Synchronizing speciation, extinction, and dispersal to island paleodynamics through Bayesian phylogenetics in a Hawaiian plant radiation

Islands are nature's most revealing laboratories. Isolated by vast stretches of ocean, they allow scientists to watch evolution unfold in real time, as species arrive, adapt, and sometimes vanish. To understand this process, researchers often look at the family trees of living creatures, tracing their ancestors back to see when and where they diverged. However, a major hurdle has long stood in the way: these family trees usually lack a reliable clock. Without fossils to mark specific moments in time, it is difficult to know if a group of species arrived on an island just a few million years ago or tens of millions of years ago. This uncertainty makes it hard to connect the history of life with the history of the land itself, which is constantly shifting, rising, sinking, and eroding.

A new study tackles this problem by creating a way to weave together the story of a plant's evolution with the geological story of the Hawaiian Islands. The researchers focused on Kadua, a group of flowering plants found only in Hawaii. By combining genetic data from modern plants with a detailed map of how the islands have changed over millions of years, they built a model that estimates when these plants arrived and how they spread without needing ancient fossils. Their work suggests that the timing of a plant's arrival and its subsequent explosion into new species is deeply tied to the life cycle of the islands themselves, rising and falling with the islands' own growth and decay.

The Hawaiian archipelago is a chain of volcanic islands stretching over a thousand miles across the Pacific. As the Pacific tectonic plate moves over a stationary hotspot of magma, new islands are born in the southeast and drift northwest, aging and sinking as they go. This creates a moving conveyor belt of land, where the youngest islands are high and volcanic, while the oldest are low, eroded, and often barren. For decades, biologists have wondered how this changing landscape shapes the life it supports. Do species arrive on the newest islands and move forward in time? Do they thrive when an island is young and full of fresh habitats, or when it is older and more complex? To answer these questions, the team turned to Kadua, a genus of about twenty-five species that has colonized every modern high island in the chain.

Previous studies had struggled to pin down exactly when Kadua first arrived in Hawaii. Estimates ranged wildly, from a recent arrival of about three million years ago to an ancient colonization over thirteen million years ago. This uncertainty made it impossible to test whether the plants followed the "progression rule," a theory suggesting that species move steadily from older islands to younger ones as new land appears. The researchers realized that to solve this, they needed a method that did not treat the family tree and the geological map as separate puzzles. Instead, they developed a new computational framework that treats them as a single, interconnected system.

This new approach, which the authors call TimeFIG, allows the model to learn from the data itself. Instead of forcing the family tree to fit a specific date based on a fossil that might not exist, the model looks at the genetic differences between the plants and the known history of the islands to figure out the most likely timeline. It asks: given the genetic relationships we see today and the way the islands have grown and shrunk over time, what is the most probable history of how these plants arrived and spread? The model also tests specific ideas about what drives evolution on islands, such as whether larger islands produce more species, whether distance makes it harder for plants to travel, and whether the age of an island affects how fast new species appear.

When the researchers applied this method to Kadua, the results offered a clearer picture of the past. The analysis suggests that the common ancestor of all modern Hawaiian Kadua likely arrived between 1.4 and 5.9 million years ago, most probably on the island of Kaua'i. However, the model also leaves open the possibility of an older arrival, perhaps on now-vanished islands that once existed to the northwest, such as Gardner and Necker. These ancient islands were once large and high, similar to the modern islands, but they have since eroded and sunk beneath the sea. The fact that no Kadua species live there today is not just missing data; it is a clue. The model treats their absence as evidence that if the plants were once there, they either went extinct or moved to younger, more hospitable islands.

The study also revealed how the physical characteristics of the islands influenced the plants' evolution. The strongest factor found was distance. The further apart two islands are, the less likely it is that a plant will successfully travel between them. This confirms a basic principle of island biology: isolation is a powerful barrier. The researchers also found evidence supporting the idea that plants tend to move from older islands to younger ones, though this pattern was not absolute. Some plants did move backward, from younger to older islands, suggesting that the flow of life is more complex than a simple one-way street.

Perhaps the most striking finding concerns the timing of when new species are born. The study suggests that the rate at which new Kadua species appear and survive is not constant. Instead, it peaks when an island is in its "middle age." When an island is very young, it is still growing and its habitats are limited. As it matures, it develops complex topography, diverse climates, and a variety of ecosystems, creating a boom in opportunities for new species to evolve. However, as the island continues to age, it begins to erode and sink, losing its height and its ecological complexity. During this decline, the rate of new species formation slows down, and extinction rates may rise. The data indicates that the highest rate of diversification for Kadua occurred on islands that were high but already beginning to show signs of decay, a sweet spot of ecological opportunity.

This pattern helps explain why the modern islands, which are at different stages of this life cycle, hold different numbers of species. The youngest island, Hawai'i, currently has the lowest rate of new species formation, but it holds the most potential for the future as it continues to grow. The oldest modern island, Kaua'i, is seeing a slight decline in its ability to generate new species, while the middle-aged islands of O'ahu and Maui Nui are currently sustaining a steady rate of diversification. The ancient Gardner and Necker complexes, which are now underwater or nearly so, show a net loss of species, a fate that likely befell any Kadua ancestors that once lived there.

The researchers acknowledge that their findings come with a degree of uncertainty, as the group of plants they studied is relatively small. However, their simulations show that the method they developed is robust enough to detect these patterns even in small groups. By integrating the history of the land with the history of life, the study provides a more rigorous way to test long-standing theories about how islands shape evolution. It moves beyond simple guesses about when species arrived, offering a dynamic view where the rise and fall of the land itself drives the rise and fall of biodiversity.

Ultimately, this work demonstrates that the story of life on an island cannot be told without the story of the island itself. The geological processes that build and destroy land are not just a backdrop for evolution; they are active participants, setting the pace for when species arrive, how they spread, and when they vanish. For the Hawaiian Kadua, the rhythm of their evolution is inextricably linked to the slow, relentless pulse of the volcanic islands, rising and falling in a dance of creation and destruction that has played out over millions of years.