Many wetland environmental gradients structure plant community composition, yet controls of plant community composition within rich fens, botanically diverse groundwater-fed wetlands, are still incompletely understood. Porewater chemistry and plant community composition were recorded for eight calcareous rich fens encompassing both calcium carbonate and calcium sulfate geological inputs in the Central New York State region. As expected, porewater sulfate and sulfide concentrations were on average higher for wetlands overlying calcium sulfate than for wetlands overlying calcium carbonate. However, within-wetland heterogeneity in porewater chemistry was high. Moss species density, moss cover, and total plant cover decreased with increased sulfide. Moss, dicot, and total cover also had a negative relationship with calcium. There were a number of species-level responses to calcium, sulfide, phosphorus, and ferrous iron. Plant height had a positive relationship with nitrogen. The strength and relative importance of some plant responses to sulfide and calcium in the current regional study differed from a previous sub-hectare scale study. The observed decrease in some metrics of site-level plant diversity with increased sulfide variability across fens highlighted the need to characterize species-environment relationships across spatial scales.

Sampling took place within eight calcareous rich fens (Fig. 1) in central New York State, USA. We established 10 sampling locations at each fen.

In summer 2009 at each of the 10 sampling locations within each of the eight sites we measured plant community composition, porewater chemistry, and depth of peat. We recorded percent cover of each vascular and bryophyte species, bare ground, litter, and open water (if applicable), and the height of the overall tallest species in a 0.1 m x 0.1 m quadrat from 24 August - 9 September 2009, with the exception of the Junius site that was sampled 6 - 14 June 2009. In the center of each vegetation quadrat we installed a porewater sipper (outer diameter 1 cm) with a sample screen extending from 8 to 12 cm below the soil surface, for an average sampling depth of 10 cm. We used a bulk interstitial porewater sipper because we sought to measure sulfide at a scale matching a large fraction of a plant’s rhizosphere.

We collected porewater from each sipper using a hand syringe equipped with an in-line 0.45 uM filter 13 - 16 September 2009, with the exception of the Junius site that was sampled 26 May - 1 June 2009. In the field, we placed subsamples in three 23 mL borosilicate glass scintillation vials. In one vial we immediately mixed a 11.5 mL sample with 11.5 mL of previously-added sulfide anti-oxidant buffer (SAOB), filling the vial to capacity to minimize headspace, and set this sample aside for sulfide analysis by ion selective electrode (ISE) in the lab as soon as possible. The SAOB was composed of NaOH, EDTA, and ascorbic acid to stabilize sulfide as S^2- to avoid oxidation to sulfate or volatile loss as H₂S. In a second vial we mixed 15 mL of sample with 5 mL of Ferrozine and HEPES reagents for ferrous iron analysis in the lab. In a third vial we collected sample for lab analysis of calcium, sulfate, and Total Dissolved Nitrogen (TDN). An additional sample was used to measure field pH, conductivity, and temperature with a portable multimeter and then discarded. Finally, we deployed a pair of 2.5 cm by 5 cm anion resin strips at each sampling location from 30 July - 14 August 2009 at a depth of 1-3 cm and retrieved the strips 31 August - 16 September 2009 for an index of phosphorus availability. As with the other measurements, the Junius resin deployment was earlier, from 10 - 11 May 2009 to 8 - 12 June 2009.

We measured porewater sulfide as S^2- with an ion selective electrode (ISE) using sodium sulfide standards. We measured porewater Fe(II) (ferrous iron) spectrophotometrically at an absorbance of 562 nm. Aqueous porewater samples were submitted for inductively coupled plasma atomic-emission spectrometer (ICP-AES) analysis of calcium, ion chromatographic (IC) analysis of sulfate, and continuous-flow analysis (CFA) of Total Dissolved Nitrogen (TDN). Resin strips were extracted with 0.5 M HCl and then the resin extract samples were reacted with Murphy-Riley reagents (sulfuric acid, ammonium molybdate, antimony potassium tartrate) and analyzed spectrophotometrically for phosphate at 880 nm.

Metadata with units and a description for each of the variables in the dataset can be found in the ReadMe.csv file.