Ecological Archives E081-003-D1
David B. Clark, and Deborah A. Clark. 2000. Tree growth, mortality, physical condition, and microsite in old-growth lowland tropical rain forest. Ecology 81: 294.
Tree species diversity reaches its maximum in tropical rain forests, where up to several hundred species may co-occur in a single hectare. There is not yet an accepted explanation for the avoidance of competitive exclusion under such conditions. In part this is due to a lack of understanding of the autecology of individual species and of the ontogenetic changes in autecology within species. This is turn is due to a lack of long-term records of environmental conditions associated with individual tree performance data. We report here data on annual height and diameter growth and associated microsite conditions for nine species of canopy and emergent trees in old-growth tropical wet forest in Costa Rica, Central America, for the period 1983-1993. The sample includes all post-establishment size classes, from plants > 50 cm tall to the largest adult sizes. These data provide the most detailed insight currently available into how environmental conditions interact with past performance and ontogenetic potential in tropical rain forest trees. The focus on an annual time step allows resolution of important aspects of regeneration that are obscured or not measurable with longer inter-census intervals. These data have been used to study the nature and diversity of tropical tree life history patterns, relationships among microsite, growth, and survival, the effects of physical damage on tropical tree regeneration, ecophysiology of tropical tree saplings, the ecological role of very large trees in old-growth forest, and interannual variations in tree growth.
Beginning in 1983 individuals of all size-classes >50 cm tall of
six species (II.B.4.c) were mapped (II.B.3.a) in an area of ca. 250 ha of old-growth
forest at the La Selva Biological Station in Costa Rica, Central America (II.B.2.a-g).
The individuals were found by walking transect lines from trails to natural
barriers like streams. All individuals sighted while walking the transects were
included, regardless of form or microsite. Although larger individuals were
more likely to be sighted, this did not affect the purpose of the sampling,
which was to accumulate a sample, unbiased by individual form or microhabitat,
of all size classes of all species. All subsequent analyses were based on individual
and size-class data at a landscape scale without regard to density or relative
abundances among size classes within a species (absolute density data were not
the focus of this study). In 1988 three additional light-loving species were
added (II.B.4.c) using the same sampling techniques. Each year survival (V28),
diameter (V28), canopy position (V93), forest phase (V118), stem condition (V83)
and height for individuals <16 m (V7) were measured. Pre-census remeasurement
comparisons (III.A.4) as well as single-blind measurements and independent field
checks of the previous year's measurements (V.A.1) were used as initial quality
controls. Data were entered from field census sheets into computer files and
checked by read-back, then screened for out of bound values.
A. DATA SET IDENTITY: Annual measurement of tree growth, mortality, physical condition, and microsite in old-growth lowland tropical rain forest (La Selva, Costa Rica), 1983-1993
B. DATA SET IDENTIFICATION CODE: 99-9002.
C. DATA SET DESCRIPTORS:
1. Originators
David B. Clark (dbclark@sloth.ots.ac.cr)
Deborah A. Clark (daclark@sloth.ots.ac.cr)
Institutional Address: Department of Biology, U.Missouri-St. Louis, St. Louis, MO
Residence Address: La Selva Biological Station, Puerto Viejo, Costa Rica
Address for Mail: O.T.S., Interlink 341, P.O. Box 02-5635, Miami, FL 33102
2. Abstract: The goal of this long-term research has been to assess the relationship between annually-measured tree performance and microsite conditions for nine canopy and emergent tree species in old-growth lowland tropical rain forest. The study site, the La Selva Biological Station in N.E. Costa Rica, is tropical wet forest (annual mean precipitation 4 m; annual mean temperature 26o C). We used unbiased selection criteria to assemble a sample of >2000 individuals of these species in all size classes except small seedlings. We annually assessed diameter and height growth as well as stem condition and indices of crown lighting and forest structure. These data cover the period 1983 - 1993, and the study is ongoing as of 1999. To our knowledge the data represent the longest-running and most highly quality-controlled measurements that combine annual tree growth and mortality with associated microsite conditions in tropical forests. The data provide the most detailed insight currently available into how environmental conditions interact with past performance and ontogenetic potential in tropical rain forest trees. The focus on an annual time step allows resolution of important aspects of regeneration that are obscured or not measurable with longer inter-census intervals. The data have been used to study the nature and diversity of tropical tree life history patterns, relationships among microsite, growth, and survival, the effects of physical damage on regeneration, ecophysiology of saplings, the ecological role of very large trees in old-growth forest, and interannual variations in tree growth.
D. KEY WORDS: Costa Rica; 1983-1993; tropical rain forest tree demography and ecology; suppression; release; life history diversity; interannual growth variation; physical damage; crown position; maximum growth rates; pioneers; shade tolerance, La Selva Biological Station, Cecropia, Simarouba, Lecythis, Minquartia, Pithecellobium, Balizia, Hyeronima, Hymenolobium, Dipteryx.
A. OVERALL PROJECT DESCRIPTION: Not applicable, this is a stand-alone project
B. SPECIFIC SUBPROJECT DESCRIPTION
1. Objectives of research. We sought to characterize the variation in life history patterns for canopy and emergent tree species in a tropical rain forest. To do this we assembled a set of nine focal tree species spanning a broad life history spectrum, and we studied the interactions among microsite characteristics and tree performance in all post-seedling life history stages of these species. Complementary goals that have developed during the study include: assessing the importance of physical damage to individuals through ontogeny, quantifying the effects of long-term suppression on different life history stages, and assessing the degree of inter-annual variation in tree performance and its relationship to climatic and atmospheric factors.
2. Site description (see McDade et al. 1994 for detailed site description)
a. Site type: Lowland tropical rain forest (Tropical Wet Forest sensu Holdridge)
b. Geography: Atlantic lowlands of Costa Rica, ca. 10o 26' N, 83o 59' W).
c. Habitat: Old-growth forest with no known history of major human interventions such as logging or post-Colombian agriculture (>90% of our study area; a small percentage is in secondary forest and disturbed habitats and is so identified in the data). Pre-Colombian settlements were common along the rivers in this region; the effects of these people on the composition of current forests in this region, if any, are unknown.
d. Geology: Part of the study area is flat old alluvial terraces (inceptisols). Most of the rest is rolling hills of ultisols with steep slopes (30-70 m long) and flatter ridgetops. Also included are several small permanent creeks with associated colluvial soils (Clark et al. 1998).
e. Watersheds, hydrology. The entire study area is within the La Selva Biological Station. All the watersheds are contained within the Station or in the adjoining Braulio Carrillo National Park.
f. Site history (McDade and Hartshorn 1994): The original 587 ha of largely old-growth forest was purchased by Leslie Holdridge to be an experimental farm in the early 1950's. The Organization for Tropical Studies purchased the property from him in 1968 and has operated it as a biological station since then, incrementally adding adjacent lands to the core reserve. La Selvas current total area is ca. 1536 ha.
g. Climate (Sanford et al. 1994). Annual precipitation averages 3962 mm and mean annual temperature is 26o C. A drier season occurs in February-May, but for all months the long-term average rainfall is >100 mm.
3. Experimental or sampling design
a. Design characteristics. We selected the original sample of study individuals in 1982-83 by walking compass bearings from trails to natural boundaries like creeks, attempting to completely cover specific areas of old growth. Without exception, we included in our sample all live individuals > 50 cm tall of our original six study species that were encountered, regardless of condition or microsite. In addition, for Lecythis ampla, Hyeronima alchorneiodes, and Hymenolobium mesoamericanum, we included all encountered individuals < 50 cm tall (young seedlings in the main). In each subsequent year, prior to each annual census, we evaluated the dataset and decided for which species-size classes increased sample sizes were needed (a minimum of > 20 individuals/species-size class currently alive was sought). For those selected species-size classes, we then added to the study sample all new individuals encountered during the course of the annual census (following the same all-inclusive criteria -- no individual of the target species/size-classes that was sighted within our study area was ever rejected). From time to time we have extended our study area (e.g., by crossing a creek). In newly included areas of forest we have followed the same procedure for adding new species/size combinations (i.e. nothing on the current "add" list is ever rejected, unless it is across some new natural boundary where we choose not to extend the study area). All individuals in our study sample have been marked and mapped with respect to trail locations and to each other (copies of these maps are stored at La Selva and off-site, under the Clarks' control). Each individual is censused annually for growth, survival, physical condition, and microsite, until they are determined in at least two consecutive censuses to be dead. We have maintained the same spatiotemporal order of working through the study area, to maintain very close to 1 yr intervals between censuses of each tree (for 1993/4, inter-census intervals were 365 + 16 d for 94% of the 2011 trees; Clark and Clark 1999).
In 1982 we selected six non-pioneer species of canopy and emergent trees (see section 3.c. below) as focal species for this study. We chose them to represent a broad gradient of shade tolerance, based on our knowledge or hypotheses at that time. All six species are also commercial timber species. In 1988 we added three species that were reported to be pioneers or high-light-demanding at other sites (two Cecropia species and Simarouba amara).
b. Permanent plots: Not applicable.
c. Data collection frequency. All individuals are censused once annually (each annual census spans the period January May/June).
4. Research methods.
a. Field/laboratory: See 2.a. and descriptions (below) of measured variables.
b. Instrumentation: Throughout the study we have used fiberglass diameter tapes marked in mm for the diameter measurements (resolution + 1 mm) of trees > 4 cm in diameter, and plastic calipers for the diameter measurements of trees < 4 cm in diameter. We used a folding ruler and a 15 m extendible measuring pole (Hastings; 1 cm resolution) for height measurements. We combined the extendible pole with the folding ruler for measurements of trees up to about 16 m tall.
c. Taxonomy and systematics: The list of study species is given below. Referenced voucher specimens are in the Herbario Nacional de Costa Rica, San José, Costa Rica (except for the voucher for Hyeronima, which is at the Instituto Nacional de Biodiversidad de Costa Rica, Santo Domingo de Heredia, Costa Rica).
Species studied since 1983 (family) [representative voucher specimen]:
Minquartia guianensis Aubl. (Olacaceae) [G.Herrera 2250]
Lecythis ampla Miers (Lecythidaceae) [R.Robles 2208]
Hymenolobium mesoamericanum H.C. Lima (Papilionoideae) [R.Aguilar 19]
Dipteryx panamensis (Pittier) Record & Mell (Papilionoideae) [R.Robles 1199]
Balizia elegans elegans (Ducke) Barneby & J. W. Grimes (Mimosoideae)
[B.Hammel 17319] (before 1999 = Pithecellobium elegans)
Hyeronima alchorneoides Allemao (Euphorbiaceae) [Chacon 751]
Species studied since 1988 (family) [representative voucher specimen]:
Cecropia insignis Liebm. (Cecropiaceae) [W. Burger 11135]
Cecropia obtusifolia Bertol. (Cecropiaceae) [R. Robles 1446]
Simarouba amara Aubl. (Simaroubaceae) [R. Robles 1670]
d. Permit history: Permits for study at La Selva from the Organization for Tropical Studies and from the Costa Rican government have been maintained current over the life of the project.
e. Legal requirements: see 3.d. above.
5. Project personnel: The project P.I.'s since its inception are David B. Clark and Deborah A. Clark. Project technicians, who have assisted in mapping and have done most of the field measurements since the late 1980s, are (with years of participation in the project) Luis Fernando Corrales (1982-1983), Gerardo Vega (1984-ca. 1990), Leonel Campos (ca. 1989-1999), and William Miranda Brenes (ca. 1990 - 1999). Only the principal investigators and these technicians have taken field data.
6. Funding sources: U.S. National Science Foundation (NSF), Organization for Tropical Studies (OTS), Andrew W. Mellon Foundation, and personal funds.
A. STATUS
1. Latest update: The data span the period 1983-1993.
2. Latest date: 1993, see A.1 above.
3. Metadata status: The metadata are complete and up-to-date for this time period.
4. Data verification: All variables have been screened for out-of-bounds values (values not possible for a variable, either by definition or greatly outside the range of values in this dataset) and have been checked by read-back from the original field data sheets by the personnel who made the field measurements. In addition, the P.I.s have annually performed a series of cross-checks after adding the most recent years data to the master database.
Before each census we carried out a remeasurement error study. A set of 50-70 trees was measured on one day, then re-measured after an interval of two to several days (data from the first measurement were not taken into the field). In 1993, for example, the average absolute difference in diameter measurements was 0.2 mm (N=54 trees) and the average absolute difference in height measurements was 11 cm (N=34); 59 of 67 crown position estimates were identical. Our standard for remeasurement acceptability was that at least 95% of the diameter remeasurements be within 1 mm, which is the precision of the diameter tape. If this standard was not reached in the first remeasurement, additional remeasurements were done after a several day interval until the standard accuracy was reached. In nearly all years, the first remeasurements met the standard.
B. ACCESSIBILITY.
1. Storage location and medium. (Ecological Society of America data archives, URL published in each issue of its journals). Original field data sheets are stored at the La Selva Biological Station, Costa Rica, with photocopies and digital file versions stored in a separate building or in San José.
2. Contact person: David B. Clark, dbclark@sloth.ots.ac.cr, tel. 506-766-6565x147, fax 506-766-6535, mail address: O.T.S., Interlink 341, P.O. Box 02-5635, Miami, FL, 33102, USA.
3. Copyright restrictions: None.
4. Proprietary restrictions: None.
5. Costs: None.
A. Data set file
1. Identity: LSTREES1.TXT (metadata are in the file METADAT1.HTML).
2. Size: 2616 records, record length 859 characters, 2.25 megabytes uncompressed.
3. Format type: ASCII, compressed with WinZip to the file LSTREES1.ZIP.
4. Header information: None (all metadata are in the file METADAT1.HTML).
5. Alphanumeric attributes: Upper case.
6. Special characters. The code for errors and missing data is 999.
7. Authentication procedures: Check-averages for five of the quantitative variables are given in section V.B. below.
B. Variable information:
The variables are listed in the order they appear in the file (V1 -V128). For each variable, the listed descriptors are: variable number, start column and length of field (in parentheses), variable name and definition, unit of measurement, data type, definition of variable codes, maximum and minimum values (excluding 999, the code for errors and missing data), precision, and data anomalies. All data are fixed format.
V1 (1,6). ID. Tree identification number. Integer. These numbers are coded to species by the first digit (see V3. SPECIES). The exceptions are a few Dipteryx that are in OTS Permanent Plot I (numbers <1000), and the two Cecropia species (C. insignis ids are in the range 4001-4499; C. obtusifolia ids are in the range 4501-4999). Minimum value: 2, Maximum value: 99369.
V2 (7,9). TRAIL. Location of a nearby point on the La Selva trail system. The letters correspond to the letters used on the La Selva trail sign posts; the numbers are distance in meters from the trail head. String.
V3 (16,3). SPECIES. Species identification codes (for more detail see B.3.c. under Metadata Class III). Integer. 1 - Cecropia obtusifolia; 2 - Simarouba amara; 3 - Balizia (formerly Pithecellobium) elegans; 4 - Cecropia insignis; 5 - Minquartia guianensis; 6 - Lecythis ampla (formerly costaricensis); 7 - Hyeronima alchorneoides (formerly oblonga); 8 - Hymenolobium mesoamericanum (formerly pulcherimum); 9 - Dipteryx panamensis.
V4 (19,5). FIRSTCENSUS. Year of the first census of each individual. Integer. Minimum value: 83, Maximum value: 93.
V5 (24,5). DEATHYEAR. Year the plant died. Integer. Value = -999 if still alive in the 1993 census. Individuals were checked for up to several years to make sure that they were in fact dead. "Dead" was defined as bark exfoliating, no leaves, no visible living tissue. Minimum value: 84, Maximum value: 93.
V6 (29,5). FIRSTHEIGHT. Height in cm at first census (see description of height measurements in V7). Trees above 1700 cm were rarely measured. Minimum value: 10, Maximum value: 1902.
V7 (34,5). HEIGHT93. Height in cm in 1993. Integer. Height was measured to the nearest cm using a 2-3 m collapsible or roll-up carpenter's ruler, or a 15-m extendible measuring pole. The tube could be lifted an additional 2-3 meters or so off the ground for occasional higher measurements. For individuals < 1.5 m tall, height was the vertical distance between the ground (at the level of the stem base) and the base of the highest meristem (for all species except Lecythis ampla, for which the height was to the highest live leaf). For individuals > 1.5 m tall, height was the vertical distance between the ground (at the level of the stem base) and the base of the highest live leaf. Minimum value: 18, Maximum value: 1803. + 1 cm.
V8 (39,5). HEIGHT92. Height in cm in 1992 (see V7). Integer. 12 - 1767 cm, +/- 1 cm.
V9 (44,5). HEIGHT91. Height in cm in 1991 (see V7). Integer. 10 - 1760 cm, +/- 1 cm.
V10 (49,5). HEIGHT90. Height in cm in 1990 (see V7). Integer. 5 - 1891 cm, +/- 1 cm.
V11 (54,5). HEIGHT89. Height in cm in 1989 (see V7). Integer. 3 - 1724 cm, +/- 1 cm.
V12 (59,5). HEIGHT88. Height in cm in 1988 (see V7). Integer. 1 - 1902 cm +/- 1 cm.
V13 (64,5). HEIGHT87. Height in cm in 1987 (see V7). Integer. 11 - 1920 cm +/- 1 cm.
V14 (69,5). HEIGHT86. Height in cm in 1986 (see V7). Integer. 5 -1850 cm, +/- 1 cm.
V15 (74,5). HEIGHT85. Height in cm in 1985 (see V7). Integer. 3 - 1704 cm, +/- 1 cm.
V16 (79,5). HEIGHT84. Height in cm in 1984 (see V7). Integer. 11 - 1611 cm, +/- 1 cm.
V17 (84,5). HEIGHT83. Height in cm in 1983 (see V7). Integer. 10 - 1274 cm, +/- 1 cm.
V18 (89,7). HGRO8384. Height change (in cm) between 1983 and 1984. Real 7.1. When HEIGHT (current year) > HEIGHT (previous year), this variable is the quantity [((HEIGHT (current year)) - (HEIGHT (previous year))/((DATE of current census) - (DATE of past census))] x 365. This gives an annualized height growth rate over the current census interval. When height has decreased, however, HGRO-- is calculated simply as (HEIGHT (current year) - HEIGHT (past year)). Most of the large negative decreases are due to physical damage, the evidence of which can usually be seen in the stem condition variable (V83-V92) for that year. Minimum value: -164, Maximum value: +167.0 cm. + 0.1 cm.
V19 (96,7). HGRO8485. Height change (in cm) between 1984 and 1985 (see V18). Real 7.1. -537.0 to +222.1 cm. + 0.1 cm.
V20 (103,7). HGRO8586. Height change (in cm) between 1985 and 1986 (see V18). Real 7.1. -628.0 to +274.8 cm. + 0.1 cm.
V21 (110,7). HGRO8687. Height change (in cm) between 1986 and 1987 (see V18). Real 7.1. -755.0 to +371.1 cm. + 0.1 cm.
V22 (117,7). HGRO8788. Height change (in cm) between 1987 and 1988 (see V18). Real 7.1. -543.0 to +216.9 cm. + 0.1 cm.
V23 (124,7). HGRO8889. Height change (in cm) between 1988 and 1989 (see V18). Real 7.1. -734.0 to +401.7 cm. + 0.1 cm.
V24 (131,7). HGRO8990. Height change (in cm) between 1989 and 1990 (see V18). Real 7.1. -689.0 to +341.9 cm. + 0.1 cm.
V25 (138,7). HGRO9091. Height change (in cm) between 1990 and 1991 (see V18). Real 7.1. -1079.0 to +410.8 cm. + 0.1 cm.
V26 (145,7). HGRO9192. Height change (in cm) between 1991 and 1992 (see V18). Real 7.1. -645.0 to +356.0 cm. + 0.1 cm.
V27 (152,7). HGRO9293. Height change (in cm) between 1992 and 1993 (see V18). Real 7.1. -346.0 to +307.8 cm. + 0.1 cm.
V28 (159,7). DIA93. Diameter in mm in 1993. Real 7.1. See V50 for description of how diameter was measured (with calipers or with a diameter tape) and V61 for the height of the point of measurement (POM). Nearly all trees < 700 mm in diameter were remeasured annually (exceptions were trees with very large, appressed lianas or with extremely malformed boles). Trees >700 mm in diameter were remeasured annually only if a remeasurement site of good form (above basal swellings or buttresses) existed below ca. 7 m. The 7 m limit was the maximum height reachable with the two 3-m sections of Swedish climbing ladders carried during the field work.
Protocols for diameter measurements depend on tree size. For individuals < 4 cm in diameter, maximum stem diameter is measured with a calipers to the nearest 0.1 mm at a marked point of measurement (POM) below the lowest branch and away from stem irregularities (the POM is indicated by a thin stripe of enamel paint). We measured these small trees at the highest acceptable point among the following: at or close to the stem base, at 40 cm above the ground, or at 130 cm above the ground. Over time, the POM is moved upward. Double measurements are taken in years when the POM changes, to enable increment calculations from both the past and succeeding growth years. In years where the POM changes, the diameter variable in this database is from the new POM, while the diameter growth variable was calculated to the old POM (see variable 40, below).
Boles > 4 cm in diameter are measured with a fiberglass fabric diameter tape to the nearest mm, rounding down. The tape is passed under any vines or roots on the stem and then is moved back and forth to clean the POM of loose bark and debris. Boles are measured at 130 cm above the ground ("breast height") when this point is not affected by buttresses or major stem deformities; otherwise, the POM is above such irregularities, often 5-7 m above the ground. Each trees POM is marked by an aluminum nail 10 cm higher on the bole (to avoid any influence on the POM form a wound reaction to the nail). In years when buttresses or other stem irregularities are found to have developed near the POM, it is moved up, and diameter is measured at both points (annual diameter growth is calculated as the difference between successive annual diameter measurements at the same POM, divided by [days between measurements/365]). Because our goal was highly accurate year to year measurements, the criterion of a round bole free of irregularities was more important than the exact height above the ground.
When each diameter measurement is taken, a second field technician checks the value against the previous years diameter on the field data forms. The tree is re-measured in cases of large increases or decreases (and this is noted, along with any supporting observations). Data quality is also repeatedly assessed by checking the master database for consistency of a trees measurements across years. In the few cases of clear contradictory data (e.g., a large diameter loss or gain bracketed by years with nearly constant diameter) the inconsistent data are either corrected (when this can be done unambiguously) or eliminated (flagged with 999 as "no data").
From 1983-1985 some large trees were measured with a tree calipers (+ 1 cm). For 10 diameter measurements in 1983 and 1984 we used 1981 diameter data from the long-term study plot where these trees occurred (see Lieberman et al. 1985; no growth increments were calculated from these data).
This variable also includes codes indicating survival or death of the individuals with no accompanying diameter measurement: 9001 (alive but only observed, not measured, as for trees along trails in the Ecological Reserve); 9003, dead but no obvious cause; 9004, clearly killed by something falling on it; 9005, possibly killed by something falling on it (e.g., damaged by litterfall in one year, then dead by the next census); 9006, dead from a cause other than litter (e.g., animal biting or digging, leafcutting ants). Minimum value: 2.2 mm, Maximum value: 1561.0 mm (this range excludes the codes >9000).
V29 (166,7). DIA92. Diameter in mm in 1992 (see V28). Real 7.1. 1.9 - 1354 mm.
V30 (173,7). DIA91. Diameter in mm in 1991 (see V28). Real 7.1. 2.2 - 1518 mm.
V31 (180,7). DIA90. Diameter in mm in 1990 (see V28). Real 7.1. 1.8 - 1511 mm.
V32 (187,7). DIA89. Diameter in mm in 1989 (see V28). Real 7.1. 1.4 - 1506 mm.
V33 (194,7). DIA88. Diameter in mm in 1988 (see V28). Real 7.1. 1.6 - 1502 mm.
V34 (201,7). DIA87. Diameter in mm in 1987 (see V28). Real 7.1. 1.7 - 1496 mm.
V35 (208,7). DIA86. Diameter in mm in 1986 (see V28). Real 7.1. 1.2 - 1492 mm.
V36 (215,7). DIA85. Diameter in mm in 1985 (see V28). Real 7.1. 1.1 - 1294 mm.
V37 (222,7). DIA84. Diameter in mm in 1984 (see V28). Real 7.1. 0.9 - 1287 mm.
V38 (229,7). DIA83. Diameter in mm in 1983 (see V28). Real 7.1. 1.0 - 1870 mm.
V39 (236,7). FIRSTDIA. Diameter at the individuals first census (see V28). Real 7.1. 1.0 - 1870 mm.
V40 (243,7). DGRO8384. Diameter change in the interval 1983-1984. When DIA (current year) and DIA (previous year) are both <> -999 and <9000, this variable is calculated as [(((DIA (current year)) - (DIA (previous year))/((DATE of current census) - (DATE of past census))) x 365] This gives an annualized diameter growth rate over the most recent census interval (all variables in mm). If either DIA (current year) or DIA (previous year) = -999, this variable is set as -999. In years where the POM was changed, DGRO was calculated to the old POM, unless that diameter was no longer usable, in which case DGRO was set to -999. Real 7.1. Minimum value: -2.3 mm, Maximum value: 26.3 mm. + 0.1 mm.
V41 (250,7). DGRO8485. Diameter change in the interval 1984-1985 (see V40). Real 7.1. -2.1 - +23.0 mm. + 0.1 mm.
V42 (257,7). DGRO8586. Diameter change in the interval 1985-1986 (see V40). Real 7.1. -1.9 - 32.9 mm. + 0.1 mm.
V43 (264,7). DGRO8687. Diameter change in the interval 1986-1987 (see V40). Real 7.1. -3.1 - +21.5 mm. + 0.1 mm.
V44 (271,7). DGRO8788. Diameter change in the interval 1987-1988 (see V40). Real 7.1. -2.0 - 22.8 mm. + 0.1 mm.
V45 (278,7). DGRO8889. Diameter change in the interval 1988-1989 (see V40). Real 7.1. -3.1 - +36.5 mm. + 0.1 mm.
V46 (285,7). DGRO8990. Diameter change in the interval 1989-1990 (see V40). Real 7.1. -2.8 - +34.2 mm. + 0.1 mm.
V47 (292,7). DGRO9091. Diameter change in the interval 1990-1991 (see V40). Real 7.1. -2.1 - +30.5 mm. + 0.1 mm.
V48 (299,7). DGRO9192. Diameter change in the interval 1991-1992 (see V40). Real 7.1. -3.3 - +26.9 mm. + 0.1 mm.
V49 (306,7). DGRO9293. Diameter change in the interval 1992-1993 (see V40). Real 7.1 -3.2 + 27.3 mm. + 0.1 mm.
V50 (313,5). HOW93. How diameter was measured in 1993. String. Codes are " CAL" (with calipers); " CIN" (with diameter tape ("cinta")); " DEN" (with a tree dendrometer, 1983-1985 only, see V28); " V-L" (10 measurements taken from an on-going long-term tree study, see V29, 1983-1984 only).
V51 (318,5). HOW92. See V50.
V52 (323,5). HOW91. See V50.
V53 (328,5). HOW90. See V50.
V54 (333,5). HOW89. See V50.
V55 (338,5). HOW88. See V50.
V56 (343,5). HOW87. See V50.
V57 (348,5). HOW86. See V50.
V58 (353,5). HOW85. See V50.
V59 (358,5). HOW84. See V50.
V60 (363,5). HOW83. See V50.
V61 (368,7). HTMEAS93. Height at which diameter was measured in 1993. In the string variable, numbers are the height above ground in cm with two exceptions. "1000" means measured at exactly 10 cm below a nail placed in the bole at ca. 140 cm above the ground (see V28 for discussion of measurement height criteria), and "1300", which means 10 cm below a nail placed at approximately 3-4 meters about the ground, but which requires 2 3-m sections of Swedish climbing ladder on opposite sides of the tree. Letter codes that follow the numbers give additional information: M (marca), measured at a painted spot; T (tubón) - a second person on the ground uses the 15 m extendible measuring rod to pass the tape around large trees; H - a very difficult tree to measure; BC (bajo clavo) - below nail; AC (arriba del clavo, above nail, when for some reason a deformity appeared at 10 cm below the nail); D - stem divided below the point of measurement. String. 0 - 600.
V62 (375,7). HTMEAS92. See V61.
V63 (382,7). HTMEAS91. See V61.
V64 (389,7). HTMEAS90. See V61.
V65 (396,7). HTMEAS89. See V61.
V66 (403,7). HTMEAS88. See V61.
V67 (410,7). HTMEAS87. See V61.
V68 (417,7). HTMEAS86. See V61.
V69 (424,7). HTMEAS85. See V61.
V70 (431,7). HTMEAS84. See V61.
V71 (438,7). HTMEAS83. See V61.
V72 (445,9). DATE93. Date the individual was measured in 1993. String. Code is Year-Month, Day, "-YEARMODY". Error code for dates is " 19770707". Minimum value:" 19930118," Maximum value: " 19930510"
V73 (454,9). DATE92. See V72. " 19920205" to " 19920601"
V74 (463,9). DATE91. See V72. " 19910131" to " 19910703"
V75 (472,9). DATE90. See V72. " 19900201" to " 19900716"
V76 (481,9). DATE89. See V72. " 19890111" to " 19890619"
V77 (490,9). DATE88. See V72. " 19880113" to " 19880708"
V78 (499,9). DATE87. See V72. " 19870109" to " 19870508"
V79 (508,9). DATE86. See V72. " 19860130" to " 19860606"
V80 (517,9). DATE85. See V72. " 19850103" to " 19850506"
V81 (526,9). DATE84. See V72. " 19840130" to " 19840630"
V82 (535,9). DATE83. See V72. " 19830218" to " 19831103"
V83 (544,5). STEMCON93. Stem condition in 1993. String. Codes are (- indicates a space, no character):"---D-": damaged, that is with an abrupt reduction in the stem diameter of >=25% of the lower section; "---H": horizontal, that is with the stem knocked horizontal to the ground and/or with the crown knocked flat to the ground; "--HR-": as H above, but with a vertical sprout now coming out of the horizontal stem; "---M-": having multiple vertical stems; "--RA-": main stem broken off below the crown (rota arriba); "---N-": normal, none of the above.
V84 (549,5). STEMCON92. See V83.
V85 (554,5). STEMCON91. See V83.
V86 (559,5). STEMCON90. See V83.
V87 (564,5). STEMCON89. See V83.
V88 (569,5). STEMCON88. See V83.
V89 (574,5). STEMCON87. See V83.
V90 (579,5). STEMCON86. See V83.
V91 (584,5). STEMCON85. See V83.
V92 (589,5). STEMCON84. See V83. [note: stem condition was not noted in 1983]
V93 (594,9). FIRSTCP. Crown position at first census. Real 9.2. The value is the mean of the two values of Crown Position Index (one each, taken independently without discussion, by the two field people doing each census). Crown Position Index codes: Lateral light only (1.5-2.5). For these cases "we subjectively subdivided this class in high, medium, and low categories based on the range of conditions at La Selva...we mentally summed all canopy openings visible from the crown of the target tree..crowns classified as being at "high lateral light" [2.5] were exposed to at least one major canopy opening or to multiple medium-sized openings; crowns at "low lateral light" [1.5] were exposed to no large or medium openings; "medium lateral light" [2.0] was used to classify intermediate situations" (from Clark and Clark 1992 and see discussion there). 3.0 = some overhead light (10-90% of vertical projection of crown area illuminated from above); 4.0 = full overhead light (>90% of vertical projection of crown area illuminated from above); 5.0 = crown completely exposed (to overhead and lateral direct light, within the 90 degree inverted cone that includes the whole crown). In 1983 data are based on only 1 value, not the mean of two independent observations. Some trees on the SOR and CCC trails are in the Ecological Reserve and could not be visited directly, so their crown position values are -999. Real 8.2. Minimum value: 1.5, Maximum value: 5.0.
V94 (603,8). CP93. Crown position in 1993. See V93. Real 8.2. 1.5 - 5.0.
V95 (611,8). CP92. Crown position in 1992. See V93. Real 8.2. 1.5 - 5.0.
V96 (619,8). CP91. Crown position in 1991. See V93. Real 8.2. 1.5 - 5.0.
V97 (627,8). CP90. Crown position in 1990. See V93. Real 8.2. 1.5 - 5.0.
V98 (635,8). CP89. Crown position in 1989. See V93. Real 8.2. 1.5 - 5.0.
V99 (643,8). CP88. Crown position in 1988. See V93. Real 8.2. 1.5 - 5.0.
V100 (651,8). CP87. Crown position in 1987. See V93. Real 8.2. 1.5 - 5.0.
V101 (659,8). CP86. Crown position in 1986. See V93. Real 8.2. 1.5 - 5.0.
V102 (667,8). CP85. Crown position in 1985. See V93. Real 8.2. 1.5 - 5.0.
V103 (675,8). CP84. Crown position in 1984. See V93. Real 8.2. 1.5 - 5.0.
V104 (683,8). CP83. Crown position in 1983. See V93. Real 8.2. 1.5 - 5.0.
V105 (691,7). FIRSTPA. Number of plants above at first census. The value = the mean of the two field peoples estimates for the number of overtopping plant crowns (crowns of ground-rooted, self-supporting plants that overtop > 50% of the crown projection of the target tree); each field person does this independently, and the two values are averaged (except in 1983, when only one observer's value was recorded). The estimates were made using clinometers to establish verticality and binoculars to distinguish distinct overtopping crowns. Beginning in 1991 we stopped doing plants above for trees already in the census, and did it only once for new trees, in the year they entered the census. Some trees on the SOR and CCC were located in the Ecological Reserve and could not be visited directly, so values for them are -999. Real 7.1. Minimum value: 0.0, Maximum value: 5.0.
V106 (698,7). PA93. See V105. Real 7.1. 0.0 - 4.5.
V107 (705,7). PA92. See V105. Real 7.1. 0.0 - 5.0.
V108 (712,7). PA91. See V105. Real 7.1. 0.0 - 5.0.
V109 (719,7). PA90. See V105. Real 7.1. 0.0 - 5.0.
V110 (726,7). PA89. See V105. Real 7.1. 0.0 - 5.0.
V111 (733,7). PA88. See V105. Real 7.1. 0.0 - 5.0.
V112 (740,7). PA87. See V105. Real 7.1. 0.0 - 5.5.
V113 (747,7). PA86. See V105. Real 7.1. 0.0 - 5.0.
V114 (754,7). PA85. See V105. Real 7.1. 0.0 - 5.0.
V115 (761,7). PA84. See V105. Real 7.1. 0.0 - 5.0.
V116 (768,7). PA83. See V105. Real 7.1. 0.0 - 5.0.
V117 (775,7). FIRSTHAB. Habitat at first census. See documentation below for V118 and V128 before reading the following. Habitat was evaluated from a multiple-year run. This means that the 83 data are converted to >83 codes if possible, and if not, are left as "7" if the tree was clearly in old-growth forest (or "6" if the tree was clearly in secondary forest). The trees whole record is examined for classification as a trail tree - when the tree is repeatedly classed as a trail tree, its FIRSTHAB is coded as "5". Likewise, if the trees first HABXX is "5" or "6", but subsequent classifications disagree, the prevalent code is used in FIRSTHAB. N.B. FIRSTHAB = 6 is the code used both for secondary habitats and for human-disturbed habitat like the arboretum (these latter habitats are coded as 9 in the HABXX variables). Thus the maximum value of FIRSTHAB is 7. Real 7.1. 1.0 - 7.0
V118 (782,7). HAB93. Habitat classification in 1993. This was assessed independently by two people. For codes from 1 - 4 the numeric values averaged to give HAB93. For cases where the two scores included a 5 or 6 and a code < 4, the higher code was used. The codes are: 1 = gap, in a "Brokaw gap" (a cylindrical hole in the vegetation extending from the sky down to within 2 m of the ground, Brokaw 1982; we additionally required this gap to be > 9 m2 in projected area) ; 2 = gap/building, when the focal tree is >2 m tall but in the middle of a Brokaw gap; 3 = building phase, defined as with highest vegetation being lower than the "average canopy height" for the immediate surrounding forest; 4 = mature phase, i.e., with the highest vegetation level being at the average full canopy height of the immediate surrounding forest; 5 = trail tree, a tree in primary forest and potentially affected by proximity to a trail, because of either higher light or possible root compaction; 6 = secondary forest; 7 = unambiguously old-growth forest but not directly accessible (as in the Ecological Reserve); 9 = other (disturbed) habitats, such as the Arboretum. Real 7.1. Minimum value: 1.0, Maximum value: 6.0.
V119 (789,7). HAB92. See V118. Real 7.1. 3.0 - 6.0.
V120 (796,7). HAB91. See V118. Real 7.1. 3.0 - 6.0.
V121 (803,7). HAB90. See V118. Real 7.1. 1.0 - 6.0.
V122 (810,7). HAB89. See V118. Real 7.1. 1.0 - 7.0.
V123 (817,7). HAB88. See V118. Real 7.1. 1.0 - 6.0.
V124 (824,7). HAB87. See V118. Real 7.1. 1.0 - 6.0.
V125 (831,7). HAB86. See V118. Real 7.1. 1.0 - 6.0.
V126 (838,7). HAB85. See V118. Real 7.1. 1.0 - 7.0.
V127 (845,7). HAB84. See V118. Real 7.1. 1.0 - 7.0.
V128 (852,7). HAB83. Habitat classification in 1983. Codes were 3: mature canopy; 2 -- building phase; 1.5 - gap edge; 1 - gap; -999 -- All other microsite conditions, including trailside and second-growth, as well as other cases when habitat was not classified. Analysis of data from subsequent years indicates that the 1983 classifications were less consistent for identifying trail-side trees. Real 7.1. 1.0 - 3.0.
A. Data acquisition
1. Data forms. Census data were taken on forms prepared by computer from the previous years annual census database. Data printed on the forms included the tree identification number and location, the previous year's height and diameter values, and the height and method of the previous diameter measurement. Because the person taking the height and diameter data could not see the previous year's values, the second person, by cross-checking the new data against the forms, provided a type of blind error-trapping in the field.
2. Location of completed field data forms. Stored in original and photocopies in several buildings at the La Selva Biological Station, Costa Rica, and personally with the Clarks.
3. Data entry/verification procedures. Data were manually digitized and checked by readback by the people who took the field measurements, within a maximum of 1 month after each census was completed.
B. QA/QC procedures. These are described under each variable. Check-averages for 5 variables are (including error codes, N=2616 for each check-average): V1. ID:+26692.23; V17. HEIGHT83:-835.51; V28. DIA93:+110.953; V94. CP93:-263.6092; V128. HAB83:-688.122.
C. Related materials. See F below.
D. Computer programs. None.
E. Archiving
1. Archival procedures. Original data sheets and digital files are stored in several locations in Costa Rica at the La Selva Biological Station and in San Jose (Clarks' residence).
F. Publications and results to 1993.
Clark, D.A. and D.B. Clark. 1994. Climate-induced variation in canopy tree growth in a Costa Rican tropical rain forest. Journal of Ecology 82:865-872.
Rhoades, C.C., R.L. Sanford, Jr. and D.B. Clark. 1994. Gender dependent influences on soil phosphorus by the dioecious lowland tropical tree Simarouba amara. Biotropica 26:362-368.
Clark, D.B., D.A. Clark, and P.M. Rich. 1993. Comparative analysis of microhabitat utilization by saplings of nine tree species in neotropical rain forest. Biotropica 25:397-407.
Oberbauer, S.F., D.B. Clark, D.A. Clark, P.M. Rich and G. Vega. 1993. Light environment, gas exchange, and annual growth of saplings of three species of rain forest trees in Costa Rica. Journal of Tropical Ecology 9:511-523.
Rich, P.M., D.B. Clark, D.A. Clark, and S.F. Oberbauer. 1993. Long-term study of solar radiation regimes in a tropical wet forest using quantum sensors and hemispherical photography. Agricultural and Forest Meteorology 65:107-127.
Clark, D.A. and D.B. Clark. 1992. Life history diversity of canopy and emergent trees in a neotropical rain forest. Ecological Monographs 62:315-344.
Clark, D.B. and D.A. Clark. 1991. The impact of physical damage on canopy tree regeneration in tropical rain forest. Journal of Ecology 79:447-457.
Clark, D.B. and D.A. Clark. 1990. Distribution and effects on tree growth of lianas and hemiepiphytes in a Costa Rican tropical wet forest. Journal of Tropical Ecology 6:321-331.
Clark, D.B. and D.A. Clark. 1989. The role of physical damage in the seedling mortality regime of a neotropical rain forest. Oikos 55:225-230.
Oberbauer, S.F., D.A. Clark, D.B. Clark, and M. Quesada. 1989. Comparative analysis of photosynthetic light environments within the crowns of juvenile rain forest trees. Tree Physiology 5:13-23.
Oberbauer, S.F., D.B. Clark, D.A. Clark, and M. Quesada. 1988. Crown light environments of saplings of two species of rain forest emergent trees. Oecologia 75:207-212.
Clark, D.B., and D.A. Clark. 1987. Population ecology and microhabitat distribution of Dipteryx panamensis, a neotropical rain forest emergent tree. Biotropica 19:236-244.
Clark, D.A., and D.B. Clark. 1987. Análisis de la regeneración de árboles del dosel en bosque muy húmedo tropical: aspectos teóricos y prácticos. Revista de Biología Tropical 35 (Suppl. 1):41-54.
LITERATURE CITED:
Brokaw, N. V. L. 1982. The definition of treefall gap and its effect on measures of forest dynamics. Biotropica 14:158-160.
Clark, D.A., and D.B. Clark. 1987. Análisis de la regeneración de árboles del dosel en bosque muy húmedo tropical: aspectos teóricos y prácticos. Revista de Biología Tropical 35 (Suppl. 1):41-54.
Clark, D.A., and D.B. Clark. 1992. Life history diversity of canopy and emergent trees in a neotropical rain forest. Ecological Monographs 62:315-344.
Clark, D.A., and D.B. Clark. 1994. Climate-induced variation in canopy tree growth in a Costa Rican tropical rain forest. Journal of Ecology 82:865-872.
Clark, D.A., and D.B. Clark. 1999. Assessing the growth of tropical rain forest trees: issues for forest modeling and management. Ecological Applications 9:981-997.
Clark, D.B., and D.A. Clark. 1987. Population ecology and microhabitat distribution of Dipteryx panamensis, a neotropical rain forest emergent tree. Biotropica 19:236-244.
Clark, D.B., and D.A. Clark. 1990. Distribution and effects on tree growth of lianas and hemiepiphytes in a Costa Rican tropical wet forest. Journal of Tropical Ecology 6:321-331.
Clark, D.B., and D.A. Clark. 1991. The impact of physical damage on canopy tree regeneration in tropical rain forest. Journal of Ecology 79:447-457.
Clark, D.B., and D.A. Clark. 1996. Abundance, growth and mortality of very large trees in neotropical lowland rain forest. Forest Ecology and Management 80:235-244.
Clark, D.B., D.A. Clark, and P.M. Rich. 1993. Comparative analysis of microhabitat utilization by saplings of nine tree species in neotropical rain forest. Biotropica 25:397-407.
Clark, D.B., D.A. Clark., and Read, J.M. 1998. Edaphic variation and the mesoscale distribution of tree species in a neotropical rain forest. Journal of Ecology 86:101-112.
Lieberman, D., Lieberman, M., Peralta, R., and Hartshorn, G. S. 1985, Mortality patterns and stand turnover rates in a wet tropical forest in Costa Rica: Journal of Ecology, 73, p. 915-924.
McDade, L.A., K.S. Bawa, H.A. Hespenheide and G.S. Hartshorn, editors. 1994. La Selva: ecology and natural history of a neotropical rain forest. University of Chicago Press, Chicago.
McDade, L.A., and G.S. Hartshorn. 1994. La Selva Biological Station. Pages 6-14 in L.A. McDade, K.S. Bawa, H.A. Hespenheide and G.S. Hartshorn, editors. La Selva: ecology and natural history of a neotropical rain forest. University of Chicago Press, Chicago.
Oberbauer, S.F., D.A. Clark, D.B. Clark, and M. Quesada. 1989. Comparative analysis of photosynthetic light environments within the crowns of juvenile rain forest trees. Tree Physiology 5:13-23.
Oberbauer, S.F., D.B. Clark, D.A. Clark, and M. Quesada. 1988. Crown light environments of saplings of two species of rain forest emergent trees. Oecologia 75:207-212.
Oberbauer, S.F., D.B. Clark, D.A. Clark, P.M. Rich and G. Vega. 1993. Light environment, gas exchange, and annual growth of saplings of three species of rain forest trees in Costa Rica. Journal of Tropical Ecology 9:511-523.
Sanford, R. L.,Jr., P. Paaby, J.C. Luvall, and E. Phillips, E. 1994. Climate, geomorphology, and aquatic systems. Pages 19-33 in in L.A. McDade, K.S. Bawa, H.A. Hespenheide and G.S. Hartshorn, editors. La Selva: ecology and natural history of a neotropical rain forest. University of Chicago Press, Chicago.
We thank Gerardo Vega, Luis Fernando Corrales and especially Leonel Campos and William Miranda Conejo for their years of careful dedication to this project. Financial support was provided by the National Science Foundation (BSR-8516371, BSR-8918185, and DEB94-07581), the Andrew W. Mellon Foundation, and the Organization for Tropical Studies. We thank OTS and NSF for making it possible to do long-term studies in old-growth tropical rain forest by operating the La Selva Biological Station.