It’s possible that I shall make an ass of myself. But in that case one can always get out of it with a little dialectic. I have, of course, so worded my proposition as to be right either way (K.Marx, Letter to F.Engels on the Indian Mutiny)
Environmental sustainability researchers at NUS found that widely used Environmental, Social and Governance (ESG) ratings often fail to reflect companies’ actual exposure to deforestation, highlighting the need to integrate satellite-based environmental data into sustainable investing frameworks.
Many investors today seek to direct their money towards companies that are responsible stewards of the environment and society. Companies are therefore given ESG ratings, which are meant to indicate whether a business is environmentally and socially responsible. These ratings increasingly influence investment decisions worldwide, but it remains unclear whether they accurately capture one of the biggest environmental impacts of businesses: deforestation.
A multi-source approach to measuring deforestation exposure
The researchers combined satellite-derived forest loss maps, company asset locations, forest-risk commodity supply-chain information, media coverage, and commercial ESG and financial characterises databases to examine the relationship between corporate deforestation and ESG scores. In total, they examined 18 overall ESG scores and 12 deforestation-related sub-scores from five major international providers alongside two independent measures of company-level deforestation.
The analyses included forest loss surrounding the assets of 13,841 companies worldwide between 2013 and 2024 and deforestation exposure of 210 companies through global supply chains for deforestation-risk commodities such as palm oil, soy, beef, pork, chicken and wood products between 2013 and 2020. The team also assessed whether deforestation-related media coverage, rather than actual deforestation, influenced ESG scores.
Ratings track public image, not forest loss
The study found no evidence that ESG ratings consistently capture companies’ actual exposure to deforestation. Instead, ESG scores responded more strongly to media attention related to deforestation, whether positive coverage of companies’ management actions, or negative coverage of deforestation scandals.
The researchers also found that companies involved in palm oil production were more likely to receive lower ESG scores for deforestation than companies associated with soy, pork, or chicken supply chains, suggesting potential commodity biases.
“This is the first study to evaluate multiple commercial ESG rating systems against independently measured corporate deforestation at the global scale. There is a need for ESG providers to integrate objective, spatially explicit deforestation information derived from satellite observations into future assessments, allowing investors to better contribute to reducing deforestation,” said Ms Zhu.
“Our study does not imply that ESG ratings are not useful as a tool to move capital away from companies behind deforestation. In fact, ESG ratings have high potential given their wide coverage of companies. The main problem is that ESG analyses do not incorporate geospatial data. If these data were to be included, ESG investing could be a gamechanger in halting deforestation globally,” added Assoc Prof Carrasco.
As global warming intensifies, extreme heat events are becoming more frequent and severe worldwide, and are increasingly occurring alongside air pollution to form compound heatwave–air pollution events. Because simultaneous exposure to extreme heat and polluted air poses substantially greater health risks than either hazard alone, these compound events have emerged as a growing global environmental concern.
India has been at the forefront of this challenge. In the pre-monsoon season of 2026, temperatures exceeded 46°C across many parts of the country, with northern and central India experiencing prolonged and intense heatwaves. At the same time, compound heatwave–PM2.5 pollution events during the pre-monsoon season have become increasingly frequent, drawing renewed attention to their impacts on public health and air quality. But why are these compound extreme events becoming more common, and what are the key mechanisms driving their formation?
A research team from Hong Kong Baptist University investigated pre-monsoon heat-PM2.5 compound extreme events across India between 2017 and 2024 by combining ground observations, satellite measurements, and atmospheric chemistry simulations. The team found that the frequency of compound extreme events increased markedly after 2020. During April 2022, more than 70% of severe PM2.5 pollution episodes across northern India occurred simultaneously with heat waves, highlighting the increasing overlap between the two environmental hazards.
Further analyses revealed that large-scale atmospheric circulation played a central role in shaping these compound extremes. An anomalous anticyclonic circulation over northern India strengthened heat conditions while generating anomalous southeasterly winds that transported additional pollution precursors into the region. These meteorological conditions accelerated atmospheric chemical processes, substantially enhancing the formation of secondary organic aerosols (SOA), a major component of PM2.5. Atmospheric chemistry simulations showed that SOA production increased by approximately 2.3 times during compound events, contributing significantly to elevated particle concentrations under extreme heat.
“Most studies have focused on heat waves and air pollution separately,” said Huibin Dai, lead author of the study. “This study shows that large-scale atmospheric circulation can simultaneously intensify heat and promote secondary aerosol formation, creating favorable conditions for compound heat-PM2.5 extremes. It highlights an important pathway through which climate variability can influence both weather extremes and air quality.”
The findings suggest that understanding future air quality requires considering not only emissions but also climate-driven changes in atmospheric circulation and atmospheric chemistry. As climate warming is expected to increase the occurrence of extreme heat, these coupled meteorological and chemical processes may further increase the risk of compound environmental extremes in densely populated regions.
“The next step is to investigate how compound heat-PM2.5 extremes will evolve under future climate change and to quantify their impacts on human health,” Dai said. “Ultimately, this work aims to better understand the mechanisms behind compound heat-PM2.5 extremes and provide scientific evidence to support early prevention, risk reduction, and climate adaptation.”
The research was conducted by a team from the Department of Geography, Hong Kong Baptist University, led by Prof. Meng Gao. This work was supported by grants from the National Natural Science Foundation of China and the Research Grants Council of the Hong Kong Special Administrative Region, China.
New study highlights the importance of lived experience, local knowledge, and community voices in climate change communication. The researchers call for climate policies and communication strategies in the Global South to move beyond broad, one-size-fits-all approaches and instead recognise the lived realities of the communities most exposed to climate change.
While people across the globe are experiencing various impacts of the changing climate, governments and communicators often use top-to-bottom, uniform communication strategies to inform people about climate change.
Challenging the assumptions behind these communication approaches, researchers found evidence that people living in different coastal socio-ecological settings can experience, understand and respond to climate change in different ways, even within a broadly similar geographic and socio-ecological settings. The findings, published in the journal Ocean & Coastal Management, from the coastal region of Kerala within southern India reveal an important gap between awareness of climate change and scientific understanding of its causes.
The researchers studied people’s perceptions, beliefs and knowledge about climate change in two coastal regions, one low-lying island and the other a mainland coast. For example, people from Vypin, a low-lying island, expressed concern about sea-level rise, recurrent flooding, and land subsidence, while people from Kozhikode mainland coast often spoke about declining fish abundance and changes in the size of sardines.
The researchers documented many such accounts, demonstrating that lived experience can provide a detailed understanding of climate change even when people may not use scientific terminologies to describe it. The study also found that people do not interpret climate change solely through scientific explanations. Participants connected climate change with deforestation, pollution, plastic waste disposal, urbanisation, and other local environmental changes. Some also explained climate change through religious and cultural worldviews, including ideas of divine control, punishment or an approaching end of the world. Other misconceptions included the idea that climate change is caused by local environmental issues, such as dumping of plastic waste.
“These perspectives should not simply be dismissed as ignorance; rather, they reveal how communities make sense of a complex and uncertain phenomenon through culturally embedded ways of understanding the world. These different understandings are also suggested to influence people's sense of agency,” says Doctoral Researcher P. Afna from the Indian Institute of Forest Management, Bhopal.
Afna also points out that some misconceptions about the causes of climate change may create an inflated sense of responsibility among individuals and local communities, who may believe that changing their own behaviour alone can address a problem driven by much larger structural processes. This can potentially shift attention away from the responsibilities of governments, high-emitting sectors, and other powerful actors to provide adaptation support, climate finance, and other forms of assistance.
The findings also show that climate vulnerability is not experienced uniformly. Participants described concerns ranging from livelihood losses and temporary relocation to threats to homes, education, health, and psychological well-being. Some participants expressed fears that rising seas could eventually take their homes, while others considered migration as a possible response to increasing vulnerability.
Importantly, communities were not simply passive victims of climate change. They described a range of responses, including afforestation, reducing plastic pollution, coastal protection, and nature-based approaches. In Vypin, for example, some participants identified mangroves as a means of reducing erosion and protecting settlements, and some had planted mangroves around their homes. The study also found a substantial difference in perceived agency: 65% of participants in Vypin said they could do something to address climate change, compared with 16.25% in Kozhikode.
Rethinking how climate change is communicated
By analysing these results, and comparing it with the current climate change policies in Kerala, India, the researchers argue that climate communication cannot rely only on broad, top-down messages or scientific terminologies. Instead, communication needs to connect scientific knowledge with what people see, experience, and value in their own environments.
“People living at the forefront of climate change already possess important and relevant knowledge about the changes occurring around them. The challenge is to connect this lived experience with scientific understanding without dismissing local knowledge, cultural beliefs or community perspectives,” says Doctoral Researcher Jithin Vijayan from the Department of Biodiversity Sciences, University of Turku in Finland.
The study calls for more place-responsive approaches that bring together scientific knowledge, local and indigenous knowledge systems, community experiences, and diverse cultural worldviews. They recommend locally adapted communication materials, participatory dialogues, experiential education, and the co-production of climate knowledge with communities, with examples. Researchers also suggests a roadmap for implementing the findings within the policy framework.
The study was carried out by a collaborative team of researchers from the Indian Institute of Forest Management, Nature Conservation Foundation-India, and the Department of Biodiversity Sciences, University of Turku, Finland.
Credit: Photo by Georgina Goodwin. Courtesy of Conservation International
As Africa prepares for the arrival of a powerful El Niño weather event, an international team of researchers led by the University of Sydney and Imperial College London has found that access to improved forecasts alone will not protect many of the region’s poorest livestock farmers from the impacts of climate change.
Published today in Nature Sustainability, the study shows that weather information must be paired with practical “risk-smoothing” measures, such as supplementary livestock feed, drought insurance, targeted financial support or improved market access, so that farmers at risk of losing livestock can act before drought strikes.
By combining improved access to climate information with supported risk-smoothing, livestock farmers can avoid environmental trade-offs that occur when either is used in isolation.
Lead author Dr Matt Clark, a Research Associate in Sustainability Science with the Thriving Oceans Research Hub in the University of Sydney’s School of Geosciences, said climate variability was becoming an “inequity multiplier”.
“If the forecasts prove correct and we see a near-record El Niño develop, we expect this will amplify inequalities across southern African rangelands,” Dr Clark said.
“The problem isn’t that poor farmers don’t have climate information. The problem is they often can’t afford to act on it. Wealthier producers can reduce herd sizes ahead of drought and rebuild when conditions improve. Many small-scale pastoralists simply don’t have that financial buffer.”
Across southern Africa, millions of pastoralists depend on cattle, sheep and goats grazing vast rangelands stretching across Botswana, Eswatini, Lesotho, Mozambique, Namibia, South Africa and Zimbabwe. Livestock are far more than a source of income, they are a family’s savings, food security and cultural identity, often passed down through generations.
As climate change brings more variability between good seasons and severe drought, those communities face increasingly difficult decisions. Wealthier livestock owners can often sell animals before drought, preserve their savings and rebuild their herds when conditions improve. Poorer pastoralists are frequently forced to keep their animals on increasingly stressed landscapes, risking devastating livestock losses that can trap families in poverty.
“It’s not that the people we work with aren’t thinking about adapting to climate change,” said Dr Gina Arena, Science Lead for Herding for Health, Conservation International.
“It’s that with greater uncertainty in climate comes greater risk. We need to help people identify the best decisions and reduce the risk of implementation. Research like this helps us do that,” Dr Arena said.
Applying detailed modelling alongside work with non-government organisations in the field, the researchers found that climate forecasts can become substantially more effective when combined with support that enables households to respond before drought takes hold.
“The encouraging finding is that this isn’t about inventing new technology. It is about applying existing knowledge in new ways,” Dr Clark said.
“Weather forecasting is becoming increasingly accurate. The opportunity now is to make sure those forecasts are paired with practical support so the people who need them most can actually benefit.”
The study suggests this combination of climate information and risk-smoothing can improve livelihoods, reduce inequality and protect rangelands at the same time, overcoming trade-offs that have long challenged climate adaptation programs.
While developed for southern African pastoral systems, Dr Clark said the framework could also help livestock producers in other climate-vulnerable grazing lands, including in Australia.
“The methodology we’ve developed for southern African pastoralists could equally be applied in Australia and other regions facing increasingly variable rainfall. As climate extremes become more common, adaptation needs to be as much about helping people act as improving the forecasts themselves.”
Testing climate adaptation at scale
The researchers developed a model representing pastoral communities across seven southern African countries, drawing on projected rainfall patterns from European climate models together with empirical data on livestock, vegetation and community wealth.
The first simulation represented a business-as-usual future, showing that increasing rainfall variability reduced household wealth, increased livestock losses, degraded grasslands and widened inequality.
The second tested adaptation strategies. Access to forecasts alone delivered only modest benefits because many households lacked the financial capacity to respond, while risk-smoothing measures on their own could force environmental trade-offs.
The third simulation examined how adaptation could spread between neighbouring communities. Based on conservation programs such as Conservation International’s Herding for Health initiative, it showed that when practical support accompanied climate information, communities learned from one another and adoption became self-reinforcing, providing a pathway for climate adaptation at scale.
“It's easy to think of mathematical modelling as abstract, but here it is doing something very practical," said Cosima Fröhner, former Mary Lister McCammon Fellow at Imperial College London. “We simulate the decisions of thousands of individual livestock farmers under different climate conditions and interventions for adaptation. As the model settles into a stable state, the wider patterns and the trade-offs between interventions emerge from all those individual decisions.”
The research was conducted through an international collaboration involving the University of Sydney, Imperial College London, Conservation South Africa, the South African Environmental Observation Network, the Max Planck Institute for Evolutionary Anthropology, the University of California Merced and the Santa Fe Institute.
Modernised herding practice, Eastern Cape, South Africa.
Credit
Photo by Will McCarry. Courtesy of Conservation International
Lead author Dr Matt Clark from the University of Sydney holding a chameleon on field work in East Africa.
Credit
Photo supplied: University of Sydney
DOWNLOAD images from Africa and of the researcher at this link. Photos from Conservation International must credit photographer by name and mention Conservation International.
RESEARCH
Clark, M. et al ‘Minimising climate change adaptation trade-offs in African rangelands’ (Nature Sustainability 2026). DOI: 10.1038/S41893-026-01938-0
DECLARATION
The researchers declare no conflicts of interest. Funding for the study was received from the Leverhulme Trust, Research England and UK Research and Innovation.
Oxford, 15 September 2026 – Nearly half of all children under 5 in developing regions are poor, undernourished, or both, according to a report published today by the Oxford Poverty and Human Development Initiative at the Universityof Oxford. The report sheds new light on deprivations in child nutrition and school attendance across the developing world revealing important information on the scale and complexity of challenges facing children.
The study is based on data from the Global Multidimensional Poverty Index (Global MPI) computed by OPHI and the United Nations Development Programme. Going beyond the single deprivations children face in nutrition andschooling, the study also examines whether they experience broader multidimensional poverty at home. Specifically, itassesses whether they live in households deprived in at least a third of the ten indicators of the Global MPI relating to health, education and living standards.
This report delivers the first global multidimensional study of children’s experiences of poverty within the home basedon observed household survey data, covering 97 countries for undernutrition and 108 for school attendance.
Scale of challenges facing children in developing countries
The report highlights the urgency of child poverty in developing regions:
Across 97 countries with available data (covering 518 million children), shockingly, nearly half of all children under 5 (252 million) are undernourished, multidimensionally poor, or both.
Around one-third of children under 5 are undernourished, meaning they are stunted or underweight.
Globally 96 million undernourished children also live in multidimensional poverty (1 in 5 children).
Each multidimensionally poor and undernourished child experiences on average four to five deprivations in addition to being undernourished themselves.
Across 108 countries with available data, 13% of the 1.04 billion school-aged children are out of school (136 million). Over 80% of all out-of-school children are also multidimensionally poor (110 million).
Each poor out-of-school child experiences an average of five additional deprivations.
Of the 1.04 billion school-aged children covered, 31% are out of school, multidimensionally poor, or both.
“We were shocked by how tightly clustered the intensity of poverty is for children,” said Sabina Alkire, Director of OPHI and co-author of the study, “the 96 million children under 5 who are both undernourished and multidimensionally poor experience, on average, four to five deprivations in addition to undernutrition. Similarly, the110 million children who are both multidimensionally poor and out of school experience, on average, five additional deprivations across the other nine global MPI indicators. The data reaffirms the need for multisectoral approaches to development programmes such as social protection”.
Over one-third of poor children experience intra-household inequality
The report examines whether all children within the same household are all disadvantaged or whether all are not disadvantaged. It finds glaring intrahousehold disparities. Over one-third of poor children experience intra-household inequality. That is, over one-third of poor children under five live in a household in which at least one child is undernourished and another is not. Similarly, over one-third of poor children live in households where some school-aged children are going to school and others are not, pointing to a poignant need to plan integrated programmes that recognise how poverty may be experienced unequally within the home.
82% of poor out-of-school children live in households where no woman has completed six years of schooling
The report provides an up to date yet stark picture of women’s educational disadvantage across developing regions. In multidimensionally poor homes, over two-thirds of poor children – whether well-nourished or undernourished – live in a household where no woman has completed six years of schooling. By comparison, just over half live in householdswhere no man has completed six years of schooling. The gap is even wider among poor children who are out of school: around 82% live in households where no woman has completed six years of schooling, compared with 62% who live in households without a man who has reached the same level.
67% of girls in Afghanistan are out of school compared with 47% of boys
The report also finds that while more boys are undernourished than girls, and globally more girls are out of school than boys, the gender disparities in school attendance vary considerably across countries. In 19 countries, representing 56% of all out-of-school children, girls face higher rates of school exclusion than boys. The largest gender gaps are found in Afghanistan, where the share of girls not attending school is 20 percentage points higher than the share ofboys (67% compared with 47%), followed by Pakistan (30.2% and 22.3%), Central African Republic (29.3% and 21.5%), and Guinea (41.3% and 33.7%).
First-generation learners
More positively in 108 countries, 71 million children aged 10–17 years old are pioneer children – the only ones in theirhousehold to complete six years of schooling. Half of these pioneer children live in South Asia (35 million), where one in nine children aged 10–17 is a pioneer child.
By country, 28 million pioneer children live in India, over 3 million in Bangladesh, 2 million in Pakistan, 837,000 in Nepal and 508,500 in Afghanistan. Supporting these young people could mobilise significant change.
India, Bhutan and Bangladesh have higher shares of pioneer girls among girls aged 10–17 than there are pioneerboys among boys aged 10–17. Only in Afghanistan are there higher levels of pioneer boys than girls among childrenof each group – 7% of boys versus 4% of girls – and half of Afghanistan’s pioneer children are also multidimensionally poor.
Yet not all children share this progress. Globally, 9% of pioneer children live in homes where at least one other school-aged child is out of school.
How this will help
The report offers a new lens for understanding child nutrition and schooling in the context of other deprivations at the global level.
Professor Sabina Alkire said of the report “this is the first study that demonstrates at scale the possibility and importance of using existing individual data alongside household data to inform evidence-based programmes that meaningfully address cycles of poverty seeded by undernourished and out-of-school children. Amartya Senrecognised nutrition and schooling to provide both instrumental and intrinsic value. We must work together at the local level in the fields of measurement and policy to address these systemic deprivations that threaten to trap generations of children in poverty”.
About OPHI: The Oxford Poverty and Human Development Initiative (OPHI) is a research centre based in the University of Oxford working on the multidimensional measurement of poverty and well-being to help guide effective policymaking around the world.
