Shifts in plant functional composition following long‐term drought in grasslands

Abstract

Plant traits can provide unique insights into plant performance at the community scale. Functional composition, defined by both functional diversity and community‐weighted trait means (CWMs), can affect the stability of above‐ground net primary production (ANPP) in response to climate extremes. Further complexity arises, however, when functional composition itself responds to environmental change. The duration of climate extremes, such as drought, is expected to increase with rising global temperatures; thus, understanding the impacts of long‐term drought on functional composition and the corresponding effect that has on ecosystem function could improve predictions of ecosystem sensitivity to climate change. We experimentally reduced growing season precipitation by 66% across six temperate grasslands for 4 years and measured changes in three indices of functional diversity (functional dispersion, richness and evenness), community‐weighted trait means and phylogenetic diversity (PD). Specific leaf area (SLA), leaf nitrogen content (LNC) and (at most sites) leaf turgor loss point (πTLP) were measured for species cumulatively representing ~90% plant cover at each site. Long‐term drought led to increased community functional dispersion in three sites, with negligible effects on the remaining sites. Species re‐ordering following the mortality/senescence of dominant species was the main driver of increased functional dispersion. The response of functional diversity was not consistently matched by changes in phylogenetic diversity. Community‐level drought strategies (assessed as CWMs) largely shifted from drought tolerance to drought avoidance and/or escape strategies, as evidenced by higher community‐weighted πTLP, SLA and LNC. Lastly, ecosystem drought sensitivity (i.e. relative reduction in ANPP in drought plots) was positively correlated with community‐weighted SLA and negatively correlated with functional diversity. Synthesis. Increased functional diversity following long‐term drought may stabilize ecosystem functioning in response to future drought. However, shifts in community‐scale drought strategies may increase ecosystem drought sensitivity, depending on the nature and timing of drought. Thus, our results highlight the importance of considering both functional diversity and abundance‐weighted traits means of plant communities as their collective effect may either stabilize or enhance ecosystem sensitivity to drought.

Is phylogenetic and functional trait diversity a driver or a consequence of grassland community assembly?

“…However, PD tempered the effect—with high PD, FD did not decrease resistance, suggesting an interplay between them such that both drive community assembly but with their impacts differing depending on each other. Griffin‐Nolan et al () showed a similar pattern, where changes in FD and PD did not consistently match; as functional traits changed from drought‐tolerance to drought‐avoidance, PD did not differ accordingly. Barber et al () showed that while PD did not change with environmental context, SD did respond to all the environmental gradients they analysed.…”

“…Barber et al () used 120 tallgrass prairie restorations to explore the PD and SD consequences of a variety of management and climatic variables in an uncontrolled field setting. Griffin‐Nolan et al () manipulated water availability to look at drought over 4 years while Miller et al () used a 19‐year time series to investigate the impacts of real droughts. Lastly, Cadotte et al () amassed results from 79 studies and 188 comparisons between FD and PD in a review that sought to glean generalities and insights into the relationship between FD and PD in studies across the globe.…”

“…Environmental context seems to play a strong role in the answer to this question, and often results in FD or PD being passengers of assembly. For example, Miller et al () and Griffin‐Nolan et al () used natural variation in precipitation and experimental drought, respectively, to investigate drought impacts on FD. Both found FD to be a passenger of community assembly changes driven by drought, though Griffin‐Nolan et al () found an increase in functional dispersion and Miller et al () found a decrease in FD mirrored a loss of species richness driven by drought.…”

“…For example, Miller et al () and Griffin‐Nolan et al () used natural variation in precipitation and experimental drought, respectively, to investigate drought impacts on FD. Both found FD to be a passenger of community assembly changes driven by drought, though Griffin‐Nolan et al () found an increase in functional dispersion and Miller et al () found a decrease in FD mirrored a loss of species richness driven by drought. Similarly, Brandt et al () found that SD and PD both changed in response to disturbance and changes in nutrient availability in a field setting, suggesting PD is a passenger to environmentally driven changes in SD.…”

Press–pulse interactions and long‐term community dynamics in a Chihuahuan Desert grassland

“…Reordering may occur among species in similar or different functional groups (grasses, forbs, legumes) under ecological presses and pulses. Moreover, species within the same functional group can differ widely in their biotic interactions, traits, and influence on ecosystem processes (e.g., Magurran and Henderson, 2010; Griffin‐Nolan et al , 2019). For example, reordering of species led to changes in total above‐ground net primary production (ANPP) in a long‐term irrigation experiment (Collins et al , 2012; Knapp et al , 2012), and reordering was shown as the primary driver of community composition change across taxa in a long‐term observational study (Jones et al , 2017).…”