In a global first, climate scientists at France’s Laboratoire des Sciences du Climat et de l’Environnement have designed a model that could allow future researchers to calculate how close tree species—or entire forests—are to irreversible damage or death from climate hazards.

Called the growth–ruin model, the new framework, recently published in Earth Systems Dynamics, has the potential to vastly increase scientists’ capacity to predict the relationship between forest loss and climate hazards like droughts and heat waves. That science has been limited to date by the fact that “most studies on tree death are based on direct or indirect observations of the behaviour and growth parameters of trees.”

To estimate the risk of forest collapse, the researchers adapted a concept from the so-called “ruin theory” used in the insurance industry “to describe the probability that a company that grows with regular income will become bankrupt due to external hazards.”

One key parameter guiding the study is that drought and heat stress have a cumulative impact on individual tree species and whole forests, thus lowering their resilience to future extreme weather events.

The other factor is that trees, unlike annual plants, contain large reserves of non-structural carbohydrates (NSCs), which allow for the rapid replacement of leaves. Critically, trees devote a certain amount of their annual productivity each year to accumulating NSCs and, when healthy, will always produce more NSCs than required to initiate leafing in the spring.

However, because there is competition between NSC production and plant growth, a tree “does not accumulate reserves indefinitely; the reserves tend to reach an optimum level.” And repeated exposure to drought and heat stress (which can cause premature leaf drop, among other things) will inexorably eat away at this NSC reserve. 

“Allocation to NSC reserves can hence be compared to an insurance system where, each year, part of the productivity is ‘paid’ to the insurance (i.e., NSCs for trees) which, in return, can be mobilized in the case of damage,” the study team explains.

Citing past studies, the authors write that “even if trees are not fully depleted of NSCs, there is a critical level of NSCs below which trees cannot recover, so they die.” Tree “ruin” can therefore be predicted at the point where “the amount of NSCs drops to a critical level below which the probability of tree death in the short term becomes very likely.” 

In a critical finding, the authors determined that, once this threshold is breached, trees will die even during extended periods of optimum weather. 

The modellers also found that more frequent extreme weather events increase the chances of a forest shifting “from a no-ruin state to probable ruin”. That shift occurs “when the return period threshold of nine years is crossed.”

Noting that “summer heat waves and droughts occurred in 2003, 2006, 2018, and 2019” across Western Europe, the study authors warn that forests in the region may be “close to the threshold of positive ruin probability.”

The duration of extreme weather events also likely plays a role in tipping forests and individual tree species over the edge. The modellers found that heat waves extending beyond two weeks made tree ruin “significantly likely” within the next 100 years. Such a heat wave, they add, occurred in Europe in 2018.