Harvard Forest

Data Archive

HF456

Microbial, Plant, and Soil Impacts on Soil Nutrient Cycling in Harvard Forest and Greater Boston 2021-2022

Related Publications

Data

• hf456-01: soil, plant and microbe data (preview)

Overview

• Lead: Jennifer Bhatnagar, Corinne Vietorisz, Nahuel Policelli
• Investigators: Abigail Li, Lindsey Adams
• Contact: Information Manager
• Start date: 2021
• End date: 2022
• Status: complete
• Location: Harvard Forest, Greater Boston
• Latitude: +42.32671 to +42.552047 degrees
• Longitude: -72.169235 to -71.173182 degrees
• Elevation: 30.5 to 330 meter
• Datum: WGS84
• Taxa: Pinus strobus (white pine)
• Release date: 2025
• Language: English
• EML file: knb-lter-hfr.456.1
• DOI: digital object identifier
• EDI: data package
• DataONE: data package
• Related links:
• Study type: short-term measurement
• Research topic: soil carbon and nitrogen dynamics
• LTER core area: mineral cycling
• Keywords: bacteria, biomass, fungi, microbes, nitrogen, nutrient cycling, phosphorus, soils, species diversity
• Abstract:

  Microbes are the driving force behind nutrient cycling within soils, secreting enzymes necessary to break down organic matter, immobilizing nutrients and C, or transferring nutrients to plant hosts. Even though nutrients would almost never move through ecosystems without microbes, we know little about how their composition and activity is related to ecosystem nutrient cycling, and their importance relative to plant and soil abiotic factors. In this study, we sought to determine which commonly measured soil microbial community characteristics best explain soil N and P cycling, and the relative contributions of microbial, plant, and abiotic factors in explaining these processes.

• Methods:

  We sampled and surveyed soils, leaf litter, and vegetation from 6 sites: 3 in Harvard Forest and 3 in suburban Boston. Within each site, we sampled from 4 forest types that span a range of tree litter qualities, which could differentially impact soil nutrient cycling: mature pine, mixed pine/hardwood, hardwood with encroaching pine saplings, and mature hardwood forests. We are especially interested in hardwood forests where young pine saplings encroach into the understory (in the absence of mature pines), because they are likely to associate with unique pine-specific mycorrhizal fungal communities, but contribute little leaf litter to the forest floor. Within each forest type, we collected 8-10 soil samples along a 60 or 90 m transect from the forest edge to interior. Along each transect, the vegetation composition remains relatively constant, while the microbial community likely changes. This allows us to isolate the effects of changes in microbial communities on soil biogeochemistry from the effects of plant communities. In each soil sample, we measured 1) soil nutrient cycling (inorganic and total N and P, net N mineralization rates, net nitrification rates, net phosphate release rates, total C), 2) fungal and bacterial community composition (16S/ITS rDNA amplicons), 3) plant community attributes (tree and understory plant composition, tree species basal area, forest floor litter composition, annual litterfall weight and chemistry), and 4) edaphic factors (total elemental composition, soil pH, % soil organic matter, soil moisture, soil temperature).

  Site description and sampling design

  To prevent complete co-correlation of variables, we designed a field system across temperate forests in Massachusetts, USA that maximized variation in soil microbial community composition, plant community characteristics, and soil nutrient cycling rates. We selected six field sites: three “urban” sites, within 15 miles of Boston’s center, including: Landlocked Forest (Lexington, MA), Whipple Hill Conservation area (Burlington, MA), and Hammond Woods (Newton, MA) and three “rural” sites in the heavily forested areas of central Massachusetts. These sites are within Petersham, MA and Phillipston, MA. Each site contains 4 forest stand types, all within 2.5 km of each other: mature white pine-dominated stands (hereafter “pine”), mature hardwood-dominated stands (hereafter “hardwood”), hardwood-dominated stand with white pine saplings encroaching into the understory (hereafter “encroachment”), and mixed mature white pine and hardwood stands (hereafter “mixed”).

  Within each forest type at each site, one transect was laid from the forest edge to interior, perpendicular to the forest edge (6 sites x 4 forest types = 24 total transects). Two replicate sampling points were established at 0m, 15m, 30m, 60m, and 90m from the forest edge. If the aboveground forest composition changed beyond 60m from the forest edge, the 90m sampling point was excluded. At each distance, one “A” sampling point was established within 5m from the transect on the left of the transect, and one “B” sampling point was established within 5m from the transect on the right of the transect. This resulted in a total of 204 sampling points (24 transects x 4 or 5 distances per transect x 2 sampling points per distance).

  Soil nutrient cycling assays

  To measure soil net ammonification and net nitrification, we used the “buried bag” method, where an initial soil sample is taken to measure ammonium and nitrate, then another soil sample is buried beneath the litter layer in the same location within a polyethylene bag and incubated in the field for a month (Durán et al., 2012; Eno, 1960; Hanselman et al., 2004; Westermann and Crothers, 1980). Initial soil sampling was conducted in mid-July 2021, and the buried bags were collected 4 weeks later in mid-August 2021. At each sampling point, the litter layer was removed before sampling and a 10 cm x 10 cm soil brownie was collected to a depth of 6 cm by cutting into the soil with a knife. 204 bulk soil samples were collected in mid-July at every sampling point, and 144 bulk soil samples were buried for net ammonification and net nitrification, 2 each at the 0m, 15m, and 60m sampling points. Four buried bags were unable to be recovered at the end of the 4 week incubation, so 140 buried soil bags were collected. Within 48 h of sample collection, soil was sieved through a 2mm sieve and 10g of soil from each sample was used for inorganic nitrogen extraction with 2M potassium chloride. 10g of soil was combined with 50 mL of potassium chloride in a plastic Nalgene bottle and shaken for 1 hr. After shaking, the solution was filtered into plastic scintillation vials through a Whatman no. 1 filter circle. The extracts were frozen at -20 ℃ for subsequent colorimetric analysis. Soil extracts were analyzed colorimetrically for NH4+ and NO3− concentrations using the microplate method as described in Caron et al. 2023. Net rates of ammonification and nitrification per gram dry soil per day were calculated as the differences in ammonium and nitrate concentrations between the initial soil sample and the buried bag soil sample divided by the total days incubated.

  To measure net soil phosphate change, a second year of soil sampling was completed in early July 2022. We followed the same soil sampling protocol as outlined above in the same sampling locations as for the net ammonification and net nitrification assay, except that we measured Olsen inorganic P instead of ammonium and nitrate. We buried 144 soil samples in the same locations as the 2021 buried bags. 8 bags were destroyed or unable to be recovered at the end of the 4 week incubation, so 136 buried soil bags were collected. Within 72 h of sample collection, soil was sieved through a 2mm sieve and 1.5g of soil was used for Olsen P extraction (Olsen et al. 1954, Frank et al. 1998). 1.5g of soil was combined with 30 mL of sodium bicarbonate in a glass flask and shaken for 1 hr. After shaking, the solution was filtered into glass scintillation vials through a Whatman no. 5 filter circle. The extracts were frozen at -20 ℃ for subsequent colorimetric analysis. Soil extracts were analyzed colorimetrically for PO43- using a microplate method. Soil extracts were pipetted into a 96-well microplate, including phosphate standard solutions ranging from 0 – 10 ppm PO4-P (Ricca Chemical). An acid molybdate solution containing ammonium molybdate, antimony potassium tartrate, sulfuric acid, and ascorbic acid was added to the extracts and standards, incubated at room temperature for 10 minutes, and the resulting color intensity was quantified in a microplate reader (Synergy H1, Biotek) at an absorbance wavelength of 882nm (Frank et al. 1998). Net rates of phosphate change per gram dry soil per day were calculated as the differences in phosphate concentration between the initial soil sample and the buried bag soil sample divided by the total days incubated.

  Additional soil abiotic variables

  During soil sampling in 2021 and 2022, at each distance from the forest edge, soil temperature was collected directly along the transect, equidistant between each A and B sampling point at all distances, with a Luster Leaf Rapitest digital soil thermometer. In 2021, the depth of the organic horizon at each distance was measured with a ruler along the transect, equidistant between each A and B sampling point at all distances. On the same 204 soil samples from July 2021 that were used for inorganic N extractions, % soil organic matter and pH were measured following the protocols described in Caron et al. 2023. Gravimetric soil moisture was measured for all soils from 2021 and 2022. On the 144 soils from the 0m, 15m, and 60m distances collected in 2021, an aliquot of sieved soil was dried at 65℃ and sent to the Ohio State Service Testing and Research Laboratory for total elemental analysis via the EPA 3051A acid digestion (U.S. EPA 2007). The same 144 soils were also ground to a fine powder in a ball-mill and combusted on an elemental analyzer for total C and N analysis.

  Leaf litter sampling and processing

  In September 2021, leaf litter depth and litter composition were measured at each distance from the forest edge along the transect. To assess leaf litter composition, a 20 cm x 20 cm square was placed upon the forest floor in between the A and B sampling points, a knife was used to cut along the edges of the square, and all litter within the square was collected. Within 12 h of collection, the litter was dried at 65℃ for 48 h and then weighed for total litter mass. Forest floor litter samples were then sorted by hand into hardwood litter, pine litter, and other litter. The weight of litter in each category was recorded for each sample.

  Falling leaf litter from the forest interior at each transect was collected for one full year from mid-September 2021 until mid-September 2022 via litter collection baskets. Each litter collection basket was 28cm x 33cm x 33cm and covered by 1.5 mm mesh to keep the leaf litter from touching the forest floor and prevent the collection of water. Two leaf litter collection baskets were installed at the A and B sampling points at either 30m or 60m from the forest edge at each transect, depending on which distance was most representative of the whole-transect vegetation. Litterfall was collected at multiple time points at all transects to capture both hardwood and pine litterfall timing: late October 2021, mid-November 2021, early December 2021, late May 2022, early August 2022, and mid-October 2022. During each collection, the leaf litter from the two baskets at each transect was pooled into the same bag, dried for at least 48 h at 65℃, and weighed. To obtain litterfall chemistry, leaf litter from the peak litterfall collection (mid-November 2021) was homogenized and ground into a fine powder for total elemental analysis using a ball-mill, resulting in one sample per transect (24 total samples). Total litterfall C, N, and P were measured according to the same protocols as soil total C, N, and P.

  Aboveground vegetation surveys

  In June 2022, at each transect, a 10m x 10m square plot was created at each distance from the forest edge centered around the sampling points. The diameter at 1.4m from the ground (Diameter at breast height, DBH) of all trees was measured for every tree greater than 5cm DBH. Each tree was identified to genus and to species if possible. The height of the tallest tree in the plot was measured using a hypsometer. In each plot, the number of white pine saplings under 5 cm DBH was counted to assess the density of white pine encroachment.

  To survey understory vegetation composition at each sampling point, we established 1m x 1m quadrats around each A and B sampling point. In each plot, the understory cover was grouped into categories: broad-leaf herbs, grassy herbs, ferns, shrubs, vines, moss, white pine seedlings, oak seedlings, other tree seedlings, leaf litter, bare soil, and rock. The coverage of each category within the plot was assessed using the Braun-Blanquet method (Braun-Blanquet 1932, Matteuci and Colma 2002): groups with 75-100% coverage were scored a 5, 50-75% = 4, 25-50% = 3, 5-25% = 2, greater than 0 - 5% = 1, 0% = 0. Groups with 1-2 individuals were scored as “rare.”

  Root density

  Root density was measured at all distances along the transect. One 10.2 cm deep, 5.1 cm diameter soil core was taken at each distance between the A and B sampling points using a hammer core. All roots that could not pass through a 2mm sieve were extracted from the core, dried for 48 h at 60℃ and weighed. Root density was calculated as the total dry weight of roots per cubic cm of the core.

  Microbial DNA extraction, amplicon sequencing, bioinformatics, and qPCR

  During the July 2021 bulk soil collection, a 50 mL tube of sieved soil was collected from each sample and frozen at -80℃ for subsequent DNA extraction. Total soil DNA was extracted from 194 of the frozen sieved bulk soil samples. Approximately 0.25 g of soil was used for extraction with the DNeasy PowerSoil Kit (QIAGEN, Hilden, Germany). To amplify microbial DNA, we used modified versions of the primer set fITS7 and ITS4 for fungi (amplifying the ITS2 region of rDNA, Anthony et al. 2017) and modified versions of the primer set 515f and 806r for bacteria (amplifying the v4 16S region of rDNA, Caporaso et al. 2011) that contained both the Illumina adapter and individual sample indexes. Amplicon quality was checked via agarose gel electrophoresis, then amplicons were cleaned with the Just-a-Plate 96 PCR Purification and Normalization Kit (Charm Biotech, MO) and quantified using the Qubit HS-dsDNA kit (Invitrogen, Carlsbad, CA). To prepare libraries for sequencing, 16S and ITS amplicons were each pooled at 25ng of DNA per sample, then both 16S and ITS amplicons were combined into a single library for sequencing, with two libraries in total: one library for all “A” sampling points, and one library for all “B” sampling points. Each library was subject to 250 base pair (bp) paired-end sequencing on an Illumina MiSeq run at the TUFTS

  Bioinformatics were performed in R (version 4.2.1), where the R package dada2 was used for sequence quality control, paired-end assembly, identification of amplicon sequence variants (ASVs), and taxonomy assignment (Caporaso et al. 2011). Taxonomy was assigned to ASVs using the naive Bayesian classifier method (Wang et al. 2007) in combination with the UNITE database (v. 9.0) as the reference for fungal ITS ASVs (Abarenkov et al. 2022, Nilsson et al. 2018), and the SILVA database (release 138.1, Quast et al. 2012) as the reference for bacterial ASVs. For ITS sequences, 13 low-read samples (with less than 8000 reads after all filtering steps) were removed from analysis, resulting in a total of 181 samples with ITS sequence data. For 16S sequences, 3 samples were determined as outliers after running a Non-Metric Multidimensional Scaling analysis using an Aitchison distance matrix calculated using a centered log ratio-transformed ASV table (Gloor et al. 2017). The 3 outliers were removed, resulting in a total of 191 samples with 16S sequence data.

  To estimate total fungal and bacterial abundances, total ITS2 and 16S v4 abundances were quantified on all DNA extracts via qPCR using the same primer sets used for amplicon sequencing: fITS7/ITS4 for fungi (White et al. 1990, Caporaso et al. 2011) and 515f/806r for bacteria (Caporaso et al. 2011). qPCR was completed following the protocol outlined in Tatsumi et al. 2023.

  Microbial community metrics

  We measured microbial community characteristics across 6 different categories of microbial community traits: functional guild composition, estimated gene proportions, estimated gene proportions derived from functional guilds, relative abundances of genera contributing the most estimated genes, relative abundances of taxa within indicator co-occurrence modules, and alpha diversity metrics.

  Functional guild composition

  To obtain the composition of fungal and bacterial functional guilds, the ITS and 16S ASV tables were normalized to relative abundances by dividing each ASV count in each sample via the ratio method, by the total number of reads in that sample. Fungal genera were assigned functional guilds, and ectomycorrhizal fungi were assigned exploration types, using the FungalTraits database (Põlme et al. 2020). Bacterial taxa were assigned as copiotrophs or oligotrophs, and into nutrient-cycling functional guilds (cellulolytic, chitinolytic, lignolytic, nitrifiers, N fixers, denitrifiers, P-cycling) using literature reviews and genomic pathway presence (Averill et al. 2020). The proportion of functional guilds in each sample was calculated by summing the relative abundances of ASVs belonging to each functional guild.

  Alpha diversity metrics

  Fungal and bacterial diversity were calculated on rarefied ASV tables using Shannon’s diversity index and fungal and bacterial evenness were calculated by dividing the Shannon’s diversity index by the natural log of the richness.

  Bacterial and fungal estimated gene proportions

  Abundances of bacterial Enzyme Commission (E.C.) numbers per ASV per sample were estimated via PICRUST2 (Douglas et al. 2020). 438 ASVs out of 20,204 were able to be assigned estimated E.C. numbers. Abundances of fungal GO terms per ASV per sample were estimated using a method similar to that described in Anthony et al (2022). Fungal genome GO annotations were downloaded from JGI’s Mycocosm All-Species Fungi Tree (Grigoriev et al. 2014) for all available published genomes belonging to fungal genera present in our dataset. The GO term gene copy numbers were standardized by the size of the species’ genome (REF). When there was an exact species genome match, each ASV of that species was assigned the standardized GO term gene copy number in the genome. When there was not an exact species genome match, but genomes from other species in the genus were available, that ASV was assigned the average standardized GO term gene copy number for all available species in the genus ASVs without genus-level taxonomic assignments were not included in the analysis. GO term copy numbers were then weighted by the normalized relative abundances of each ASV in each sample. The relative abundances of GO term copy numbers per sample were calculated by summing the relative abundances of GO terms in all ASVs in each sample.

  To calculate the gene proportions encoding enzymes involved in N and P cycling in each sample, we summed the relative abundances of all GO terms or E.C. numbers encoding the enzymes. To calculate the gene proportions of these enzymes derived from specific functional guilds in each sample, we summed the relative abundances of all GO terms or E.C. numbers encoding these enzymes that are derived from the functional guilds listed.

  Relative abundances of genera contributing the most genes

  To identify the fungal and bacterial genera that contributed the most GO term or E.C. number abundances for chitinases, proteases, phosphatases, and nitrification enzymes across all samples, we summed the total GO terms or EC numbers from each ASV in all samples, grouped by genus. Using our ASV tables normalized to relative abundances, in each sample, we then calculated the cumulative relative abundance of the top 3 genera contributing the most GO term or E.C. number abundances for each enzyme category.

  Indicator co-occurrence modules

  To identify clusters of commonly co-occurring taxa that are associated positively or negatively with nutrient cycling rates (“indicator taxa modules”), we ran a Weighted Gene Correlation Network Analysis (WGCNA) using the WGCNA R package (cite), on our 16S and ITS ASV tables. We input net ammonification, net nitrification, and net phosphate change rates as the trait data. All ASVs that occurred in less than 2 samples were removed (REF), leaving 2688 fungal ASVs remained and 6459 bacterial ASVs. R package flashClust (cite) was used for detecting outliers of samples with the clustering method “average.” No outliers were detected for ITS or 16S. Modules of commonly co-occurring ASVs were identified using the dynamic tree cut method based on dissTOM hierarchical clustering, with deepSplit = 2 and minModuleSize = 4. Modules with dissimilarity coefficients of 0.9 were merged for ITS and coefficients of 0.8 were merged for 16S. The moduleTraitCor function was used to correlate rates of nutrient cycling with co-occurrence modules. Significant correlations between a co-occurrence module first principal component and net ammonification, net nitrification, or net phosphate change were identified with the function corPvalueStudent.

  For the modules with highly significant relationships with nutrient cycling rates (p less than or equal to 0.001), a list of ASVs that most strongly belonged in the module (kME greater than 0.66) was compiled. Then, for all the modules that were significantly positively or negatively associated with each nutrient cycling metric, the cumulative relative abundances were summed of all ASVs with a kME greater than 0.66 in those modules, using an ASV table normalized to relative abundances. This created a metric of “module abundance” for bacteria and fungi associated with high ammonification, low ammonification, high nitrification, low nitrification, high phosphate release, and low phosphate release modules.

• Organization: Harvard Forest. 324 North Main Street, Petersham, MA 01366, USA. Phone (978) 724-3302. Fax (978) 724-3595.

• Project: The Harvard Forest Long-Term Ecological Research (LTER) program examines ecological dynamics in the New England region resulting from natural disturbances, environmental change, and human impacts. (ROR).

• Funding: National Science Foundation LTER grants: DEB-8811764, DEB-9411975, DEB-0080592, DEB-0620443, DEB-1237491, DEB-1832210. Other funding: Department of Energy Biological and Environmental Research grants: DE-SC002040 and DE-SC0012704; Charles Bullard Fellowship

• Use: This dataset is released to the public under Creative Commons CC0 1.0 (No Rights Reserved). Please keep the dataset creators informed of any plans to use the dataset. Consultation with the original investigators is strongly encouraged. Publications and data products that make use of the dataset should include proper acknowledgement.

• License: Creative Commons Zero v1.0 Universal (CC0-1.0)

• Citation: Bhatnagar J, Vietorisz C, Policelli N. 2025. Microbial, Plant, and Soil Impacts on Soil Nutrient Cycling in Harvard Forest and Greater Boston 2021-2022. Harvard Forest Data Archive: HF456 (v.1). Environmental Data Initiative: https://doi.org/10.6073/pasta/a387b147236481983365b6c01a47c2b1.

Detailed Metadata

hf456-01: soil, plant and microbe data

1. ID: unique sample ID
2. Site: site from which soil was sampled
3. Transect_Lat: latitude of the transect sampled (unit: degree / missing value: NA)
4. Transect_Long: longitude of the transect sampled (unit: degree / missing value: NA)
5. Urbanization: urbanization status of the site (either rural or urban)
   • rural: rural
   • urban: urban
6. Forest_type: dominant overstory plant community type
   • Encroachment: hardwood with encroaching pine saplings
   • Hardwood: mature hardwood forests
   • Mixed: mixed pine/hardwood
   • Pine: mature pine
7. Distance: distance from the forest edge of the sampling location (unit: meter / missing value: NA)
8. NH4i_final: ammonium concentration of the soil sample (unit: microgramPerGram / missing value: NA)
9. NO3i_final: nitrate concentration of the soil sample (unit: microgramPerGram / missing value: NA)
10. Ammonification: net soil ammonification rate (unit: microgramPerGramPerDay / missing value: NA)
11. Nitrification: net soil nitrification rate (unit: microgramPerGramPerDay / missing value: NA)
12. N_min: net soil nitrogen mineralization rate (unit: microgramPerGramPerDay / missing value: NA)
13. Al: soil total aluminum content (unit: microgramPerGram / missing value: NA)
14. As: soil total arsenic content (unit: microgramPerGram / missing value: NA)
15. B: soil total boron content (unit: microgramPerGram / missing value: NA)
16. Ba: soil total barium content (unit: microgramPerGram / missing value: NA)
17. Ca: soil total calcium content (unit: microgramPerGram / missing value: NA)
18. Cd: soil total cadmium content (unit: microgramPerGram / missing value: NA)
19. Co: soil total cobalt content (unit: microgramPerGram / missing value: NA)
20. Cr: soil total chromium content (unit: microgramPerGram / missing value: NA)
21. Cu: soil total copper content (unit: microgramPerGram / missing value: NA)
22. Fe: soil total iron content (unit: microgramPerGram / missing value: NA)
23. K: soil total potassium content (unit: microgramPerGram / missing value: NA)
24. Li: soil total lithium content (unit: microgramPerGram / missing value: NA)
25. Mg: soil total magnesium content (unit: microgramPerGram / missing value: NA)
26. Mn: soil total manganese content (unit: microgramPerGram / missing value: NA)
27. Mo: soil total molybdenum content (unit: microgramPerGram / missing value: NA)
28. Na: soil total sodium content (unit: microgramPerGram / missing value: NA)
29. Ni: soil total nickel content (unit: microgramPerGram / missing value: NA)
30. P: soil total phosphous content (unit: microgramPerGram / missing value: NA)
31. Pb: soil total lead content (unit: microgramPerGram / missing value: NA)
32. S: soil total sulfur content (unit: microgramPerGram / missing value: NA)
33. Sb: soil total antimony content (unit: microgramPerGram / missing value: NA)
34. Se: soil total selenium content (unit: microgramPerGram / missing value: NA)
35. Si: soil total silicon content (unit: microgramPerGram / missing value: NA)
36. Sr: soil total strontium content (unit: microgramPerGram / missing value: NA)
37. Tl: soil total thallium content (unit: microgramPerGram / missing value: NA)
38. V: soil total vanadium content (unit: microgramPerGram / missing value: NA)
39. Zn: soil total zinc content (unit: microgramPerGram / missing value: NA)
40. pH: soil pH (unit: dimensionless / missing value: NA)
41. soil_temp: soil temperature (unit: celsius / missing value: NA)
42. litter_depth: litter layer depth (unit: centimeter / missing value: NA)
43. O_depth: soil organic layer depth (unit: centimeter / missing value: NA)
44. soil_moisture: proportion soil water content by weight (unit: dimensionless / missing value: NA)
45. SOM: proportion soil organic matter content by weight (unit: dimensionless / missing value: NA)
46. prop_pine_litter: proportion of forest floor litter that is pine litter, measured in a 20 x 20cm square above the sampling point (unit: dimensionless / missing value: NA)
47. prop_hw_litter: proportion of forest floor litter that is hardwood litter, measured in a 20 x 20cm square above the sampling point (unit: dimensionless / missing value: NA)
48. total_floor_litter: total weight of forest floor litter, measured in a 20 x 20cm square above the sampling point (unit: gram / missing value: NA)
49. pine_floor_litter: total weight of forest floor pine litter, measured in a 20 x 20cm square above the sampling point (unit: gram / missing value: NA)
50. hw_floor_litter: total weight of forest floor hardwood litter, measured in a 20 x 20cm square above the sampling point (unit: gram / missing value: NA)
51. Broad_leaf_herbs: understory cover of broad-leaf herbs, measured in a 1m x 1m square around the sampling point using the Braun-Blanquet method on a scale of 0-5
    • 0: 0% coverage
    • 1: trace-5%
    • 2: 5-25% coverage
    • 3: 25-50% coverage
    • 4: 50-75% coverage
    • 5: 75-100% coverage
52. Grassy_herbs: understory cover of grassy herbs, measured in a 1m x 1m square around the sampling point using the Braun-Blanquet method on a scale of 0-5
    • 0: 0% coverage
    • 1: trace-5%
    • 2: 5-25% coverage
    • 3: 25-50% coverage
    • 4: 50-75% coverage
    • 5: 75-100% coverage
53. Ferns: understory cover of broad-leaf herbs, measured in a 1m x 1m square around the sampling point using the Braun-Blanquet method on a scale of 0-5
    • 0: 0% coverage
    • 1: trace-5%
    • 2: 5-25% coverage
    • 3: 25-50% coverage
    • 4: 50-75% coverage
    • 5: 75-100% coverage
54. Shrubs: understory cover of woody shrubs, measured in a 1m x 1m square around the sampling point using the Braun-Blanquet method on a scale of 0-5
    • 0: 0% coverage
    • 1: trace-5%
    • 2: 5-25% coverage
    • 3: 25-50% coverage
    • 4: 50-75% coverage
    • 5: 75-100% coverage
55. Vines: understory cover of vines, measured in a 1m x 1m square around the sampling point using the Braun-Blanquet method on a scale of 0-5
    • 0: 0% coverage
    • 1: trace-5%
    • 2: 5-25% coverage
    • 3: 25-50% coverage
    • 4: 50-75% coverage
    • 5: 75-100% coverage
56. Moss: understory cover of moss, measured in a 1m x 1m square around the sampling point using the Braun-Blanquet method on a scale of 0-5
    • 0: 0% coverage
    • 1: trace-5%
    • 2: 5-25% coverage
    • 3: 25-50% coverage
    • 4: 50-75% coverage
    • 5: 75-100% coverage
57. White_pine_seedlings: understory cover of white pine seedlings, measured in a 1m x 1m square around the sampling point using the Braun-Blanquet method on a scale of 0-5
    • 0: 0% coverage
    • 1: trace-5%
    • 2: 5-25% coverage
    • 3: 25-50% coverage
    • 4: 50-75% coverage
    • 5: 75-100% coverage
58. Oak_seedlings: understory cover of oak seedlings, measured in a 1m x 1m square around the sampling point using the Braun-Blanquet method on a scale of 0-5
    • 0: 0% coverage
    • 1: trace-5%
    • 2: 5-25% coverage
    • 3: 25-50% coverage
    • 4: 50-75% coverage
    • 5: 75-100% coverage
59. Other_tree_seedlings: understory cover of tree seedlings other than white pine or oak, measured in a 1m x 1m square around the sampling point using the Braun-Blanquet method on a scale of 0-5
    • 0: 0% coverage
    • 1: trace-5%
    • 2: 5-25% coverage
    • 3: 25-50% coverage
    • 4: 50-75% coverage
    • 5: 75-100% coverage
60. Leaf_litter: understory cover of exposed leaf litter, measured in a 1m x 1m square around the sampling point using the Braun-Blanquet method on a scale of 0-5
    • 0: 0% coverage
    • 1: trace-5%
    • 2: 5-25% coverage
    • 3: 25-50% coverage
    • 4: 50-75% coverage
    • 5: 75-100% coverage
61. Bare_soil: understory cover of exposed soil, measured in a 1m x 1m square around the sampling point using the Braun-Blanquet method on a scale of 0-5
    • 0: 0% coverage
    • 1: trace-5%
    • 2: 5-25% coverage
    • 3: 25-50% coverage
    • 4: 50-75% coverage
    • 5: 75-100% coverage
62. Rock: understory cover of exposed rock, measured in a 1m x 1m square around the sampling point using the Braun-Blanquet method on a scale of 0-5
    • 0: 0% coverage
    • 1: trace-5%
    • 2: 5-25% coverage
    • 3: 25-50% coverage
    • 4: 50-75% coverage
    • 5: 75-100% coverage
63. Max_height: maximum height of understory vegetation (unit: meter / missing value: NA)
64. No_white_pine_seedlings: number of white pine seedlings in a in a 1m x 1m square centered around the sampling point (unit: dimensionless / missing value: NA)
65. No_oak_seedlings: number of oak seedlings in a in a 1m x 1m square centered around the sampling point (unit: dimensionless / missing value: NA)
66. All_plants: cumulative sum of all understory plant scores (specifically Broad_leaf_herbs, Grassy_herbs, Ferns, Shrubs, White_pine_seedlings, Oak_seedlings, Other_tree_seedlings) using the Braun-Blanquet method (unit: dimensionless / missing value: NA)
67. AM_basal: basal area of all arbuscular mycorrhizal trees greater than 5cm diameter at breast height in a 10m x 10m square plot centered between the A and B sampling points (unit: centimeterCubed / missing value: NA)
68. ECM_basal: basal area of all ectomycorrhizal trees greater than 5cm diameter at breast height in a 10m x 10m square plot centered between the A and B sampling points (unit: centimeterCubed / missing value: NA)
69. hw_basal: basal area of all hardwood trees greater than 5cm diameter at breast height in a 10m x 10m square plot centered between the A and B sampling points (unit: centimeterCubed / missing value: NA)
70. pine_basal: basal area of all white pine trees greater than 5cm diameter at breast height in a 10m x 10m square plot centered between the A and B sampling points (unit: centimeterCubed / missing value: NA)
71. num_stems: number of trees larger than 5cm diameter at breast height in a 10m x 10m square plot centered between the A and B sampling points (unit: dimensionless / missing value: NA)
72. total_basal: cumulative basal area of all trees greater than 5cm diameter at breast height in a 10m x 10m square plot centered between the A and B sampling points (unit: centimeterCubed / missing value: NA)
73. ECM_AM: ratio of ectomycorrhizal:arbuscular mycorrhizal tree basal area of all trees greater than 5cm diameter at breast height in a 10m x 10m square plot centered between the A and B sampling points (unit: dimensionless / missing value: NA)
74. no_pine_saplings: number of white pine saplings smaller than 5cm diameter at breast height in a 10m x 10m square plot centered between the A and B sampling points (unit: dimensionless / missing value: NA)
75. no_oak_saplings: number of oak saplings smaller than 5cm diameter at breast height in a 10m x 10m square plot centered between the A and B sampling points (unit: dimensionless / missing value: NA)
76. PO4: soil phosphate concentration (unit: microgramPerGram / missing value: NA)
77. PO4_release: net soil phosphate release or uptake rate (unit: microgramPerGramPerDay / missing value: NA)
78. root_density: root density in the top 10.2cm of soil (unit: gramPerCentimeterCubed / missing value: NA)
79. ECM_abundance: relative abundance of ectomycorrhizal fungi in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
80. SAP_abundance: relative abundance of saprotrophic fungi in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
81. soilSAP_abundance: relative abundance of soil saprotrophic fungi in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
82. AM_abundance: relative abundance of arbuscular mycorrhizal fungi in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
83. root_endophyte_abundance: relative abundance of root endyphytic fungi in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
84. contact_expl: relative abundance of contact exploration type ectomycorrhizal fungi in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
85. short_dist_expl: relative abundance of short distance exploration type ectomycorrhizal fungi in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
86. med_dist_expl: relative abundance of medium distance exploration type ectomycorrhizal fungi in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
87. long_dist_expl: relative abundance of long distance exploration type ectomycorrhizal fungi in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
88. mat_expl: relative abundance of mat exploration typeectomycorrhizal fungi in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
89. Nitrifier_abundance: relative abundance of nitrifying bacteria in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
90. Decomposer_abundance: relative abundance of decomposer bacteria (summed cellulolytic, lignolytic, and chitinolytic bacteria) in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
91. Cellulolytic_abundance: relative abundance of cellulolytic bacteria in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
92. Lignolytic_abundance: relative abundance of lignolytic bacteria in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
93. Chitinolytic_abundance: relative abundance of chitinolytic bacteria in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
94. Copiotroph_abundance: relative abundance of copiotrophic bacteria in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
95. Oligotroph_abundance: relative abundance of oligotrophic bacteria in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
96. Ndecomp_EC_abund: relative abundance of chitinase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
97. oxidation_EC_abund: relative abundance of oxidase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
98. celluloseDecomp_EC_abund: relative abundance of cellulose decomposition gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
99. carbDecomp_EC_abund: relative abundance of carbohydrate decomposition gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
100. Pdecomp_EC_abund: relative abundance of phosphatase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
101. peptidase_EC_abund: relative abundance of peptidase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
102. Ndecomp_GO_abund: relative abundance of chitinase gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
103. oxidation_GO_abund: relative abundance of oxidase gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
104. celluloseDecomp_GO_abund: relative abundance of cellulose decomposition gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
105. carbDecomp_GO_abund: relative abundance of carbohydrate decomposition gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
106. orgNuptake_GO_abund: relative abundance of organic nitrogen uptake gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
107. PO4uptake_GO_abund: relative abundance of phosphatase uptake gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
108. nitrification_GO_abund: relative abundance of nitrification gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
109. NH4uptake_GO_abund: relative abundance of ammonium gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
110. Pdecomp_GO_abund: relative abundance of phosphatase gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
111. peptidase_GO_abund: relative abundance of peptidase gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
112. ECM_Ndecomp_GO: relative abundance of ectomycorrhizal chitinase gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
113. SAP_Ndecomp_GO: relative abundance of saprotrophic fungal chitinase gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
114. ECM_orgNuptake_GO: relative abundance of ectomycorrhizal organic nitrogen uptake gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
115. SAP_orgNuptake_GO: relative abundance of saprotrophic fungal organic nitrogen uptake gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
116. ECM_PO4uptake_GO: relative abundance of ectomycorrhizal phosphate uptake gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
117. SAP_PO4uptake_GO: relative abundance of saprotrophic fungal phosphate uptake gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
118. ECM_NH4uptake_GO: relative abundance of ectomycorrhizal ammonium uptake gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
119. SAP_NH4uptake_GO: relative abundance of saprotrophic fungal ammonium uptake gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
120. ECM_Pdecomp_GO: relative abundance of ectomycorrhizal phosphatase gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
121. SAP_Pdecomp_GO: relative abundance of saprotrophic fungal phosphatase gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
122. ECM_peptidase_GO: relative abundance of ectomycorrhizal peptidase gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
123. SAP_peptidase_GO: relative abundance of saprotrophic fungal peptidase gene copies in the fungal community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
124. Oligotroph_Ndecomp_EC: relative abundance of oligotrophic bacterial chitinase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
125. chitinolytic_Ndecomp_EC: relative abundance of chitinolytic bacterial chitinase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
126. cellulolytic_Ndecomp_EC: relative abundance of cellulolytic bacterial chitinase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
127. lignolytic_Ndecomp_EC: relative abundance of lignolytic bacterial chitinase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
128. Oligotroph_Pdecomp_EC: relative abundance of oligotrophic bacterial phosphatase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
129. Oligotroph_peptidase_EC: relative abundance of oligotrophic bacterial peptidase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
130. chitinolytic_peptidase_EC: relative abundance of chitinolytic bacterial peptidase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
131. cellulolytic_peptidase_EC: relative abundance of cellulolytic bacterial peptidase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
132. lignolytic_peptidase_EC: relative abundance of lignolytic bacterial peptidase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
133. copiotroph_peptidase_EC: relative abundance of copiotrophic bacterial peptidase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
134. peptidase_phosphatase_top_fun_abundance: relative abundance of top 3 fungal genera contributing most peptidase and phosphatase genes (unit: dimensionless / missing value: NA)
135. Ndecomp_top_fun_abundance: relative abundance of top 3 fungal genera contributing most chitinase and peptidase genes (unit: dimensionless / missing value: NA)
136. chitinase_top_bac_abundance: relative abundance of top 3 bacterial genera contributing most chitinase genes (unit: dimensionless / missing value: NA)
137. peptidase_top_bac_abundance: relative abundance of top 3 bacterial genera contributing most peptidase genes (unit: dimensionless / missing value: NA)
138. NPdecomp_top_bac_abundance: relative abundance of top 3 bacterial genera contributing most phosphatase genes (unit: dimensionless / missing value: NA)
139. fun_amm_neg_module_abundance: cumulative relative abundance of fungal taxa in low ammonification module as identified by Weighted Gene Network Correlation Analysis (unit: dimensionless / missing value: NA)
140. fun_nitr_pos_module_abundance: cumulative relative abundance of fungal taxa in high nitrification module as identified by Weighted Gene Network Correlation Analysis (unit: dimensionless / missing value: NA)
141. fun_nitr_neg_module_abundance: cumulative relative abundance of fungal taxa in low nitrification module as identified by Weighted Gene Network Correlation Analysis (unit: dimensionless / missing value: NA)
142. fun_pmin_pos_module_abundance: cumulative relative abundance of fungal taxa in high phosphate change module as identified by Weighted Gene Network Correlation Analysis (unit: dimensionless / missing value: NA)
143. fun_pmin_neg_module_abundance: cumulative relative abundance of fungal taxa in low phosphate change module as identified by Weighted Gene Network Correlation Analysis (unit: dimensionless / missing value: NA)
144. bac_amm_pos_module_abundance: cumulative relative abundance of bacterial taxa in high ammonification module as identified by Weighted Gene Network Correlation Analysis (unit: dimensionless / missing value: NA)
145. bac_amm_neg_module_abundance: cumulative relative abundance of bacterial taxa in low ammonification module as identified by Weighted Gene Network Correlation Analysis (unit: dimensionless / missing value: NA)
146. bac_nitr_pos_module_abundance: cumulative relative abundance of bacterial taxa in high nitrification module as identified by Weighted Gene Network Correlation Analysis (unit: dimensionless / missing value: NA)
147. bac_nitr_neg_module_abundance: cumulative relative abundance of bacterial taxa in low nitrification module as identified by Weighted Gene Network Correlation Analysis (unit: dimensionless / missing value: NA)
148. bac_pmin_pos_module_abundance: cumulative relative abundance of bacterial taxa in high phosphate change module as identified by Weighted Gene Network Correlation Analysis (unit: dimensionless / missing value: NA)
149. bac_pmin_neg_module_abundance: cumulative relative abundance of bacterial taxa in low phosphate change module as identified by Weighted Gene Network Correlation Analysis (unit: dimensionless / missing value: NA)
150. Nitrosomonodaceae_abundance: relative abundance of bacteria in the Nitrosomonodaceae family in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
151. bac_richness: bacterial richness in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
152. bac_shannon: bacterial shannon diversity in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
153. fun_richness: fungal richness in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
154. fun_shannon: fungal shannon diversity in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
155. bac_evenness: bacterial evenness in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
156. fun_evenness: fungal evenness in high-throughput amplicon sequence data (unit: dimensionless / missing value: NA)
157. edaphic_PC1: first principal component of a Principal Coordinates Analysis encompassing all soil abiotic variables: pH, soil_temp, litter_depth, O_depth, soil_moisture, SOM (unit: dimensionless / missing value: NA)
158. bac_copy_number: copy numbers of the v4 16S DNA region per 1ul DNA extract as measured by quantitative PCR (unit: dimensionless / missing value: NA)
159. ITS_copy_number: copy numbers of the ITS2 DNA region per 1ul DNA extract as measured by quantitative PCR (unit: dimensionless / missing value: NA)
160. fun_amm_pos_module_abundance: cumulative relative abundance of fungal taxa in high ammonification module as identified by Weighted Gene Network Correlation Analysis (unit: dimensionless / missing value: NA)
161. Cmetabolism_EC_abund: relative abundance of central carbon metabolism gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
162. denitrification_EC_abund: relative abundance of denitrification gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
163. soil_temp2022: soil temperature in 2022 (unit: celsius / missing value: NA)
164. soil_moisture2022: proportion soil water content by weight in 2022 (unit: dimensionless / missing value: NA)
165. Copiotroph_Ndecomp_EC: relative abundance of copiotrophic bacterial chitinase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
166. Copiotroph_Pdecomp_EC: relative abundance of copiotrophic bacterial phosphatase gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
167. Nitrifiers_nitrification_EC: relative abundance of nitrifying bacterial nitrification gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
168. Copios_nitrification_EC: relative abundance of copiotrophic bacterial nitrification gene copies in the bacterial community estimated from amplicon sequence data (unit: dimensionless / missing value: NA)
169. soil_percN: soil total percent nitrogen (unit: dimensionless / missing value: NA)
170. soil_percC: soil total percent carbon (unit: dimensionless / missing value: NA)