You are here

Home » Data & Archives

Harvard Forest Data Archive

HF439

Soil Carbon at Forest Edges along an Urban to Rural Gradient in Massachusetts since 2018

Related Publications

Data

Overview

  • Lead: Sarah Garvey, Pamela Templer, Jennifer Bhatnagar, Lucy Hutyra
  • Investigators:
  • Contact: Information Manager
  • Start date: 2018
  • End date: 2021
  • Status: ongoing
  • Location: Central Massachusetts, Eastern Massachusetts
  • Latitude: +42.21217 to +42.48042 degrees
  • Longitude: -72.50868 to -71.11603 degrees
  • Elevation: 45 to 285 meter
  • Datum: WGS84
  • Taxa:
  • Release date: 2023
  • Language: English
  • EML file: knb-lter-hfr.439.3
  • 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: organic matter movement, land use and land cover change, human-environment interactions
  • Keywords: biogeochemistry, calcium, carbon, land use, magnesium, microbes, organic matter, ph, soil moisture, soil respiration, soil temperature, urban
  • Abstract:

    Global proliferation of forest edges through anthropogenic land-use change and forest fragmentation is well documented, and while forest fragmentation has clear consequences for soil carbon (C) cycling, underlying drivers of belowground activity at the forest edge remain poorly understood. Increasing soil C losses via respiration have been observed at rural forest edges, but this process was suppressed at urban forest edges. We offer a comprehensive, coupled investigation of abiotic soil conditions and biotic soil activity from forest edge to interior at eight sites along an urbanization gradient to elucidate how environmental stressors are linked to soil C cycling at the forest edge. Despite significant diverging trends in edge soil C losses between urban and rural sites, we did not find comparable differences in soil % C or microbial enzyme activity, suggesting an unexpected decoupling of soil C fluxes and pools at forest edges. We demonstrate that across site types, soils at forest edges were less acidic than the forest interior (p less than 0.0001), and soil pH was positively correlated with soil calcium, magnesium and sodium content (adj R2 = 0.37), which were also elevated at the edge. Compared to forest interior, forest edge soils exhibited a 17.8% increase in sand content and elevated freeze-thaw frequency with probable downstream effects on root turnover and decomposition. Using these and other novel forest edge data, we demonstrate that significant variation in edge soil respiration (adj R2 = 0.46; p = 0.0002) and C content (adj R2 = 0.86; p less than 0.0001) can be explained using soil parameters often mediated by human activity (e.g., soil pH, trace metal and cation concentrations, soil temperature), and we emphasize the complex influence of multiple, simultaneous global change drivers at forest edges. Forest edge soils reflect legacies of anthropogenic land-use and modern human management, and this must be accounted for to understand soil activity and C cycling across fragmented landscapes.

  • Methods:

    Field data collection

    This study was conducted along an urbanization gradient from urban Boston to rural central Massachusetts (MA) as part of the Urban New England (UNE) project. Eight field sites were established at forest edges in temperate, mixed-deciduous forests from May to June 2018.We characterized the degree of urbanization intensity surrounding each forest site using impervious surface area (ISA; Mass GIS Data: Impervious Surface 2005, 2007), population density (US Census Bureau, 2010), and distance to Boston Common in downtown Boston (sensu Raciti et al., 2012), as well as whether the field site fell within the boundaries of a Census Designated Place (CDP) (US Census Bureau, 2021). Four sites were classified as rural and four as urban.

    At each of the eight sites, we established a single transect orthogonal to the forest edge (defined by the tree stem-line) and extending 90 m into the forest interior. Along each transect, we installed a pair of PVC soil respiration collars at 0, 15, 30, 60, and 90 m from the forest edge (n=10 collars per transect; 80 collars total). All field data provided reflects the mean of the two replicates at each distance from the edge.

    Soil sampling: extracellular enzymes and other soil properties

    Soils samples were co-located with soil respiration collars and collected during the 2018 and 2019 growing seasons. Organic horizon soils assayed for potential extracellular enzyme activity (EEA) were collected once in the early growing season (May 21 – 23, 2019), twice in the mid-season (August 15 – 23, 2018 and July 15 – 17, 2019), and once in the late season (Sept 17 – 20, 2019). Samples were assayed for the following enzymes: cellobiohydrolase (CBH; an exocellulase), β-glucosidase (BG; hydrolyzes cellobiose into glucose), β-xylosidase (BX; degrades xylose in hemicellulose), β-glucuronidase (BGLU; degrades glucuronic acid in hemicellulose), α-glucosidase (AG; degrades starch), β-N-acetyl- glucosaminidase (NAG; degrades chitin), acid phosphatase (AP; mineralizes phosphorus), leucine aminopeptidase (LAP; degrades polypeptides), polyphenol oxidase (PPO), and peroxidase (PER). Principle component analysis was applied to mean potential EEA per g soil across sampling events to reduce the 10 enzyme variables to two factors, principle components 1 and 2 (soil enzyme PC1 and soil enzyme PC2, respectively). This was repeated for resource-specific enzyme activity per g SOM (resource-specific enzyme PC1 and resource-specific enzyme PC2, respectively).

    August 2018 soil samples were also subsampled for soil % C, % N and trace element analysis, and July 2019 samples were subsampled for soil pH measurements. % C and N content were determined using an elemental analyzer (NC 2500 Elemental Analyzer, CE Elantech, Lakewood, NJ, USA). The Cornell Nutrient Analysis Lab determined total elemental concentrations of Mg, Ca, Na, Pb, and As using EPA method 3050 for soil sample digestion on a hot block with nitric acid and 6010B for ICP-OES.

    Soil cores were collected in July 2019, and mineral horizon soil was used for soil texture analysis (hydrometer method).

    In situ soil monitoring

    Sensors at seven UNE field sites (four urban, three rural) logged soil temperatures (5 cm depth) half-hourly from July 6, 2018 to November 22, 2019 using Onset HOBO Pendant Data Loggers. Soils were considered frozen if the 24 h mean soil temperatures were less than -0.5 ˚C. Freeze-thaw transitions were defined as shifts where soils were frozen one day, and then not frozen the following day.

  • 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: USDA NIFA grants: 2017-67003-26615 and 2017-67003-26487; NSF Traineeship grant to Boston University DGE-1735087; Garden Club of America Zone VI 2020 Fellowship in Urban Forestry; Boston University Biogeoscience Program.

  • 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: Garvey S, Templer P, Bhatnagar J, Hutyra L. 2023. Soil Carbon at Forest Edges along an Urban to Rural Gradient in Massachusetts since 2018. Harvard Forest Data Archive: HF439 (v.3). Environmental Data Initiative: https://doi.org/22445ee40b1ee92dc95db50757a02463.

Detailed Metadata

hf439-01: forest edge site data

  1. site: name of site
  2. site_type: site classification
    • Rural: rural
    • Urban: urban
  3. edge_type: adjacent land type
    • Non-Road: non-road
    • Road: road
  4. dfe: distance from forest edge (unit: meter / missing value: NA)
  5. pct_C_mean: mean organic horizon soil % C for a given distance from the edge. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  6. pct_C_ste: standard error on the mean soil % C for a given distance from the edge. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  7. pH_mean: mean organic horizon soil pH for a given distance from the edge. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (pH = -log10[H+]) (unit: dimensionless / missing value: NA)
  8. pH_ste: standard error on the mean soil pH for a given distance from the edge. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  9. As_mg_kg_mean: mean organic horizon soil arsenic (As) for a given distance from the edge in mg per kg soil. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  10. As_mg_kg_ste: standard error on the mean soil As for a given distance from the edge in mg per kg soil. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  11. Pb_mg_kg_mean: mean organic horizon soil lead (Pb) for a given distance from the edge in mg per kg soil. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  12. Pb_mg_kg_ste: standard error on the mean soil Pb for a given distance from the edge in mg per kg soil. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  13. Ca_mg_kg_mean: mean organic horizon soil calcium (Ca) for a given distance from the edge in mg per kg soil. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  14. Ca_mg_kg_ste: standard error on the mean soil Ca for a given distance from the edge in mg per kg soil. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  15. Na_mg_kg_mean: mean organic horizon soil sodium (Na) for a given distance from the edge in mg per kg soil. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  16. Na_mg_kg_ste: standard error on the mean soil Na for a given distance from the edge in mg per kg soil. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  17. Mg_mg_kg_mean: mean organic horizon soil magnesium (Mg) for a given distance from the edge in mg per kg soil. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  18. Mg_mg_kg_ste: standard error on the mean soil Mg for a given distance from the edge in mg per kg soil. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  19. Sand_Pct_mean: mean mineral horizon soil sand content (%) for a given distance from the edge. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  20. Sand_Pct_ste: standard error on the mean soil sand content (%) for a given distance from the edge. Reflects measurements from 2 replicates (A & B) at each distance from the edge. (unit: dimensionless / missing value: NA)
  21. number_freeze_thaw_transitions: number of 24 h freeze-thaw transitions logged at a given distance from the edge from July 6, 2018 to November 22, 2019 (unit: dimensionless / missing value: NA)
  22. num_days_frozen: number of days (24 h) frozen logged at a given distance from the edge from July 6, 2018 to November 22, 2019 (unit: dimensionless / missing value: NA)
  23. Soil_enzyme_PC1: soil enzyme PC1 resulting from principle component analysis of mean potential EEA per g soil across sampling events (unit: dimensionless / missing value: NA)
  24. Soil_enzyme_PC2: soil enzyme PC2 resulting from principle component analysis of mean potential EEA per g soil across sampling events (unit: dimensionless / missing value: NA)
  25. Resource_specific_PC1: resource-specific enzyme PC1 resulting from principle component analysis of mean potential EEA per g soil organic matter (SOM) across sampling events (unit: dimensionless / missing value: NA)
  26. Resource_specific_PC2: resource-specific enzyme PC2 resulting from principle component analysis of mean potential EEA per g SOM across sampling events (unit: dimensionless / missing value: NA)