Fig 1.
(a) Location of the study site in northeast Germany.
(b) Spatial distribution of the twelve measurement spots at the four clusters (aerial photograph 2009, NW = northwest, SE = southeast, + = more inundated, − = less inundated, b = rush stands, c = sedge stands p = reed stands). (c) micro-relief and inundation intensities in November 2009 (NW = northwest, SE = southeast). Own creation. Map in a) is reprinted from Bundesamt für Kartografie und Geodäsie under a CC BY license, with permission from the German Ordinance to Determine the Conditions for Use for the Provision of Spatial Data of the Federation (GeoNutzV), original copyright 2014.
Fig 2.
Closed chamber set-up to measure the CH4 exchange at inundated conditions.
a) Scheme of a measurement spot with the chamber placed at the collar and an attached vacutainer for gas sampling. b) Photograph of a sedge spot from the study site in July 2010.
Table 1.
Average peat water properties at the NW and SE clusters in the first year after flooding.
Fig 3.
Environmental variables at the more (NW+, SE+) and less (NW-, SE-) inundated clusters after flooding.
From November 2009 until January 2011, we recorded a) water level (WL), b) peat temperature (PT), c) pH-value, d) electric conductivity (EC), e) concentration of chloride (Cl-), f) sulfate (SO42-), g) total organic carbon (TOC), and h) total bound nitrogen (TNb) in the peat water at 10 cm peat depth. Generalized additive models were fitted to the data to display the overall trends during the measurement period (solid line: modelled value; dashed lines: 95% confidence interval); the goodness of model fit is given by R2 and p-value. Extreme values (≥1.5 x interquartile range) were not included into modelling; they are marked by arrows and values are given next to the symbol of the particular cluster.
Fig 4.
Properties of the a) peat water and b) peat (0-20cm) before (pre) and after (post) flooding.
Spots which shared one dip well are represented by one data point. Only parameters that contributed significantly to the NMDS (p < 0.01) were included. Parameters included in peat water NMDS were water level (WL), the contents of total organic carbon (TOC) and nitrogen(TN), chloride (CL) and sulfate (SO4). Stress of the peat water NMDS was 0.012. Flooding significantly (p<0.001***) altered peat water characteristics. For pre-flooding peat water properties from the 25.11.2009 and for post-flooding those from the 25.11.2010 were used (n = 14).
Table 2.
Peat properties before (“PRE”) and after (“POST”) flooding in the peat zones dominated by living roots (0–10 cm, D1) and the more humified peat (10-20cm, D2).
Fig 5.
Exchange of CH4 at the more (NW+, SE+) and less (NW-, SE-) inundated clusters.
The CH4 exchange was measured from November 2009 until November2010 using the closed chamber approach.
Fig 6.
Impact of peat temperatures (°C) on CH4 fluxes (mg m-2 h-1) in the four seasons (blue = winter, green = spring, yellow = summer, red = autumn) in the northwest (NW) and the southeast (SE) of the study site at the more (+) and less (-) inundated measurement spots.
Data are interpolated by generalized additive models. Error bars indicate the standard error of the interpolation.
Fig 7.
Interaction of site characteristics with the exchange of CH4 in the first year after flooding.
Measurement spots are arranged according to Bray-Curtis-dissimilarity regarding the average site characteristics (median) of the measurement spot (for abbreviations see Fig 4) (a). The size of the symbols represents the relative values of the annual emissions of CH4 (b). All parameters contributed significantly (p<0.001) to the ordination. The polygons display the cluster (i.e. location and inundation level, see Fig 1c) of the particular measurement spot. Clusters differed significantly regarding their site characteristics (p<0.05). Stress of the NMDS was 0.012.