Light in the forest

If a tree falls in the forest, it can create an opening for more incoming light. And that makes a significant impact on the surrounding environment, according to new research. 

An international science team used supercomputer simulations to model forest population dynamics of tree diversity in tropical forests. The researchers hope their work will help support forest management efforts and aid in conservation.

“Our work shows that competition for light following gaps created by the death of large trees alone can support the critical diversity of trees found in tropical forests,” said Damla Cinoglu, a postdoctoral researcher at the O'Dwyer lab at the University of Illinois at Urbana-Champaign.

Roughly equal environmental conditions of soil and sunlight give rise to a wide variety of trees in terms of fast and slow growth, short and long lifespans, and more. Reporting in the July 2025 issue of the British Ecological Society’s Journal of Ecology, Cinoglu and colleagues reveal what drives the persistence of this remarkable diversity.

An Ecologist’s Dream

The researchers developed a model to assess whether random, small-scale gap disturbances and subsequent competition for light can support long-term coexistence. They found that the competition for light provides a competitive relief for fast growing tree species to fulfill their demographic advantage over shade-tolerant slow growing tree species. 

Cinoglu and colleagues based their model on data observed in demographic groups in the Barro Colorado Island forest dynamics plot, located in Panama and having the distinction as “the most intensively studied tropical forest in the world," according to the Smithsonian Tropical Forest Institute. Like in an ecologists dream, every tree above waist height in a 50-hectare plot has been tagged and mapped since 1980.

Seeing the Forest Through the Tree Models

“Our model was a simple, mechanistic forest dynamics model that relied on the assumption that trees are able to move their crowns to reach sunlight,” Cinoglu explained. 

In addition to modeling the basic processes of growth, death, recruitment, and dispersal, the researchers simulated a "community of communities" landscape, introducing random canopy gaps disturbances and the resulting competition for light that shapes how tree crowns are arranged within each patch. 

The researchers relied on forest dynamics census data and empirical parameters previously calculated from data by Rüger et al., 2020 from the forest dynamics plot in Barro Colorado Island, Panama. 

Their simulation experiments include niche and neutral versions of the model, where the niche models relied on the differences in life history among demographic groups, and the neutral models assumed each species had the same exact demography. 

In demonstrating that life history differences led to longer persistence times compared to neutral models, the researchers faced a major computational hurdle — to run the model long enough and as many times as needed, to be able to identify the distribution of extinction and persistence

“We had a fantastic experience working with the TACC team,” she added. “Access to TACC systems allowed us to run multiple models simultaneously through parallelization, greatly speeding up our simulations. The TACC staff also provided technical support in setting up and optimizing our modeling environment.”

The Path Forward

According to Cinoglu, researchers in the Farrior Lab at UT Austin are building on this work by investigating what drives the different life strategies seen in the data, and whether that diversity follows rules scientists can predict.

 “Trees are long-lived organisms, and many forest processes take centuries or even millennia to unfold," Cinoglu said. "This makes it difficult to test ecological questions through experiments and observe their results in our lifetime. Supercomputers allow us to run these simulations at the scale and replication needed to draw robust conclusions."

The study, "Small disturbances and subsequent competition for light can maintain a diversity of demographic strategies in a neotropical forest: Results from model–data integration," was published July 2025 in the British Ecological Society Journal of Ecology. The study authors are Damla Cinoglu of the University of Illinois Urbana-Champaign; Nadja Rüger of the University of Leipzig and the Smithsonian Tropical Research Institute; Robin R. Decker and Caroline E. Farrior of The University of Texas at Austin. Funding was awarded by the National Science Foundation DEB 1939559.