The work was led by Sharan Majumdar, a professor of atmospheric sciences at the University of Miami’s Rosenstiel School of Marine, Atmospheric, and Earth Science, in collaboration with scientists from ECMWF, the NSF National Center for Atmospheric Research, and the University of Bonn. Among the collaborators was Quinton Lawton, a former doctoral student of Majumdar’s who is now an assistant professor in the Department of Earth, Atmosphere and Environment at Northern Illinois University. The team focused their analysis on the Atlantic basin from 2020 through 2024, a five-year window that captures both steady improvements in conventional forecasting and the dramatic arrival of AI-based alternatives.

At the heart of the study is a comparison between two ECMWF systems. The first is the Integrated Forecasting System, or IFS, the center’s traditional numerical weather prediction model, which simulates the atmosphere by solving the physical equations that govern fluid motion, thermodynamics, and moisture on a three-dimensional grid. The second is the Artificial Intelligence Forecasting System, or AIFS, a machine-learning model trained on decades of historical atmospheric data to learn the patterns of atmospheric evolution directly. Rather than stepping forward through time with explicit physics calculations, the AIFS produces forecasts in a fundamentally different way, and ECMWF made it operational in 2025, marking a significant expansion of machine-learning-based forecasting alongside its physics-based numerical models.

The specific targets of the analysis were African easterly waves, the westward-moving atmospheric disturbances that form over sub-Saharan Africa during the boreal summer. These waves are the seedlings of many Atlantic hurricanes. As they roll off the West African coast and cross the tropical Atlantic, only some of them manage to organize into tropical cyclones, and identifying which ones will succeed is a central challenge of hurricane forecasting. The timing and location of this transition, known as tropical cyclogenesis, can determine how much warning coastal populations receive, which makes any improvement in lead time potentially life-saving.

The researchers first established a baseline by examining how the conventional IFS performed across the study period, and the results show genuine progress. In 2023, the grid spacing of the IFS ensemble was reduced from 18 kilometers to 9 kilometers, effectively doubling the model’s resolution and allowing it to represent smaller-scale tropical features with greater fidelity. Over the same years, the average probabilities of tropical cyclone formation produced by the conventional system generally increased, indicating growing confidence and skill in detecting developing systems. These improvements reflect the steady, incremental gains that operational forecasting centers have delivered through better observations, more sophisticated model physics, and finer computational grids.

But the more striking findings emerged when the team turned to the AI system. Among the 18 tropical cyclones that developed in the Atlantic in 2024, the AIFS ensemble frequently produced higher probabilities of development than the IFS at lead times of roughly 84 to 120 hours, which corresponds to about three and a half to five days before formation. This advantage was particularly pronounced for stronger tropical waves, the very systems that forecasters most want to flag early. In other words, when a disturbance showed signs of being a serious candidate for development, the AI model often recognized that potential days sooner than the traditional ensemble did.

The picture reversed at shorter lead times. At 36 to 48 hours before formation, the AIFS ensemble generally produced lower probabilities of development than the IFS, especially for weaker systems. This contrast highlights one of the fundamental challenges of forecasting tropical cyclogenesis: the signal that a disturbance will become a tropical cyclone can change substantially as the event approaches. A wave that looked promising five days out may struggle to consolidate, while a seemingly innocuous disturbance can rapidly organize. The study also found that conventional IFS probabilities often increased sharply when forecasts moved from three days to two days before a storm was officially named, and that there was considerable variation from storm to storm, depending in part on the wind-speed threshold used to define development. Cyclogenesis, it turns out, is not a single crisp event but a process whose detectability depends on both the system and the definition applied to it.

The AI system also showed advantages in forecasting where developing tropical systems would be located. The average position error of the AIFS ensemble mean was often smaller than the errors from the AIFS single forecast, the deterministic IFS, and the IFS ensemble mean. Position accuracy matters enormously in practice, because a disturbance’s location relative to warm ocean waters, wind shear, and dry air strongly influences its chances of survival, and because forecasters use position guidance to decide where to focus monitoring resources, aircraft reconnaissance, and public communications. An ensemble mean that consistently pins down the location of a developing system gives forecasters a more reliable anchor for their assessments.

Importantly, the findings do not suggest that AI should replace traditional numerical weather prediction. Majumdar emphasized that AI forecasts can complement the established IFS, providing forecasters with another source of information when assessing the potential for tropical cyclone formation. This framing reflects a growing consensus in the operational meteorology community: machine-learning models trained on historical reanalyses can capture large-scale patterns with remarkable efficiency and at a fraction of the computational cost of physics-based simulation, but the two approaches have different failure modes, and agreement or disagreement between them is itself informative. A forecaster seeing high development probabilities from both systems at a four-day lead time has a very different situation than one seeing a divergence between the two.

The study was supported by a National Science Foundation grant, AGS-2438140, and by the European Centre for Medium-Range Weather Forecasts, and it arrives at a moment of rapid institutional change. ECMWF’s decision to make the AIFS operational in 2025 signals that machine-learning forecasting has moved from research curiosity to production tool at one of the world’s leading centers. For Atlantic hurricane forecasting, Majumdar pointed out, the potential significance is substantial. Better identification of developing systems several days in advance, he said, could give forecasters more time to monitor disturbances, assess possible tracks and communicate emerging risks. The study results point toward a future in which traditional physics-based forecasting and AI-generated guidance work side by side or are combined, potentially giving forecasters a more complete picture of which Atlantic tropical waves are most likely to become the next named storm. As the 2020s progress and both classes of models continue to improve, the practical question is shifting from whether AI belongs in the hurricane forecaster’s toolkit to how best to blend its strengths with the physics-based systems that have anchored tropical cyclone prediction for decades.

Subject of Research: Artificial intelligence and numerical model forecasting of Atlantic tropical cyclogenesis from African easterly waves

Article Title: AI’s eye on the tropics

Article References: AI’s eye on the tropics. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: tropical cyclones, hurricane forecasting, artificial intelligence, ECMWF, African easterly waves, tropical cyclogenesis, ensemble forecasting, machine learning, Atlantic basin, Weather and Forecasting, numerical weather prediction, University of Miami

The first of the two studies set out to answer a question that has long interested developmental scientists: what actually happens inside a baby when a joyful exchange with a parent unfolds? To find out, the researchers measured changes in the heart rates of nearly 1,000 four-month-old infants as they engaged in positive interactions with their mothers. Electrodes were attached to the babies’ chests and stomachs, while trained observers watched the mother-infant pairs and recorded, moment by moment, whether each member of the pair displayed positive, neutral, or negative engagement. This combination of continuous physiological recording and fine-grained behavioral observation allowed the team to link what was happening between the mother and baby with what was happening inside the baby’s body.

The physiological measure at the center of the study was respiratory sinus arrhythmia, a form of heart-rate variability that is coupled with breathing. Scientists understand this variability to be a biological indicator tied to emotion regulation and to a person’s adaptability to different environments. In adults and in children, higher heart-rate variability in this range is generally associated with a more flexible capacity to respond to social and emotional demands. By calculating this measure from the infants’ heart-rate data, the researchers could ask whether moments of shared positivity were accompanied by shifts in the biological systems that support emotional self-regulation.

The answer was striking. The researchers found increases in the babies’ heart-rate variability precisely when the infants and their mothers were positively engaged at the same time. Crucially, the same physiological changes did not appear when only the baby was smiling, or when only the mother was showing positive emotion. Both partners had to be on the same positive wavelength for the effect to emerge. Daniel Messinger, a professor in the Department of Psychology at the University of Miami College of Arts and Sciences and a senior author on the paper, put it plainly: “They both have to be positive. When they’re both positive, particularly for longer periods of time, you see this increase in the baby’s heart rate.”

That mutual requirement may be the study’s most important insight. It suggests that emotional development is not simply a matter of a happy baby or an affectionate parent acting alone. Instead, the physiological states associated with emotion regulation appear to be recruited during genuine two-way exchanges, in which both the caregiver and the infant are sharing positive affect simultaneously. “Play is not just fun and games. It’s influencing the baby’s physiology,” said Messinger, who is also a professor of pediatrics. “There’s something about positive sociality that’s associated with this internal baby state. The baby’s physiology is linked to what’s going on in the interaction.” In other words, the interaction itself, not the mood of either partner in isolation, is what seems to matter.

The scale and design of the study lend weight to its conclusions. The research was part of a national study on drug exposure conducted at four sites across the country, including Miami, and involved several universities. When the team analyzed the heart-rate data, they controlled for prenatal drug exposure and social adversity, helping to ensure that the observed link between shared positivity and heart-rate variability was not simply an artifact of those factors. Among the members of the research team was Samantha G. Mitsven, a University of Miami alumna who worked on the study while completing her Ph.D. in developmental and child psychology. With nearly 1,000 infants contributing data, the study represents one of the larger physiological investigations of early mother-infant interaction.

The second study, also published in Philosophical Transactions of the Royal Society B, approached infant smiling from a different angle by asking how babies and adults influence each other’s smiling over time. The researchers observed 58 infants at two ages, 4 months and 8 months, while the babies interacted with their mothers, their fathers, and a stranger. This design allowed the team to compare smiling dynamics across different relationship partners within the same infants, revealing patterns that would be invisible in studies focused on mothers alone.

The results showed a clear division of labor in early smiling exchanges. Adults initiated and reciprocated smiling interactions at higher rates than infants did, and seeing an adult smile significantly increased the rate at which babies began smiling themselves. Babies, on the other hand, were more likely to be the ones who ended shared smiling moments, a pattern the researchers interpret as a possible sign that infants are learning to moderate their own excitement. The team also found developmental change: at 8 months, the babies were both more likely to start a smiling interaction on their own and more likely to reciprocate an adult’s smile compared with when they were 4 months old. “That suggests that the dyad becomes more dynamically responsive to one another at older ages,” said Messinger, who was a first author on the paper along with Eliala Alice Salvadori, who conducted the research for her dissertation at the University of Amsterdam, and Cristina Colonnesi, an associate professor at the University of Amsterdam. Another first author, Alan B. Cobo-Lewis, completed his postdoctoral fellowship and undergraduate studies at the University of Miami and is now an associate professor at the University of Maine.

Perhaps the most socially revealing finding came from the comparison across partners. At both ages, babies were most likely to return their mother’s smiles, compared with smiling interactions involving their fathers or an unknown adult. The adults who were strangers actually initiated and sustained smiles more often than the mothers did, yet the infants responded preferentially to their mothers. “This was, importantly, the first paper that involved not only moms and dads, but also strangers,” Messinger noted. “The strangers initiate and sustain smiles more than moms, but infants are more likely to smile back at the mom. So, the relationship really counts there.” The finding underscores that infant smiling is not a reflexive response to any friendly face, but a selectively tuned behavior shaped by the specific relationship between the baby and the person in front of them.

Together, the two studies sketch a picture of early emotional development as a genuinely interactive process. The heart-rate study shows that mutual positive engagement, sustained over time, is associated with increases in a physiological indicator linked to emotion regulation, while the smiling study shows that these exchanges become more fluid, reciprocal, and relationship-specific as infants grow from 4 to 8 months of age. For parents, the practical message is encouraging rather than demanding: ordinary moments of shared laughter and smiling are not idle pleasures but may be part of the biological work of early development. For scientists, the results open a window onto how social experience and physiology intertwine in the first months of life, suggesting that the foundations of emotional well-being are built not in solitude, but in the back-and-forth of a shared smile.

Subject of Research: The role of shared positive caregiver-infant interactions in infants' physiological and emotional development

Article Title: Shared smiles may shape infants’ emotional development

Article References: Shared smiles may shape infants’ emotional development. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: infant development, emotion regulation, heart-rate variability, respiratory sinus arrhythmia, mother-infant interaction, smiling, caregivers, developmental psychology, physiology, social interaction, University of Miami, Philosophical Transactions of the Royal Society B