8/12/2019 1989 Ecology Schupp

1/4

Arrival and Survival in Tropical Treefall Gaps

Eugene W. Schupp; Henry F. Howe; Carol K. Augspurger; Douglas J. Levey

Ecology, Vol. 70, No. 3. (Jun., 1989), pp. 562-564.

Stable URL:

http://links.jstor.org/sici?sici=0012-9658%28198906%2970%3A3%3C562%3AAASITT%3E2.0.CO%3B2-5

Ecologyis currently published by Ecological Society of America.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtainedprior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content inthe JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/journals/esa.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academicjournals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community takeadvantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.

http://www.jstor.orgMon Oct 15 15:21:26 2007

http://links.jstor.org/sici?sici=0012-9658%28198906%2970%3A3%3C562%3AAASITT%3E2.0.CO%3B2-5http://www.jstor.org/about/terms.htmlhttp://www.jstor.org/journals/esa.htmlhttp://www.jstor.org/journals/esa.htmlhttp://www.jstor.org/about/terms.htmlhttp://links.jstor.org/sici?sici=0012-9658%28198906%2970%3A3%3C562%3AAASITT%3E2.0.CO%3B2-5

8/12/2019 1989 Ecology Schupp

2/4

5 6 2 SPECIAL FEATURE-TREEFALL GAPS AND FOREST DYNAMICS Ecology Vol. 70 No. 3

Therefore the selection that takes place beneath the ACKNOWLEDGMENTSclosed canopy has a great influence on the suite of Several NSF grants have contributed to the developments~e ci esavailable to take advantage of the short burst of the ideas contained In this essay among them: BSR83-of light in a gap, we need to study the mechanisms 071 19 OCE84-08610. and OCE86-08829. I thank mq col-

leagues at Santa Barbara and P. Cochranc J . Robinson C.affecting the population dynamics of individuals both Uhl . B Wellington C, Zammit. and anonymous review-inside and outside gaps to achieve a fuller understand- ,,, for comm,nts on this manuscrioting of forest ecology. For repnnts of t h ~ sspecla]Feature. see footnote page 535

E co l o f l , 70(3), 1989. pp. 562-564i 1989 b) the Ecolog~cal oclety of Amenca

ARRIVAL AND SURVIVAL IN TROPICAL TREEFALL GAPSEUGENEW. SCHUPPEstacrBn BrolBgrca de Doriana, .4partado 105 6, E-41 08 0 Sevrlla, Sp ain

HENRYF. HOWEDepar tmen t o f Brolog~ca l crences, Cn r~ er s l t ~ ,f Illlnors at Chrcago,Box 4348, Chlcago, Illrnors 60680 C'S,4CAROLK. AUGSPURGERDepa rtment o f Plant Brology, Cnrversrty of Illrnors,Crbana. Illrnors 61801 C'SA

ANDDOUGLAS. LEVEYDepartment of loo lo^: Cnr~ ersrrl, f Florrda,Garnesl rlle, Florrda 326 11 C SA

Most tropical tree species require light from a treefall Martinez-Ramos et al. 1989). The probability that agap at some time during their lives to reach maturity. tree of a given species will enter the forest canopy is aResponses to light condi tions have been dichotomized function of the joint probabilities of arriving and sur-as shade-intolerant pioneers or shade-tolerant climax viving in particular habitats. We emphasize three is-species (e.g., Whitmore 1975, 1982, 1989). The former sues that, for any species, define probable regenerationtypically have small, widely dispersed seeds from which at a given site: (1) pattern of seed arrival in gaps an djuveniles establish only in gaps, while the latter typi- beneath the canopy, (2) proportion of forest area incally have larger seeds that can germinate beneath the gap vs. closed canopy, and (3) survival to reproductiveforest canopy and can persist as suppressed juveniles maturi ty of seeds landing in gaps and beneath the can-or grow slowly until a gap forms. According to this opy. This view enables recruitment of tropical trees toframework. a new gap promotes shade-intolerant re- be interpreted from the perspective of relative advan-generation through germination and shade-tolerant re- tages ofgiven characteristics within the context of thosegeneration through release of suppressed juveniles. environments in which individuals with those char-Truly shade-tolerant species can grow to maturity be- acteristics are located.neath the forest canopy, but even these are likely tobenefit from any increases in light levels beneath thecanopy (Uhl et al. 1988,Canham 1989, Lieberman et Most animal seed vectors avoid recently created gaps.al. 1989b, Martinez-Ramos et al. 1989). New gaps offer few resources to frugivores and are

Although useful, we believe this dichotomy limits dangerous sites for birds to perch or bats to fly. Weviews of gap dynamics by implying that each species expect lower animal-mediated seedfall in recent gapsis constrained to a specific pathway to the forest can- than beneath the forest canopy (see Brokaw 1986,opy. In reality, all species recruit to differing degrees Charles-Dominique 1986). The degree of discrepancyfrom dispersal into new gaps and from release of dor- in seedfall density between gaps and beneath the forestmant seed or juvenile banks beneath the canopy (see canopy will vary with seed size and dispersal agent

8/12/2019 1989 Ecology Schupp

3/4

June 1989 SPECIAL FEATURE-TREEFALL GAPS AND FOREST DYNAMICS 563

(Fig. IA). Large monkeys, toucans, and guans are un-likely to deposit the large seeds they disperse in gaps.Instead, they drop seeds under canopy trees that theyuse as perches. The more numerous birds and batsdispersing small-seeded plants are only slightly morelikely to carry seeds to recent gaps. The activity ofthese animals is highest in and adjacent to maturinggaps. Shade-intolerant trees bearing fleshy fruits andunderstory shrubs responding to increased light withenhanced fruit production draw dispersal agents to ma-turing gaps (Levey 1988). Seedfall densities of small-seeded species should be higher in maturing gaps thanin either recent gaps or beneath the forest canopy, butmay be highest in the 5-10 m wide band of forestsurrounding the gap, where, after feeding, small fru-givores take refuge from predators to which they areexposed in gap centers (Howe 1979, Snow and Snow1986).

Wind may disperse seeds into gaps more efficientlythan animals (Fig. IA). Turbulence created by air flowacross the broken canopy and convectional currents inheated gaps (Kimmins 1987) should increase the prob-ability of seed deposition in treefalls by altering thewind speed and aerodynamic behavior of seeds and bytrapping seeds in eddies (Burrows 1975). Furthermore ,air sinking into the leading edge of a gap (Kimmins1987) might pull adjacent winds with their seeds intothe canopy opening, increasing the area over which atreefall captures seeds. We predict that a dispropor-tionate number of wind-dispersed seeds should fall inrecent gaps. Densities of seeds of wind-dispersed species1.6times greater in gaps than in paired adjacent canopysites in a Panamanian forest are consistent with thisprediction (Augspurger and Franson 1988).

Newly formed gaps cover only 1% of the area oftropical forests (Hartshorn 1978, Brokaw 19856, Dens-low 1987, Lawton and Putz 1988). Thus, the vast ma-jority of seeds ofall dispersal modes should fall beneatha closed canopy (also see Brokaw and Scheiner 1989).Even with disproport ionate capture of wind-dispersedseeds by gaps, Augspurger and Franson (1988) estimatethat only 4 .1 of the seeds fall in gaps 3 yr of age.Because new openings are rare and disproportionatearrival is subtle, survival in gaps relative to areas be-neath the canopy must be extremely high if dispersaldirectly into recent gaps is to be the priman pathwayto maturity. Nonetheless, a slight change in relativearrival probabilities can have a major impact on re-cruitment.

Survival depends on both physiological attributesand biotic interactions. Important biotic interactionsinclude disease an d seed and seedling predation. Patho-gen resistance influences dormant seed survival and

Und er- M aturin g Understory Recenta story Gap Gap

Large-Seeded Small-Seeded Wind-DispersedAnimal-Dispersed Animal-DispersedFIG. 1. Predicted patterns of arrival, survival. and re-

cruitment in tropical forests. The > axis is scaled as a pro-portion of the maximum value for each dispersal categorq..Arrows indicate the direction of change as gap conditionsreturn to those beneath the forest canopy. (A) Density ofseedfall. Seedfall of large-seeded animal -dispersed species isalmost exclusively beneath the canopy. that of small-seededanimal-dispersed species is highest adjacent to maturing gapsand least in recent gaps. Wind-dispersed seedfall is highest ingaps. (B) Probability of surviving disease and seed and seed-ling predation. Large-seeded animal-dispersed survival islowest in and near gaps because of the concentration of seedand seedling predators. Small-seeded animal-dispersed andw~nd-dispersed urvival are highest in recent gaps because ofthe reduced incidence of d~se as e.C) Probabllit) of a d~spersedseed producing a reproductive adult. Large-seeded animal-dispersed species survival to reproduction is virtuall) limitedto seeds germinating beneath the canopy. Small-seeded ani-mal-dispersed recruitment is highest around maturing gapsdue to the high seed input and short-term residency in theseed and seedling banks, and lowest in maturing gaps due tocompetition, low light levels, and the long interval until thegap reopens. Wind-dispersed species recruitment is highestin recent gaps and lowest in the forest.

8/12/2019 1989 Ecology Schupp

4/4

56 SPECIAL FEATURE-TREEFALL GAPS AN D FOREST DYNAMICS Ecology. Vol. 70. No. 3

seedling establishment in the shade Fig. 1B). In bothfield Augspurger 1 9 8 4 ) and greenhouse Augspurgerand Kelly 1984) studies of seedlings of 18 Panamanianwind-dispersed species, mortality from disease was al-most always higher up to 7 2 times) in shade than insun. Higher light and lower humidity in forest openingsare inimical to plant pathogens Weber 1973 , Rotem1978) , and faster growth in gaps reduces the period ofsusceptibility Augspurger 1 9 8 3 b ) .Disease rather thanlight limitation may be the proximate source of mor-tality for shaded seedlings of most shade-intoleranttropical tree species. Seedlings in general may havegreater establishment-phase mortality in the shade thanin sun Ganvood 1982 , Sork 1985) . but large-seeded,animal-dispersed species seem to be less susceptible topathogens than are seedlings of small-seeded animal-dispersed or wind-dispersed species Brokaw 1985 b,Sork 1987) . For the vast majority of species, however,the relative roles of pathogens, physiology and animalsin limiting regeneration are unknown.

Large seeds and seedlings are most vulnerable topredation in gaps see Fig. B) where rodents seek shel-ter in limb and vine tangles Rood and Test 1968 ,Emmons 1982) . Predation by rodents of seeds and ju-veniles of Faramea occidentalis in Panama Schupp1 9 8 8 a , b ) and We lfia georgii in Costa Rica Schuppand Frost 1989) is far greater in gaps than benea th thecanopy. Similarly, Virola sur inam ensis seedlings aremore likely to be killed or heavily browsed in gaps thanin the forest ( . 05 I ; H . F. Howe, personalobservation). Preda tion on small seeds and seedlings isprobably not as disproportionately high in gaps be-cause rodents are not as predominant a source of mor-tality.

Regeneration in a particular place is determined bythe interplay of probabilities of arrival and survival.Large-seeded animal-dispersed species will establishalmost entirely beneath the forest canopy because fewsuch seeds reach a gap and most that do are eaten bymammals Fig. 1C). These seedlings apparently areshade tolerant and resistant to disease. Many of thesespecies benefit from higher light in gaps if they avoidseed predation Platt and Hermann 1986, Uhl et al.1 9 8 8 ) ,but their establishment is not restricted to gaps.Despite attrition beneath the forest canopy, highermammalian seed predation in gaps makes survival tomaturity more likely for seeds deposited beneath thecanopy than for seeds dispersed directly into gaps.

In contrast, both small-seeded animal-dispersed andwind-dispersed species often depend on gaps for es-tablishment Fig. C ) Augspurger 1 9 8 4 , Brokaw1985b) .A number of these species have dormant seedsthat wait for a new gap to form, but dormancy also

involves the risk of high attri tion. Mortality of Cordiaelaeagnoides Sarukhan 1 9 8 0 )and Cecropia obtu s fol iaMartinez-Ramos and Alvarez-Buylla 1986 , Martinez-

Ramos et al. 1 9 8 9 ) seeds in the soil was >95 in 1yr. Long-term accumulation of dormant seeds makesthe seed bank numerically more importa nt than freshseed rain in gap regeneration Putz and Appanah 1987,Lawton and Putz 1 9 8 8 ) , but the loss of dormant seedsto animals, fungi, and deep burial Ganvood, in press)means that the low-density seed rain in recent gaps willbe disproportionately important for recruitment of thesesmall-seeded species. A small seed landing in a recentgap will be much more likely to become an adult thanwill a seed landing beneath the canopy, where it maytake decades for a gap to form.

Although the greatest proport ional survival of small-seeded species is in new gaps, we hypothesize that thegreatest likelihood of regeneration may occur aroundextant maturing gaps Fig. C ) , where new treefalls aremost likely to occur Hubbell and Foster 1986c, Law-ton and Putz 1988) . Enlarging of matur ing gaps wouldrelease seeds and seedlings that had accumulated asdispersers carry fruits out of the gaps and into theadjacent forest for processing. Release within a rela-tively short period of time also would decrease attrit ionin the seed or seedling banks relative to those locatedbeneath the forest canopy or in maturing gaps wherelight levels are low Denslow 1 9 8 7 ) and are likely toremain so for long periods of time. This postulatedcoupling of high seedljuvenile densities with an in-creased probability of canopy opening suggests that gapexpansion may be more important in tropical forestregeneration than is generally thought.

In summary, locations and frequencies of recruit-ment will be determined by joint probabilities of ar-rival and survival in the habitat mosaic of a tropicalforest. Probability of arrival is determined by dispersalmode, of survival by do rmant seeds and/or juveniles,physiology, and the interactions of each species withpathogens and predators in gaps and the surroundingforest. These combined factors, along with growth ofjuveniles in the different light microenvironmentswithin the forest and in gaps, will determine the relativeprobabilities that each species will reach reproductivematurity after dispersal to a particular site in a forest.The size, number, and distribution of gaps will interactwith these patterns of arrival and survival to determinecommunity levels of regeneration.

We acknowledge support from the National Science Foun-dation, the Noyes Foundation, the Organization for TropicalStudies. and the Smithsonian Tropical Research Institute. N.Ganvood. E. Leigh. W. Platt, and two anonymous reviewershelped improve previous versions of the manuscript.For r pnnts of th i s Special Feature see footnote I page 5 3 5