Features

On World Mosquito Day, Dr Lucy Harrison, postdoctoral researcher at Oxford’s Infectious Diseases Data Observatory (IDDO) at the Centre for Tropical Medicine and Global Health, reflects on the global impact of the mosquito and her research into malaria drug resistance.

A small insect, a global impact

Dr Lucy Harrison. Credit: James Harrison.

Every year on 20 August, World Mosquito Day marks the extraordinary role of one tiny insect in shaping human health. Mosquitoes are responsible for transmitting some of the world’s most devastating diseases, including malaria, dengue, Zika, and yellow fever.

Malaria alone causes more than 260 million cases and nearly 600,000 deaths annually. Around 95% of this burden is borne by people in Sub-Saharan Africa. Despite investments of over USD 4 billion in malaria control efforts in 2023, global funding still falls short of what is needed to meet the World Health Organization's Global Technical Strategy.

The relationship between humans and malaria is ancient. Evidence suggests the disease afflicted populations from the time of the Egyptians and may even have been described by Hippocrates.

From discovery to mathematics

World Mosquito Day commemorates the discovery by Sir Ronald Ross, on 20 August 1897, that female Anopheles mosquitoes transmit malaria. Ross also pioneered the first mathematical model of vector-borne disease, showing how infected mosquitoes create infected people and vice versa.

His insights laid the foundation for malaria control strategies: if mosquito numbers are reduced, the opportunities for transmission fall. George Macdonald later refined this work, introducing the concept of the ‘reproduction number’ or 'R number' — a measure familiar today from its use during the COVID-19 pandemic.

Fighting back

Mosquito-borne diseases can be tackled by controlling mosquito populations, reducing human exposure, and treating infections. Approaches include removing stagnant water, spraying insecticides, releasing genetically modified mosquitoes that reduce reproduction, using repellents and bed nets, and deploying effective medicines.

Several key malaria drugs come from natural sources. Quinine, derived from the bark of the cinchona tree, was used for centuries and even gave rise to tonic water. More recently, artemisinin, discovered in 1972 from sweet wormwood (Artemisia annua), revolutionised malaria treatment. Its discovery earned Chinese pharmacologist, Tu Youyou, the Nobel Prize in Physiology or Medicine in 2015.

But widespread drug use also fuels drug resistance. Artemisinin-resistant parasites were first documented in 2008, and are linked to specific mutations in the parasite’s genes. To preserve treatment effectiveness, the WHO now recommends combining artemisinin with a partner drug to slow the evolution of resistance.

Mapping resistance

At Oxford, my work focuses on mapping the spread of genetic mutations in malaria parasites across Sub-Saharan Africa. These mutations are linked to resistance against frontline drugs such as artemisinin.

With the power of modelling, my maps use the data that is available to predict what proportion of malaria parasites may have mutations linked to drug resistance in locations where we don’t have any data.

In many regions where malaria transmission is most intense, there is little or no genetic data. Running clinical trials to test drug effectiveness is costly and resource-intensive. To overcome this, we use geospatial models that can predict the likely distribution of resistance even in areas without data.

These models combine available genetic data with information on the distance and time between data collections, and environmental conditions such as malaria prevalence. By doing so, we can predict the prevalence of resistant parasites in areas where we don’t have any data.

The maps produced at the University of Oxford will be made freely available through the Infectious Diseases Data Observatory’s Artemisinin Molecular Surveyor. The Surveyor is a living systematic review which can be used by researchers and policy-makers to visualise the current state of global drug resistance in the malaria parasite.

A screenshot of the Infectious Disease Data Observatory’s Artemisinin Molecular Surveyor, which visualises published genetic data of the malaria parasite.

This World Mosquito Day reminds us that mosquitoes remain one of humanity’s most persistent threats. At Oxford, researchers are combining field data, genetics, and advanced modelling to provide the evidence needed to guide global health decisions, helping to ensure that life-saving drugs remain effective for the communities that need them most.

The Community Impact Lab links the research talent of Oxford University graduates with local community organisations tackling environmental, economic and social inequality across Oxfordshire. Impact Lab fellows outline what it's like to be involved

The Community Impact Lab is one of four programmes set up by Oxford SDG Impact Lab to harness the skills and knowledge of the University's graduate students to advance the UN sustainable development goals. 'We recognised a gap in graduate students' experience,’ says Alex Betts, Co-founder of SDG Impact Lab and Pro Vice-Chancellor for External Engagement, Sport and Community at Oxford University. ‘They get an amazing academic education when they come to Oxford. We wanted to complement this by offering them the opportunity to use their talents to make a difference to society and the planet.'

Some of this year’s 14 Community Impact Lab Fellows outline how they feel they have made a difference to the communities and organisations they have supported – and to themselves.  

Breaking down barriers

Sana Shah is completing a DPhil in History at Somerville College and has been part of the Impact Lab team supporting Oxfordshire Football Association. Her task has been to understand how football can promote better health, equality and community cohesion across the county. 

‘I chose this project because I felt it echoed my own research as an historian and my love of stories. It was about bringing people together and really listening to them – particularly to voices often left out of formal consultation.’

Sana teamed up with her research partner Ciao Anchi, a member of Wolfson College who is taking an MPhil in Global and Area Studies. They conducted in-person interviews and community engagement sessions with people from diverse backgrounds across Oxfordshire.

‘We also attended local matches, training sessions, and community events, where we spoke informally with parents, players, coaches, and volunteers. We reached upwards of 50-60 individuals.’

They found a massive enthusiasm for the beautiful game but also financial and cultural barriers to taking part, from the cost of kit to the pub culture among some adult teams.

Sana wants the FA to do ‘more listening’ to players and the people running local football groups, when it comes to allocating resources. She also suggests that Oxford colleges could get more involved in community football – perhaps sponsoring local teams. ‘Working on this project has completely changed how I see football – I now understand it not just as a sport, but as a community lifeline for many people.’

Promoting social justice

Arden Jaeger is pursuing an MSt in History of Art and Visual Culture and is a member of Wadham College. As a Community Impact Fellow, he has been working with the Tap Social Movement, which offers employment and training to people with convictions via the brewery, bakery, café and tap bars it runs across Oxfordshire.

‘I came to Oxford with the attitude that I would say yes to opportunities that came my way. I wanted to find a more measurable way of making a social impact.’

Arden particularly appreciates the interdisciplinary nature of being a member of the impact Lab. ‘It felt slightly strange at first. I was surrounded by scientists and social scientists tackling big problems while I was looking at pictures of whales.

‘But I have been able to demonstrate that the visual arts can have a huge impact, for example as a means of communicating the complexity or urgency of an issue.’

Tap Social has worked with around 60 prisoners or prison leavers and nearly a third of its staff have prison experience. Arden’s task has been to find out what impact secure employment has on people with convictions, their families and their friends. He also needed to consider how the Tap Social Movement could be scaled for broader social impact.

With his research partner, Diya Ramful – a DPhil Environmental Research student, also at Wadham College – Arden reached out to local arts organisations and criminal justice charities working with marginalised groups as well as Tap Social staff and customers. ‘There is huge support for Tap Social’s mission. Among staff with prison experience, a secure job is more than an income. It affects their housing, social connections and their mental health.

‘This project showed me just how interlinked problems of social justice can be. Tap Social is a positive case study that should be used to inspire other employers.’

Supporting local growers

Ali Elhassan and Clara Cecil already had an interest in sustainable food systems when they opted to work on a project with OxFarm2Fork, which links the county’s institutions, including 18 Oxford colleges, with 20 local agroecological food producers.

Clara, an MSc student in Sustainability, Enterprise and the Environment and a member of Mansfield College, has worked with a small start-up looking at funding and technology investment in regenerative farming. Developing measurement frameworks was very theoretical and I wanted to be more involved at a community level,’ she says.

Ali is an experienced civil and environmental engineer and is taking an MSc in Water Science, Policy and Management and is a member of Hertford College. Before joining the Impact Lab he carried out his own research into food waste in Oxford and wanted to do more. ‘The OxFarm2Fork project has been a good counterbalance to my academic work, although I found it hard fitting it all in at first.’

The pair agree that it was fun too. ‘We didn’t know each other before but have become good friends,’ Ali says.

OxFarm2Fork was set up by Good Food Oxfordshire which hopes to use Ali and Clara’s findings to help extend the Farm2Fork collaboration to other institutions.

Clara and Ali visited four local farms, identified more than 80 plant and insect species and tested the nutrient content of more than 20 different crops.

‘Some crops definitely have a much higher nutrient density than supermarket equivalent,’ Ali says. ‘But it is not a consistent picture and ideally we need to find a more detailed means of measuring nutrient content.’

Clara and Ali also interviewed farmers and students and chefs in participating Oxford colleges. They think there is more mileage to be had from partnerships with Oxford colleges such as piloting contracted growing around particular crops – possibly more unusual crops that cannot be sourced elsewhere.

They certainly feel that the inspirational growers they met have the commitment and work ethic to rise to the challenge. ‘The farmers were awesome,’ Clara says.

The Community Impact Lab is supported by the University of Oxford’s ESRC Impact Acceleration Account.

Dr Zakiyya Adam, Research Associate at the Transport Studies Unit within the University of Oxford’s School of Geography and the Environment, outlines recommendations for the implementation of cycling-promoting initiatives in mid-sized European cities.

Dr Zakiyya Adam, Research Associate, Transport Studies Unit

When a city provides only cycle lanes, this comes with the implicit belief that individuals can attain a bike without any financial aid, are confident in their cycling proficiency, and that they have a secure location at which to store their bike.

This is not the case for many, especially those from lower socioeconomic backgrounds.

Truly inclusive policy should strive to remove barriers to cycling - both objective and perceived - for all segments of society.

Likewise, cycling provisions should not only focus on individuals; cargo bikes make it possible for businesses to swap out their car or van trips and for children to be transported around as passengers. To encourage this use, the larger profile and heavier frame of e-cargo bikes needs to be accounted for when designing cycle infrastructure.

Whilst it is admirable that many cities aspire to be cycle-friendly and encourage people to shift from car travel to bikes, good will and cycle lanes alone will not instigate behaviour change.

Research as part of the SPECIFIC project, led by the Transport Studies Unit (TSU), and conducted in collaboration with academics in Austria, the Netherlands, Poland and Switzerland, has sought to understand what exactly is needed to see this ambition be realised in mid-sized European cities.

Truly inclusive policy should strive to remove barriers to cycling - both objective and perceived - for all segments of society.

A thorough analysis of cycling-promoting initiatives was conducted in Bristol by TSU researchers as well as in Graz (AT), Maastricht (NL), Poznań (PL) and Bellinzona (CH) by the other teams.

In 2008, Bristol was the first city in the UK to gain Cycling City status, which secured significant investment for cycling schemes. Since then, Bristol has continued its commitment to increase the number of cyclists through the creation of dedicated cycle lanes, better cycling facilities, and more cycle training. Despite the hilly terrain, Bristol has a strong cycling culture and good infrastructure.

In each of the five cities, the researchers interviewed a comprehensive array of individuals from across local government, consultancy, advocacy groups and academia who were directly involved in such schemes, and also engaged with publicly available literature.

Policy briefs were produced for each of the five cities, highlighting the key factors that enabled or hindered pro-cycling initiatives and outlining learnings for governance and planning.

Common threads emerged across the five mid-sized cities spanning the UK, Switzerland, Austria, the Netherlands and Poland, with the following five recommendations for encouraging participation and long-term behaviour change.

1. Safe, Continuous, and Inclusive Infrastructure

If we want more people to cycle, we need to build for the bikes - and riders - we have now and want in the future.

Cycle lanes should be physically separated from other road users where possible.

Gaps, fragmentation, or poorly integrated cycle lanes are major deterrents to cycling due to safety concerns, especially for families and new riders. And all cycling infrastructure – including, for example, bollards and chicanes, and cycle parking - should be designed to be inclusive, ensuring accessibility for cargo bikes and e-bikes that are heavier and larger than conventional bikes.

If we want more people to cycle, we need to build for the bikes - and riders - we have now and want in the future.

2. ‘Cycling Support’ Beyond Paths

Cycle paths alone are not enough. Supporting infrastructure - such as secure parking, maintenance facilities, and cycle training - is essential.

Cycle paths alone are not enough. Supporting infrastructure - such as secure parking, maintenance facilities, and cycle training - is essential.

On-street cycle hangars are particularly important for people living in apartments or without access to home storage. And secure parking near mobility hubs and workplaces often determines whether people choose to ride – the risk of your bike not being there at the end of the day is not one many can afford to take.

Only by addressing the full range of rider needs does cycling become a viable and attractive option for all.

3. Community Engagement Early and Often

Public participation, especially in planning and prioritising investments, is essential. Cities that actively consult residents on infrastructure tend to see higher uptake and less resistance.

Public participation, especially in planning and prioritising investments, is essential. Cities that actively consult residents on infrastructure tend to see higher uptake and less resistance.

In Bristol, for example, cycle hangars that enable six bikes to be parked in the space of one car were heavily subscribed to at just the planning stage, and faced little public resistance as the locations were proposed and voted on by Bristolians. Citizen-led initiatives also help foster lasting behavioural change.

People back what they help to build.

4. Address Cultural and Behavioural Barriers

Campaigns, community rides, gamified apps, and ambassador programmes can play a key role in shifting perceptions and normalising cycling.

Even with adequate cycling infrastructure, social norms and personal habits can discourage people from making the switch.

Perceived safety risks, discomfort in bad weather, status quo bias, and deeply embedded car-centric mindsets all hinder uptake.

Campaigns, community rides, gamified apps, and ambassador programmes can play a key role in shifting perceptions and normalising cycling. In Bellinzona, for example, the Bellidea and Bikecoin mobile apps reward users with points redeemable for vouchers or discounts, reinforcing cycling as a socially approved and economically rewarding practice.

Changing streets is important but changing minds is essential.

5. Build Long-Term Capacity

In the UK, the spending review in June 2025 saw cuts to the level of funding for Active Travel England, who many local authorities are reliant on for delivering cycle schemes. Volunteer-led efforts are valuable, but they are not sustainable without institutional backing.

Projects must be properly resourced, with long-term investment in training, evaluation, and leadership.

In the UK, the spending review in June 2025 saw cuts to the level of funding for Active Travel England, who many local authorities are reliant on for delivering cycle schemes. Volunteer-led efforts are valuable, but they are not sustainable without institutional backing.

Cities also need robust monitoring and feedback systems in order to adapt and scale successful schemes. Only with long-term investment can promising initiatives become permanent solutions.

-

International best practice offers a clear roadmap for action: invest in connected infrastructure, provide inclusive cycling support, embed public engagement in planning, focus on encouraging behavioural change, and commit to long-term investment and leadership.

And it is important that we get this right, as cycling initiatives have the potential to not just change how we move, but transform how we feel and live.

Dr Emily Warner, from Oxford University’s Nature-based Solutions Initiative, discusses the challenges of measuring biodiversity and capturing its complexity. She introduces a new framework aiming to simplify this process for practitioners, which was developed in collaboration with Dr Licida Giuliani and Dr Grant Campbell from the University of Aberdeen as part of an Agile Sprint on scaling up nature-based solutions in the UK.

Biodiversity supports the very fundamentals of human life, but its multi-faceted nature means it is easy for aspects of it to be in decline without us even realising.

Biodiversity supports the very fundamentals of human life, but its multi-faceted nature means it is easy for aspects of it to be in decline without us even realising.

Across the UK, abundance of all species has declined by an average of 19% since 1970 and nearly one in six species are at risk of extinction. The July 2025 assessment of progress on the Environmental Improvement Plan highlights the many habitat-based measures being implemented to tackle UK biodiversity loss, from four new National Nature Reserves to planting over 5,500 has of new woodland in England.

To understand whether these efforts are supporting progress towards the apex goal of thriving plants and wildlife, we need to assess how biodiversity is responding. Thinking about how we monitor these changes might seem boring, but it is important, and we won’t solve the biodiversity crisis without it!

Dr Emily Warner

Why measuring biodiversity is so hard

From an increasing interest in biodiversity credits to national and international commitments to reverse biodiversity loss, the need for effective biodiversity monitoring methods is clear.

The challenge is that measuring biodiversity is notoriously complex. The Convention on Biological Diversity’s definition of biodiversity highlights how expansive a concept biodiversity is: 'the variability among living organisms from all sources and the ecological complexes of which they are part; this includes diversity within species, between species, and of ecosystems'. 

With an increasing need to demonstrate success from conservation projects, the question of how to measure biodiversity is increasingly at the forefront of practitioners’ minds.

For example, nature-based solutions projects, which work with nature to tackle societal challenges, such as restoring a wetland to mitigate flooding, must also deliver benefits for biodiversity at their core. Similarly, multiple biodiversity credit systems – which allow the trading of tokens representing improved biodiversity – are in development in the UK alone, emphasising the critical need to be able to document increasing biodiversity.

With the rush to come up with a simple, tractable method of measuring biodiversity there is a simultaneous risk of oversimplifying, and we need to ask whether measuring something inadequately could be worse than not measuring it at all.

UK invertebrates are declining faster than plants and birds, threatening the foundation of ecosystems and direct benefits they provide to humans, such as food security, which is underpinned by pollination and pest control. 

For example, biodiversity net gain in the UK aims to ensure any habitat lost during development is replaced by more or better quality habitat. Biodiversity units are estimated based on habitat size, quality, location, and type, however, this approach overlooks many habitat attributes crucial to invertebrates, running the risk that invertebrate biodiversity will not be protected. UK invertebrates are declining faster than plants and birds, threatening the foundation of ecosystems and direct benefits they provide to humans, such as food security, which is underpinned by pollination and pest control. 

In contrast to monitoring carbon sequestration associated with conservation projects, where the focal unit of measurement – a tonne of carbon – is unequivocally defined, biodiversity’s complexity requires a much more nuanced approach. It is perhaps unrealistic to expect to reduce biodiversity down to a single measurable variable, without acknowledging that doing so will inevitably lose a huge amount of information on changes in biodiversity. 

A better way to measure what matters

To measure something diverse and complex we need to accept that the monitoring approach should reflect that diversity and complexity, while balancing this with feasibility. One way to increase the measurability of biodiversity is to structure the concept, breaking it down into component parts.

In contrast to monitoring carbon sequestration associated with conservation projects, where the focal unit of measurement – a tonne of carbon – is unequivocally defined, biodiversity’s complexity requires a much more nuanced approach.

In 1990, conservation biologist Reed Noss developed a hierarchical framework, organising biodiversity into three axes: composition, structure, and function, which can be assessed at four scales (genetic, population, community, landscape). If each axis represents a different aspect of biodiversity, then measuring metrics across the different axes should more widely capture biodiversity.

However, for each axis there are still many possible metrics that can be measured. Returning practitioners - or anyone else who wants to measure biodiversity - back to their original predicament of selecting the best metrics to effectively assess biodiversity.

Our recent research developed an ecological monitoring framework for nature-based solutions projects, seeking to overcome this problem.

We reviewed 71 possible biodiversity metrics, ranking them based on how informative they are and how feasible they are to measure. Of these, 30 metrics scored highly enough on both informativeness and feasibility to enter our framework. These metrics were grouped into Tier 1, Tier 2, and Future metrics.

Tier 1 are the highest priority metrics in terms of informativeness and represent all three axes of biodiversity. Future metrics are equally informative but currently too technically challenging or costly to measure. Tier 2 metrics are informative but often less widely applicable than Tier 1 metrics.

These metrics are now freely available in a searchable database, allowing practitioners to identify suitable metrics for their projects based on criteria such as cost, technical expertise required, and availability of a standardised methodology for data collection. 

As assessing biodiversity requires investment of time, expertise, and money, we want its results to be as impactful as possible.

Our database will channel the energy put into biodiversity monitoring towards cohesive, effective data collection, that widely captures change across the complexity of biodiversity, encouraging measurement of the different axes and scales of biodiversity. 

Dr Emily Warner measuring biodiversity in the field. Credit: Ella Browning

We hope our database will help to navigate the huge pool of possible biodiversity metrics, highlighting the most useful metrics for assessing biodiversity and giving a clearer understanding of what information they provide. 

The next step in any biodiversity monitoring plan is then getting out and collecting the data, ideally in a standardised way that will allow comparison between projects or to existing datasets.

The 'how' of biodiversity monitoring unmasks another layer of complexity, as for most of the metrics in the database there are multiple potential methods for data collection and decisions need to be made about a sampling plan. In some cases, there are even different ways of calculating the final metric.  

A large part of the research underpinning the development of our metrics database involved identifying existing standardised methodologies that could be used to collect data.

The increased interest in monitoring biodiversity could lead to a boom in biodiversity data, representing a huge opportunity to better understand the trajectory of biodiversity across a wider range of UK contexts, but also the potential risk of a missed opportunity to maximise the outcomes of this data collection effort.

By helping make these standardised metrics and methodologies available, we hope to encourage coordinated, large-scale biodiversity data collection to support effective biodiversity action and also highlight where more guidance is needed to support data collection on the ground. 

Effective monitoring to turn the tide on biodiversity loss

Our monitoring tool aims to provide shortcuts to developing a monitoring approach, highlighting what different metrics tell us about biodiversity, connecting these to available methods and allowing practitioners to search these metrics based on key criteria.

If we want to bend the curve of biodiversity loss we need effective monitoring to understand how well our efforts to restore nature are working.

If we want to bend the curve of biodiversity loss we need effective monitoring to understand how well our efforts to restore nature are working.

We have been aware that biodiversity has been declining since before I was born and this continues to escalate. My hope is that I will see the transition to a positive trend in biodiversity over the rest of my career and that this monitoring tool could be one small step on this pathway.

Dr Marcia Zilli, Postdoctoral Research Assistant in Climate Dynamics, and Dr Neil Hart, UKRI Future Leaders Fellow, School of Geography and the Environment, explore the likelihood of increases in intense rainfall events alongside heatwaves in Brazil. 

It has been a little over one year since the devastating floods in Rio Grande do Sul. While Brazil has been spared a second flood of the same unprecedented scale, extreme rainfall events continue to hit.

Our results show that while the frequency of all tropical-extratropical cloud bands will reduce by a third, the most intense rain events - those ranked as a one-in-five event in the present climate - triple in likelihood by the end of the century for the highest greenhouse gas emissions scenario. 

For example, on 11^th October 2024 the city of São Paulo was hit by an intense rain storm with winds over 100km/h leading to an electricity outage effecting more than two million people. 

The outage lasted for 35 hours, but parts of the city remained without power for more than three days. 

The economic losses ranged from damaged household appliances and spoiled food to hospitals having to discard large amounts of medication that must be kept under refrigeration. The business and entertainment sector also faced large losses, with the financial impact of what was already a catastrophic event multiplied by the fact that the outage happened over the weekend. In total, the economic losses are estimated in more than BRL 1.65 billion (equivalent to £200 million).

Our latest paper, ‘Threefold increase in most intense South Atlantic convergence zone events by 2100 in convection-permitting simulation’, produced in collaboration with Ron Kahana and Kate Halladay, senior researchers from the Met Office, investigates changes in the frequency and intensity of the weather system responsible for these severe storms under continued planetary warming.

In a typical Brazilian summer, large bands of cloud develop across South America spreading from the tropical Amazon forest southeastwards into the southern Ocean. These are called tropical-extratropical cloud bands and produce up to two-thirds of annual rainfall across much of southeast Brazil. 

However, intensely raining clusters of thunderstorms are embedded within these cloud bands. It is these embedded rainstorms which create the natural disasters such as the São Paulo storm mentioned above and the massive landslides in Rio de Janeiro state that took place in January 2011, killing more than 200 people.

In our paper, we evaluated future climate projections based on a high-resolution state-of-art climate model. Traditional climate models, such as those used in the latest IPCC report, have a spatial resolution of about 100km (think of these as pixels which the model can simulate) - too large to correctly represent intensely raining clusters of thunderstorms. 

The new generation of models have a much finer spatial resolution - 4.5km - resulting in a more realistic representation of the localised heavy rainfall clusters. This improvement happens because the physical equations governing thunderstorm development can be explicitly used by the model rather than the statistical approximations of thunderstorms used in traditional 100km resolution climate models.

Our results show that while the frequency of all tropical-extratropical cloud bands will reduce by a third, the most intense rain events - those ranked as a one-in-five event in the present climate - triple in likelihood by the end of the century for the highest greenhouse gas emissions scenario. 

Crucially, this cloud band intensification rate is far higher than that estimated in typical climate models, suggesting that this future rainfall risk has been underestimated in previous work. These results imply a greater risk of flooding events as the planet continues to warm. 

Furthermore, the reduced total frequency of cloud band events could result in both more frequent droughts and more frequent heat waves. 

Focusing on drought-heatwave risks is the next step in our research. Our team has recently started a new project under the Climate Science for Services Partnership – Brazil, managed by the Met Office, to diagnose the climate processes driving recent droughts in the northern Amazon and unprecedented compound drought-heatwaves extended across the eastern Amazon down to the coastal cities of Sao Paulo and Rio de Janeiro. 

This work is being done in close collaboration with Brazilian national forecast and disaster warning agencies to support their crucial efforts to enhance forewarning and preparedness, with the goal of avoiding the devastation caused by growing risks in which weather flips from heatwaves to floods and back again.