Simulated Hurricane Experiment - Vegetation Response

• Investigators: Audrey Barker-Plotkin, Sarah Cooper-Ellis, David Foster, Ann Lezberg
• Contact: Audrey Barker-Plotkin
• Start date: 1990-01-01
• End date: ongoing
• Location: Tom Swamp Tract (Harvard Forest)
• Latitude: +42.49
• Longitude: -72.20
• Elevation: 310 meters
• Taxa:
• Keywords: forest dynamics, hurricane, tree maps, wind damage
• Abstract:

  The following abstract is taken from a published paper (Cooper-Ellis, S., D. R. Foster, G. Carlton and A. Lezberg. 1999. Forest response to catastrophic wind: results from an experimental hurricane. Ecology 80: 2683-2696).

  "Infrequent, intense wind disturbance is an important factor in northeastern U.S. forests, yet little is known about the early stages of vegetation reorganization, or the processes that facilitate biotic regulation of ecosystem function after such storms. We designed an experiment, based on a simulated hurricane blowdown, to examine the relationship of tree damage patterns to mortality and regeneration, community dynamics, vegetation recovery and ecosystem processes. In October 1990, selected canopy trees in a 50 x 160 m area within a 75-yr-old Quercus rubra-Acer rubrum forest in central Massachusetts were pulled over by a winch, using records from the 1938 hurricane to determine the number of trees and direction of fall. The resulting damage to 65% of trees closely approximated effects of the 1938 storm on New England forests. Damage and mortality varied by tree species and size, indicating the importance of pre-disturbance forest structure and composition in determining the range and severity of impact. Measurements of vegetation and environment in the experimental area and control indicated that, although the manipulated stand sustained dramatic damage and structural reorganization, resilience of trees and understory vegetation provided tight biotic control of ecosystem processes, including nutrient cycling. Continued leaf-out and induced sprouting by damaged trees, increased growth by saplings and understory plants, and seedling establishment on disturbed microsites stabilized the microenvironment. Our findings are in contrast to studies of disturbances in which mortality was higher when damaged trees were removed from the site. This suggests that salvage logging following wind disturbance may have serious long-term implications."

• Methods:

  A. Methods for manipulation and tree damage and mortality assessment

  The Tom Swamp Hurricane Experiment was initiated in 1990 in order to study long-term vegetation and ecosystem response to wind disturbance. The 0.8 ha experimental and 0.6 ha control (untreated) areas were surveyed and marked at 10m by 10m intervals prior to the manipulation to form grids with long axes (experiment: 160m; control: 120m) running 85.5 degrees. Half of each area was surrounded by electrical fencing to exclude deer. Untreated red oak-red maple forest of similar age and structure occupied the 30 m buffer between and around the manipulation and control areas.

  In both areas, all trees of diameter greater than or equal to 5 cm dbh (experimental: 888 trees; control: 776 trees) were mapped prior to the manipulation using a mapping technique in which inter-tree distances were measured to the nearest centimeter (Boose, E.R., E.F. Boose and A.L. Lezberg. 1998. A practical method for mapping trees using distance measurements. Ecology 79:819-827). For each tree, species, canopy position, health, existing damage and diameter were recorded. Trees with multiple stems originating from a single base were measured as separate individuals.

  In the first week of October, during peak hurricane season, 235 dominant and codominant trees (76.5% of all live canopy trees) in the experimental area were individually pulled over toward the northwest to approximate the direction of hurricane winds. A choker wrapped around the main trunk at heights of 2.5m to 4.5m was fastened to a cable from a winch on a logging skidder, and pulling force was applied until the weight of the crown pulled the tree over. Trees closest to the operating site but just beyond the 30m buffer of intact forest were pulled first and trees farther away were pulled onto previously felled trees. The skidder operated from a clearcut 10-m strip to the north of the manipulation creating swaths of damage less than 15m wide and moving westward. Trees damaged by other trees were left where they fell and no effort was made to pull trees beyond their final point of repose if they were caught in other standing trees, fallen crowns, mounds, or other fallen boles.

  Tree Damage Assessment: During the winter of 1990 to 1991, all 888 trees in the experimental area were relocated and classified into one of eight primary damage classes. Details of damage were recorded, and pits and mounds measured and mapped. Damage was also assessed for the 776 trees in the control plot. New damage occurring in subsequent years of the study was occasionally recorded. Details of this can be found in the tree damage metadata file.

  Leafout and Sprouting Assessment: In the experimental and control areas all trees were assessed in mid-summer for degree of crown survival and for presence (1991), numbers (1992, 1993, 1994, 1995, 1997, 2000), and location (basal, trunk, or branch) of sprouts. Trees with no leafout of residual crowns were designated as "dead" and any sprouts from these trees were counted as new individuals.

  At each tree determine a LEAFOUT class: L4 = mostly leafed out or more than 90% branches leafed out; L3 = more than 50% but less than 90% leafed out; L2 = less than 50% but more than a single branch leafed out; L1 = only a few leaves or small branches leafed out, or 1 large branch; less than 10% leafout. F = false, no leafing out.

  At each tree determine if sprouts are present for each SPROUT class: BASP = new growth (dating to damage) emerging from immediately below ground, from base or root collar; TRSP = new growth (epicormic shoots) from major stem; BRSP = new growth from branch (not just buds).

  For basal sprouts, count individual sprouts for individuals less than 15. For trees with sprouts greater or equal to 15, give a stem class with 15-24=15+, 25-34=25+, etc. Branch sprouts and trunk sprouts are generally only tallied for presence/absence because they are too difficult to count on standing trees. Count branch sprouts below the major crown only on standing undamaged trees. If there is unusually prolific sprouting please add to comments an estimate of abundance. In 1991, basal sprouts were not counted and only presence/absence is indicated. In 1992, trunk sprouts were counted in the same way as basal sprouts. Count leafed out sprouts originating after the manipulations only.

  B. Sapling and sprout regeneration survey

  In the hurricane manipulation experiment, twenty-four 2-m x 5-m plots were randomly established on each of three transects running parallel to the long axis in the experimental area (~west to east) and on one transect running parallel to the long axis in the control, for a total of 96 plots.

  Individual saplings and sprouts were measured and tagged in these plots in 1990 (before the manipulation), 1991, 1993, 1996 and 1999. All stems greater than 0.3m total height were included in this survey. Previously tagged saplings and sprouts may exceed 5cm in diameter and should still be measured. Tags should be moved if they are girdling the tree and this damage to the tree should be noted. The 1999 sampling was the final measurement of regeneration; future work will focus on regeneration that has grown above a 5 cm dbh threshold (see methods in part D, below).

  For each stem, several variables are measured, and are described below:

  LOCATION. C = center of plot; E = eastern section of plot; EC = east central section of plot; N = northern section of plot; NC = north central section of plot; NE = northeastern section of plot; NW = northwestern section of plot; S = southern section of plot; SC = south central section of plot; SE = southeastern section of plot; SW = southwestern section of plot; W = western section of plot; WC = west central section of plot. Saplings/sprouts on the edge of the plot are labeled "*" in the location column.

  SPECIES. The first two letters of the genus and species names of the tree are used. Black birch and yellow birch were lumped together until old enough to distinguish with certainty.

  AGE. If the age of a sapling or sprout cannot be determined (because of age, browsing, etc.), this will be noted in the comments and the age column will be left blank. Stems that are residual (present before the pulldown) are given an age of "R". Others are recorded as N91 (new in 91), N92 (new in 92), etc. Often the age determination was somewhat uncertain; in these cases the age is enclosed in brackets (e.g., [N91]).

  STEM TYPE. SAP = sapling; BASP = basal sprout; TRSP = trunk sprout; RTSP = root sprout; BRRT = rooted branch.

  HEIGHT. Height is the vertical distance from the ground to the highest leaf-tip. If a stem is taller than 3 m, height is measured to the nearest 0.5m (0.1m if possible), using a telescoping height pole. If less than 3m tall, measure height to nearest 0.1m, using a folding meter stick. If a stem originates from a mound, or in a pit, the height should be adjusted to be the distance to the GROUND (this may have to be estimated somewhat). In general, the height should measure the position in the canopy of the crown of the tree.

  STEM LENGTH. Stem length is length of the stem up to, but not including, the new growth. It is often different from the height, as stems are sometimes bent or originate above or below ground level. Stem length is recorded only for smaller stems.

  Diameters are taken with calipers at the point indicated with paint on the "old" stems when it can be relocated. If the old paint cannot be relocated, new spots are painted and it is noted in the comments section. To relocate an old bole diameter, we looked at 1993 data sheets and measured at one half of the 1993 stem length. Measurements are taken with the paint spot aligned between the forks of the calipers, showing on the open side where possible. Old spots were repainted. "New" stems were marked where measured.

  BASAL DIAMETER. Measure above basal swelling or 1cm from the ground, whichever is greater. (should be marked on tree). Recorded in cm. Some of these are becoming quite hard to reach as the trees grow. Diameter tape is sometimes easier to use. Any unusual basal swelling or scars are also noted in the comments. Some basal diameter markings are really buried and should be relocated and marked above ground. If a new basal diameter is marked, this is noted in the comments.

  BOLE DIAMETER. If a tree is less than 2m, measure half way up the stem length. If tree is more than 2 m, measure at 1.33m (dbh). Measure to nearest 0.1mm Recorded in cm. If we could not find paint from the previous years, we relocated the bole by taking half of the 1993 stem length (or height if no SL is noted). In comments, "new BOLE @____" was noted.

  DBH. Measure only if stem is longer than 2 m. Recorded in cm. If the last measurements for a stem were basal and bole diameters, but the stem is now long enough to get a DBH, we will take all three measurements. The height of DBH is measured from the base of the stem, not from the ground, even on trunk sprouts. This makes it too high to reach on many sprouts with an elevation.

  In deciding whether to give a stem a bole diameter or a dbh, refer to the previous year's data set. If the stem had a bole diameter and no dbh previously, but is now large enough for a dbh, both bole diameter and dbh should be recorded. If a bole diameter and a dbh were recorded previously, only a dbh is needed. If the basal diameter is inaccessible, both bole diameter and dbh should be recorded if possible.

  STEM OF ORIGIN. For sprouts and rooted branches only. Examples: SAP? for unidentified sapling , TR0820, or SAP0240 (Include PULLDAM in comments: type of damage to stem of origin: BENT, UPRT, CRSN, RTBR, TPSN, LEAN, STND) . We relied on the previous years designations of PULLDAM to stems of origin unless the damage clearly occurred since the pulldown.

  HEIGHT OF EMERGENCE. Location along originating stem from which sprouts emerge. RC= root collar, 0013 = 13 cm along bole, 2412 = 24 meters or 2412 cm along bole. BG = below ground. Measure from root crown (where bole connects to roots). For previously tagged trunk sprouts that are a long way from the base of the tree, we will rely on previous years' data.

  ELEVATION. Height from ground to the base of sprout or sapling (which is not necessarily the lowest point on the tree). For basal sprouts, this is 0, and for most saplings, this is also 0, but if a sapling is on mound or in a pit, it may have elevation. Measure in m, to the nearest 0.01m (cm). If the stem is growing in a pit, the elevation may be negative. For some TRSPs, elevation and height of emergence are equal (if the stem of origin is standing). For others, they may be different (if the stem of origin is bent, leaning, etc). Both elevation and height of emergence should be recorded where appropriate, even if they are equal.

  #SPROUTS. Number of sprouts in a clump (not always the number of sprouts on the tree). Only for sprouts. Clumps emerge from the same point on the root collar.

  SUBSTRATE. See the diagram for details. GR = ground; ST = stump; MT = mound top; RP = root plate; MO = mound; MB = mound base or mound edge; RO = rock; PT = pit; TR = tree; PE = pit edge.

  COMMENTS. Include damage, if there are multiple sprouts from same tree, browsing, and whether the stem is dead, down, broken, bent, leaning, buried, leader dead, new leader, or gone. In the data file, columns D1991, D1993, and D1996 record whether a stem is dead, living or gone in each of these years of measurement. If damage was detailed in previous years, it will not be repeated, likewise on shared stems (unless there are additional saplings/sprouts). New basal markings will be noted, as will new bole markings (and location of the new bole).

  Leaning stems were classified as follows in 1996 and 1999: Class 1 Lean = 0 - 44 degrees from vertical. Class 2 Lean = 45-90 degrees from vertical (including those "around 45 degrees" and those "around 90"). Class 3 Lean more than 90 degrees from vertical.

  C. Understory vegetation methods

  Twenty-four 2-m x 5-m plots were randomly established on each of three transects running parallel to the long axis in the experimental area (~west to east) and on one transect running parallel to the long axis in the control, for a total of 96 plots.

  Composition and structure of understory vegetation were assessed before (1990) and after (1991, 1992, 1995, 2000) the experiment in the 96 plots described above (shrubs) and in 1-m2 plots nested within the 2-m x 5-m plots (herbs). Cover of shrub, herb, fern, graminoid and lycopod species was estimated and total cover of life-form categories (saplings, seedlings, shrubs more than 1 m, shrubs less than 1 m, ferns, herbs/graminoids, clubmoss, and moss) was tallied in the control and experimental areas in all four sampling years. Height of the tallest individual shrub for each species on each plot was also measured and cover of substrates (leaf litter, woody debris, rock, exposed level soil, pit, and mound) was estimated in broad cover classes (+ = less than 1%; 1 = less than 5%; 2 = 5-25%; 3 = 25-50%; 4 = 50-75%; 5 = more than 75%).

  Herbs: At each 1m2 subplot plot a circular plastic Delvin frame divided into 8 pie shaped divisions was centered around the center flag to facilitate judgments of cover percentages. The plot frame lies parallel to the ground (vertical on uproots). Cover is defined as the projection (on the ground within the sampling frame) of all above-ground portions of each species, whether that species individual is rooted within the plot or not. Cover estimates were made to the nearest 0.01 percent (for cover less than 0.15 or the nearest 0.1% for individual species; cover estimated only to the nearest 0.1% in 2000). Plastic cutouts of known size were used to estimate cover. Fruits and large flowers were not included in cover estimates. Total subgroup cover was estimated for the following at each plot: 1) Seedlings less than 0.3 m tall; 2) Shrubs less than 1 m tall; 3) Herbs/graminoids; 4) Ferns; 5) Moss; 6) Clubmoss. Shrubs more than 1 m tall, saplings/sprouts more than 0.3 m but less than 5 cm dbh, and overhanging materials were given a cover class between 1 and 4 (1 = less than 16%; 2 = 17-49%; 3 = 50-83%; 4 = 84-100%).

  Substrates were also given a cover class in the following categories: 1) Leaf litter; 2) Disturbed soil on mound; 3) Bare (exposed) soil on flat ground; 4) Bare soil in pit; 5) Vertical intact ground on mound top; 6) Vertical exposed ground-organic soil; 7) Exposed rocks/stones; 8) Woody debris -- large boles, stumps, other fragments; 9) Tree base or exposed roots of live and dead trees. Substrates should add up to 100%.

  Shrubs: Cover of shrub species was estimated in the same way as herb species cover inside 10m2 rectangular plots (2m wide by 5m long). Maximum heights for each species present on each plot were taken in meters to the nearest 0.1m. Shrubs taller than 3m were recorded as greater than 3m. Debris and substrate were estimated in cover classes as follows: 5 = more than 75%; 4 = 50-75%; 3 = 25-50%, 2 = 5-25%; 1 = 1-5%; + = less than 1% for the following: prostrate branches and crowns up to 3 m high, boles, mound/pit complexes, large rocks more than 0.5m diameter, and tree stumps. For analysis purposes, these cover classes are translated to their cover range midpoint as follows: + = 0.5%; 1 = 3%; 2 = 15%; 3 = 38%; 4 = 63%; 5 = 88%.

  D. Recruitment survey

  The objective of this survey is to document recruitment of trees into the experimental and control sites at the hurricane manipulation site. Regeneration via seedlings, advance regeneration and sprouts has been studied in small (2x5m) plots along transects in these sites, but these plots are likely too small to capture the dynamics of stems growing into the more than 5cm diameter class. Therefore, a complete survey will be conducted. At this point, individual tracking and mapping of stems will not be done. This survey will allow us to track gross changes in species composition, density and basal area over time. Individuals in the new cohort may be tagged and mapped in the future. The first survey was done in summer 2000 and will be revisited on 3 year interval.

  Each 10m x 10m grid cell in the pulldown and control sites will be examined for live stems that have grown into the greater than 5cm dbh class since the original survey in 1989 or 1990. In the pulldown site, it may help to delineate the cell boundaries with a 50m tape. As new stems are encountered, record for each: grid cell number, species, diameter at breast height (1.37m), stem type (sapling or sprout), stem of origin if a sprout (e.g., sapling less than 5cm dbh, Tree #356), notes (such as forked stems, dead leader, basal sprouts, damage). Mark completed stems with railroad chalk or a logger's crayon so that we'll know it has been done. In 2000, we did not note whether the stem was growing on a mound or pit, but we will record this in future surveys.

  The following species are considered trees: Acer rubrum, A. saccharum, Betula allegheniensis, B. lenta, B. papyrifera, Castanea dentata, Carya glabra, C. ovata, Fagus grandifolia, Fraxinus Americana, Ostrya virginiana, Pinus strobus, Populus tremuloides, Prunus pensylvanica, P. serotina, Quercus rubra, Sorbus americana, and Tsuga canadensis. Species including Corylus sp., Amelanchier sp., Viburnum sp. are shrubs.

  Methodological details: (1) Live stems only. (2) Trunk sprouts are included if: a) on a prostrate bole, or b) are less than 1.37m up the stem of a standing/leaning/bent tree. (3) Bent or leaning stems counted if stem is more than 5cm at 1.37m from ground, not 1.37m along the stem. (4) Height to measure diameter is 1.37 from ground (for example, a tree growing on a mound or stump would be measured at 1.37m from level ground, not 1.37m up the stem). (5) Diameter measurements are to the nearest 0.1cm. (6) Some stems may have originated as sprouts, but if stem of origin cannot be seen, the stem will be noted as a sapling (e.g., most Castanea dentata look like saplings, but obviously are old basal sprouts.

• Related datasets: HF050 HF052