"Our findings suggest that rainfall patterns drive the reproduction of many species, depending on the species' traits," Jesse Lasky of Pennsylvania State University and Columbia University, both in the US, told environmentalresearchweb. "Changes in rainfall patterns due to climate change may thus result in a shift in which species are performing best, and the traits that comprise the trees that persist."

Lasky and colleagues from Columbia University and Aarhus University in Denmark studied the Guánica State Forest in southwest Puerto Rico. This tropical dry forest experiences a mean annual temperature of around 25°C and yearly rainfall averaging 860 mm in two bursts – an early rainy season in April–May and a late rainy season in August–November. The mid-summer drought between these rains is expected to intensify with climate change; in 2015 the seasonal rains failed.

"Tropical forests are special in that you can find plants reproducing at any time of the year," said Lasky. "A big question is, why is there such diversity in the timing of reproduction, known as phenology? In dry forests, rainfall and drought are highly variable, and sometimes unpredictable, and probably play a major role in the timing of reproduction."

With that in mind, Lasky and colleagues looked for evidence of mechanisms that could influence the timing of tree reproduction: species interactions, between trees and other species like pollinators, and moisture limitation. Abiotic factors such as rainfall could encourage synchronization of plant reproduction, whereas interactions with other living things may favour synchronization or lead trees to flower at different times to create their own niche.

"Our study was notable in that it included high frequency – biweekly – monitoring of a whole dry forest community, i.e. all species that dropped flowers and seeds in our sampling baskets," Lasky said. "Our findings suggest that tropical trees may have distinct reproductive periods and that this separation may allow them to coexist."

The team measured the numbers of flowers, fruits and seeds falling from 45 species of tree by collecting them on a mesh, then used wavelet analysis to investigate the timing of the flower and seed production.

"We found that different drivers of phenology are likely acting at different temporal scales," said Lasky. "So that if you only focus on one scale of variation in reproduction over time, you might miss the impacts of an important driver, like species interactions."

The tree community flowered in synch on a seasonal scale (every 5–6 months) and on a short scale (around one month, after rain), the researchers found. On an intraseasonal scale of roughly 3 months, however, seed production occurred at different times for different species.

Some species traits were closely associated with the timing of reproduction in response to rainfall, suggesting they allow plants to reproduce in harsh periods of the year when conditions are dry.

Species of tree with large leaves, indicating that they’re sensitive to lack of water, had a reproductive peak in synch with the peak in seasonal rainfall. Species with dense wood, which are typically drought resistant, tended to flower when it was drier. And flowering of tall species and those with large leaves was most tightly linked to intraseasonal rainfall fluctuations on a two month scale.

Lasky and colleagues concluded that the tree community’s phenology has synchronous, asynchronous and compensatory dynamics depending on which timescale you look at. And some of these variations are due to seasonal fluctuations in abiotic constraints such as rainfall, as plants have different strategies for dealing with seasonal moisture shortages.

"However, these responses exhibit limits in response to extreme drought events," writes the team in Environmental Research Letters (ERL). "Change in seasonality is a major feature of global climate change. To better predict climate impacts on dry forest ecosystems, ecologists should build a community-level understanding of multi-scale phenological diversity and its functional basis."

The 2015 drought dramatically reduced reproduction across the whole community but the size of the drop for individual tree species did not seem linked to their properties.

Previous studies had tended to look at only one scale, or only one plant species. "In the 1970s there was a lot of interest in determining if the phenological diversity of tropical plant communities was patterned, that is, non-random," said Lasky. "Researchers were very interested in whether species reproducing at different times promoted their coexistence because they did not compete for a shared pollinator or frugivore [fruit-eater]. However, that research was largely focused on single timescales of analysis, which may have caused them to miss patterns in reproduction."

There’s also been study of the functional traits associated with reproduction at different times of year, Lasky says, but it has typically not been done using samples of whole tree communities.

Lasky is currently pursuing research to understand "the genetic and physiological mechanisms by which populations of dry forest trees adapt to their local precipitation and drought regimes".

Related links

Related stories