Heterogeneous trait responses of Páramo plant species and community to experimental warming

Understanding climate change impact on the functional trait composition, and hence ecosystem functioning of tropical alpine regions is critical for predicting biodiversity responses. We tested the effects of a decade of warming on the morphological, chemical, and genomic traits of Páramo species using open-top-chambers (OTC). We conducted vegetation surveys and collected samples from individuals inside and outside the OTC plots to estimate differences between treatments (warming vs control). Vegetation cover decreased over time in both treatments suggesting a potential decline in soil moisture in our study area. Warming led to a reorganisation of the trait space and trait network structure. Nevertheless, we did not find significant differences in trait values or the direction of change between species whose % vegetation cover increased in OTC (or decreased less) compared to control over time. Community-weighted mean values of plant height, leaf area, leaf dry matter content, genome size, leaf C and P, significantly increased over time only in OTC plots– i.e. traits associated with carbon storage and decomposition. Whilst warming -and reduced soil moisture - lead to heterogeneous species responses without a clear winning trait strategy, changes at the community-level may have important implications for Páramo ecosystem functioning.

Methods
Study area and experimental design

The study area is located in the Páramos of the Yanacocha Reserve, Ecuador, at 4,200 m a.s.l (0.13518°S, 78.57458°W). The passive warming experiment was installed in 2012 and consists of five parallel monitoring blocks, each of them containing 40 1 m2 plots delimited for different treatments. Plots assigned to a warming treatment were surrounded by a 3 x 3 m hexagonal open-top chamber (OTC). Control plots and those under warming treatment (OTC plots) were randomly chosen across the five blocks. Our analyses are based on the data from 17 OTC plots and 20 control plots.

For details on the climate in the study area refer to the published paper.

Field data collection

Vegetation surveys: Before installing the OTCs, a baseline survey was conducted in 2012, recording vegetation cover (%) of each seed plant species per plot. A further five vegetation surveys were conducted on the control and OTC plots in 2013, 2014, 2016 (a few plots for this survey were measured in December 2015), 2017, and 2019. The OTCs of five warming plots were blown away by the wind after the 2017 survey, therefore, for 2019 we used the data from the remaining OTCs. A total of 71 taxa were recorded in the selected plots but we chose to work with the 40 most dominant species based on % vegetation cover.

Collection of leaves for trait analysis: All trait data were collected in June 2022. To measure the morphological traits for each species whenever possible (i.e., we were limited by the number of individuals present in OTCs), we collected five leaves each from three individuals inside the OTC plots and from three individuals outside them. We also measured the vegetative plant height of 1 to 10 individuals per treatment (OTC mean = 6, control mean = 8). For chemical traits, due to the small size of most species, collecting enough material for analysis was challenging and thus, for a given treatment (warming or control) and for each species, we pooled the leaves from more than one individual together. Lastly, for the genome size analyses, we collected leaf material from one individual per species and stored them in a plastic bag with a wet tissue to preserve them until analysed. Given the variable availability of plant material it was not possible to collect samples for all traits and all 40 species.

Laboratory analyses

Samples for morphological and chemical analyses were taken to the Universidad de las Américas in Ecuador, where the following morphological traits were measured: (i) leaf area (cm2), (ii) leaf blade thickness (mm), (iii) specific leaf area (SLA in cm2/g) which is the fresh leaf area divided by its dry mass, and (iv) leaf dry matter content (LDMC in mg/g) which is the dry leaf mass divided by the fresh leaf mass.

Samples for characterising the chemical traits were processed to estimate: (i) the amounts of macro- (i.e., K, P, Mg, Ca in ppm) and (ii) micro-nutrients (i.e., Al, B, Fe, Na in ppm) in the leaves, and (iii) the leaf C and N content (% and C/N ratio). Finally, the genome size (number of base pairs in the DNA in gigabases per unreplicated gametic nucleus (i.e. the 1C-value), Gb/1C) of each species was estimated at the Royal Botanic Gardens, Kew, by propidium iodide flow cytometry.