Teresa Rosas: single traits are not enough to predict tree growth

In our new post Teresa Rosas, Talent and Gender officer at CREAF (Spain), presents her work ‘Are leaf, stem and hydraulic traits good predictors of individual tree growth?’, discusses the complexity of ecological relationships and shows that there is life for a PhD beyond academia.

About the paper

In the 21st century, humanity faces the huge challenge to adapt to rapid global change.

As ecologists, one of our main contributions is to increase our ability to predict changes in forest structure, composition and function. Since we do not have a crystal ball, we need to provide the ecological theory and tools to better understand these changes.  This paper is our five cents in that direction.

Rosas Teresa working in their COVID-19 office
Teresa Rosas working in their COVID-19 office

In this study, we assess whether commonly used traits, describing the functioning of leaves and the water transport capacity of wood, can be used to predict tree growth rates along a regional water availability gradient in a Mediterranean forest. We resample 90 Spanish forest inventory plots dominated by six of the most common trees in the region: Scots pine (Pinus sylvestris), black pine (Pinus nigra), Aleppo pine (Pinus halepensis), beech (Fagus sylvatica), downy oak (Quercus humilis) and holm oak (Quercus ilex). Our results show that individual traits, including some of the most widely used functional traits, are poor predictors of tree growth. Even when trait-growth relationships exist, they do not necessarily conform to simple hypotheses based on our understanding of organ-level processes and how they affect growth. The reason for this lack of relationship is because traits covary with each other in complex ways (for example, in our case more acquisitive leaves were associated with a lower capacity of the wood to withstand drought, which is a key stress factor in the study area) and frequently show compensatory responses to changes in the environment.

Overall, our study highlights that to understand trait-growth (and by extension trait-performance) relationships it is not enough to select traits closely related to physiological mechanisms and context-specific environmental drivers. We should also account for how traits covary to understand the variation of whole-tree performance along environmental gradients and, ultimately, to predict changes in forest demography in response to changing environmental conditions.

About the research

Over the last two decades, the classic ‘comparative ecology’ renamed in the eighties as ‘functional ecology’ has emerged as a refreshed discipline under the name of ‘trait-based ecology’, with the promise to turn ecology from a primarily descriptive science into a more mechanistic and predictive discipline.

Sampling the canopy of one of the study beech (Fagus sylvatica) forests. Credit: Carles Batlles
Sampling the canopy of one of the study beech (Fagus sylvatica) forests. Credit: Carles Batlles

When I started my thesis several years ago, it was surprising for me to realize that several foundational assumptions of trait-based ecology had not been rigorously tested, and yet if I wrote ‘functional traits’ in Google Scholar I got thousands and thousands of papers. The good news was that I had the opportunity to join Jordi Martínez-Vilalta and Maurizio Mencuccini’s lab to start my PhD journey under their supervision and I was able to join a project to examine one of those poorly tested assumptions: the expectation that traits are ‘functional’ to the degree that they determine individual fitness and can be used in the field to predict organismal responses to changes in the environment. In 2015, and after nine months of intense field and lab teamwork, we got the data to start exploring whether plant traits could simplify the complexity of ecology to the point that we could make (predictable) sense of it.

One of the main conclusions of our paper is that expecting that one single trait will predict individual performance across temporal, spatial and taxonomic scales is pretentious. Firstly, our results stressed that understanding the role of traits on growth will benefit by moving from single-trait approaches to a whole-tree approach, in which a wider set of traits are integrated. However, it should be noted that the fact that the main axes of plant variation do not necessarily translate into main plant ‘functional’ axes, together with the evidence that different trait combinations can provide similar individual fitness in a given environment, may limit our interpretations. Secondly, traits and their associated functions are invariably context-specific. Thus, adding environmental data and their covariation with traits into analyses can be an essential step towards a more predictive discipline.  Therefore, we should improve our efforts by selecting traits that are closely tied with the main environmental drivers in the study system of interest, although they can be more technically challenging to obtain.  Thus, perhaps it is time to step back a bit in order to have a closer look at the conceptual definition of ‘function’ in trait-based studies and limit its use to cases where the link between function and traits has really been tested.

About The Author

I have a degree in Biology and a PhD in Terrestrial Ecology. The challenge of understanding new things has always been my motor and ecology offered me the opportunity to go in-depth into the highly interconnected diversity of life that will never stop to fascinate me. I guess this is why I decided to pursue a career in research, and specifically, in ecology. However, during my thesis, I realized that what really makes my day are people and that I am more passionate about pushing for a healthier research culture, rather than the scientific details of a particular research project. For this reason, I decided to move away from the academic career but not from science and since I defended my thesis I‘ve been working as a Science manager. So yes, I’m a proud former researcher that although I spent an incredible time doing my thesis, I decided to happily quit academia (I feel that all efforts to remove the prejudice attached to a non-academic career will not be never in vain).

Currently, I am the Talent and Gender Officer at CREAF, the same public research centre located near Barcelona where I conducted my PhD. We have urgent challenges in front of us. We need to redefine the way in which we define excellence and success in academia because researchers’ wellbeing and research integrity, and thus, the knowledge that we are generating, are at stake. As ecologists, we all understand why diversity is important in an ecosystem and why introducing an endangered species in an ecosystem without creating favourable conditions for their survival makes no sense. A similar thing happens with gender and other diversity issues. I think we should stop encouraging women to do leadership training and fit in a hostile research system and put the attention on male allies that need to take action to build more inclusive workplaces where half of the population also feels welcome.

Read the research in full here

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s