Data on vegetation composition, soil edaphic variables and fungal communities in 1-13 year old clearcuts in central Sweden DOI: 10.5878/vfre-f585 Methodology ------------ 1. Study area and design The research was conducted in central Sweden (59-60° N) in two study areas in the transition between the hemiboreal and boreal zone. In total, 36 clearcuts were selected: 18 clearcuts of previously fertilized forests paired with 18 clearcuts of unfertilized forests with similar characteristics (site index, time since clearcutting, soil type). All forests were property of Sveaskog AB, who provided the data necessary to select them. In the fertilized sites, 150 kg N ha-1 had been applied once (n = 14) or twice (n = 4) between 1973 and 2006 in the form of the commonly used Skog-CAN, which is ammonium nitrate with added dolomite (CaMg(CO3)2), to reduce the risk of acidification, and 0.2% boron (B). A space-for-time substitution approach was used to determine the change over time in soil conditions, plant and fungal communities after clearcutting. The forests were clearcut between 2009 and 2018 and sampled between May and September 2022, i.e. 4-13 years after clearcutting. All clearcuts had underwent mechanical soil preparation and had been planted with tree seedlings (Pinus sylvestris and Picea abies). On each clearcut, three circular plots with a radius of 10 m were delineated away from any retention trees. In these plots soil samples were taken, soil respiration was measured, ground vegetation and tree layer were surveyed and young trees were sampled to estimate tree growth rate. Soil samples and soil respiration measurements were taken in the parts of the plot that were undisturbed during soil preparation, i.e. where the soil organic layer was intact. After analysing the results from this first field campaign, an additional sampling campaign in August 2023 was set up to collect soil samples and measure soil respiration in more recent clearcuts, i.e. within a year before sampling (clearcut in 2022 or the beginning of 2023, before the start of the growing season). In this campaign, 6 clearcuts of previously fertilized forests were selected and paired with 6 clearcuts of unfertilized stands with similar characteristics. In the fertilized sites, 150 kg N ha-1 had been applied once between 2004 and 2012. After clearcutting, the soil was mechanically prepared in two of the six pairs and no tree planting had taken place yet in any of the clearcuts. The soil sampling and soil respiration measurements were conducted in the same way as during the first sampling campaign. 2. Tree and ground vegetation survey Composition of ground vegetation was determined using a 1x1 m frame divided into 25 quadrats. The number of quadrats in which a taxon was present was recorded for each vascular plant species and for mosses, lichens and vascular plants as a group. For tree species, a distinction was made between individuals belonging to the tree layer (> 1.2 m) or the ground vegetation (< 1.2 m). This was repeated six times across each 10-m-radius plot. In each plot, a circular subplot with radius 3 m was delineated, in which the species and height of all trees (> 1.2 m) were recorded, from which tree density and tree layer composition were later calculated. In the clearcuts from 2009-2015, three individuals (> 1.2 m) of both Picea abies and Pinus sylvestris were sampled for estimation of tree growth rate. If less than three individuals of a species were present in a plot, the species was not sampled. Sampling was done by cutting the tree at the base and collecting a disc of the stem. These cross-sections were taken back to the laboratory where they were sanded and scanned. Tree ring widths were measured using the measuRing R package (Lara et al., 2015), and growth rates (yearly diameter increase, in mm) were extracted from linear regressions of cumulative tree ring width against year, plotted for each individual tree separately. 3. Soil respiration Soil CO2 flux was measured on rain-free days in two rounds: one in spring (16 May – 9 June 2022) and one in summer (22 August – 12 September 2022). During the second sampling campaign in 2023, only one round of soil CO2 flux measurements was conducted (11-17 August 2023). In each plot, respiration was measured at five locations: one in the middle and four closer to the edge, using a closed chamber constructed from a dark, non-transparent PVC-collar (diameter = 23.5 cm, height = 15 cm) equipped with a portable infrared CO2 gas analyzer (Vaisala GMP343) and a humidity and temperature meter (Vaisala HM70). All living ground vegetation was removed before gently pushing the chamber 0-1 cm into the soil, minimizing soil disturbance while making sure that no gaps were present between the collar and the soil surface. CO2 concentration was then recorded for 3 min at 15s intervals. A quadratic function was fitted between CO2 concentration and time, and CO2 flux was calculated from the linear term on a per area basis (mg C m-2 h-1), accounting for chamber temperature and volume according to standard equations (Kutzbach et al., 2007). After each CO2 measurement, the soil water content and temperature, as well as the depth of the organic layer were determined. Soil water content was measured four times with a soil moisture sensor (Meter GS3 with a Pro-Check reader) and the mean value was used in further analyses. Soil temperature was measured at 3 cm depth. 4. Soil sampling 25 soil cores (diameter 3 cm) were taken in a grid pattern across each plot, the mineral layer was removed and the organic layer, including litter, was pooled into one soil sample. Parts of the plot were the soil organic layer was removed during soil preparation were avoided. Soil samples were stored on ice until frozen at -20° C. 5. Soil CN and pH analyses After weighing and homogenisation in a freeze-mill, a subsample was freeze-dried, weighed (before and after to determine % dry weight), and assessed for carbon and nitrogen content using a combustion elemental analyser (TruMac CN; LECO, St. Joseph, MI, USA). A 5 g subsample of freshly frozen, homogenised soil was shaken in 25 mL of deionised water for 10 min at 650 rpm and left to equilibrate for 15 min before measuring pH with a PHM93 pH meter (Radiometer, Copenhagen). The carbon and nitrogen stocks of the organic layer were calculated by multiplying the dry weight of the soil sample by the carbon and nitrogen content, respectively, and scaling it up to tonnes ha-1 based on the total area of the soil cores. 6. Soil enzyme assays (only for the soil samples from 1yo clearcuts) From the frozen homogenised soil samples of the 2023 sampling campaign, potential enzymatic activities of five hydrolytic enzymes (cellobiohydrolase, β-glucosidase, β-xylosidase, β-N-acetyl-glucosaminidase and acid phosphatase) and of manganese peroxidase were determined (Kyaschenko et al., 2017; Saiya-Cork et al., 2002). Soil suspensions were made by shaking a volume of frozen soil equivalent to 2 g dry soil in 200 mL sodium acetate buffer (50 mM, pH 5) for the hydrolytic enzyme assay, and the equivalent to 5 g dry soil in 50 mL sodium acetate buffer for the manganese peroxidase assay. For the hydrolytic enzymes, the soil suspensions were further diluted 10 times and 50 µL fluorogenic umbelliferyl substrate was added to 200 µL soil suspensions (0.001 g dry weight soil ml-1). After incubating in the dark for 2h, 10 µL 0.5 M NaOH was added to stop the reaction, and fluorescence was measured, controlling for background fluorescence (assays without the incubation step). The soil suspensions were also incubated with a standard methylumbelliferone solution as a quenching control. Soil suspensions with too high quenching were further diluted and the assay was repeated. Net fluorescence was converted to enzyme activity expressed per min and g organic matter. For the manganese peroxidase assay, 50 µL of clear supernatant of soil suspensions (0.1 g dry weight soil ml-1) were added to a buffer solution with 3-dimethylaminobenzoic acid and 3-methyl-2-benzothiazolinone hydrazone hydrochloride and either MnSO4 or EDTA (which chelates Mn). Four combinations were done: one with Mn and H2O2 (peroxidase activity including Mn-dependent), one with EDTA and H2O2 (Mn-independent peroxidase activity), one with EDTA (negative control) and one with Mn, H2O2 and a commercial horseradish peroxidase (Sigma-Aldrich, Burlington, MA, USA) (positive control). Immediately after mixing the reagents, plates were put in the plate reader and absorbance was measured every 3 min for 30 min. Mn-dependent activity was calculated as total peroxidase activity minus Mn-independent peroxidase activity and expressed as absorbance per minute and g organic matter. 7. soil metabarcoding A 0.25 g subsample of freeze-dried and ball milled soil was used for DNA extractions with the Nucleospin Soil kit (Macherey-Nagel) following the manufacturer’s instructions. About 1000 bp long markers, including the ITS2 region together with parts of the large subunit, were amplified from diluted DNA extracts (1 ng/µl), using the forward primer gITS7 and the reverse primer TW13 with unique identification tags attached to both primers (Ihrmark et al., 2012; Tedersoo et al., 2018). Amplification was done in 50 µl reactions consisting of 0.5 µM forward primer, 0.3 µM µl reverse primer, 0.25 µl DreamTaq polymerase, 5 µl dNTPs, 5 µl DreamTaq buffer, 1.5 µl MgCl2, 3.25 µl H2O and 25 ng template DNA under the following conditions: 5 min at 94° C, 21-25 cycles of 30 s at 94° C, 30 s 56° C and 1 min at 72° C and finally 8 min at 72° C. PCR products were equimolarly pooled and cleaned with the E.Z.N.A. Cycle Pure Kit (Omega Bio-Tek). After a quality control by Bioanalyzer (Agilent tech), the amplicon pool was sequenced on the PacBio Sequel II platform (Pacific Biosciences) at SciLifeLab NGI (Uppsala, Sweden). Sequence data was submitted to the NCBI Sequence Read Archive under BioProject PRJNA1191207. 8. Processing of the sequence data Quality filtering and OTU clustering were conducted with the SCATA bioinformatics pipeline (https://scata.mykopat.slu.se). Sequences containing both primer and identification tag sequences, with minimum length of 200 bp, average quality > 20 and single base quality > 3 were used in single-linkage OTU clustering. Clustering was done at four different similarity thresholds (99.5%, 99%, 98.75% and 98.5%). OTUs were identified using the Species Hypothesis (SH) matching service, based on the UNITE database (Nilsson et al., 2019) and integrated in the PlutoF platform (Abarenkov et al., 2010). Identifications were double-checked against NCBI Genbank. After comparing the identifications at the various similarity thresholds, the optimal threshold for this dataset, with the greatest correspondence between OTUs and species, was determined to be 98.75%. Only sequences attributed to the fungal kingdom were used in further analyses. The FungalTraits database (Põlme et al., 2020) was used to attribute OTUs to a specific lifestyle. OTUs that were attributed a saprotrophic lifestyle were split up into saprotrophic ascomycetes, saprotrophic agaricomycetes and other saprotrophs. For ascomycete OTUs with potentially versatile saprotrophic and root-associated lifestyles (e.g. root endophyte, dark septate endophyte, ericoid mycorrhizal fungus) or with an unknown primary lifestyle, the SH to which they were attributed was searched in UNITE to check whether it had been found in root samples before. If so, they were attributed to root-associated ascomycetes and otherwise either to saprotrophic ascomycetes (saprotrophic primary or secondary lifestyle according to FungalTraits) or as unknown (unknown lifestyle according to FungalTraits). 9. qPCR quantification Copy numbers of the ITS2 region were quantified from diluted DNA extracts (0.5 ng per reaction) on a CFX Connect Real-Time System (Bio-Rad) using the forward primer gITS7 (Ihrmark et al., 2012) and reverse primers ITS4 and ITS4arch (Sterkenburg et al., 2018; White et al., 1990) in duplicates. The ITS2 copy numbers were converted to ITS2 copy number mg-1 organic matter and corrected to fungal ITS2 copy number mg-1 organic matter by multiplying total copies with the ratio of fungal sequences in that sample, based on the metabarcoding data (to correct for non-target amplification e.g. of plant DNA). ITS2 copy numbers of the four dominant ecological groups (root-associated ascomycetes, saprotrophic ascomycetes, ectomycorrhizal fungi and saprotrophic agaricomycetes) and of individual OTUs were estimated by dividing the number of sequences from each group or OTU by the total number of fungal sequences (both from metabarcoding data) and then multiplying by the fungal ITS2 copy numbers for that sample to end up with a copy number-corrected OTU table, which was used in further analyses. Overview of the data -------------------- The data consists of 8 tab delimited files and an R script. Data from different files is linked to each other through the Plot name. 1. PlotData This tab contains metadata on the sampled plots: Nr (simple number used throughout lab analyses) Plot: ID of each of the plots: this consists of - a letter referring to the area: S = Västmanland sampled in 2022; U = Uppland sampled in 2022; Y = Västmanland sampled in 2023 - a number referring to the pair within the sampled area (see also variable "Pair") as a clearcut of a previously fertilized forest was paired up with a clearcut of an unfertilized forest - a letter (f or a) referring to whether it was the fertilized clearcut or ambient clearcut - a number (1 to 3) referring to the plot within the clearcut This is the same for the following two variables: Clearcut and Pair (Plot is nested within Clearcut, which is nested within Pair) geographical data (latitude and longitude in WGS84) Year in which the sampling took place: either 2022 or 2023 Year in which the forest was clearcutted carbon and nitrogen content (%) soil pH Soil moisture (ratio) Soil organic matter (SOM, also ratio) Soil carbon stock (in tonnes per hectare) Soil nitrogen stock (in tonnes per hectare) fungal ITS copy numbers as a measure for fungal abundance (in fungal ITS2 copy numbers per milligram of organic matter) year_fert: the year the forest was fertilized during the previous rotation period (only for previously fertilized clearcuts) Nr_fert: the number of times the forest was fertilized during the previous rotation period (only for previously fertilized clearcuts) yrs_since_fert: the number of years between when the forest was fertilized during the previous rotation period and the clearcutting (only for previously fertilized clearcuts) age_at_harvest: how old the tree layer was at the time of clearcut harvesting (in years) Pine_%_before_clearcutting: the cover of Pinus sylvestris before clearcutting (in %) Spruce_%_before_clearcutting: the cover of Picea abies before clearcutting (in %) Birch_%_before_clearcutting: the cover of Betula before clearcutting (in %) 2. SoilRespirationData This file contains data from the soil respiration measurements: five measurements per plot, per season. The variables Pair, Clearcut, Plot and SamplingPoint indicate where sampling took place. There are five sampling points per plot, three plots per clearcut, two clearcuts in each pair. fert: whether the clearcut was fertilized or not (ambient) during the previous rotation period year_cut: in which year the forest was clearcutted Round: whether the measurement was done during the spring or during the summer SamplingDate (YYYY-MM-DD) and SamplingTime indicate when the measurement took place geographical data (latitude and longitude in WGS84) SoilTemp: temperature of the soil at the time of sampling (in degrees Celcius) OrganicLayer: depth of the soil organic layer in cm Moisture: % soil moisture flux.quad is the CO2 flux calculated from the linear term of a quadratic function between CO2 concentration and time on a per area basis (mg C m-2 h-1). 3. Enzyme assay data Data from the enzyme assays. It is expressed as enzyme activity per min and g organic matter. CBH: cellobiohydrolase BG: β-glucosidase BXD: β-xylosidase NAG: β-N-acetyl-glucosaminidase aP: acid phosphatase 4. Vegetation_plot Data on the vegetation composition (sum of the subplots). The number indicates the number of quadrats in which a taxon was recorded (max is 150). The first column gives the plot name. The next 5 columns indicate the cover of trees above 1.2 m (TreeLayer), the cover of vascular plants (VascularPlants), the cover of mosses, of lichens and of Sphagnum mosses. All other columns indicate the cover at species level (except for tree species, which are split every time between tree layer (i.e. above 1.2 meter) and ground vegetation (below 1.2 meter)) This data is a sum of the data that was recorded in subplots. Each subplot consisted of 25 quadrats and there were 6 subplots per plot. 5. Vegetation_subplot Data on the vegetation composition as measured in the field. Similar data to Vegetation_plot, but at the level of the subplot instead of the plot. The number indicates the number of quadrats in which a taxon was recorded (max is 25). 6. Tree_height Height measurements from trees. First column gives the plot name Second column gives the tree species third column gives the height of the tree in centimeters forth column gives the data as it was recorded in the field fifth column gives remarks that were made in the field 7. Tree_growth_rate Growth rate of sampled trees (yearly diameter increase, in mm), extracted from linear regressions of cumulative tree ring width against year. Sample name consists of the plot ID, an abbreviation of the tree species name and the replicate number. Plot ID, species name and replicate number are also given 8. OTUtable+taxonomy OTU table + taxonomic and ecological information (see methods). Raw sequence data was submitted to the NCBI Sequence Read Archive under BioProject PRJNA1191207. The numbers are the number of sequences per OTU (in the rows) and per sample (in the columns) The last 11 columns give information on the total number of sequences per OTU, the taxonomic information of the OTU and to which lifestyle the OTU was attributed.