More is not always better: Peat moss mixtures slightly enhance peatland stability

Early September 2021, we collected 40 cores (diameter 22.5 cm, 15 cm depth) of acrotelm peatmoss communities from a Sphagnum­-dominated peatland in the Store Mosse National Park, Sweden (57°17’54 N, 14°00’39 E). Our sampling strategy was set up to create a single-density replacement series (Jolliffe, 2000) with two naturally co-occurring Sphagnum mosses: S. cuspidatum Ehrh. ex Hoffm and S. medium *Limpr. These species grow in close proximity to each other in the field and make up >70% (cover) of the field layer in the field site. Our sampling strategy resulted in cores with monocultures (100% cover) of *S. cuspidatum or S. medium, and three mixtures of S. cuspidatum and *S. medium *– 75%/25%, 50%/50%, 25%/75% (n = 8). Final cover ratios may differ by ± 5%. Vascular plants were carefully removed

Following and acclimation period, the eight mesocosms for the five compositionally distinct Sphagnum communities were randomly assigned an experimental water table drawdown treatment: mild or deep drawdown (n = 4). The mild water table drawdown consisted of a water table drawdown of 5 cm followed by a seven-day drought period in which no precipitation was added. The deep water table drawdown was similar except that the initial seven-day mild drought was followed by a water table drawdown of 20 cm ­– hence a full drainage ­– and an additional two-week drought. Every drawdown event was preceded by mimicking a heatwave, which under natural conditions precedes a drop in water table­. The heatwave was realized by increasing the climate room temperature to *c. *35°C for 10 hrs. using twelve infrared lamps (Philips IR 250 RH IR2 230-250V 250W) that were installed 70 cm above the surface of the mesocosms. Both drought treatments were trailed by a rewetting period (= recovery period) which lasted eight weeks.

Carbon dioxide (CO2) concentrations were measured throughout the experiment using an acrylic, cylindrical, transparent chamber (ID 29 cm, height 30 cm) that could be placed air-tight on the mesocosms using a rubber inner sealing. The chamber was equipped with an internal fan. On measurements days – two times per week in the acclimation period; four times per week in the drought period, two times per week in the recovery period – the transparent chamber was connected in a closed loop to a laser-based gas analyser that uses Optical Feedback – Cavity-Enhanced Absorption Spectroscopy to measure gases in air (LI-7810 CH4/CO2/H2O Trace Gas Analyzer, LI-COR Biosciences). Gas measurements were performed with a frequency of 1 hz over a 120 s interval. The chamber was vented between measurements.

Measurements with the transparent chamber represent Net Ecosystem Exchange (NEE). CO2 fluxes were calculated using the R package FluxCalR (Zhao, 2019), which makes use of the change in gas concentration in the chamber over time. Due to instability in the gas flux measurements, we used a dead band of 30 s for the NEE calculations. The ecological sign convention was used for the NEE data; hence, positive flux values indicate CO2uptake, while negative flux values indicate CO2 loss to the atmosphere.

We calculated ecosystems stability measures for all individual mesocosms based on the framework described by Isbell et al., 2015 (Nature, 526, 574–577. https://doi.org/10.1038/nature15374). 

## Description of the data and file structure

- NEE_acc.txt; containing the Net ecosystem CO2 exchange data from the acclimation period

- NEE_dd.txt; containing the Net ecosystem CO2 exchange data from the deep water table drawdown period

- NEE_md.txt; containing the Net ecosystem CO2 exchange data from the mild water table drawdown period

- NEE_rec.txt; containing the Net ecosystem CO2 exchange data from the recovery period

- Stability.txt; containing the calculated values

day : [numeric] time in days in the water table drawdown period which starts at t = 0; negative values indicate the acclimation period

day_rec : [numeric] [in NEErec only] time in the recovery period, baselined for the mild and deep water table drawdown treatment

id : individual identifier for the mesocosms; used as the error variable nested within day in Linear Mixed Effects Model

treat : [factor w/ 2 levels] MD = mild water table drawdown, DD = deep water table drawdown

rep : [factor w/ 4 levels] replicates

flux : [numeric] Net ecosystem exchange in µmol CO2 m-2 s-1

comp: [factor w/ 5 levels] Sphagnum community composition: # 1 = 100 % S. cuspidatum;

                                                              # 2 = 75% S. cuspidatum / 25% S. medium

                                                              # 3 = 50% S. cuspidatum / 50% S. medium

                                                              # 4 = 25% S. cuspidatum / 75% S. medium

                                                              # 5 = 100% S. medium

For the Stability.txt file:

id : individual identifier for the mesocosms; used as the error variable nested within day in Linear Mixed Effects Model

treat : [factor w/ 2 levels] MD = mild water table drawdown, DD = deep water table drawdown

rep : [factor w/ 4 levels] replicates

NEEacc : [numeric] Mean net CO2 exchange during the acclimation period

NEEd : [numeric] Net CO2 exchange at the last day of the water table drawdown treatment

NEErec : [numeric] Net CO2 exchange at the last day of the recovery period

resis : [numeric] resistance [dimensionless], see main text for calculation

resil : [numeric] resilience [dimensionless], see main text for calculation

rec.rate : [numeric] the change in net CO2 exchange per day, during the recovery period

absresis : [numeric] absolute values of resistance

## Code/Software

An R script that was used for analysis has been made available