The influence of community composition and trait distribution on biomass production in grassland ecosystems

(1)

The influence of community composition

and trait distribution on biomass

production in grassland ecosystems

Bachelor thesis

Author: Tom Scheltema

Student number: 12383074

Date: 30-05-2021

(2)

Abstract

Ecosystem services are dependent on the functioning of ecosystems, which in turn are highly dependent on biodiversity and species composition. The composition of plant communities is affected by the changing climate and more extreme weather events. Increasing evidence suggests that altered plant community composition influences plant functional traits and biomass production. However, a knowledge gap on how and to what extent plant community composition determines these parameters is indicated. This research aims to fill this knowledge gap by integrating the aspects of plant community composition, plant functional traits and biomass production. This is achieved by comparing five different plant community compositions with respect to evenness and plant identity. The research was carried out at the Science Park at the University of Amsterdam where 20 grassland mesocosms were harvested after a growth period of five months. The expression of six different root traits (average root diameter, root surface area, root volume, specific root length, root tissue density and root length density) and the aboveground, belowground and total biomass were determined per mesocosm. Links between community composition and root traits, as well as between root traits and biomass production and between community composition and biomass production were tested through a statistical analysis. The results suggest that the communities dominated by fast-growing species produced more biomass than communities dominated by slow-growing species, which is related to a higher root volume, root length density and root surface area. Furthermore, communities dominated by the forbs produced less biomass than communities dominated by the grasses, which is linked to a higher average root diameter. Lastly, the results show that communities with relative evenness show balanced values for root traits and biomass production relative to the dominated communities. Therefore, this research contributed to the further understanding of the fundamental links between community composition, root traits and biomass production.

(3)

Introduction

Ecosystem services are important for human welfare (Fisher, Turner, & Morling, 2008), e.g. through agriculture and carbon sequestration. These ecosystem services depend on the functioning of ecosystems, which stability is dependent on multiple factors including biodiversity and species composition (Deyn & Putten, 2005; Streitwolf-engel, Boller, Wiemken, & Sanders, 1998; Wallace, 2007). With the global climate changing and causing more extreme weather events, the structure of these plant communities changes (Orwin et al., 2010) while they endure increasing stresses. This shows the importance of plant

community composition, since it modifies plant productivity under extreme weather events (Kreyling, Wenigmann,

Beierkuhnlein, & Jentsch, 2008). But this also holds for less extreme circumstances. Naeem et al. (1996) found that decreased biodiversity in an ecosystem can lead to local changes in net primary production and therefore also for ecosystem properties. Biodiversity refers to the abundance and variety of species, plant species in this case, in ecosystems. It can be expressed in species evenness, which is a measure for how equal the numerical distribution of different species in an ecosystem is. The distribution of traits in a community is dependent on the identity and abundance of these present species, which influences ecosystem functioning (Long et al., 2019). Increased understanding of the processes that affect the functioning of ecosystems enhances the understanding of the behaviour of ecosystems and how they should be treated (Comas, Becker, Cruz, Byrne, & Dierig, 2013).

Community composition

Growing evidence suggests that plant functional traits are linked to belowground response to community composition changes (Fry et al., 2018). Furthermore, Grigulis et al. (2013) found that the plant traits that determine aboveground processes are strongly linked to microbial traits that determine belowground processes. This shows the complexity of these systems, since factors are intertwined and therefore influence each other. Thus, looking at plant traits at the level of individuals instead of the community level is insufficient (Mclaren & Turkington, 2010). However, a knowledge gap is indicated on which of these plant traits are relatively important for the functioning of the soil, and on the way these traits play a role in plant growth (Fry et al., 2018). Assessing the exact impact of the processes between plant traits on soil productivity also remains difficult because of this knowledge gap (De Deyn, Cornelissen, & Bardgett, 2008; Kalluri, Yang, & Wullschleger, 2020). For a further understanding of these processes, an integrated approach on the different factors is needed (Mahaut, Fort, Violle, & Freschet, 2020).

Previous studies indicated that plant traits belonging to a specific species can predict soil properties and ecosystem functioning in a community with only one species. These traits however don’t predict the influence on ecosystem functioning by change in composition in a mixed community (Long et al., 2019). This shows that community composition is a determinant of how plant traits affect ecosystem processes. Vitra et al. (2019) showed that for root traits the community weighted-mean is an indicator for community drought responses. Thus, a community as a whole influences ecosystem functioning by determining the expression of traits. Adjusting a plant community, e.g. by removing a functional group therefore also affects ecosystem properties, regardless of the species’ dominance (Mclaren & Turkington, 2010). Grigulis et al. (2013) also indicate ecosystem service trade-offs because plant traits and microbial properties affect ecosystem functioning through plant community composition. Plant community composition is an important factor for ecosystem functioning, because plant species do not only differ from each other, but also vary significantly within species due to the interactions between species and environmental factors (Bardgett, Mommer, & Vries, 2014). The findings by Mahaut et al. (2020) which suggest that direct effects of plant functional group identity (e.g. grasses or forbs) on biomass production is weak may therefore be different from this research because communities consisting of different functional group identities are considered. Moreover, when plant species with different traits are combined through coexistence, they can enhance ecosystem functioning compared to individual plant traits that are frequently conflictive (De Deyn et al., 2008).

Plant traits and biomass production

As plants are adapting to their environment, tradeoffs in root traits have to be made (Comas et al., 2013). For example under drought stress, a smaller root diameter, higher root tissue density and larger specific root length are favourable traits because they enhance water acquisition (Ibid.). Therefore, these traits are strongly linked to a plant’s biomass production. Plant traits are however not only reactive on the changing environmental conditions, since regional climate changes influence plant trait distributions (Madani et al., 2018), but also strong predictors for plant biomass production (Da et al., 2007; Madani et al., 2018). Moreover, Lipowsky, Weigelt, Buchmann, Schmid, & Schulze (2012) emphasize that, to comprehend how community

composition influences biomass production, identifying relevant root traits is indispensable. For roots in particular, the specific root length, with components root tissue density and root diameter (Ostonen et al., 2007), is a strong predictor for plant productivity (Mahaut et al., 2020). Root traits like root volume (Rose, Atkinson, Gleason, & Sabin, 1991), root diameter (Prieto, Stokes, & Roumet, 2016; Peter Ryser & Lambers, 1995) and root length density (Lynch, 1995) are examples of root traits which determine nutrient acquisition and therefore aboveground biomass production (Sadanandan Nambiar, 1990). However, aboveground and belowground productivity don’t necessarily correlate. This is because the diversity belowground can differ from the aboveground diversity, even though they are linked (Deyn & Putten, 2005). Therefore, effects of root traits on both aboveground and belowground biomass production will be considered separately in this resarch. The root traits furthermore determine the community composition of microorganisms like fungal species in the rhizosphere in temperate grasslands, which are highly important for e.g. plant productivity (Bergmann et al., 2020) and play a key role in maintaining plant biodiversity and therefore ecosystem functioning (Streitwolf-engel et al., 1998). New information on this complex interconnected system will

(4)

In sum, already a significant amount of relations have been indicated about which traits play a role in biomass production, but how exactly community composition and these traits affect biomass production together is still discussed. This leads to the following research question:

● How do plant community composition and root trait distribution affect biomass production in grasslands? For the plant community composition, species evenness and identity are considered. Root traits that will be considered are specific root length, root tissue density and root diameter, since they are known to be related to plants’ biomass production (Comas et al., 2013; Mahaut et al., 2020), and furthermore root length density, root volume and root surface area. A distinction is made between the community composition and the root trait distribution to firstly indicate relations with each other and consequently the relation of root traits to biomass production. Lastly, the relation between community composition and biomass production is explored. The research question can therefore be divided into three sub-questions:

1. How are root trait distributions affected by plant community composition? 2. How does root trait distribution affect biomass production?

3. How does community composition relate to biomass production in general?

As stated before, it is necessary to identify relevant root traits for understanding how community composition influences biomass production (Lipowsky et al., 2012). To what extent this influence is present is however highly questionable because of the current knowledge gap (Deyn & Putten, 2005; Kalluri et al., 2020). Because combined traits can enhance soil functioning through coexistence of plants (De Deyn et al., 2008), it is expected that community composition is a significant predictor for trait distributions. Since growing evidence suggests that plant traits influence both belowground (Bergmann et al., 2020) and aboveground biomass production (Da et al., 2007), it is expected that plant trait distribution is a significant predictor for biomass production. Particularly specific root length is expected to strongly predict biomass production (Mahaut et al., 2020). Plant functional group identity, which is considered for community composition, is found to be a weak predictor for plant productivity (Ibid.). The direct effects from plant community composition on biomass production are therefore expected to be limited.

Methods

Experimental setup

In order to compare different community compositions, 20 grassland mesocosms with different community compositions have been installed and grown from October 2020 under the same environmental conditions at the Science Park, an instance of the University of Amsterdam (Fig. 1). For the experiment, 4 different typical Dutch plant species have been used: Anthoxanthum

odoratum, Dactylis glomerata, Leontodon hispidus and Rumex acetosa. These species differ in growth strategy and functional

group type. A. odoratum is a slow growing grass, D. glomerata a fast growing grass, L. hispidus a slow-growing forb and R.

acetosa a fast-growing forb. The community composition can therefore be compared by plant identity and relative dominance

(Table 1). The species are present with different evenness but same species richness. For each species, a relative dominance community is set up with a 30 : 2 : 2 : 2 composition, as well as a relative evenness community with a 9 : 9 : 9 : 9 composition. This means 5 different community compositions. The plants are grown in the same natural clay soil sourced close to

Amsterdam and the mesocosms stand in the same area to prevent different climatic conditions. The communities are separated by pots to prevent eventual interactions between mesocosms. Mesocosms are a simplification of natural ecosystems and can be seen as representative for testing ecological hypotheses (Dzialowski et al., 2014). The pots have a volume of 43 L per pot. The seeds have been propagated in the greenhouse of the University of Amsterdam for ten weeks until the seedlings could be transplanted to the pots outside and are obtained from Cruydt-Hoeck’s Wild Plant Seeds & Flower Meadow Mixtures. This research is divided into 2 experiments, the first one to determine the aboveground and belowground biomass and the second one to determine the trait distribution of the different communities.

(5)

Table 1: Plant species, strategy and plant functional group identity and abbreviation

Plant community dominance Strategy + functional group identity Abbreviation

Anthoxanthum odoratum Slow growing grass DAO

Dactylis glomerata Fast growing grass DDG

Leontodon hispidus Slow growing forb DLH

Rumex Acetosa Fast growing forb DRA

Species evenness relative evenness E

Fig. 1

Separated mesocosms at the Science Park from the University of Amsterdam, experimental setup.

Experiment 1: Biomass

The aim of this experiment is to compare the biomass production in different plant communities. In order to do so, the aboveground and belowground biomass will be measured. Since different measurement methods apply to aboveground and belowground biomass, this section is divided into subsections. The total biomass production is calculated afterwards.

(6)

Aboveground biomass

To indicate aboveground biomass production, the shoot biomass is measured. First, the seedlings are cut to the soil level. The shoot fresh weight is measured immediately after harvest and dried at 60-70 degrees Celsius for three consecutive days. Finally, the shoot dry weight is measured.

Belowground biomass

For the belowground biomass, firstly three soil cores per pot are removed from the pots and are put in a storage at 4 degrees Celsius. To precisely measure the root biomass, the roots are then washed to remove all soil particles. Consequently, the roots are stored in 20% ethanol. This is because the roots are also used for measuring plant traits in experiment 2. It is important to mention that experiment 2 should be carried out before continuing with experiment 1 after the roots are stored in the 20% ethanol. The roots are dried by paper towels and consequently weighed to find the root fresh weight. Afterwards, the roots are dried at 60-70 degrees Celsius for three consecutive days. Finally, the root dry weight is measured.

Experiment 2: Traits

The aim of this experiment is to compare different root traits in different community compositions. This can be done through comparing community weighted means as stated before in the introduction. To measure these root trait values, a WinRHIZO root scanner is used. This apparatus performs analyses on root morphology (Seedlings, 2021). In this experiment, root length, root surface area, average root diameter and root volume are measured after the roots have been stored in 20% ethanol (see previous section), whereas root tissue density, specific root length and root length density will be calculated with these measurements afterwards. For the root volume, the software assumes the roots to be cylyndric (Ibid.). Since the same sample volume of soil is used for each pot, the distribution of root length and root length density will be identical. Therefore, only root length density will be discussed.

After washing the roots with demi water, the roots are spread out in a 25x20 cm tray. Clustering of roots is prevented by spreading them with tweezers, crossing roots are recognized by the root scanner. The tray is then put on the STD 4800 scanner, with EPSON Perfection software (Fig. 2). Consequently, the root image is acquired via WinRHIZO software with specific image acquisition parameters (Table 2). Consequently, the root image is acquired. The root scanner makes sure to avoid shadows and to do precise calibration. The sample positioning frame is chosen and the analysis is made, which takes a few seconds. The data is saved automatically after the analysis as ASCII text format which is easily readable by a lot of programs including Excel. The data is processed further in Excel.

Table 2: Image acquisition parameters WinRHIZO root scanner

Image type Grey levels

Resolution High (600 dpi)

Dust removal Medium

Fig. 2

(7)

Statistical analysis

After the data is collected it is converted to RStudio (RStudio Team, 2020) where statistical analyses are executed. An analysis of variance (ANOVA) is executed to compare community compositions in terms trait distributions (sub-question 1) and biomass production (sub-question 3). To determine which community compositions differ significantly from each other, a Tukey’s post hoc analysis is executed. For the plant community composition, plant identity and relative dominance are considered. Furthermore, both simple and multiple linear regression, using linear models, have been used to identify how root traits, the independent variables, predict biomass production (sub-question 2). The biomass production is hereby divided in aboveground, belowground and total biomass. The root traits are tested for predicting biomass production. Because all these traits are connected with each other, they can’t be seen as independent predictors. Therefore, they have all been compared with biomass production separately.

Results

Community composition and root traits

A correlation was found between community composition and root trait distributions. For root surface area, root volume, average root diameter, root length density, specific root length and root tissue density the assumptions for an ANOVA were met and were therefore able to be used for this research. It can be seen that fast growing species have larger values for the root traits than slow growing species, regarding root volume, root surface area and root length density (Fig. 3). The communities dominated by the fast-growing forb R. Acetosa (DRA) for example differed from the communities dominated by the slow-growing forb L. Hispidus (DLH) for root volume, surface area and root length density with a higher value for DRA (F(4,15) = 4.65, p < 0.0088; F(4,15) = 4.47, p = 0.013; F(4,15) = 5.11, p = 0.026). Roots in the DRA community have a higher root volume than those in the communities dominated by slow growing grass A.Odoratum (DAO) (F(4,15) = 4.65, p = 0.035). Furthermore, DLH and communities dominated by the fast growing grass D. Glomerata (DDG) are different from each other regarding surface area and root length density with a higher value for DDG for these traits (F(4,15) = 4.47, p = 0.044; F(4,15) =5.11, p = 0.0059). It can furthermore be seen that forbs have larger values for root average diameter than grasses. DDG, DAO and communities with relative evenness (E) for example differ from DLH regarding average root diameter, with a higher value for DLH (F(4,15) = 14.93, p = 0.00059; p = 0.00071; p = 0.043). Moreover, DRA has a significantly larger average root diameter than DAO (F(4,15) = 14.93, p = 0.00053). Interestingly, the evenly distributed community lies in between the values from the other communities for all traits. No significant differences were found between community compositions regarding specific root length and root tissue density (F(4,15) = 1.31, p = 0.31; F(4,15) = 0.99, p = 0.44).

(8)

Fig. 3

Plant community composition affects the expression of community root traits. Root surface area (a), average root diameter (b), root volume (c), root length density (d), specific root length (e) and root tissue density (f). DAO = slow growing grass, DDG = fast growing grass, DLH = slow growing forb, DRA = fast growing forb, E = relative evenness of species. On top of the diagrams, the letters indicate significant differences between communities per root trait. Boxplots with the same letter are not significantly different using Tukey’s comparison of all means. The other traits mentioned in the methods section did not meet the assumptions for normality and are therefore not included.

(9)

Root traits and biomass production

The same root traits as from the previous section have been used to determine the relationship with aboveground, belowground and total biomass production (Table 3). For aboveground biomass production, approximately 30-47% of the variance can be significantly explained by root volume, root length density and root surface area. Approximately 16% of the variance is

explained by average root diameter, which is however only marginally significant. For belowground biomass production, roughly 35-60% of variance is significantly explained by all root traits, apart from average root diameter, specific root length and root tissue density. Specific root length and root tissue density explain roughly 30% of variance, where root tissue density is however insignificant. For average root diameter, less than 1% of variance is explained and is moreover insignificant. For total biomass production, approximately 30-51% of variance is explained by all root traits apart from average root diameter and specific root length, which only explain roughly 5% of variance and are furthermore insignificant. Average root diameter and specific root length are furthermore negatively related to biomass production, meaning that a higher average root diameter and/or specific root length mean a lower biomass production. The impact from specific root length on biomass production is however found to be negligible. In addition, average root diameter and root tissue density show to relatively have most impact on biomass production (Fig. 4). The other linear models are in the Appendix. Root volume and root length density relatively only show little impact, whereas root surface area and specific root length project the least impact on biomass production.

(10)

Table 3: The role of root traits in determining biomass production as measured by simple linear regression. Coefficient is the result

of a linear model regressing the biomass production against the root traits, indicating the direction and impact of the relationship. Aboveground biomass

Independent variable Coefficient R^2 Significance

Average root diameter -1.1 0.16 0.079

Root volume 0.088 0.30 0.012

Root length density 0.041 0.46 0.00097

Specific root length 0.0002 0.022 0.55

Root surface area 0.0007 0.40 0.0026

Root tissue density 1.28 0.18 0.38

Belowground biomass

Root volume 0.086 0.51 0.00043

Specific root length -0.0006 0.26 0.026

(11)

Total biomass

Root volume 0.17 0.46 0.0011

Specific root length 0.0004 0.024 0.46

(12)

Fig. 4

Relationships according to simple linear regression between biomass production with root tissue density and average root diameter, which showed relatively the highest impact on biomass production. Biomass production is separated in aboveground, belowground and total biomass production.

(13)

Community composition and biomass production

The different community compositions have been used to explore relations with biomass production (Fig. 5). DLH and DRA differ between total biomass (F(4,15) = 3.16, p = 0.045). This means that regarding total biomass, the fast-growing forb produces more biomass than the slow-growing forb. DAO produces more aboveground biomass than DLH (F(4,15) = 3.52, p = 0.032). This means that the slow growing grass produces significantly more aboveground biomass than the slow growing forb. Furthermore, DLH also differs from DDG, DRA and E for aboveground biomass, however not significantly (F(4,15) = 3.52, p = 0.079; p = 0.089; p = 0.43). It can be seen that DLH produces the least biomass with the greatest difference between the other communities for aboveground biomass. A trend relation can furthermore be seen between slow-growing and fast-growing species, where slow-growing species seem to produce less biomass. In contrast, DAO, DDG and DRA are producing relatively much aboveground biomass. Moreover, for biomass in general, E lies approximately in between the other communities.

Fig. 5

Aboveground biomass, belowground biomass and total biomass productivity per community, in figure a, b and c respectively. DAO = slow growing grass, DDG = fast growing grass, DLH = slow growing forb, DRA = fast growing forb, E = relative evenness of species. On the top of the diagrams, the letters indicate significant differences between communities per root trait. Boxplots with the same letter are not significantly different using Tukey’s comparison of all means.

(14)

Discussion

This study aims to indicate links between plant community composition, root trait distribution and biomass produciton. How these factors interact is however yet to be substantiated. The different sub-questions are answered and integrated with each other to indicate these interactions in order to answer the main research question.

Community composition and root traits

Regarding sub-question 1, the results support the hypothesis that root trait distriubtions are affected by plant community composition. Fast-growing species are found to have larger values for root volume, root length density and root surface area compared to slow-growing species. Apparently, fast-growing species invest a lot in these root traits. High root length density is found to be an adequate strategy for plants for local competition for nutrients and water (Lynch, 1995; Ravenek et al., 2016). Since root surface area is also larger for fast-growing species, they seem to produce more roots to collect a lot of nutrients. The high root volume is also in line with research which indicated that a population with large root volumes grow faster than a popultion with lower root volumes (Rose et al., 1991). Furthermore, a larger average root diameter was found for forbs compared to grasses. Thicker roots have been associated with resource conservation (Prieto et al., 2016). In addition, a positive trend is indicated for root diameter and nutrient availability (Ryser & Lambers, 1995). This could indicate a difference in strategy regarding the use of resources between grasses and forbs, where forbs invested in thicker roots for higher resource conservation instead of investing in resource acquistion under conditions where nutrients are not a limiting factor. Further research should point out whether a larger average root diameter is a favourable trait under non-limiting conditions. The values for the root traits for communities with relative evenness were approximately in between the values of the other communities. Apparently, the root traits are balanced out under relative abundance of species. Since studies indicating effects of species evenness on plant traits are often contradicting each other (Orwin, Ostle, Wilby, & Bardgett, 2014), further research is needed to indicate specific relations. For root tissue density and specific root length, no differences between community compositions were indicated. This contradicts other research, which found root tissue density to be positively related to growth rate (Kramer-Walter et al., 2016), which would mean a difference between fast-growing and slow-growing species. The conditions under which the results from this research are gathered might have however been more favourable than the natural conditions under which the plants in this study have been grown (e.g. favourable average temperatures, watering regularly), which may have been a determing factor for root tissue density expression. In addition, Ryser (1996) related a high root tissue density to fast-growing species, explaining that this is necessary for the roots and leafs to expand fast. Further research should point out the role of community compositions on this strategy. Specific root length was found to be independent of plant growth strategy (Kramer-Walter et al., 2016), which is in line with these results.

Root traits and biomass produciton

With regards to sub-question 2, the results support the hypothesis that trait distribution affects biomass production. Root surface area, root length density, root tissue density and root volume were found to be able to predict biomass production and were moreover positively related, while specific root length showed a more or less negligible relation. As stated before, a larger root surface area and root length density are found to be effective traits in competition for nutrients and water which are fundamental elements for biomass production (Sadanandan Nambiar, 1990), while higher root volume was also found to be positively correlated to biomass production. Zhang et al. (2017) found a positive relation between root surface area and root length density with biomass production, which is also in line with the results. This suggests the positive relations are consistent with other studies. How exactly these traits affect biomass production however remains unclear (Yang, Zhang, & Zhang, 2012). For specific root length, the negligible relation is in contrast with other studies linking specific root length to increased biomass produciton (Ostonen et al., 2007; Mommer et al., 2011) and furthermore with the hypothesis that specific root length is an important predictor of biomass produciton. Kramer-Walter et al. (2016) however state that specific root length is independent from relative biomass production. Further research should therefore point out the exact role of specific root length on biomass production. Average root diameter negatively predicted aboveground biomass production with relatively large impact. This is in line with the previous section, where thick roots were linked to a relatively small investment in resource acquisition, which may explain the negative relation with aboveground biomass production since less nutrients are collected. Other studies indicated a positive relation between belowground biomass production and average root diameter (Cai, Li, & Jin, 2019; Cougnon et al., 2017). This is in line with the idea that aboveground and belowground biomass do not necessarily correlate, because of differences in aboveground and belowground diversity (Deyn & Putten, 2005).

Community composition and biomass production

A trend relation is indicated which suggests that fast-growing species produce more biomass than slow-growing species. In addition, Wilsey & Wayne Polley (2006) established that aboveground biomass production was higher on average for grasses compared to forbs. These findings are consistent with the result that the slow growing forb produces the least amount of biomass. With respect to sub-question 3, the results are in not line with the hypothesis that plant functional identity is a weak predictor for biomass production (Mahaut et al., 2020). Further research should point out whether the trend relation indeed projects reality by specifically assessing the difference in biomass production between fast-growing and slow-growing species. Moreover, albeit fast-growing species are projected to produce more biomass in this research, it has been indicated they have a smaller life span compared to resource conservative species (Ryser, 1996). Since slow-growing species are related to resource conservation, as has been stated in the previous sections, research with a longer time frame could indicate different

(15)

results regarding growth strategy and biomass production. The differences between aboveground and belowground biomass production within communities are also in line with research by Deyn & Putten (2005) who state that aboveground and belowground biomass production don’t necessarily correlate due to relative diversity, as stated in the previous section. Also, a study indicated a negative relation between species evenness and biomass production, which affect is found to be partly the result of the presence of large species (Mulder et al., 2004). This is in line with the results, since the communities of relative evenness seem to produce less biomass than dominated communities, apart from communities dominated by the slow-growing forb.

Linking community composition, root trait distribution and biomass produciton

Communities dominated by fast-growing species were found to have larger values for root volume, root length density and root surface area compared to communities dominated by slow-growing species, which traits are furthermore found to be positively related to biomass production. In addition, a trend relation indicated that communities dominated by fast growing-species produce more biomass than communities dominated by slow-growing species. Linking these results suggests that the

communities dominated by fast-growing species produce more biomass than communities dominated by slow-growing species, which is related to a higher root volume, root length density and root surface area. Furthermore, communities dominated by the forbs are found to have larger values for average root diameter than communities dominated by the grasses, which is therewith found to be negatively related to biomass production. Moreover, a trend relation is indicated showing that communities dominated by the forbs produce less biomass than communities dominated by the grasses. Comparing these results suggests that communities dominated by the forbs produce less biomass than communities dominated by the grasses, which is linked to a higher average root diameter. Finally, communities with relative evenness were found to be balanced out regarding root trait distribution and biomass production relative to the dominated communities. These results show that communities with relative evenness show balanced values for root traits and biomass production relative to the dominated communities. Further research should point out whether these indicated links are appliccable to a larger range of plant species and, most importantly, why these potential relations exist. The scale of the experiments should therefore be much larger in future research to point out potentially more significant differences and stronger links. Also, the chosen time frame could be expanded in order to to compare results for growth periods of different length. With the changing climate which increases stresses on these ecosystems, further research should also point out how the factors of community composition, root trait distribution and biomass production respond to these stresses.

Conclusion

Concluding, this research explored the links between community composition, root trait distribution and biomass production in grassland ecosystems. The results suggest that the communities dominated by fast-growing species produce more biomass than communities dominated by slow-growing species, which is related to a higher root volume, root length density and root surface area. The results furthermore show that communities dominated by the forbs produce less biomass than communities dominated by the grasses, which is linked to a higher average root diameter. The results aslo show that communities with relative evenness show balanced values for root traits and biomass production relative to the dominated communities. With these findings, more insight is created in the role of community compositions related to root traits and biomass production. This enhances the understanding about the way ecosystems should be treated in different ways. The first way is that this research helps choosing for species composition in managed grasslands and for ecosystem restoration in generial. Another way is that, since links have been found between root trait distribution and biomass production, root samples can serve as an indication for biomass production potential. Enhanced understanding of these processes also contributes to the idea that community composition is important for the functioning of ecosystems and therefore ecosystem services (Deyn & Putten, 2005; Streitwolf-engel et al., 1998; Wallace, 2007). This furthermore contributes to the idea that looking at plant traits at the level of communities instead of at the level of individuals is necessary (Mclaren & Turkington, 2010). To conclude, for research on such complex intertwined systems it is difficult to draw hard conclusions on relations between different factors. Indicating relations is however very valuable for finding trends and patterns, which induces increased understanding of these systems.

(16)

References

Bardgett, R. D., Mommer, L., & Vries, F. T. De. (2014). Going underground : root traits as drivers of ecosystem processes.

Trends in Ecology & Evolution, 29(12), 692–699. https://doi.org/10.1016/j.tree.2014.10.006

Bergmann, J., Weigelt, A., Plas, F. Van Der, Laughlin, D. C., Kuyper, T. W., Guerrero-ramirez, N., … Mommer, L. (2020). The

fungal collaboration gradient dominates the root economics space in plants. 1–10.

Cai, H., Li, F., & Jin, G. (2019). Fine root biomass, production and turnover rates in plantations versus natural forests: effects of stand characteristics and soil properties. Plant and Soil, 436(1–2), 463–474. https://doi.org/10.1007/s11104-019-03948-8 Comas, L. H., Becker, S. R., Cruz, V. M. V, Byrne, P. F., & Dierig, D. A. (2013). Root traits contributing to plant productivity

under drought. 4(November), 1–16. https://doi.org/10.3389/fpls.2013.00442

Cougnon, M., De Swaef, T., Lootens, P., Baert, J., De Frenne, P., Shahidi, R., … Reheul, D. (2017). In situ quantification of forage grass root biomass, distribution and diameter classes under two N fertilisation rates. Plant and Soil, 411(1–2), 409–422. https://doi.org/10.1007/s11104-016-3034-7

Da, L., Pontes, S., Soussana, J.-F., Louault, F., Andueza, D., & Carrère, P. (2007). Leaf traits affect the above-ground productivity and quality of pasture grasses. Functional Ecology, 21, 844–853. https://doi.org/10.1111/j.1365-2435.2007.01316.x

De Deyn, G. B., Cornelissen, J. H. C., & Bardgett, R. D. (2008). Plant functional traits and soil carbon sequestration in contrasting biomes. In Ecology Letters (pp. 516–531). https://doi.org/10.1111/j.1461-0248.2008.01164.x Deyn, G. B. De, & Putten, W. H. Van Der. (2005). Linking aboveground and belowground diversity. 20(11).

https://doi.org/10.1016/j.tree.2005.08.009

Dzialowski, A. R., Rzepecki, M., Kostrzewska-szlakowska, I., Kalinowska, K., Palash, A., & Lennon, J. T. (2014). on m er ci e

on on m er ci. 73(3), 603–612. https://doi.org/10.4081/jlimnol.2014.721

Fisher, B., Turner, R. K., & Morling, P. (2008). Defining and classifying ecosystem services for decision making. Ecological

Economics, 68(3), 643–653. https://doi.org/10.1016/j.ecolecon.2008.09.014

Fry, E. L., Savage, J., Hall, A. L., Oakley, S., Pritchard, W. J., & Ostle, N. J. (2018). Soil multifunctionality and drought

resistance are determined by plant structural traits in restoring grassland. 99(10), 2260–2271. https://doi.org/10.1002/ecy.2437

Grigulis, K., Lavorel, S., Krainer, U., Legay, N., Baxendale, C., Dumont, M., … Clément, J. (2013). Relative contributions of

plant traits and soil microbial properties to mountain grassland ecosystem services. 47–57. https://doi.org/10.1111/1365-2745.12014

Kalluri, U. C., Yang, X., & Wullschleger, S. D. (2020). Perspective Plant Biosystems Design for a Carbon-Neutral Bioeconomy.

2020.

Kramer-Walter, K. R., Bellingham, P. J., Millar, T. R., Smissen, R. D., Richardson, S. J., & Laughlin, D. C. (2016). Root traits are multidimensional: specific root length is independent from root tissue density and the plant economic spectrum.

Journal of Ecology, 104(5), 1299–1310. https://doi.org/10.1111/1365-2745.12562

Kreyling, J., Wenigmann, M., Beierkuhnlein, C., & Jentsch, A. (2008). Effects of Extreme Weather Events on Plant Productivity

and Tissue Die-Back are Modified by Community Composition. 752–763. https://doi.org/10.1007/s10021-008-9157-9 Lipowsky, A., Weigelt, A., Buchmann, N., Schmid, B., & Schulze, E. (2012). Using Plant Functional Traits to Explain Diversity –

Productivity Relationships. 7(5). https://doi.org/10.1371/journal.pone.0036760

Long, J. R. De, Jackson, B. G., Wilkinson, A., Pritchard, W. J., Oakley, S., Mason, K. E., … Bardgett, R. D. (2019).

Relationships between plant traits , soil properties and carbon fluxes differ between monocultures and mixed communities in temperate grassland. (January), 1704–1719. https://doi.org/10.1111/1365-2745.13160

Lynch, J. (1995). Root Architecture and Plant Productivity ’. In Plant Physiology (Vol. 109, pp. 7–13).

Madani, N., Kimball, J. S., Ballantyne, A. P., Affle, D. L. R., Peter, M., Bodegom, V., … Matthew, O. (2018). Future global

productivity will be affected by plant trait response to climate. (January), 1–10. https://doi.org/10.1038/s41598-018-21172-9

Mahaut, L., Fort, F., Violle, C., & Freschet, G. T. (2020). Multiple facets of diversity effects on plant productivity: Species richness, functional diversity, species identity and intraspecific competition. Functional Ecology, 34(1), 287–298. https://doi.org/10.1111/1365-2435.13473

Mclaren, J. R., & Turkington, R. (2010). Ecosystem properties determined by plant functional group identity. (Grime 1998), 459– 469. https://doi.org/10.1111/j.1365-2745.2009.01630.x

Mommer, L., Visser, E. J. W., van Ruijven, J., de Caluwe, H., Pierik, R., & de Kroon, H. (2011). Contrasting root behaviour in two grass species: A test of functionality in dynamic heterogeneous conditions. Plant and Soil, 344(1), 347–360. https://doi.org/10.1007/s11104-011-0752-8

Mulder, C. P. H., Bazeley-White, E., Dimitrakopoulos, P. G., Hector, A., Scherer-Lorenzen, M., & Schmid, B. (2004). Species evenness and productivity in experimental plant communities. Oikos, 107(1), 50–63.

(17)

https://doi.org/10.1111/j.0030-1299.2004.13110.x

Naeem, S., Håkansson, K., Lawton, J. H., Crawley, M. J., Thompson, L. J., Naeem, S., … Thompson, L. J. (1996). Biodiversity

and Plant Productivity in a Model Assemblage of Plant Species Published by : Wiley on behalf of Nordic Society Oikos Stable URL : https://www.jstor.org/stable/3546198 REFERENCES Linked references are available on JSTOR for this article : refe. 76(2), 259–264.

Orwin, K. H., Buckland, S. M., Johnson, D., Turner, B. L., Smart, S., Oakley, S., & Bardgett, R. D. (2010). Linkages of plant

traits to soil properties and the functioning of temperate grassland. (C), 1074–1083. https://doi.org/10.1111/j.1365-2745.2010.01679.x

Orwin, K. H., Ostle, N., Wilby, A., & Bardgett, R. D. (2014). Effects of species evenness and dominant species identity on multiple ecosystem functions in model grassland communities. Oecologia, 174(3), 979–992.

https://doi.org/10.1007/s00442-013-2814-5

Ostonen, I., Püttsepp, Ü., Biel, C., Alberton, O., Bakker, M. R., Lõhmus, K., … Brunner, I. (2007). Specific root length as an indicator of environmental change. Plant Biosystems, 141(3), 426–442. https://doi.org/10.1080/11263500701626069 Prieto, I., Stokes, A., & Roumet, C. (2016). Root functional parameters predict fine root decomposability at the community level.

Journal of Ecology, 104(3), 725–733. https://doi.org/10.1111/1365-2745.12537

Ravenek, J. M., Mommer, L., Visser, E. J. W., van Ruijven, J., van der Paauw, J. W., Smit-Tiekstra, A., … de Kroon, H. (2016). Linking root traits and competitive success in grassland species. Plant and Soil, 407(1–2), 39–53.

https://doi.org/10.1007/s11104-016-2843-z

Rose, R., Atkinson, M., Gleason, J., & Sabin, T. (1991). Root volume as a grading criterion to improve field performance of Douglas-fir seedlings. New Forests, 5(3), 195–209. https://doi.org/10.1007/BF00028111

Ryser, P. (1996). The Importance of Tissue Density for Growth and Life Span of Leaves and Roots: A Comparison of Five Ecologically Contrasting Grasses. Functional Ecology, 10(6), 717. https://doi.org/10.2307/2390506

Ryser, Peter, & Lambers, H. (1995). Root and leaf attributes accounting for the performance of fast- and slow-growing grasses at different nutrient supply. Plant and Soil, 170(2), 251–265. https://doi.org/10.1007/BF00010478

Sadanandan Nambiar, E. K. (1990). Interplay between nutrients, water, root growth and productivity in young plantations.

Forest Ecology and Management, 30(1–4), 213–232. https://doi.org/10.1016/0378-1127(90)90138-2

Seedlings, A. (2021). WinRHIZOTM_2021.

Streitwolf-engel, R., Boller, T., Wiemken, A., & Sanders, I. R. (1998). Mycorrhizal fungal diversity determines plant biodiversity,

ecosystem variability and productivity. 74(1994), 69–72.

Vitra, A., Deléglise, C., Meisser, M., Risch, A. C., Signarbieux, C., Lamacque, L., … Mariotte, P. (2019). Studies Responses of

plant leaf economic and hydraulic traits mediate the effects of early- and late-season drought on grassland productivity.

1–15. https://doi.org/10.1093/aobpla/plz023

Wallace, K. J. (2007). Classification of ecosystem services : Problems and solutions. 9.

https://doi.org/10.1016/j.biocon.2007.07.015

Wilsey, B. J., & Wayne Polley, H. (2006). Aboveground productivity and root-shoot allocation differ between native and introduced grass species. Oecologia, 150(2), 300–309. https://doi.org/10.1007/s00442-006-0515-z

Yang, J. chang, Zhang, H., & Zhang, J. hua. (2012). Root Morphology and Physiology in Relation to the Yield Formation of Rice. Journal of Integrative Agriculture, 11(6), 920–926. https://doi.org/10.1016/S2095-3119(12)60082-3

Zhang, Y., Liu, M., Saiz, G., Dannenmann, M., Guo, L., Tao, Y., … Lin, S. (2017). Enhancement of root systems improves productivity and sustainability in water saving ground cover rice production system. Field Crops Research, 213(August), 186–193. https://doi.org/10.1016/j.fcr.2017.08.008

(18)

Appendix

Relationships according to simple linear regression between biomass production with average root diameter, root volume, root surface area, root length density, specific root length and root tissue density. Biomass production is separated in aboveground (a), belowground (b) and total biomass production (c).

(19)

(20)