people experiencing malnutrition. In 2017, flood-driven landslides in Sierra Leone claimed the lives of more than 1,000 people and left thousands homeless, many to this day. Torrential rains that hit Somalia in 1997 caused more than 2,000 deaths and displaced 250,000 people. The Ethiopian drought of 1983–84, which caused at least 300,000 casualties and wrecked millions of lives, is considered one of the worst hydrometeorological disasters of the past 50 years.

In such experiences, Africa stands out from other regions. During 2000–22, it had more floods and flood-related deaths than did North America and Europe combined (see ‘Africa’s disproportionate toll’). Only South and southeast Asia experienced more, owing in part to dense populations in flood-prone countries such as low-lying Bangladesh. With Africa hosting many low- and middle-income nations, it faces its own set of challenges, notably a lack of disaster preparedness and response.

To see why this matters, compare the effects of two similar storms: Intense Tropical Cyclone Idai, which wreaked havoc across Madagascar, Malawi, Mozambique and Zimbabwe in 2019; and Hurricane Ida, which swept across the eastern United States in 2021. Both were category 4 events with wind speeds of more than 200 kilometres per hour — in fact, Ida was slightly more intense, with higher maximum winds. Whereas US residents were alerted to evacuate before Hurricane Ida made landfall, Cyclone Idai caught African populations by surprise. The death toll in Africa was more than 1,000, compared with just under 100 in the United States.

Preparedness and response capabilities are fundamental to managing hydrological and meteorological (or hydromet) risks. They rely on multi-layered hydromet systems — consisting of weather monitoring, forecasting and nowcasting (local forecasts up to 6 hours in advance) and early warnings. Crucially, each layer cannot function without the preceding one. Whereas hydromet systems have been taken for granted by wealthy countries in the global north for decades, these layers are missing, outmoded or malfunctioning across Africa — more so than in any other global region.

The radar database run by the World Meteorological Organization (WMO) depicts a stark contrast in monitoring capacity, for example (see go.nature.com/3qufhdj). In Europe and the United States, there are 636 radar stations for a population of 1.1 billion and a landmass of 20 million square kilometres. In Africa, there are just 37 stations for a comparable population of 1.2 billion and a greater landmass of 30 million km2 (see ‘Weather station deficit’). Of the weather stations that do exist across Africa, more than 50% do not provide data that are accurate enough to enable forecasting and nowcasting.

Until recently, this deficiency has not been prioritized by intergovernmental agencies. The Intergovernmental Panel on Climate Change, for example, failed to recognize these gaps in its Sixth Assessment Report. A glimmer of hope came from the Early Warnings for All initiative (go.nature.com/3trrsvt), led by the WMO and the United Nations, which in November 2022 pledged US$3.1 billion for hydromet systems in 30 countries, including 13 African nations.

This is a good start, but it remains lacklustre. The United States, for example, has in the past 30 years installed, maintained and modernized 122 next-generation weather radar stations at a cost of $3.1 billion — equivalent to the whole Early Warnings for All pledge. Africa is more than three times the area of the United States, so from a purely geographical perspective, the pledge should be tripled. And the infrastructure deployment and training priorities need to be much better defined.

Three climate policies that the G7 must adopt — for itself and the wider world

Globally, disaster-risk reduction, too, is vastly underfunded. In 2020, total international aid amounted to $194 billion, with $95 billion targeted at Africa. For every $100 of such aid spent between 2010–18, just $0.47 was allocated to disaster-risk reduction (see go.nature.com/3dtavvh). Importantly, gaps in hydromet systems often render other development investments redundant: there is no point investing in smallholder farms, for example, if floods are simply going to wash them away.

This underfunding is unacceptable, given the potential savings that could come from investment. Climate change is expected to cost the continent more than $50 billion annually by 2050. Meanwhile, hydromet systems are expected to save African countries $13 billion in asset losses and $22 billion in livelihood losses per year (see go.nature.com/3fw6bdp). The Early Warnings for All pledge is too little, too late. African countries deserve a more nuanced and comprehensive road map. Here we provide five priorities for such work.

Identify the most at-risk

Risk is best thought of as a function of the magnitude and location of a hazard, a given population’s exposure to that hazard and its intrinsic vulnerability (its degree of defencelessness) — such as those caused by socioeconomic factors. This approach rightly underpins the Australian Government’s framework for climate and disaster-risk reduction, for example.

Some Western narratives tend to portray Africa as a single homogeneous hydromet unit. It is not. There are notable differences in risk across the continent, and these are expected to change as time goes on. For example, in terms of future climate hazards, East Africa is expected to experience warmer temperatures and more-protracted droughts; the Niger River Delta in western Africa is expected to experience delays in the onset of the wet season; and both the frequency and intensity of cyclones is expected to increase south of Madagascar.

In terms of vulnerability, at present, the Sahel region (the semiarid part of western and north-central Africa) hosts many of the continent’s most vulnerable citizens, with populations already distressed and displaced by disease, conflicts and climate change. In future, vulnerability is expected to increase in the south and on the west coast.

Maps of hyperlocal risk, evaluated as a function of hazard, exposure and vulnerability, should be used to prioritize action. Risk maps have been produced for Africa, but these are not nearly detailed enough. Methodical examinations to identify the highest-risk populations are currently lacking in the literature3. The research community needs to help assess exposure and vulnerabilities to risks at greater granularity and in various climate-change scenarios. For example, researchers could identify farming communities that rely on rain-fed agriculture and might face greater exposure to hydromet hazards and rainfall variability.

Research funding should address these gaps in data collection and analysis, and data must be publicly available, easily accessible and shared with African governments and intergovernmental organizations. One example is the Global Flood Partnership, an interdisciplinary team of scientists and risk managers established in 2014 to develop flood monitoring, forecasting and impact-assessment tools, including hazard maps (www.globalfloodpartnership.org). It has helped to strengthen preparedness and response; more work such as this can and should be done across other kinds of hazard.

Invest in weather monitoring and forecasting

There is a crucial need to boost the number of weather stations in Africa, in particular those with weather-surveillance radars that have dual-polarization capabilities. These can detect storm clouds, evaluate storm structures and help to forecast severe weather. More will be needed. These radars can ‘see’ only about 1,800–3,000 metres above ground, and cannot detect lower convective systems, including tornadoes, which are strongest below about 1,500 metres. Radar can also be disrupted by mountainous terrain, and these systems are expensive: a single weather radar station can cost more than $1 million.

Why we need a new economics of water as a common good

To fill gaps in radar coverage, ground observations can be used. Software applications such as the Meteorological Phenomena Identification near the Ground (mPING) project run by the US National Oceanic and Atmospheric Administration could be extended to Africa to crowdsource weather reports from trained weather spotters and untrained local citizens. Yet, such observations provide only partial coverage and are limited during weather-related disasters, when people on the ground are at risk and telecommunications are disrupted.

Satellite-based weather monitoring and forecasting is also crucial and must be improved. Indeed, unlocking the potential of existing satellite coverage could enable Africa to leapfrog capital-intensive ‘bricks and mortar’ infrastructure approaches such as radar stations. Most national hydromet services in Africa already have access to satellite imagery through collaboration with the European Organisation for the Exploitation of Meteorological Satellites (EUMETSAT). Some of these data are less granular for Africa than for the United States and Europe. But the more important problem is that there aren’t enough skilled meteorologists in African hydromet services to interpret this satellite imagery for forecasting.

In addition, global ground-based lightning detection networks (such as GLD360) already offer good data coverage of Africa, and can and should be used to detect markers of patches of severe weather.

This all requires significant investment in local infrastructure and upskilling. Policymakers will have to seek out financing solutions and funds, such as the WMO’s Systematic Observations Financing Facility and the Africa Climate Change Fund.

Promote computational nowcasting

Monitoring and forecasting are not enough on their own. A sophisticated form of forecasting called numerical weather prediction, which involves real-time weather simulation and computer modelling, is needed to get good information with a lead time of 24 hours or less. Computational nowcasting is necessary to pick up extreme weather with local detailed forecasts of just minutes to a few hours, and is crucial when it comes to tracking unfolding disasters.

For example, despite increasing weather monitoring on Africa’s Lake Victoria, without advanced nowcasting, thousands of people still die each year from thunderstorms that capsize fishing vessels4.

All African national hydromet services should have access to basic, short-range numerical weather prediction models, such as EUMETSAT’s Nowcasting and Very Short-Range Forecasting software packages (called the Nowcasting Satellite Application Facility, NWC SAF). The £9-million (US$11-million) African Science for Weather Information and Forecasting Techniques (SWIFT) project, funded by the Global Challenges Research Fund (GCRF) and led by the University of Leeds, UK, in partnership with the WMO, made a good start. From 2017 to 2022, a team of UK and African atmospheric scientists, social scientists and operational forecasters (including M.G., P.N., J.G.O. and A.T.G.) worked together to understand how best to utilize NWC SAF for users in Senegal, Ghana, Nigeria and Kenya. More work such as this is sorely needed.

The advent of artificial intelligence (AI) and machine learning could advance the automation and speed of meteorological nowcasting. Furthermore, it could provide real-time predictions of disaster dynamics and magnitude by coupling weather-related data with cellular location services and social-media activity. Globally, AI-based nowcasting has not been deployed at scale, but AI for Good initiatives (https://ai4good.org) could advance these capabilities, including in Africa.

As for weather monitoring, this requires money to develop infrastructure and skills — particularly for computational capacity, Internet bandwidth and mobile network coverage, as well as training in advanced nowcasting software. Funds that address digital divides, such as the Africa Digital Rights Fund, could offer stop-gap financing.

Improve early warnings

Several African nations have early-warning systems: good services are provided, for example, in Ghana, Kenya, Rwanda, Uganda and South Africa. But there is much room for improvement. To be effective, warnings must state not just what the weather will be, but also what it will do to local populations, and provide clear direction if action is required. People need to know whether they should evacuate, and if so, they need clearly defined evacuation routes, well-built shelters and equipped emergency staff.

Communications must use local and understandable language and dialects, particularly given the varied literacy rates in Africa. The Forecasting African Storms Application (FASTA), which is being used in Kenya, was developed by GCRF African SWIFT to translate nowcasting outputs into consumer-friendly weather updates. It should be rolled out to other countries.

Warnings can be transmitted by multiple channels, including radio, sirens and social media, to use existing infrastructures for maximum reach and redundancy. With 82% of Africa’s population having access to a mobile broadband network, text messaging is a powerful means of delivering targeted warnings5, yet no African country currently does this inclusively and consistently. This is a missed opportunity.

Early-warning services need regular community engagement, practice and consistent evaluation and improvement. Short-term financing assistance could come through initiatives such as the Climate Risk and Early Warning Systems Initiative (CREWS), which has distributed more than $8 million to Burkina Faso, Cameroon, Chad, Mali, Niger and Nigeria to strengthen their early-warning services.

Earmark money for modernization, maintenance and security

Once established, multi-tiered hydromet systems require maintenance to remain functional and modern. The World Bank has estimated that some $1.5 billion is needed annually for the modernization and maintenance of a fully fledged hydromet system. But this expense is ten times smaller than the tens of billions of annual asset and job losses from extreme weather. A budget must also be earmarked for wear and tear.

Vandalism is another concern. For example, South Africa loses millions of dollars’ worth of copper cables each year to theft. We expect this to extend to hydromet gear, which should be protected by measures including security lighting, surveillance cameras and perimeter fences, patrols and alarms. Research is needed to identify the highest-risk areas and the best theft-prevention methods.

Next steps

With the African population expected to reach more than 2 billion by 2050, and with the accelerating impacts of climate change, the damage and death toll in Africa will balloon if current gaps in hydromet systems remain. Policymakers must keep pushing the Early Warnings for All initiative into the spotlight, and advocate for its extension. Specifically, the parties preparing for Africa Climate Week and the Africa Climate Action Summit, both in September, should lead discussions on boosting the pledged finances by at least threefold, while following the road map we propose.

Before more lives are lost, let’s give all Africans a fair chance to reduce their exposure to climate risks by fixing this hydrometeorological hazard blind spot.

[This Article is written by Asaf Tzachor, Catherine E. Richards, Masilin Gudoshava and was published in 'nature.com']