Survey of bracken biomass, soil chemistry and other environmental factors in Snowdonia and the Lake District in 2014 [LTLS]

References to accompany lineage are available from supporting information. Mean annual precipitation data were obtained for each location from the CEH-GEAR dataset, an interpolated grid based on a network of meteorological stations (Tanguy et al., 2014). The altitude of each location was determined in the field using a GPS device. Site-specific N deposition estimates were obtained from APIS (APIS, 2015). Plots of 1x 1 m were demarcated, and all plant species occurring in the plot were recorded. Bracken fronds that emerged from within the plot were cut at ground level, bulked and weighed. A subsample of approximately 500g was weighed, dried at 60°C to constant weight, and re-weighed to determine moisture content for calculating total dry bracken biomass. All other vegetation in the plot was cut at ground level and at the plot boundaries, and dried at 60°C to constant weight. A 25g subsample of dry bracken fronds was ground for chemical analysis. Carbon and N contents were determined by oxidative combustion followed by thermal conductivity detection using a Vario EL elemental analyser (Elementar Analysensysteme GmbH, Hanau, Germany). Subsamples were digested using a method based on EPA (2007) in a 3:1 mixture of hydrochloric acid and nitric acid in sealed Teflon vessels, heated to 175°C over a period of 7 minutes and held at 175°C for 4.5 minutes. After cooling, the vessel contents were filtered and diluted to volume. Concentrations of P, K, Ca and Mg were measured by Inductively Coupled Plasma Optical Emission Spectrometry using an Optima 7300DV analyser (Perkin Elmer, Waltham, Massachusetts USA).

Soil was sampled from three places within the plot using a 5 cm diameter split-tube auger. Plant litter was removed, and the auger was inserted to 15 cm depth. In a few plots this was not possible due to the presence of stones, and instead holes were excavated to 15 cm depth and their dimensions recorded. The fresh sample was weighed, and a subsample removed and weighed. Moisture content was determined by measuring the weight of the subsample after drying for three days at 60 oC, and the organic C concentration was determined by measuring weight after ignition at 350 oC for three days. The total dry weight of soil was determined, and used to calculate bulk density on the basis of the total volume of the auger or excavated holes. The pH of fresh soil was measured in a slurry of 10 g soil in 25 mL deionised water. The remaining fresh soil sample was air-dried, and stones and roots not passing through a 2 mm sieve were removed. The soil was then ground, and subsamples were taken for chemical analysis. Concentrations of soil total C, N, P, K, Ca and Mg were determined using the same methods as for plant tissue analysis. Organic P was obtained as the difference between total P and inorganic P concentrations. A 0.5 g milled sub-sample was used for the extraction of inorganic P using 25 ml of 0.5 M sulphuric acid, which was shaken for 16 hours, centrifuged at 10 000 rpm for 30 minutes and filtered into 50 ml test tubes. The filtrate was analysed for total inorganic P using molybdate colorimetry with a Seal Analytical AQ2 discrete analyser.

Soil total N/C ratio was used as an indicator of N availability. The other nutrient elements were expressed as stocks, since mass concentration may under-represent availability in dense mineral soils and element/C ratio may underrepresent availability in more organic soils (Rowe et al., 2012). Stocks (g m-2) in the top 15 cm of soil were calculated using the bulk density measurement. The stock of organic P, and total element stocks of K, Ca and Mg, were used as indicators of availability.

Species occurrence was used to characterise environmental conditions on the basis of indicator scores (Diekmann, 2003). Environmental trait scores for European species were defined by Ellenberg et al. (1992) on axes corresponding to fertility ('Ellenberg N'), alkalinity ('Ellenberg R'), moisture ('Ellenberg F') and light availability ('Ellenberg L'). In the current study, plot-specific values on each of these axes were calculated as the mean values for the species present of indicator scores as recalculated for UK species by Hill et al. (2000). Cover-weighting was not applied, in part to prevent the scores for bracken itself dominating the mean scores, but also since basing indicator scores on presence-absence data reduces the effect of short-term environmental variation (Kafer and Witte, 2004).

This dataset is included in the following collections

The Long Term Large Scale Project [LTLS]