In this Insight, Monique Weemstra, a postdoc at the University of Michigan talks about her latest research looking at the different root strategies of trees and shows how a model approach helps unveil the black box of forests.

About the paper

What’s the paper about?

In this paper, we examine how variation in several root functional traits affects tree performance. It is often assumed that an increase in specific root length (SRL; root length per unit root mass) is a key strategy to enhance soil resource uptake and consequently tree performance, as high-SRL roots have a large surface area at relatively low carbon (C) costs. However, high-SRL roots generally have shorter lifespans leading to resource losses to the plant, and the link between SRL and tree performance is not yet well established. In fact, recent studies emphasize that besides a higher SRL, there are several alternative root trait strategies to enhance resource uptake and hence, tree performance; for instance, the root surface area can also be enlarged by increasing root mass density (i.e. root mass per soil volume), but which requires C for root production. Here, we studied how tree performance is affected by increases in root mass density, SRL and concurrent changes in root lifespan.  

We used a mechanistic model that simulates the whole tree’s daily C balance (used as a proxy for performance) based on changes in SRL, root mass density and lifespan, the C costs and resource uptake benefits involved in changing these traits, and their interactions with aboveground functional processes (e.g. photosynthesis and evaporation). In addition, we tested some of our model outcomes against tree root trait data collected in a common garden forest in the Netherlands.

How did you develop this study?

This work originates from a field study that unexpectedly showed that trees increased their root mass density rather than their SRL on a resource-poor compared to a resource-rich soil. A following literature review revealed a trade-off between SRL and root lifespan across tree species. Given these results, we started wondering about the actual net uptake benefits and C costs of having high-SRL but short-lived roots, compared to having a high root mass density, at the whole-tree level.

Our question would have been difficult to address in the field, because root traits and tree growth are affected by multiple biotic and abiotic factors that are virtually impossible to control or correct in forests. Therefore, we incorporated root trait information in an existing aboveground tree performance model (Sterck et al. 2011,Tree Phys.), which then allowed us to test the sole effects of variation in SRL, root mass density and lifespan (via their impacts on crown processes) on plant performance by keeping confounding biotic and abiotic variables (e.g. tree size) constant. We tested whether our model simulations matched any patterns observed in our empirical root data. Our work illustrates how model and empirical approaches can complement each other to test and generate new hypotheses.

What is new about this paper?

A main novelty of our paper is its integrative approach. Plants functionally balance their above- and belowground traits and resources. Therefore, many plant trait studies examine above- and belowground interactions by correlating a single leaf and a single root trait, as species with acquisitive leaves (i.e. for example having a high SLA) must have acquisitive roots (i.e. high SRL) to match aboveground resource demands. However, above- and belowground resource supply and demand can be balanced through various, simultaneous trait adaptations at the phenotype level, rather than through changes in a single trait at the organ level. In contrast to such a correlative approach, our study integrates simultaneous changes in different root traits and crown properties to simulate tree performance. Although our model does not capture the complexity of trees in natural systems, it offers a new opportunity to explore how above- and belowground traits can be integrated at the whole-tree level and how this drives tree performance.

About the research

What are the next steps in this field?

First, more and more root trait studies call for a critical look at root traits and their actual function for plant performance. Fortunately, different research initiatives are on their way that link root traits and functioning and help (root) researchers decide which traits are relevant for answering different research questions, and how to measure them. In line with this, despite the development of large-scale root trait databases, there are only few data available on some important plant functional processes. For instance, root respiration and uptake rates and mycorrhizal traits – especially in natural systems. Getting more data would not only serve as model input but could also contribute to defining the mechanistic interactions among and between root and leaf traits that underlie our models.

To me, one of the most exciting next steps in (trait-based) ecology lies in finding ways to study (multiple) traits in an integrative way and establish their impacts at the phenotype level. Our study shows how models in combination with field data may prove a relevant approach to do so.

About the Author

How did you get involved in ecology?

Besides loving nature, I have always been interested in people, culture and human societies which is why I studied anthropology after high school. For my MA thesis, I studied the role of customary activities in remote indigenous economies in northern Australia. During this (amazing!) field work in Australia, I learned how much indigenous economies (as well as cultures, societies and personal wellbeing) are linked with ecology – for instance, the arts and crafts sector strongly depends on natural resources. This truly sparked my interest in ecological science, which is how I ended up doing research on inter- and intraspecific differences in root and other functional traits in relation to plant performance and environmental variation.  

What are you currently working on?

My previous postdoc ended last December (2019), where we studied (or better yet: are still studying…) intraspecific (co-)variation in leaf and root traits across multiple species along an elevation gradient in the French Alps. As of Mid-January 2020, I am working as a postdoc at the University of Michigan in the department of Ecology and Evolutionary Biology, with Dr Natalia Umaña and Dr Jenny Zambrano (University of Washington). In this project, we study interspecific differences in root and leaf traits along a soil fertility gradient in a tree community in Michigan and how these differences relate to tree performance.

What article are you most proud of?

I am proudest of this article and although it has frustrated me at times, it is now my favourite paper. In the first place because I think that our approach really adds something new to plant ecology: the use of an integrative model, the whole-tree perspective and showing the potential relevance of root mass density. Second, not being a modeler myself, I struggled getting this paper started towards the end of my PhD project. With my PhD deadline coming up, I was not too confident about being able to understand the model and interpret its outcomes. But with great support from, and discussions with my PhD supervisors, I really started to enjoy writing this paper about it.

Read the article here and the free plain language summary here.