clark & clark 1995
Post on 14-Apr-2018
Embed Size (px)
7/27/2019 Clark & Clark 1995
ELSEVIER Forest Ecology andManagement 0 (1996) 235-244Pores;;;ologyManagement
Abundance, growth and mortality of very large treesin neotropical lowland rain forestDavid B. Clark *, Deborah A. Clark
Department of Biology, Uniuersity of Missouri-St. Louis. St. Louis, MO 63121-4499, USA
AbstractVery large trees, arbitrarily defined as those over 70 cm diameter above buttresses, account for a major portion of theabove-ground biomass in neotropical rain forests. Owing to the scarcity of indiv iduals of a given species and the diff iculty ofaccurate measurement, there are few species-level data on the growth, mortality, and abundance of species that regularlyreach emergent status. We report such data for very large individuals from old-growth tropical wet forest at the La SelvaBiological Station in the Atlantic lowlands of the Republic of Costa Rica. The landscape-scale abundance of a ll speciesreaching over 70 cm diameter was assessed using 515 O.Ol-ha quadrats located at grid points in a 500 ha area of old-growthforest. In the total sample of 2301 stems 10 cm or more in diameter, very large individuals accounted for 2% of the stems,
23% of the basal area, and 27% of the estimated above-ground biomass. Growth and survival for five species that regularlyattain emergent status were measured in a 150 ha area within the 500 ha plot. Survival of 282 very large individuals of thefive species was measured over 6 years. The mean annual mortality rate of the total sample was only 0.6% year-. Meanannual diameter growth increments varied from 1.9 to 5.2 mm year- among species, and were negatively correlated withdiameter in four of the five species. For a sample of 193 indiv iduals measured over 7 years, growth almost exactly equalledlosses in basal area and biomass due to mortal ity. Because al l of these species are regularly recruiting new trees into the over70 cm diameter class, the amount of biomass in the large-indiv idual size class is increasing over the 150 ha old-growth studyarea. Historic disturbance and/or current climatic change are hypothesized to account for the increase. We identify lack ofstandard diameter measurement criteria, and small and potentially unrepresentative plot locations as two problems inassessing the role of very large trees in other neotropical forests. Future studies should sample larger areas; this will increasethe generality of the conclusions and will make possible a species-level comparison of the ecology of very large tropicaltrees.Keywords: Emergent;Carbon storage;Tropical ree demography1. Introduction
The largest organisms in any forest are the bigtrees. This is particularly true in forests where occa-
* Correspondingauthor at: La Selva Biological Station, IN-TERLINK-341, P.O. Box 02-5635,Miami, FL 33152,USA.
sional emergents tower over the surroundingcanopy-level trees. These massive individuals aremuch more important ecologically than their densitysuggests. For example, in tropical wet forest thebiomass of one tree 150 cm in diameter equals thatof approximately 607 trees 10 cm in diameter(calculated from Brown and Iverson, 1992). If verylarge trees are still growing at only modest diameter
0378.1127/96/$15.00 0 1996Elsevier ScienceB.V. Al l rights reservedSSDI 0378-l 127(95)03607-5
7/27/2019 Clark & Clark 1995
236 D.B. Clark, D .A. Clark/ Forest Eco logy and Management 80 (1996) 235--244
increments, volume and biomass increments can besubstantial. In addition to the carbon stored in livingtrees, the enormous trunks of fal len emergents canact as important sinks for carbon for decades tocenturies after a tree falls (Harmon and Hua, 1991).
Very large trees can also play a key role in forestdisturbance regimes. Treefall gap size is generallycorrelated with the size of the gap-making tree (cf.Tyrrell and Crow, 1994). In forests where the distur-bance regime is driven primarily by tree-falls, deathsof emergents may cause the largest disturbances. Thelongevity of large individuals also increases theirpotential to generate small-scale heterogeneity inmicroclimate or soil (cf. Rhoades et al., 1994). Ifvery large trees live substantially longer than theircanopy leve l neighbors, they may create special mi-croenvironments around them due not only to theirmassive size, but also to greater persistence throughtime. Growth and death rates of very large treestherefore can affect major ecosystem attributes suchas net rate of carbon storage (Lug0 and Brown,1992; Brown, 1995) and the frequency of largeendogenous disturbances.
Very large trees are also ecophysiologicallyunique. Lacking the shelter of neighboring canopy-level crowns, emergents are exposed to the hottestand most evaporative canopy environments. In thisexposed and potentia lly stressed position, very largetrees are likely to be more t ightly coupled to weatherand climate conditions than smaller individuals. Forthese reasons, one might expect the first effects ofchanging global c limate on canopy tree crowns to beexpressed in emergent indiv iduals.
In spite of the inherent interest in very large trees,there are few species-level data on their ecology.This is particularly true for tropical rain forests,where the absolute density of very large trees is lowand the total diversity of canopy tree species isextremely high. The number of tree species perhectare in these forests commonly exceeds 100 andcan pass 300 (Gentry, 1990; Gentry and Ortiz, 1993),and for individuals 10 cm or more in diameter, mostspecies are represented by less than one individualha-. Even plots of substantial area will seldomcontain enough very large individuals of a givenspecies to produce a statistically useful sample.
A further reason for the lack of good data on verylarge tropical trees is the difficulty of obtain ing good
measurements. Buttresses reaching 10 m or higherare common, necessitating the use of multiple sec-tions of ladder to measure diameter above buttresses.The upward growth of buttresses frequently rendersold points of measurement unusable in studies withlong remeasurement intervals. The frequent occur-rence of large lianas and hemiepiphytes further hire-ders accurate measurements.
In this paper we report data on the growth, mor-tali ty and landscape-scale abundance of very largetrees in a neotropical lowland rain forest. We alsocontrast the growth and mortality of very large treesto that of smaller trees in the same forest. Finally, wecompare our results with data from other tropicalsites, and discuss what we consider to be the majoroutstanding questions in the ecology of very largetropical trees.
To study very large trees one must first make asubjective decision as to How large is very large?.The size of the largest l--2% of the trees in anyforest varies with many factors, including soil. slope,altitude, latitude, disturbance history, and biogeogra-phy. For purely practical reasons based on ease ofmeasurement (see Methods) we used a cutoff of 70cm diameter. Subsequently we discovered that otherresearchers had also used the same limits (Lug0 andBrown, 1992; White, 1994; Brown, 1995) or strati-fied their data by tree size such that the over 70 cmsize class could be analyzed (Gentry and Terborgh,1990; Rankin-de-Marona et al., 1990). For the speciesdescribed here most of the individuals over 70 cm indiameter have emergent crowns (Clark and Clark,1992). We realize that any division point is at leastpart ially arbitrary. However, based on practicality ofmeasurement, facility of data comparison, relativeabundance and crown position we believe that 70 cmdiameter criterion is a useful division point forneotropical wet or moist forest data.
2. Study siteThe study was carried out at the La Selva Biologl-
cal Station in the Atlantic lowlands of the Republicof Costa Rica. A detailed description of the site isgiven in McDade et al. (1994). The old-growth forestis classified as Tropical Wet Forest in the Holdridgesystem (Hartshom and Peralta, 1988). Rainfall aver-
7/27/2019 Clark & Clark 1995
D.B. Clark, D.A. Clark/ Forest Ecology and Management 80 (1996) 235-244 237
ages about 4000 mm annually, and no month aver-ages less than 100 mm (Sanford et al., 1994).
3. MethodsThe trees measured for growth and survival are
part of a long-term ongoing study of tree demogra-phy in a 150 ha area of non-swamp old-growthtropical rain forest. Here we report data on the fivespecies (Table 1) in the long-term study which regu-larly reach emergent status. A full description of theprotocols for sampling and annual measurement isgiven in Clark and Clark (1992). The first individu-als in the long-term study were measured in 1983,and new individuals have been added annually ac-cording to a standard set of protocols (described inClark and Clark, 1992). We used al l indiv iduals over70 cm diameter above buttresses at first census tocalculate annual mortality (m, Sheil et al.,, 1995)over a 6 year period for trees that entered the studyin 1983, as well as for the subsequent cohorts addedeach year through 1988. Hence, mortality for the1983 individuals was assessed over the period 1983-1989, and the 6 year measurement period was shiftedone year for the cohorts added in successive years upthrough 1988. Data from all six 6-year periods werecombined for each species. To compare death ratesbetween large and small size classes of the samespecies, we also calculated 6-year mortality ratesusing the same methods for al l trees lo-70 cm indiameter at first encounter.
or irregularities to over 7 m height are excluded.Trees over 70 cm in diameter are measured annuallyonly if a cylindrical site of measurement is presentwithin about 7 m of the ground (i.e. reachable fromtwo joined 3-m sections of tree climbing ladders)