EMS - Biomass Inventories

Data

• hf069-01: plot map
• hf069-02: compass distance map
• hf069-03: LAI plot-level
• hf069-04: LAI site-level
• hf069-05: litter chemistry
• hf069-06: canopy chemistry
• hf069-07: litterfall
• hf069-08: litter decomposition
• hf069-09: soil moisture
• hf069-10: soil respiration
• hf069-11: tree data
• hf069-12: coarse woody debris
• hf069-13: fine woody debris
• hf069-14: understory summary
• hf069-15: annual summary
• hf069-16: allometric equations and species names

Overview

• Lead: William Munger, Steven Wofsy
• Investigators: Carol Barford, David Bryant, Victoria Chow, Daniel Curran, Elizabeth Hammond-Pyle, Lucy Hutyra, Christine Jones, Kathryn McKain
• Contact: Leland Werden
• Start date: 1993-01-01
• End date: ongoing
• Location: Prospect Hill Tract (Harvard Forest)
• Latitude: +42.54
• Longitude: -72.17
• Elevation: 340 meters
• Taxa:
• Research topic: plot
• Study type: long-term measurement
• LTER core area: primary production
• Keywords: biomass, CO2, CWD, litter, soil moisture, soil respiration
• Release date: 1999
• EML version: knb-lter-hfr.69.11
• Revisions:
• Related links:
• Project Website
• EMS - Canopy-Atmosphere Exchange of Carbon, Water and Energy
• Abstract:

  In 1993, we installed 40 circular, 10 m radius biometric plots in the footprint of the EMS tower on Prospect Hill. We randomly placed the plots within 100 m increments along ten 500 m transects that extend from the tower in the northwest and southwest directions. In 2001, we removed three plots (G3, H3, H4) from the datasets and ceased measurements there due to their inundation by a beaver pond. In 1999, we installed 6 additional circular, 10 m radius biometric plots on the the Simmes Lot, adjacent to Prospect Hill to study the effects of a selective harvest that occurred there in the winter of 2000-01. In the summer of 2001, we expanded the harvested plots in size to 15 m radius and ceased measurements at one plot (X4) because it was unaffected by the harvest. The harvest also affected three of the original tower plots (A4, A5, B5), which were expanded in size as a part of the harvest plot group. Consequently, there are 34 tower plots and 8 harvest plots. We have taken the following ecological measurements at each site: tree growth, woody debris, litter, leaf area increment (LAI), leaf chemistry, and soil respiration and moisture.

• Methods:

  LAI

  Leaf area index (LAI) data was collected at the biometric plots of the EMS site in 1998, 1999, 2005, 2006, 2007, 2008.

  An LAI2000 sensor was used to make 5 measurements at each plot: one at the center, and four more at points 2 meters away from the center, in the 4 cardinal directions (N,E,S,W). All measurementswere made at sunrise or sunset when the sun was close to the horizon. Plot-based LAI values were calculated using the standard methods of the LiCor program C2000. LAI values were calculated by averaging the 5 subplot values for each plot, and by masking the outer ring of the sensor (ring 5) so as to block out possible sunspots which may have hit the outer ring as the sun rose up from the horizon.

  Leaf Chemistry

  Foliar chemistry data was collected from leaf litter and green canopy leaf samples in 1998-2000, 2006, 2007 and 2008 at the Harvard Forest EMS biometric plots. Samples were dried at 65 degrees C for 3 days, ground, and analyzed at the Harvard Forest on a Fison Nitrogen Analyzer NA 1500 in 2000 and on a Elementar Vario Micro CHNS analyzer in 2007 and 2008.

  Green leaf samples were collected by shooting them down from the canopy with buckshot.

  Litter has been collected, dried, and weighed annually at all plots since 1998. A portion of the collected leaves, fruits and flowers, and wood pieces from 1998, 1999, and 2000 were analyzed for N and C content.

  Litterfall

  Three baskets with a radial area of 0.13 m^2 were randomly placed in each plot at the EMS tower footprint and Harvested sites. Litter collection began at the EMS site (plots A-H, 1-5) in 1998 and the Harvested site (plots X 1-6) in 1999.

  Each year litter was collected throughout the fall and once the following spring, dried and weighed. Each spring's measurement is included in the previous year's dataset.

  The litter was sorted in the following ways:

  1998-1999: Leaf litter was sorted by tree species. The unknown category refers to leaf litter from unidentified species, including leaves that we too fragmented or decayed to be identified. EMS data from 1999 onward excludes the harvested plots (A4, A5, B5), and from 2000 onward, excludes plots lost to a swamp (G4, H3, H4). There was no spring collection in 1999. Rather, the first collection of that year was in mid August. We estimated that 30% of the total litter collected on doy.229, 1999 fell over the winter and was thus summed as part of the 1998 data and 70% of the total litter collected on doy.229 fell during the summer and was thus included in 1999 annual sum.

  2000-2003: Litter was sorted by leaf species, fruits, flowers, and buds, and twigs and barks. Measurements in the Harvested plot, X4, were discontinued after 2001 because it was unaffected by the harvest.

  2004: Litter was sorted by leaves (including needles), and non-leaves (including twigs, bark, fruits, flowers, buds, etc.)

  2005: Litter was sorted by leaves (including needles), twigs and bark, acorns, and (non-acorn) fruits, flowers, and buds.

  2006: Litter was sorted by leaf species, fruits, flowers, and buds, and twigs and barks. "Decomposed matter" was included in the "unknown" category when litter mass was calculated.

  2007: Litter was sorted by leaf species, fruits, flowers, and buds, and twigs and barks by lkw at Boston University. Decomposed matter was sorted as a component of the "unknown" category and all oak species were included in the same category.

  Plots A4, A5, B5, G3, H3, H4 are excluded from all calculations for the EMS site, and the plot X4 is excluded from all calculations for the Harvested site, although they remain in the raw data for years when measurements were taken there.

  At some points in the litter collection, plots and baskets were missed: doy.229 1999: plots B1, C2, and E1 contained birds' nests that had been built in the traps, which thus contained materials that had not fallen directly from the canopy into the baskets. doy.229 and 531 1999: Out of all the Bradys, only the plots A4, A5, and B5 were visited. From doy.299, 2001 through 2005, X1 only had two baskets. On doy.250 2001 there was one less basket in the EMS site. In 2003 H1 only had two baskets. From 2004-05 B2 only had two baskets. 4/23/07: Marsha had 1 basket, A5 had 2 baskets, C2 had 2 baskets, all b/c some baskets had been blown over.

  All of these exceptions were taken into account when calculating annual sums.

  Litter Decomposition

  During the first week of December, 2001, mesh bags (w/ a 1 sq mm mesh size) containing oak, maple, and beech leaves were set out on the ground in five plots.

  The litter samples were oven dried, but then sat on a shelf before they were weighed again and put out for this particular study. Therefore, any moisture the leaf samples may have accumulated between when they were oven dried and when they were weighed is accounted for in the processing.

  Samples were collected on several dates over the following years. Eventually, the bags accumulated too much dirt and roots growing through them that the leaves could not be separated for weighing. This was the case for the samples that have no end weight recorded.

  Two samples, B5 24 and B3 74, gained weight over the time period that they were outside, which means there was either a mistake in collecting or recording the data, or the mesh bags actually gained material while they were outside that was not part of the original sample. These samples were removed from the data in the processing.

  Soil Moisture

  Two TDR probes were located at each plot where soil respiration was measured. Many more measurements were made at the cut-site plots than the EMS plots in the early years because the primary goal at that time was to track a potential change in soil moisture due to the selective harvest that occurred in 2001.

  Two different instruments were used to generate the TDR waveforms. For 1999-2001 a Tektronix cable tester (1502B TDR) was used and in 2002-03 a soil moisture system by Campbell Scientific (TDR100 and CS605 probes) was used.

  Soil Respiration

  Soil respiration was measured throughout the growing season in 1998-2000, 2002-2003, and once in 2006. It was measured in 9 plots in EMS tower footprint (B3, B4, C3, D5, E3, G3, H3, H4, G5) and in 9 plots on the adjacent harvest site (A4, A5, B5, X1-X6). See the plot map for plot locations. There are 5 permanently installed soil collars in each plot. Each collar has a unique number, 1-90. Plots for measuring soil respiration in the EMS tower footprint were chosen to be distributed across a range of soil drainage classes - 3 poorly-drained sites, 3 moderately drained sites, and 3 well-drained sites. The 3 poorly drained plots (G3, H3, H4) were inundated with water from a beaver pond and measurements ceased there in 2001. Measurements at the cut site began in 1999 and a selective harvest occurred on the site in 2001. Measurements at plot X4 ceased in 2001 because the plot was unaffected by the harvest.

  Respiration was measured using a portable soil pack with a LICOR 6252 in a closed chamber system. CO2 was recorded every 5 seconds for between 3.5 and 4.5 min. In 1998, some of the measurements were made at night, after 6pm.

  Soil CO2 flux was calculated assuming a cylindrical chamber, negligible volume in the sample flow lines and Licor cells, and a constant air pressure. Chamber height was taken as the sum of the chamber top (11.3 cm) and the average height of individual collars, as measured in the field. Soil and air temperatures were measured in the field at the time of each soil respiration measurement. Air temperature data was filled with half hourly air temperature data measured at a height of 2.5 m on the EMS tower.

  To estimate the slope of the CO2 vs. t curve, we first smoothed the curve with a Savitsky-Golay filter, and then took the mean and standard deviation of the 1st derivative where it flattened out at the back side of the curve. The portion of the 1st derivative that was used in the calculation was selected by hand for each individual curve. If the 1st derivative never flattened out, a soil flux was not calculated. Soil fluxes were also not calculated for curves where the relative standard deviation of the first derivative > 1. The mean and standard deviation of the 1st derivative of each curve, along with the number of points used to calculate them, are printed with the flux data, giving the user of the data the option of filtering the data with more or less rigourous standards.

  Trees

  All live trees greater than or equal to 10 cm DBH were measured in the tower plots and greater than or equal to 5 cm dbh in the harvested plots. Each tree's DBH (cm) was calculated from measurements of their dendrometer's expansion (or contraction) (mm), which were taken with digital calipers. Species specific allometric equations were used to calculate each tree's weight (kg) from their DBH. Each tree's carbon content was assumed to be one half of its weight. The trees' weight in C was summed to yield aboveground woody biomass (AGWB) and increment (AGWI), and biomass lost to mortality and gained through recruitment, all in MgC/ha/yr, for each year and site. Net primary production (ANPP) was calculated as the sum of the increment of surving trees (AGWI) and the sum of the increment of ingrowth + the sum of fine litterfall using the methods described in Clark et al. 2001. Note that measurements for years 1994-1997 at the tower site represent multi-year averages.

  The understory data includes saplings and shrubs greater than or equal to 1 cm DBH and less than 10 cm DBH at the tower plots and less than 5 cm DBH at the harvested plots. Saplings and shrubs were measured in the tower plots in November 2004 and 2006 and in the harvested plots in early June and November 2006. Diameters were measured with digital calipers and recorded in centimeters. For the 2004 survey, trees were measured in 1 cm increments (ie. greater than or equal to 1 to less than 2, greater than or equal to 2 to less than 3, etc.) and the number of trees in each plot, spp, and size group were tallied in the column labelled "nindivs". For the 2006 surveys all stem measurements were listed individually. Note that yellow and black birch trees are difficult to tell apart when they are small and should be considered interchangeable in the understory data.

  Woody Debris

  Coarse woody debris is all pieces greater than or equal to 7.5 cm in diameter. CWD surveys involved measuring all CWD pieces in our biometeric plots in the footprint of the EMS tower and at our cut site on the Simmes Trust land.

  Fine woody debris is all pieces greater than or equal to 2 cm in diameter and less than 7.5 cm in diameter. FWD was measured using the line transect method from Harmon and Sexton 1996, which gives the following formula for calculating volume per unit area (m^3*m^-2): V=9.869 * sum(d^2/8*L), where d is he piece diameter in m and L is the transect length in m. Line transects were randomly placed in different locations across the tract of land for each survey and every other 10 m segment on the transect was surveyed for FWD.

  Annual means and 95% confidence intervals (where plots and segments are the unit of measurement)of volume and biomass are summarized in the files labeled "summary". Biomass was calculated from volumes using decay-class specific densities from Liu et al 2006. For CWD pieces categorized as whole tree snags, biomass was calculated using spp-specific allometric equations for live trees, which are listed for our live tree data. Volumes of whole tree snags were calculated from biomass estimates using the inverse of Liu et al's densities.

  References:

  Harmon ME and Sexton J. 1996. Guidelines for measurements of woody detritus in forest ecosystems. Publication No. 20. U.S. LTER Network Office: University of Washington, Seattle, WA, USA. 73 pp.

  Liu WH, Bryant DM, Hutyra LR, Saleska SR, Hammond-Pyle E, Curran D, Wofsy SC. 2006. Woody debris contribution to the carbon budget of selectively logged and maturing mid-latitude forest. Oecologia 108(1): 108-117.

• Use:

  This dataset is released to the public and may be freely downloaded. Please keep the designated Contact person informed of any plans to use the dataset. Consultation or collaboration with the original investigators is strongly encouraged. Publications and data products that make use of the dataset must include proper acknowledgement. For more information on LTER Network data access and use policies, please see: http://www.lternet.edu/data/netpolicy.html.

• Citation:

  Munger W, Wofsy S. 1999. EMS - Biomass Inventories. Harvard Forest Data Archive: HF069.

Detailed Metadata

hf069-01: plot map

1. plot map: EMS plot map

hf069-02: compass distance map

1. compass distance map: EMS compass distance map

hf069-03: LAI plot-level

1. year: year (YYYY)
2. jday: Julian day (DDD)
3. plot: plot
4. LAI: leaf area index. An LAI2000 sensor was used to make 5 measurements at each plot: one at the center, and four more at points 2 meters away from the center, in the 4 cardinal directions (N,E,S,W). All measurements were made at sunrise or sunset when the sun was close to the horizon. Plot-based LAI values were calculated using the standard methods of the LiCor program C2000. LAI values were calculated by averaging the 5 subplot values for each plot, and by masking the outer ring of the sensor (ring 5) so as to block out possible sunspots which may have hit the outer ring as the sun rose up from the horizon. (dimensionless )
5. SE: standard error of leaf area index (dimensionless )

hf069-04: LAI site-level

1. year: year (YYYY)
2. jday: Julian day (DDD)
3. site: site
   • ems: EMS tower footprint site
   • cut: selectively harvested site
4. all.mean: mean for each plot and Julian day of measurement. The data reported in this file is of plant area index (PAI), and has not been corrected for woody material. (squareMeterPerMeterSquared )
5. all.stdev: standard deviation for each plot and Julian day of measurement. The data reported in this file is of plant area index (PAI), and has not been corrected for woody material. (squareMeterPerMeterSquared )
6. all.conf.int: 95% percent confident intervals (squareMeterPerMeterSquared )
7. decid.mean: mean deciduous LAI (squareMeterPerMeterSquared )
8. decid.stdev: deciduous LAI standard deviation (squareMeterPerMeterSquared )
9. decid.conf.int: deciduous LAI 95% confidence intervals (squareMeterPerMeterSquared )

hf069-05: litter chemistry

1. year: year (YYYY)
2. jday: Julian day (DDD)
3. plot: plot
4. spp: species. Leaf samples were collected by shooting them down from the canopy with buckshot. See the allometries file for a key of spp abbreviations.
5. N.percent: percent nitrogen. Samples were dried at 65 degrees C for 3 days, ground, and analyzed at the Harvard Forest on a Fison Nitrogen Analyzer NA 1500 in 2000 and on a Elementar Vario Micro CHNS analyzer in 2007 and 2008. (dimensionless )
6. C.percent: percent carbon. Samples were dried at 65 degrees C for 3 days, ground, and analyzed at the Harvard Forest on a Fison Nitrogen Analyzer NA 1500 in 2000 and on a Elementar Vario Micro CHNS analyzer in 2007 and 2008. (dimensionless )

hf069-06: canopy chemistry

1. sample.date: sample date (YYYY-MM-DD)
2. plot: plot
3. sample.number: sample number (number )
4. tree.tag: tree tag
5. spp: species. See the allometries file for a key of spp abbreviations.
6. N.percent: percent nitrogen. Samples were dried at 65 degrees C for 3 days, ground, and analyzed at the Harvard Forest on a Fison Nitrogen Analyzer NA 1500 in 2000 and on a Elementar Vario Micro CHNS analyzer in 2007 and 2008. (dimensionless )
7. C.percent: percent carbon. Samples were dried at 65 degrees C for 3 days, ground, and analyzed at the Harvard Forest on a Fison Nitrogen Analyzer NA 1500 in 2000 and on a Elementar Vario Micro CHNS analyzer in 2007 and 2008. (dimensionless )
8. vial.label: vial label

hf069-07: litterfall

1. year: year (YYYY)
2. doy: day of year (DDD)
3. site: site
   • ems: EMS tower footprint site
   • harv: selectively harvested site
4. plot: plot
5. spp: species. See the allometries file for a key of spp abbreviations.
6. wt-g: weight (gram )

hf069-08: litter decomposition

1. spp: species.See the allometries file for a key of spp abbreviations.
2. tag: tag number
3. plot: plot. Whereas the first four plots are located in the Simmes Trust land, which was cut in the winter of 2000/2001, the control plot, B3, is located on Prospect Hill in the EMS tower footprint, just across the road from the cut site. The upland & lowland descriptions refer to the rocky topography of the plot and whether the bags were in a low or high spot.
   • A5: heavy cut, edge of skidtrail
   • B5: heavy cut, edge of skidtrail, light/moderate cut
   • X2: light/moderate cut, light skidtrail
   • X1: heavy skidtrail
   • B3: upland, lowland
   • NA: NA
4. skid.1.yes.0.no: whether the sample was in a skid area
   • 1: skid trail
   • 0: no skid trail
   • NA: NA
5. cut.1.heavy.0.light: whether it was in an area that had been moderately or heavily cut
   • 1: heavily cut
   • 0: moderately cut
   • NA: NA
6. upland.1.lowland.0: upland or lowland, refers only to samples in the control plot
   • 1: upland
   • 0: lowland
   • NA: NA
7. end.doy.yr: end day of year (DDD)
8. start.weight.g: start weight (gram )
9. end.weight.g: end weight (gram )

hf069-09: soil moisture

1. date: date (YYYY-MM-DD)
2. plot.probe: plot and probe number for unique identification of each probe
3. Ka: dialectric constant calculated from each TDR waveform. To calculate a diaelectric constant (Ka) from each waveform, we used an automated script which first applied a Savitsky-Golay filter to the waveform data and then used a tangent-line method to locate the two inflection points. (number )
4. WC: water content calculated using Topp's equation (Topp et al. 1980) (cubicCentimetersPerCubicCentimeters )

hf069-10: soil respiration

1. date: date (YYYY-MM-DD)
2. jday: Julian day (DDD)
3. time: time (hhmm)
4. plot: plot
5. collar: collar number (1-90)
6. range.min: minimum CO2 concentration measured (ppm) (dimensionless )
7. range.max: final [maximum] CO2 concentration measured (ppm) (dimensionless )
8. x.cutoff: time (s) on the curve after which the mean of the 1st derivative was calculated (second )
9. mean.slope: mean of the 1st derivative (ppm/s) (dimensionless )
10. stdev.slope: standard deviation of the 1st derivative (ppm/s) (dimensionless )
11. N.pts: N points used to calculate the mean of the 1st derivative (number )
12. flag.stdev: flag for when the relative standard deviation > 1
13. Tair: air temperature (celsius )
14. Tsoil: soil temperature (celsius )
15. collar.height.m: average collar height (meter )
16. flux: soil CO2 flux (micromole CO2/m2/s) (micromolesPerMeterSquaredPerSecond )

hf069-11: tree data

1. year: year (YYYY)
2. jday: Julian day (DDD)
3. tree.type: tree type
   • live: live
   • daed: dead
   • und: understory
   • rcrt: recruited
4. site: site
   • harv: selectively harvested site
   • ems: EMS tower footprint site
5. plot: plot
6. nindivs: for understory, tally of the number of trees in each plot, spp, and size group. (number )
7. tag: tage number
8. spp: species. See the allometries file for a key of spp abbreviations.
9. dbh: diameter at breast height taken with digital calipers. (centimeter )

hf069-12: coarse woody debris

1. year: year (YYYY)
2. site: site
   • harv: selectively harvested site
   • ems: EMS tower footprint site
3. Plot: plot name where piece was measured
4. tag: tag number nailed to piece
5. log: piece type
   • 0: not a log
   • 1: log
6. log.snag: piece type
   • 0: not a log snag
   • 1: log snag
7. whole.tree.snag: piece type
   • 0: not a whole tree snag
   • 1: whole tree snag
8. stump: piece type
   • 0: not a stump
   • 1: stump
9. spp: the species of each piece was identified where possible. Some pieces were only recorded as hardwood or conifer. Spp identification was not possible for many pieces. See the allometries file for a key of spp abbreviations.
10. length.m: length of the piece in meters (meter )
11. baseD: diameter of the base of the piece in cm (centimeter )
12. midD: diameter of the center of the piece in cm (centimeter )
13. DBH: diameter at breast height in cm; measured only for whole tree snags (centimeter )
14. topD: diameter at the top of the piece in cm (centimeter )
15. decay: decay class (1-5) assigned to the piece
16. bark: estimated percent of the piece covered in bark (dimensionless )
17. ground: estimated percent of the piece in contact with the ground. For example a horizontal log would be 100% in contact with the ground, whereas a snag would be 0% in contact with the ground. (dimensionless )
18. phys.char: physical characteristics
    • 1: leaves present
    • 2: twigs present
    • 3: branches present
    • 4: bark on branches
    • 5: bark on bole
    • 6: sapwood sloughing
    • 7: sapwood friable
    • 8: heartwood friable
    • 9: hollow
19. biol.char: biological characteristics
    • 1: conks
    • 2: moss
    • 3: lichen
    • 4: termites
    • 5: other bug signs

hf069-13: fine woody debris

1. year: year (YYYY)
2. site: site
   • ems: EMS tower footprint site
   • harv: selectively harvested site
3. transect: transect number
4. segment: segment number
5. diameter.mm: the diameter of each piece was measured in mm where it intersected the transect (millimeter )
6. decay: a decay class 1-5 was assigned to each piece, using the same classification system as for CWD

hf069-14: understory summary

1. site: site
   • cut: selectively harvested site
   • tow: EMS tower footprint site
2. DBH: diameter at breast height
   • 1-5: 1 to 5 cm
   • 5-10: 5 to 10 cm
3. trees.per.ha: mean and 95% confidence intervals for trees per hectare for 2006 (number )
4. ba: mean and 95% confidence intervals for basal area (m^2/ha) for 2006 (metersSqauredPerHectare )
5. awb: mean and 95% confidence intervals for aboveground woody biomass (MgC/ha) for 2006 (megagramsPerHectare )

hf069-15: annual summary

1. year: year (YYYY)
2. site: site
   • harv: selectively harvested site
   • ems: EMS tower footprint site
3. AGWI: aboveground woody increment (megagramsPerHectarePerYear )
4. AGWB: aboveground woody biomass. (megagramsPerHectarePerYear )
5. recruit: biomass gained through recruitment (megagramsPerHectarePerYear )
6. mort: biomass lost to mortality (megagramsPerHectarePerYear )
7. ANPP: net primary production. calculated as the sum of the increment of surving trees (AGWI) + the sum of the increment of ingrowth + the sum of fine litterfall using the methods described in Clark et al. 2001. (megagramsPerHectarePerYear )

hf069-16: allometric equations and species names

1. allometries: allometric equations used to calculate aboveground woody biomass (BM) from tree DBH measurements at HFEMS biometric plots
2. species: species names