In Insights we discover the story behind and beyond a recent publication in Functional Ecology: What inspired the authors to do the research, how the project developed, leading to the final publication and what implications their results might have on the scientific community and on society.

This week, Ellie Goud of the Department of Ecology and Evolutionary Biology at Cornell University, Ithaca, USA, discusses her recent paper on the use of non-destructive plant traits as a proxy for carbon fluxes from peatland ecosystems.

Can you briefly explain your field of expertise and why it is important?

I am primarily a plant physiological ecologist and biogeochemist, but I also draw from community ecology and phylogenetics. I am fundamentally interested in the causes and consequences of plant functional diversity and I work at individual plant, community and ecosystem scales. Understanding the underlying physiological mechanisms behind patterns of plant productivity, species distributions, and plant responses to environmental change are important to our knowledge of ecosystem processes and biogeochemical cycles, but also to our general understanding and appreciation of the amazing ways that plants have evolved to do what they need to do across the diversity of Earth’s environments.

Can you give a brief summary of the paper in layman’s terms?

In this study, we assessed the ability of four plant traits to predict carbon dioxide (CO₂) and methane (CH₄) fluxes, which are important greenhouse gases. We measured traits that did not require the plant material to be removed from the plant (‘non-destructive’ traits): leaf area, leaf longevity, growth form, and aerenchyma tissue (conduits for gas exchange in plant roots and stems). We did this study in a temperate peatland in Ontario, Canada. Peatlands, including bogs and fens, are key features of the northern landscape that support an impressive diversity of plants adapted to the harsh, acidic wetland conditions, such as Sphagnum peat mosses, dwarf blueberry shrubs, sedges and carnivorous pitcher plants. Peatlands are also critical in the global carbon cycle, storing approximately one-third of the global organic soil carbon pool. However, the contribution of plant traits to peatland carbon cycling is relatively unexplored. We found that CO₂ fluxes positively related to leaf area and persistence, and negatively related to the proportion of woody species. CH₄ fluxes positively related to aerenchyma tissue and leaf area of sedges and rushes. The significance of trait-flux relationships differed based on whether data were averaged at the level of plot, species or microsite. We found that leaf area was the strongest predictor of CO₂ and CH₄ fluxes and is recommended in other systems where it is not ideal to measure traits destructively.

As you already mention in your introduction, the use of plant traits to explain processes in ecosystems has been taking flight recently. What inspired your research question and hypotheses for this publication? And why peatlands?  

This publication is part of a larger, ongoing research program at the Mer Bleue bog to understand peatland biogeochemical cycling under past, present and future conditions. This particular project was part of my master’s degree research and sought to understand the controls on CO₂ and CH₄ fluxes along the bog – margin gradient, which integrates changes in water table depth, water chemistry and plant species composition from the bog centre to beaver ponds along the edge of the peatland. My advisors Tim Moore and Nigel Roulet are experts in peatland carbon cycling and hydrology and I brought the plant biology perspective to the table. I was interested in the role of individual plant species and their characteristic traits in CO₂ and CH₄ fluxes along this gradient, which had not been previously explored at Mer Bleue. Why peatlands? I was first introduced to peatlands during an undergraduate Flowering Plant Diversity course at McGill (taught by Dr. Marcia Waterway), where we learned to identify plants in the field. I was immediately captivated by the plants that could survive in the harsh bog environment, especially the Ericaceous shrubs, and the beauty of peatlands as a whole. Being able to combine my personal affection for peatlands with my interest in plant physiological ecology made this project really fun and meaningful for me.

How did your ideas for this research evolve during the project? Did you run into difficulties, and if yes, how did you overcome them?

We were originally going to measure gas fluxes and plant species cover, not plant traits. The idea was to relate species composition, water table depth and peat temperature to carbon fluxes over two growing seasons. However, during the first two months in the bog measuring CO₂ and CH₄ fluxes from different microsites, it became very clear that species differed substantially in their flux rates and I wanted to know why. More specifically, I wanted to know if variation in plant traits was responsible for these differences in carbon fluxes, or if it was just due to differences in water table depth or temperature among sites. Destructive sampling of the plants in our sites was not an option, since we measured the same sites every other week (over two seasons). That is where the non-destructive traits came in! I would say the biggest difficulties that we had to overcome were the typical things you run into during field work – equipment not working, batteries dying and those unexpected rainy days that prevent you from measuring CO₂ exchange. The nice thing about plant ecology and measuring anatomical traits is that a lot can be done with a notebook, a pencil and a ruler!

Which new knowledge gaps did your research expose?

Generally speaking, we showed that leaf area is not only a good predictor of CO₂ flux, but also CH₄, which is pretty exciting. A lot of trait-based work has focused on CO₂ exchange, but being able to predict other greenhouse gases, such as CH₄ from plant traits is still being developed. For the peatland community specifically, we show that carbon cycling rates are not only controlled by abiotic variables (water, temperature), but are strongly tied to plant species composition and, importantly, the specific traits of those species.

Your research touches on the problem of scale in linking plant traits to processes. At some scale you find clear links with one of your parameters, whilst on another scale you do not. How would you think this problem could be solved?

I don’t know if there is any one solution to this problem. As we mentioned in the paper, the objectives of any particular study will dictate whether traits and other variables of interest are averaged across space and/or time or not. In this study, it was more accurate to use total leaf area to predict carbon fluxes, yet the community-weighted mean of leaf area was still ecologically meaningful. In other cases, discrepancies among scales of data aggregation could be more problematic depending on how those results will be interpreted and applied. For example, averaging trait values across very different plant species (mosses, rushes, shrubs) in this study was a source of discrepancy among our levels of data aggregation. Perhaps the best defense is a good offense – carefully thought out experimental design that specifically takes these issues into account before the data is collected.

At the end of your paper, the main conclusion if you like, you call for a better understanding on the relationships between above- and belowground traits. Could you develop a bit further on that? How would you see such research to evolve?

Predicting ecosystem respiration (ER) and CH₄ from plant traits is not as straight forward as predicting photosynthetic CO₂ uptake from traits because ER and CH₄ are influenced almost equally by both plants and microbes. These microbial communities are very sensitive to changes in their environment, which in turn influences soil and peat CO₂ and CH₄ emissions. In this study, we found leaf area and aerenchyma to be strong predictors of CO₂ and CH₄ fluxes, respectively, but suggest that leaf persistence and growth form do not sufficiently capture belowground plant and microbial processes that are important for ER and CH₄ fluxes. I think there is real potential in finding the links between measureable plant traits and below ground microbial processes at various scales, but a major challenge will be separating the abiotic (e.g., water, temperature) from plant effects (e.g., root exudates, competition). One way forward would be establishing a set of below ground traits that can either be feasible to measure in the field or that have reliable above ground proxies. Fortunately there is great work being done in this area in peatlands and other ecosystems! (see current research at Oak Ridge National Laboratory)

How do your findings translate to other ecological research disciplines?

Our findings contribute to the more general field of understanding the role of plant traits in regulating ecosystem services and can inform hypothesis testing in other systems. It’s valuable to demonstrate when traits behave as predicted (e.g., leaf area, aerenchyma), but the unexpected relationships open up new lines of inquiry. For example, the dichotomy between ‘evergreen’ and ‘deciduous’ as predictors of carbon cycling is useful and established in many systems, but did not hold up in our study because of evergreen rushes that were far more productive than the deciduous shrubs. Our study is also a nice example of merging plant ecology principles with biogeochemistry – these disciplines are often pursued independently of each other, but I think combining the two approaches is of mutual benefit to these fields and can shed light on the ‘bio’ side of biogeochemistry.

­­What would be your advice/message to conservationists or policy makers based on your results? Where should they put their effort if protecting or restoring peatlands?

Peatlands play an interesting role in the carbon economy – they can sequester significant amounts of atmospheric carbon over time, but can also release large amounts of stored carbon if they are perturbed. This balance between carbon sink and source is what drives so much of current peatland research, and can be of particular interest to land managers and policy makers. One application of our results could be in the selection of plant species for managing for methane emissions, as we identified certain species and plant traits that positively relate to CH₄ fluxes.

Although not explicitly discussed in our paper, protecting and restoring peatlands relies heavily on maintaining the right hydrology to support the unique peatland plant species and environmental conditions (acidic, saturated peat). Our results support a large body of work that demonstrate that peatland carbon cycling is particularly sensitive to changes in both temperature and hydrology (water table depth). Hydrologic disturbances, either draining or flooding, are major threats to peatland ecological integrity and a challenge in their restoration and conservation. The good news is that peatland restoration is active and successful in Canada and many other countries, both in terms of restoring plant communities and their carbon sink function.

As a final question, a bit more of a personal one. If you are not at work, researching peatlands, what do you like to do?

I balance my personal time between being a mom, spending time with friends and family, and being a traditional folk musician. Before pursuing a career in science, I was a music teacher and music still plays a very large role in my day to day. I have played the fiddle my whole life and am always involved in some sort of musical project. Since moving to Ithaca to pursue doctoral work at Cornell, I have had the great fortune of playing music with some excellent people. The main project that occupies my free time is the Celtic band Arise & Go that performs Scottish, Irish and French-Canadian music. When I’m not in Ithaca, I like to be back home in Atlantic Canada enjoying the sea, and helping out on my folks’ apple orchard.

Read the full article on predicting peatland carbon fluxes from non-destructive plant traits or the free plain language summary here.