The study, conducted between 2022 and 2023 using a constructivist grounded theory approach, did not simply ask clinicians to describe their routines. Researchers used semi-structured interviews lasting 45 to 60 minutes, supplemented with hypothetical patient vignettes that allowed participants to reason through realistic cases without feeling personally exposed. Data collection and analysis ran in parallel, with each interview transcribed verbatim and coded line by line using constant comparison, a process supported by MAXQDA software. From this iterative analysis a core category emerged that captures the entire dilemma: healthcare professionals are striving to enhance cardiovascular risk awareness in their patients while balancing responsibility and uncertainty.

That balancing act begins with assessment itself. The clinicians described monitoring HbA1c, blood pressure, lipid levels and lifestyle habits, but they openly admitted that it was difficult to know which parameter best reflected a patient’s true cardiovascular risk. Some focused primarily on blood sugar; others prioritised cholesterol before turning their attention to glycaemic control. Almost all relied on risk visualisation tools from the Swedish national diabetes register, and physicians additionally used standardised instruments such as SCORE2 to classify risk as high or low. Yet the tools cut both ways. Some professionals refrained from formal risk assessment altogether because navigating multiple systems was perceived as time-consuming and cumbersome, meaning that cardiovascular risk in those cases went inadequately communicated.

Even when assessment succeeded, translation into understanding proved far harder. Participants described painstaking efforts to identify what each patient actually knew, asking individuals to explain in their own words why they were attending the visit and what they understood about their disease. They built step-by-step visual pictures, used metaphors and explanatory models to illustrate blood sugar and lipid levels, and checked comprehension by asking patients to recount what had been agreed. One diabetes nurse explained the stakes bluntly: poorly adapted information is wasted, so she tried to be clear about risks of vision changes, stroke, heart attack and kidney failure without wrapping the message too vaguely. Notably, some clinicians communicated microvascular complications while hesitating over macrovascular ones, sometimes splitting the risk message across different moments in the same visit.

The emotional dimension of this work emerged as perhaps the most striking finding. Clinicians reported genuine frustration and even feelings of failure when they could not reach certain patients, and some described the painful necessity of prioritising those willing to attempt change over those who were not, a decision one nurse called very difficult to make. Patients who expressed shame at missing treatment targets posed a particular challenge, as did patients perceived as fearful, overwhelmed by information, or struggling with mental health difficulties, in whose cases some professionals admitted avoiding risk discussions altogether. The clinicians walked a constant tightrope between encouragement and demand, wanting to convey seriousness without resorting to scare tactics or blame, and recognising that fear alone rarely produces sustainable lifestyle change.

Trust emerged as the currency that made any of this possible. Participants described building professional yet personal relationships, drawing on their own experiences, offering advice anchored in the patient’s actual life, and reinforcing progress to instil hope. Continuity was identified as crucial, because trust accumulates over repeated encounters rather than single appointments. Yet the organisational reality worked against this ideal. Many patients received only one visit per year instead of the guideline-recommended two, appointments became checklist-oriented and conveyor-belt-like, and limited time for follow-up questions hindered meaningful dialogue. Some clinicians encouraged patients to write down questions for their next visit as a workaround, a small and telling adaptation to systemic shortage.

Beneath the individual struggles lay a deeper structural problem: the diabetes team, in any meaningful sense, often did not exist. Diabetes specialist nurses described feeling alone in carrying overall responsibility for patients with type 2 diabetes, absorbing duties that would ordinarily fall to physicians because of chronic doctor shortages. Patients rarely met a physician in a planned manner, and nurses and doctors sometimes applied different strategies to the same patient, creating uncertainty about how treatment and risk communication should proceed. Some nurses admitted they did not always trust the clinical decisions of the physicians they worked with. Both professions called for shared responsibility, a common working model, structured diabetes-focused meetings and greater consensus, so that patients would hear consistent information from every member of the team.

The study’s authors argue that these findings reframe cardiovascular risk communication not as a simple transfer of information but as an ongoing, dynamic dialogue shaped in real time by the tension between objective medical data and patients’ subjective interpretations. Risk calculators and visual tools, however sophisticated, proved insufficient on their own to raise awareness or change behaviour, which helps explain why patients so often report not receiving or not understanding risk information. The researchers suggest that person-centred communication requires more than adapting facts to individual patients; it demands creating shared meaning about what risk actually signifies for a particular life, listening to patient narratives and addressing emotional and existential concerns before expecting behavioural change to follow.

The implications reach well beyond northern Sweden. The authors acknowledge limitations, including recruitment difficulties that forced reliance on snowball and convenience sampling, a predominantly female participant pool, and digital interviews that may have limited rapport, but they note that similar themes emerged across participants of varying age, sex, experience and setting, suggesting robust consistency. Their recommendations are pointed: healthcare systems must provide clinicians with the time, continuity, staffing and team structures without which even the most skilled communicator cannot succeed, and future research should explore collaboratively with patients how cardiovascular risk can be conveyed without vagueness or threat. Until then, the professionals on the front line of the diabetes epidemic will continue doing what this study documents so vividly: shouldering an enormous communicative burden alone, hoping their words reach patients before a heart attack makes the message unmistakably clear.

Subject of Research: Cardiovascular risk communication to people with type 2 diabetes in primary healthcare

Article Title: Primary Healthcare Professionals' Perspectives on Cardiovascular Risk Communication to Persons With Type 2 Diabetes: A Grounded Theory Study

Article References: Stenlund, A.-L., Hellström Ängerud, K., Lilja, M., Otten, J., & Jutterström, L. (2026). Primary Healthcare Professionals' Perspectives on Cardiovascular Risk Communication to Persons With Type 2 Diabetes: A Grounded Theory Study. Nursing Open, 13(9), Article e70828. https://doi.org/10.1002/nop2.70828

Image Credits: AI Generated

DOI: 10.1002/nop2.70828

Keywords: type 2 diabetes, cardiovascular risk, risk communication, primary healthcare, grounded theory, diabetes specialist nurses, person-centred care, patient self-management, healthcare teams, Sweden, qualitative research, HbA1c

The research team, led by Victor Marchezini of Brazil’s National Center for Monitoring and Early Warning of Natural Disasters, CEMADEN, argues that warning systems are not merely technical installations but long-term social processes embedded in public institutions. The Sendai Framework for Disaster Risk Reduction and the Early Warnings for All initiative have driven measurable progress at the national level: the number of countries reporting multi-hazard early warning systems rose from 56 in 2015 to 119 in 2024. Yet 40 percent of countries, 76 in total, still report having no such systems, and far less is known about how warnings are implemented below the national level, where most disasters actually strike. More than 25 years ago, a United Nations working group concluded that in many countries the institutional capacity to develop local early warning systems simply does not exist. The new study demonstrates that this diagnosis remains painfully accurate.

Scientifically, the innovation lies in treating local early warning systems as public policy that requires two complementary forms of institutional capacity. Technical-administrative capacities cover the internal functioning of local disaster risk management agencies: their human, financial and material resources, risk maps, and access to forecasts and monitoring systems. Political-relational capacities capture governance, including coordination across sectors and levels of government, risk communication strategies, contingency planning, and the engagement of civil society. The ICLEWS framework measures both across seven dimensions: human resources, financial resources, risk mapping, weather and subseasonal forecasting with rainfall monitoring, risk communication, social participation, and local and regional warning system governance. These dimensions map directly onto the four internationally recognized warning subsystems: risk knowledge, monitoring and forecasting, communication and dissemination, and preparedness and response.

The data came from an online survey administered between March and July 2025 in partnership with civil defense units across Brazil’s 26 states. The questionnaire contained 41 questions on disaster preparedness, of which 16 were selected for the ICLEWS analysis. Responses were converted to ordinal values, normalized to a common zero-to-one scale using the min-max method, and aggregated with equal weights into dimension and capacity scores, following an adaptation of the OECD Handbook on Constructing Composite Indicators. Kendall’s tau-b rank correlation was then used to explore the relationship between the two capacity types, an approach suited to ordinal data with non-normal distributions and tied values.

The central statistical result is a moderate positive association between technical-administrative and political-relational capacities, with a tau-b coefficient of 0.374. In practical terms, municipalities that score higher on staffing and resources also tend to score higher on communication, participation and governance, but the relationship is far from deterministic, confirming that the two capacities retain distinct analytical identities. The aggregate picture is sobering. Technical-administrative capacity recorded a median of just 0.356, with the middle half of municipalities ranging from 0.165 to 0.477, while political-relational capacity fared slightly better at a median of 0.422.

Beneath the aggregates, the human resource findings are striking. One quarter of local civil protection managers have less than one year of experience, and 8 percent report no prior experience at all. Some 22 percent of municipal civil defense units consist of a single officer, 16 percent have two members, and only about one third have more than five. Across the responding municipalities the survey identified 13,516 civil defense officers, but only one quarter hold permanent public service positions, meaning the rest are exposed to the job turnover that fragments institutional memory and disrupts warning operations. Financial capacity is equally fragile: only one quarter of civil defense units have their own budgets, with most relying on transfers from other municipal departments, federal or state programs, parliamentary amendments or donations. The authors note that similar patterns of fiscal dependence and unstable staffing have been documented in Indonesia, India, Malawi and Zambia, suggesting a global structural problem rather than a Brazilian peculiarity.

Risk mapping emerged as the most developed technical dimension, with 57 percent of civil defense units using municipality-generated risk maps, though one quarter of cities lack such maps entirely and only 14 percent use geotechnical suitability maps that could steer new construction away from hazard zones. Forecasting use is limited: 38 percent of respondents report using weather forecasts, but only 15 percent use subseasonal forecasts covering 10 to 30 day horizons, and 41 percent use rainfall monitoring. Notably, many respondents answered that they did not know whether their municipality used these tools, which the authors interpret as a sign that local officers may be unfamiliar with how to access available forecast products. On the political-relational side, only 32 percent of municipalities report operating a formal alert system, yet 69.5 percent rate their alert-issuing capacity as high or medium, a discrepancy the researchers attribute to reliance on informal channels such as social networks, WhatsApp groups and radio that officers do not count as official systems.

Perhaps most damning for the vision of people-centered early warning are the participation figures. Nearly one quarter of municipalities take no action at all to support people living in at-risk areas, and among those that do, only 11 percent act frequently. Just 6.4 percent of municipalities reported that their populations frequently participate in civil defense activities. Governance scores were the lowest within the political-relational dimension: 30 percent of cities have no contingency plan, and while plans exist for riverine floods in 47.5 percent of cases and droughts in 35.6 percent, only 7.9 percent have plans for heatwaves, an emerging and growing hazard. Even in a country with an extensive coastline, only about a quarter of surveyed coastal cities reported contingency plans for coastal erosion. Intermunicipal coordination, essential when hazards cross jurisdictional boundaries, is maintained by just 61 percent of municipalities.

The authors conclude that the Early Warnings for All initiative often disregards the profound institutional, territorial and fiscal inequalities among municipalities, particularly in the Global South, where global standards like multi-hazard and people-centered warning systems are formally adopted without the conditions for substantive implementation. Without stable financing, permanent staffing, access to forecasts, intermunicipal coordination and genuine social participation, they warn, local early warning systems will remain fragmented and operate below their potential to reduce losses, strengthen prevention and save lives. Institutional capacity, they argue, should be understood not as a fixed state but as a dynamic process, and international guidelines must be localized, adapted to local social realities and backed by sustained investment, or the universal early warning promise will falter precisely where it matters most: in the communities where disasters actually happen.

Subject of Research: Implementation capacity of local early warning systems and the localization of the Early Warnings for All initiative

Article Title: Implementation Capacity of Local Early Warning Systems: The Need for Localism in the EW4All Initiative

Article References: Marchezini, V., Oda, P. S. S., Ferreira, A. M., da Fonseca, M. N., Cotting, A. L. M., Dias, K. G. C., Sampaio, M. R. P., & Coura, C. M. (2026). Implementation Capacity of Local Early Warning Systems: The Need for Localism in the EW4All Initiative. International Journal of Disaster Risk Science. https://doi.org/10.1007/s13753-026-00765-3

Image Credits: AI Generated

DOI: 10.1007/s13753-026-00765-3

Keywords: early warning systems, EW4All, disaster risk reduction, ICLEWS framework, institutional capacity, Brazil, municipal civil defense, multi-hazard early warning, risk communication, social participation, Sendai Framework, localism

The stakes could hardly be higher. Indonesia’s tropical climate, with relatively consistent temperatures and distinct wet and dry seasons, has historically masked the growing threat of extreme heat. Yet heatwaves in Southeast Asia are becoming more frequent, with rising numbers of warm days and nights often compounded by high humidity, which intensifies physiological heat stress and impairs both physical and cognitive function. A 2024 report from Climate Central found that roughly 6.3 billion people, about 78 percent of the global population, experienced at least 31 days of extreme heat exceeding 90 percent of historical temperatures between 1991 and 2020. For Indonesia, the consequences are measurable in lost labor: the country ranks among those incurring the largest losses in work capacity, estimated at 4 to 6 percent of annual gross domestic product, with each worker losing an average of 71.8 hours in a single year, some 15 billion hours nationwide. More than half of Indonesia’s 270 million people already live in heat-exposed urban coastal areas, and by 2035 two-thirds of the population is projected to be urban.

At the heart of the problem lies a definitional bottleneck. Indonesia’s official extreme heat definition, codified in BMKG’s legal documentation, uses a single fixed threshold: a daily dry-bulb temperature anomaly exceeding the climatological average by 3 degrees Celsius. That metric has never been calibrated against health outcomes in the Indonesian context, and it overlooks several factors that matter enormously in the tropics. Warm nights, which are warming faster than daytime temperatures globally, are ignored. Humidity, which dramatically amplifies heat stress, is not captured. And Indonesia’s naturally low temperature variability means a 3-degree anomaly may be far rarer, and far more anomalous, than in temperate climates where such thresholds were first developed. International guidance from the World Meteorological Organization and World Health Organization recommends that heat early warning thresholds be grounded in epidemiological evidence linking temperature and humidity to mortality, hospital admissions, and heat illness, and that composite indices such as the Wet Bulb Globe Temperature or Universal Thermal Climate Index, which combine temperature, humidity, wind, and radiation, be used to capture physiological strain. Indonesia’s current approach does none of this.

The research team, led by Yoshua A. Nugroho of BMKG and the University of Copenhagen together with Emmanuel Raju, Agie W. Putra, and Carolina P. Marghidan, designed their study around a conceptual framework adapted from work on United States tornado warning decisions, translating it carefully to the very different physics and perception of heat. Twenty senior operational meteorologists with at least five years of service helped refine the survey through semi-structured discussions before it was distributed to 300 eligible forecasters in June 2024, one of Indonesia’s hottest months. The final sample of 140 valid responses closely matched BMKG’s workforce demographics, with 64.3 percent male participants, a majority aged 25 to 34, and over half working at local stations. The team then applied hierarchical ordinary least squares regression, entering predictors in blocks that mirrored the operational reasoning process: internal challenges first, then external challenges, then data-driven decisions, personal judgment, and communal judgment.

The results on dissemination challenges were striking in their selectivity. Of all the potential obstacles the researchers modeled, only two internal factors emerged as significant. The first was terminology ambiguity, meaning the lack of a universally accepted, context-appropriate definition of extreme heat for tropical settings, which showed medium-sized effects on both technical and non-technical challenges. Without a shared, credible definition, forecasters reinterpret guidance individually, producing inconsistent warnings and delayed action. The second was reliance on weather station data as the primary validation measure. Most stations have limited instrumentation, and urban neighborhoods, particularly informal settlements, are demonstrably hotter than what nearby stations record, meaning dangerous heat events can simply go unmeasured. External challenges, such as the absence of collaboration mechanisms with health departments and local governments, did not independently predict dissemination difficulties, but they mattered in a subtler and more troubling way.

That subtlety surfaced in the analysis of perceived effectiveness. The interaction between heavy reliance on station data and weak external collaboration was the strongest predictor of whether forecasters believed the current system was working well. In agencies with few ties to health authorities or municipal governments, forecasters who leaned on station data were significantly more likely to view the existing threshold-based system as adequate. The authors interpret this as institutional self-reinforcement: the system appears effective not because it truly captures heat risk, but because it aligns with operational norms under constraint, and forecasters’ views are rarely corrected because they seldom engage with the agencies that see heat’s health impacts firsthand. The researchers also argue that the apparently insignificant effect of missing vulnerability data reflects insufficient operationalization rather than irrelevance. Indonesia already holds census data, disaster agency risk maps, and community health profiles, but none are integrated into heat warning operations as decision-support tools.

The decision-making findings carry the sharpest implications. The full model explained 48 percent of the variance in whether a forecaster chose to issue a heat warning. Data-driven factors dominated, contributing 23 percent of explained variance, with the forecaster’s own interpretation of meteorological analyses, such as model outputs and forecast trends, showing a significant medium effect. Field reports from weather observers did not matter statistically. Among personal judgment factors, the most influential single predictor, with the largest effect in the entire model, was recognition of past experiences with successful warnings. In practice, forecasters consult a mental library of previous events, recalling which combinations of temperature, humidity, and wind once produced impactful outcomes, and match current conditions against that archive. This is what cognitive scientists of forecasting call tacit expertise, built over hundreds of hours of operational work, and the study shows it functioning as the primary decision substrate precisely where formal guidance is weakest.

The second significant personal factor was warning philosophy. Forecasters who adopt a liberal approach, issuing warnings more readily than colleagues, were significantly more likely to warn, a pattern the authors link to ambiguous terminology forcing individual adaptation and to defensive, self-protective behavior aimed at avoiding blame. The liberal philosophy minimizes the risk of failing to warn, but it inflates the false alarm ratio, and repeated false alarms erode public urgency in the phenomenon known as the cry wolf effect. Notably, communal judgment, such as seeking second opinions from colleagues, and interaction effects between human judgment and meteorological input showed no significant influence, suggesting that heat warning decisions in Indonesia remain largely solitary rather than collaborative. For a hazard whose impacts unfold across health systems, labor markets, and urban infrastructure, that individualization of risk is itself a structural vulnerability.

The authors close with concrete recommendations for BMKG and policymakers. They call for co-developing adaptive, locally calibrated heat thresholds with epidemiologists, urban planners, and public health experts so that exposure, sensitivity, and adaptive capacity enter the warning criteria directly. They urge integration of real-time health surveillance streams into the impact-based forecasting platform, deployment of additional measurement tools in dense urban areas, and contextual vulnerability data as reference layers for forecasters. They also recommend operational training built on experiential learning, reflective practice, and scenario-based simulation to strengthen the demonstrated interplay between data and experience. Theoretically, the study extends the sociology of forecasting by showing that when formal guidance and context-specific thresholds are underdeveloped, tacit expertise ceases to be a supplementary refinement of data and becomes the foundation of the warning itself. As extreme heat accelerates across the tropics, the study suggests that the most important upgrade to any heat early warning system may not be a new algorithm, but the institutional scaffolding that lets human judgment, good data, and cross-agency trust reinforce one another.

Subject of Research: Operational challenges and decision-making in heat early warning systems in Indonesia

Article Title: Heat Risk Communication by Operational Meteorologists in Indonesia: Current Operational Challenges and Decision-Making Dilemmas in Heat Early Warning System

Article References: Nugroho, Y. A., Raju, E., Putra, A. W., & Marghidan, C. P. (2026). Heat Risk Communication by Operational Meteorologists in Indonesia: Current Operational Challenges and Decision-Making Dilemmas in Heat Early Warning System. International Journal of Disaster Risk Science. https://doi.org/10.1007/s13753-026-00760-8

Image Credits: AI Generated

DOI: 10.1007/s13753-026-00760-8

Keywords: heat early warning systems, operational meteorologists, Indonesia, BMKG, extreme heat, heat thresholds, risk communication, decision making, impact-based forecasting, tacit expertise, climate change, disaster risk science

The review, led by Zecheng Li and Chng Saun Fong of the Institute for Advanced Studies at Universiti Malaya, together with colleagues spanning software engineering, public health, and chemical engineering, synthesised sixty-nine studies on how compound environmental events are defined, forecast, and communicated within early warning system frameworks. The team followed established systematic review methodology, including PRISMA reporting guidelines, and mapped every study against a purpose-built taxonomy that organises the field into three connected stages: how compound events are operationally defined, how they are jointly forecast, and how the resulting risk reaches the public as an actionable warning.

The first stage, event definition, turns out to be far more consequential than it might appear. Most existing systems identify a compound event simply by overlapping fixed or percentile-based thresholds, for example declaring a compound hot-and-polluted episode when both temperature and ozone or particulate matter exceed chosen cut-offs simultaneously. This approach is transparent and easy to automate, but the review shows it is also arbitrary: studies of mortality in Wuhan, Montreal, Beijing, and dozens of other cities demonstrate that the health impact of a given pollutant concentration shifts substantially with temperature, meaning a threshold defined without reference to measured health outcomes can systematically misclassify genuinely dangerous days. Impact-based interaction models, which use epidemiological relationships to weigh the joint effect of heat and pollution, offer a more defensible alternative, and the review documents a rapidly growing body of work fitting distributed lag non-linear models and machine learning methods to hospital admissions and mortality records to derive compound definitions grounded in actual harm.

On the forecasting side, the technological picture has changed dramatically in just a few years. Coupled meteorology-chemistry models such as WRF-Chem and operational systems built on the NOAA Global Forecast System can now simulate how heat, boundary-layer dynamics, and atmospheric chemistry interact, capturing for instance how a heat dome suppresses ventilation and allows pollutants to accumulate. More striking still is the arrival of artificial intelligence in numerical weather prediction. The review highlights machine learning systems that have achieved skillful medium-range global forecasts and, more recently, probabilistic ensemble forecasts generated entirely by neural networks, alongside hybrid approaches that combine physical chemical transport models with machine learning bias correction to deliver high-resolution particulate matter predictions. Interpretable models such as random forests equipped with SHAP explainability are also being used to unpick which environmental variables drive ozone formation and health outcomes, giving forecasters both accuracy and a measure of transparency.

These forecasting advances feed into a third strand of technology: unified risk indices. Rather than issuing a heat alert and an air quality alert independently, several research groups have constructed combined indices that merge temperature and multiple pollutants into a single health-relevant number, validated against mortality in places as varied as Monterrey, Taiwan, and Beijing. Studies building graded heat-health risk forecasts with full-season coverage across China demonstrate that such integrated products can be produced at national scale. The review treats these indices as a crucial bridge between the raw machinery of forecasting and the blunt reality of public communication, because a single number with a clear protective message is far easier to act on than two parallel warnings that may never be reconciled.

The final stage, risk communication and warning triggers, is where the review finds both encouraging innovation and stubborn weaknesses. Impact-based forecasting, now promoted by the World Meteorological Organization, shifts the emphasis from describing what the weather will be to describing what it will do, and evaluation studies from South Korea and New Zealand show that warnings framed around expected impacts measurably improve risk perception and protective behaviour. Probabilistic trigger rules drawn from forecast-based financing allow humanitarian agencies to release funds before a disaster strikes, and research on visual and verbal communication of uncertainty shows that well-designed probabilistic messages help decision makers rather than confuse them. Newer experiments, including digital heat warning platforms for older adults and generative AI chatbots tailored to multi-lingual communities, hint at how compound warnings might one day reach vulnerable individuals directly.

Yet despite this rapid progress at every stage, the review’s central finding is a persistent integration gap. Definitions are developed by epidemiologists, forecasts by atmospheric modellers, and warning triggers by disaster management agencies, usually without shared targets or common validation. A compound event may be defined one way in a research paper, forecast another way in an operational model, and trigger a warning under a third set of rules entirely. Uncertainty estimates produced by ensemble forecasts rarely survive the journey into the public warning product. The authors argue that this fragmentation means the impressive technical capabilities documented across the literature are not being converted into coherent, end-to-end systems, and that the missing ingredient is a shared compound-risk target that all three stages are explicitly designed to serve.

To remedy this, the review proposes both a taxonomy and a research agenda. The taxonomy classifies compound-event generation methods by how they combine hazards, whether through threshold overlap, statistical interaction models, coupled physics-based simulation, or AI-driven fusion, and by how they connect to impacts and triggers. The research agenda calls for uncertainty-aware model fusion so that probabilistic information flows unbroken from forecast to warning; for impact-based warning thresholds calibrated against health and service outcomes rather than convenient percentiles; and for end-to-end validation that tests not whether a model predicted a hazard accurately, but whether the resulting warning changed behaviour and reduced harm. The authors also stress equity considerations documented in the underlying literature, including evidence that the health effects of compound heat and pollution fall disproportionately on historically marginalised neighbourhoods, and that people-centred design remains essential even as the technology becomes more sophisticated.

The timing of this synthesis is significant. Under the United Nations Early Warnings for All initiative, governments worldwide are being pressed to expand multi-hazard early warning coverage, and climate change is steadily increasing the frequency with which extreme heat and air pollution coincide over populated regions, from the Pearl River Delta to California, Delhi, and London. The review’s message to policymakers is that simply adding more hazards to existing single-hazard systems will not suffice; the architecture itself must become compound-aware. If the integration gap can be closed, the authors conclude, the technological ingredients for warnings that genuinely reflect the compound risks people face already exist. What remains is the harder work of connecting them.

Subject of Research: Technological advances in generating compound heat and air pollution events within early warning system frameworks

Article Title: A systematic review of technological advances in compound-event generation within early warning system frameworks

Article References: Li, Z., Fong, C. S., Ab Hamid, S. H., Aghamohammadi, N., Jamali, S. N., & Sulaiman, N. M. (2026). A systematic review of technological advances in compound-event generation within early warning system frameworks. Air Quality, Atmosphere & Health, 19(9), Article 204. https://doi.org/10.1007/s11869-026-02091-5

Image Credits: AI Generated

DOI: 10.1007/s11869-026-02091-5

Keywords: compound events, early warning systems, extreme heat, air pollution, impact-based forecasting, machine learning, risk communication, public health, climate risk, heat-health warning, coupled meteorology-chemistry models, systematic review

What unfolded across China in the summer of 2026 was not a string of isolated bad luck. It was, according to the scientists and commentary accompanying the events, a demonstration of a new normal: extreme events once labeled “once-in-a-century” are increasingly compound, concurrent, and geographically dispersed, arriving faster than the institutions and infrastructure built to manage them. A warming trend that outpaces the global average, as officially acknowledged by Chinese meteorological authorities, is raising both the frequency and the intensity of extremes. When a single season delivers typhoon remnants to a northeastern inland province, catastrophic urban flooding to central China, and near-record heat to the northwest simultaneously, the premise that hazards arrive one at a time, separated by decades of recovery and planning, collapses entirely.

The term “once-in-a-century” is shorthand for an event with a 1 percent chance of occurring in any given year, a statistical construct known to hydrologists as the hundred-year return period. Specialists have long recognized that this language is easily misinterpreted: it describes a probability, not a promise that such an event will occur only once every hundred years. In any given century, a 1-percent-annual-chance flood may strike twice, three times, or not at all, and independent river basins can each experience their own hundred-year event in the same year. Yet the intuitive public reading — that a community has earned a century of safety after surviving one such flood — persists, and it shapes how residents, officials, and even engineers respond when the waters return far sooner than the label implies.

Climate change has rendered the historical baseline on which those return periods rest effectively obsolete. The statistics embedded in flood maps and design standards assume stationarity, the idea that the range of natural variability observed in the past will continue into the future. That assumption no longer holds. On 20 July 2021, Zhengzhou recorded 201.9 millimeters of rainfall in a single hour, a new national record, with three-day cumulative rainfall of 617.1 millimeters, approaching the city’s entire annual mean. The deluge overwhelmed a city that had, in part, been built under China’s “sponge city” programme, an ambitious urban design initiative intended to absorb and reuse stormwater rather than shunt it into overburdened drainage networks. Four years later, in August 2025, Zhengzhou’s flood-control emergency response was again raised to Level III, with work, business, classes, and public transport suspended citywide. The hundred-year flood had taken four years to make its second appearance.

These events carry direct and severe health consequences, from drowning and traumatic injury to the disruption of routine and emergency medical care. Globally, public health systems have developed genuine capacity to respond to the acute aftermath of a single flood. China’s Centers for Disease Control and Prevention, for example, issues detailed post-flood guidance on indoor mold identification and remediation, with explicit precautions for vulnerable groups including pregnant women, children under 12, adults over 65, and people with asthma. This is well-designed, evidence-based guidance. But it is guidance calibrated to a single, bounded event: dry the building, remove saturated porous materials within 48 hours, ventilate, and protect vulnerable residents during cleanup. It says nothing about a household, or a health system, absorbing a second or third such event within years rather than decades, nor about heat and flood advisories issued in the same week to the same population. Adaptation guidance built for isolated disasters, however rigorous, is necessary but insufficient for compound, recurring ones.

The problem is not merely communicative but structural. Reservoirs, drainage systems, evacuation routes, and emergency stockpiles are sized against historical exceedance probabilities, and insurance and recovery budgets are paced to the expectation of long intervals between catastrophic losses. When extremes cluster, recovery is truncated. Communities still repairing one flood face the next before foundations have dried, and the cumulative physical and psychological burden of repeated displacement and rebuilding accumulates in ways that single-event frameworks never anticipated. Meanwhile, simultaneous hazards compete for the same response capacity: the personnel, shelters, and supply chains needed for flood evacuation in one province are the same resources a heat emergency in another might demand, and heat itself compounds flood risk by pre-conditioning drought, hardening soils, and altering atmospheric moisture available for subsequent downpours.

Three shifts follow from this pattern, argued in commentary published in The Lancet Regional Health – Western Pacific by John S. Ji and Qihong Deng. First, risk communication should retire probabilistic century-scale language in public-facing materials in favor of terms that convey trend and recurrence, such as “increasingly frequent severe rainfall” rather than “once-in-a-century flood,” so that recurrence within a decade is not received as a statistical anomaly requiring no institutional response. The words officials choose shape whether a second hundred-year flood in four years is read as a freak coincidence or as evidence of a trend demanding action. Second, infrastructure and early-warning design standards, including within the sponge city and reservoir safety programmes, should incorporate non-stationary risk models that use recent-period distributions rather than full-record historical baselines, which mathematically dilute the recent shift toward higher-intensity rainfall. A century of rain-gauge data in which the wettest years cluster in the last decade tells a very different story than the same record averaged across its full length.

Third, public health guidance should evolve from single-event response protocols toward compound-event and cumulative-exposure frameworks, addressing populations facing sequential or simultaneous heat and flood risks within the same season, as well as the mental and physical health burdens of repeated displacement and recovery. That means modeling the health effects of mold exposure after repeated inundation, planning clinically for populations that experience flooding and extreme heat within weeks of one another, and treating the psychological toll of chronic disaster exposure as a first-order public health concern rather than an afterthought. The tools exist — epidemiological surveillance, seasonal forecasting, and increasingly sophisticated artificial intelligence-based forecasting and digital simulation systems that Chinese cities are already deploying — but they must be aimed at the compound problem, not the singular one.

Experts and meteorological authorities have officially acknowledged the warming trend and the rising frequency of extremes, and there are encouraging signs of technical adaptation in forecasting and alerting. Public health guidance and infrastructure standards, however, continue to speak as if these events remain rare. This summer made the mismatch impossible to ignore. Flood warnings in the northeast and heat alerts in the northwest arrived in the same week, and a city rebuilt to absorb the once-unthinkable watched the once-unthinkable return within a political term. The climate has changed. The language, and the standards built upon it, need to change with it.

Subject of Research: Compound climate extremes and the obsolescence of once-in-a-century risk communication

Article Title: Compound climate extremes and “once-in-a-century” risk communication

Article References: Ji, J. S., & Deng, Q. (2026). Compound climate extremes and “once-in-a-century” risk communication. The Lancet Regional Health – Western Pacific, 74, Article 101973. https://doi.org/10.1016/j.lanwpc.2026.101973

Image Credits: AI Generated

DOI: 10.1016/j.lanwpc.2026.101973

Keywords: compound climate extremes, once-in-a-century flood, risk communication, non-stationary risk, sponge city, Zhengzhou rainfall, Typhoon Bavi, Tropical Storm Maysak, public health preparedness, extreme heat, climate change, flood resilience

The research team, led by Lei Lin of the Southwest University of Political Science and Law, together with Jing Tan of Chongqing University and Di Zheng of the Sichuan Academy of Social Sciences, recruited 599 urban residents in China for two randomized online experiments. The choice of population was deliberate. China has more than 900 million urban permanent residents, a group that overlaps heavily with the audiences of official government microblogs, weather applications, and other digital warning channels, and these users are among the most frequent adopters of AI-enabled services. The experimenters used Credamo, a survey platform with verified users, and applied strict quality controls: eligibility was restricted to the highest credit ratings, duplicate submissions from the same IP address or device were blocked, minimum completion times were enforced, and attention and comprehension checks filtered out careless respondents. Of 636 initial responses, 37 were excluded under preregistered criteria, leaving a final sample whose demographic profile shifted negligibly after cleaning.

The scenario at the heart of both experiments was a heavy rainfall warning issued during peak flood season in a Chinese city, a hazard chosen for its familiarity, its low political sensitivity, and the fact that both government agencies and AI platforms plausibly issue such alerts. Warning severity was communicated through the standardized color-coded signals of the China Meteorological Administration, ranging from blue for general risk through yellow and orange to red for extreme risk. All experimental materials were identical in content, length, and style; only the source cue varied. Pilot testing confirmed that participants could accurately identify who was said to have issued each warning and perceived the scenarios as realistic.

The first experiment employed a three-group between-subjects design comparing a government-only warning, an AI-only warning, and a consistent joint warning issued by both. Three hundred valid participants were randomly assigned, 101 to the government condition, 100 to the AI condition, and 99 to the joint condition, and balance checks confirmed no pretreatment differences across groups on gender, age, education, income, occupation, or disaster experience. A one-way analysis of variance revealed a substantial effect of warning source on information trust, with a partial eta squared of 0.215, a medium-to-large effect. Post hoc comparisons using Tukey’s honestly significant difference test showed that trust in government warnings significantly exceeded trust in AI warnings, with a mean difference of 0.29 points on the five-point scale and a Cohen’s d of roughly 0.45.

More striking still was the performance of the joint condition. Warnings attributed jointly to the meteorological station and an AI-driven warning platform produced the highest trust of all, exceeding the government-only condition by 0.26 points and the AI-only condition by 0.55 points, the latter corresponding to a Cohen’s d of approximately 0.85, a large effect. The authors interpret this as evidence that credibility emerges not from any single source but from the combination of institutional authority and technological channels. When AI alerts appear alongside official government communication, they are more likely to be perceived as complementary rather than autonomous, easing concerns about algorithmic opacity and unclear responsibility. The finding echoes the long-standing ‘speak with one voice’ principle in risk communication, which emphasizes coherent messaging across sources to sustain public confidence.

The experiment also confirmed that trust translates into action. In a hierarchical regression controlling for gender, age, education, prior disaster experience, AI usage, institutional trust, and risk perception, adding information trust to the model raised the explained variance in protective intentions from a nonsignificant 2.9 percent to 12.7 percent. Information trust was the strongest predictor of intentions to monitor hazard information, move vehicles and belongings to safety, identify emergency shelters, and stockpile supplies, with a standardized coefficient of 0.416. Risk perception retained a smaller but significant effect. Protective intention items showed acceptable internal consistency, with Cronbach’s alpha of 0.72, while the trust and anticipated regret composites reached 0.82 and 0.84 respectively.

The second experiment turned to the harder question: what happens when government and AI disagree? Four conditions were constructed. In one, both sources issued a low-level blue warning; in a second, both issued an orange warning, serving as a reference and replicating the joint condition from the first study. In the two conflict conditions, one source issued orange while the other issued blue, with the order of severity reversed between the two versions. After excluding invalid responses, 387 participants were distributed almost evenly across the four groups, and randomization checks again showed no meaningful imbalances. A pretest had verified that a one-level discrepancy was perceived as a genuine conflict and that orange was consistently read as more severe than blue.

Conflict, it turned out, produced a powerful emotional response. A one-way analysis of variance revealed a large effect of warning consistency on anticipated regret, with a partial eta squared of 0.349, and a nonparametric Kruskal-Wallis test corroborated the result. Participants in both conflict conditions reported significantly higher anticipated regret than those in either consistent condition, exceeding the consistent-low group by roughly 0.85 to 0.91 points and the consistent-high group by 0.33 to 0.39 points, all differences highly significant. Crucially, the two conflict groups did not differ from each other, meaning it did not matter whether the government or the AI issued the more severe alert. When sources diverged, people temporarily set aside questions of institutional authority and focused instead on avoiding the cost of underestimating the threat, a pattern consistent with worst-case thinking described in regret theory and loss aversion research.

Anticipated regret also proved to be a formidable driver of behavior. In a hierarchical regression, adding anticipated regret to the controls raised explained variance in protective intentions from a nonsignificant 1.7 percent to 26.8 percent, with a standardized coefficient of 0.502, meaning a one-unit increase in expected regret corresponded to roughly half a unit more protective intention, all else held constant. Together, the two experiments support a dual-pathway framework of decision making in multi-source warning environments: under consistent messaging, responses are structured by cognitive trust grounded in institutional authority and technological support; under conflicting messages, responses are instead guided by the emotional motivation to avoid future regret.

The implications for warning system design are substantial. Because joint government-AI releases enhance confidence, the authors suggest that policymakers develop coordination mechanisms between meteorological agencies and AI service providers, aligning update schedules, severity thresholds, and release timing to reduce discrepancies before they reach the public. At the same time, they caution against exploiting regret as a persuasive tool, which could breed anxiety or distrust; instead, communication strategies should explain why different systems produce different alerts and offer graded recommendations for how to respond. The study acknowledges limitations: the simulated scenarios cannot fully reproduce the urgency of real disasters, the sample of urban online users limits generalizability, and the two experiments examined trust and regret in separate randomized conditions rather than estimating a unified causal pathway. Nonetheless, as AI becomes an increasingly visible participant in public warning dissemination, the research makes clear that institutional credibility remains the anchor of public confidence, and that when machines and governments speak in harmony, people listen most closely of all.

Subject of Research: Public trust and behavioral responses to disaster warnings issued by governments versus artificial intelligence systems.

Article Title: Public Responses to Disaster Warnings from Government Versus AI: Evidence from Behavioral Experiments

Article References: Lin, L., Tan, J., & Zheng, D. (2026). Public Responses to Disaster Warnings from Government Versus AI: Evidence from Behavioral Experiments. International Journal of Disaster Risk Science. https://doi.org/10.1007/s13753-026-00766-2

Image Credits: AI Generated

DOI: 10.1007/s13753-026-00766-2

Keywords: disaster warnings, artificial intelligence, government trust, anticipated regret, risk communication, behavioral experiments, early warning systems, protective behavior, algorithm aversion, flood warnings, China, decision making