Baker Growth And Func Composition

Perspectives in Plant Ecology, Evolution and Systematics

Vol. 6/1,2, pp. 21–36 © Urban & Fischer Verlag, 2003 http://www.urbanfischer.de/journals/ppees

Variation in tropical forest growth rates: combined effects of functional group composition and resource availability Timothy R. Baker1,2,3,*, Michael D. Swaine1 & David F.R.P. Burslem1 1

School of Biological Sciences, University of Aberdeen, Aberdeen, UK Max-Planck-Institut für Biogeochemie, Jena, Germany 3 Centre for Biodiversity and Conservation, School of Geography, University of Leeds, UK 2

Received: 12 February 2003 · Revised version accepted: 28 March 2003

Abstract Rates of tree growth in tropical forests reflect variation in life history strategies, contribute to the determination of species’ distributional limits, set limits to timber harvesting and control the carbon balance of the stands. Here, we review the resources that limit tree growth at different temporal and spatial scales, and the different growth rates and responses of functional groups defined on the basis of regeneration strategy, maximum size, and species’ associations with particular edaphic and climatic conditions. Variation in soil water availability determines intra- and inter-annual patterns of growth within seasonal forests, whereas irradiance may have a more important role in aseasonal forests. Nutrient supply limits growth rates in montane forests and may determine spatial variation in growth of individual species in lowland forests. However, its role in determining spatial variation in stand-level growth rates is unclear. In terms of growth rate, we propose a functional classification of tropical tree species which contrasts inherently fastgrowing, responsive species (pioneer, large-statured species), from slow-growing species that are less responsive to increasing resource availability (shade-bearers, small-statured species). In a semi-deciduous forest in Ghana, pioneers associated with high-rainfall forests with less fertile soils, had significantly lower growth rates than pioneers that are more abundant in low-rainfall forests with more fertile soils. These results match patterns found in seedling trials and suggest for pioneers that species’ associations with particular environmental conditions are useful indicators of maximum growth rate. The effects of variation in resource availability and of inherent differences between species on stand-level patterns of growth will not be independent if the functional group composition of tropical forests varies along resource gradients. We find that there is increasing evidence of such spatial shifts at both small and large scales in tropical forests. Quantifying these gradients is important for understanding spatial patterns in forest growth rates. Key words: irradiance, maximum size, nutrient supply, pioneer, regeneration groups, water availability

*Corresponding author: Centre for Biodiversity and Conservation, School of Geography, University of Leeds LS2 9JT, UK; e-mail: [email protected]

1433-8319/03/6/01-02-021 $ 15.00/0

22

T. R. Baker et al.

Introduction An understanding of the patterns of tree growth is a fundamental goal of ecological research in tropical forests. Interspecific variation in maximum potential growth rate consistently emerges as one of the most important factors in the definition of robust functional groups, as growth rate integrates numerous traits that underlie trade-offs among strategies for resource acquisition, defence against natural enemies, and allocation to reproduction. Such groups provide practical and meaningful classifications of tropical forest species, which are needed both by foresters, for modelling growth and yield, and by ecologists, to explain the life history diversity in tropical forest trees (Vanclay 1994; Richards 1996; Whitmore 1998; Turner 2001). In addition to these inherent differences between species, growth also varies with resource availability. At the species level, this variation may help to explain the limits to species distributions (e.g. Gunatilleke et al. 1996; Veenendaal et al. 1996a). At the stand level, understanding the environmental factors that control tropical forest productivity is critical for quantifying the carbon balance of tropical forests. Spatial variation in stand-level growth rates will depend on both variation in resource availability and any differences in the functional composition of tropical forests. The large variation in forest composition and dynamics at both small and large scales (Phillips et al. 1994, in press; Burslem & Whitmore 1999, in press; ter Steege et al. 2000) suggests that there may be

important differences between tropical forests in the relative abundance of different functional groups. These differences may be as important as gradients in resource availability for determining current and future patterns of forest productivity. Building on the writings of Budowski (1965), Tim Whitmore made an important contribution both to the description of tropical tree functional groups and to highlighting their significance to central questions of tropical forest ecology (Whitmore 1974, 1975, 1998). He also drew attention to the relationship between plant functional traits, including growth rate, and tropical forest silviculture (Whitmore 1975). Whitmore’s promotion of a fundamental dichotomy between fast-growing, gap-demanding tree species (labelled ‘pioneers’) and slow-growing, non-gap-demanders (‘non-pioneers’ or ‘climax’ species) is, perhaps, one of his best known and most frequently cited contributions (Whitmore 1975, 1984, 1998; Swaine & Whitmore 1988). This perspective was undoubtedly influenced by Whitmore’s own research on the dynamics and growth rates of seedlings, saplings and large trees of the twelve most common large tree species growing on his plots on Kolombangara in the Solomon Islands during the period 1964–1971 (Whitmore 1974). Based on an initial seven-year study of seedling responses to canopy openings, Whitmore (1974) classified these twelve species into four groups (Table 1) and emphasized the importance of differences in growth rate between species as a determinant of group membership. A link between the four groups

Table 1. Characteristics of the twelve species studied since 1964 in lowland tropical rain forest on Kolombangara, Solomon Islands. The four species’ groups among the twelve common timber tree species are classified according to the conditions required for seedling establishment and onward growth and are based on observations over 6.6 years over the interval 1964–1971 (Whitmore 1974). Nomenclature follows sources described in Burslem & Whitmore (1999). Shade-tolerance class

Species

Conditions to establish

Conditions to grow up

Wood density (kg m–3)1

I

Dillenia salomonensis Maranthes corymbosa Parinari papuana ssp. salomonensis Schizomeria serrata

High forest High forest High forest High forest

High forest High forest High forest High forest

550 720 660 490

II

Calophyllum neo-ebudicum Calophyllum peekelii Pometia pinnata

High forest or small gaps High forest High forest or disturbed

High forest/gaps High forest/gaps High forest or ?small gaps

500 480 590

III

Campnosperma brevipetiolatum Elaeocarpus angustifolius

High forest or gaps High forest

Gaps Gaps

330 350

IV

Endospermum medullosum Gmelina moluccana Terminalia calamansanai

Mostly gaps Mostly gaps High forest, soon dying except in gaps

Gaps Gaps Gaps

370 410 460

1

Data from Anonymous (1976, 1979).

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

Variation in tropical forest growth rates

defined for the twelve species growing on Kolombangara and the dichotomy proposed in his later writings can be made if the three species in Group IV of Table 1 are treated as pioneers, and the remaining species as non-pioneers. Since the mid 1970s, there has been a proliferation of research on the characteristics that define the main functional groups of tropical trees and their relationship to forest dynamics and regeneration, expanding and extending the ideas of Whitmore and his predecessors and contemporaries. Some authors have questioned the existence of a fundamental dichotomy of life history types among tropical trees (e.g. AlvarezBuylla & Martinez-Ramos 1992; Grubb 1996), and the issue has been comprehensively reviewed several times (Brokaw 1985; Denslow 1987; Brown & Jennings 1998; Brokaw & Busing 2000; Turner 2001; Burslem & Swaine 2002). Here, we specifically examine how useful these concepts are in understanding variation in tree growth rates over gradients in resource availability. We consider the following specific questions: 1. What is the evidence for resource limitation of tropical tree growth? 2. How do functional groups defined in various ways differ in their growth responses to resource availability? 3. Are there gradients in the functional-group composition of tropical forests, and are they important in determining variation in stand-level growth? We consider each of the major resources (water and irradiance) or group of resources (nutrients) known to limit plant growth over spatial gradients, both separately, and, where possible, in combination. We adopt a broad definition of growth rate in order not to impose severe limitations on the scope of the review, and emphasize evidence from field-based studies of adult trees. Thus we include studies of litter fall, fine-root production, leaf-level photosynthesis, and phenology as well as direct measures of trunk growth based on measurements of diameter change or the width of annual rings.

Factors limiting tree growth Water The total amount of rainfall is one of the most important factors setting the limits to the distribution of forests as opposed to woodland or thicket in the tropics (Holdridge 1967; Walter 1979; White 1983; Woodward 1987). However, the forested regions experience considerable spatial and temporal variation in the availability of soil water that acts as a major limiting

23

factor on overall rates and temporal patterns of growth. The importance of soil water availability is most apparent within forests with strongly seasonal climates. Over ten years in dry forest in Mexico (mean annual rainfall 707 mm) the annual increment of two deciduous species correlated with rainfall during the mid wet-season (Bullock 1997). Within the same forest, over a five-year period when annual rainfall ranged from 485 to 1331 mm, Whigham et al. (1990) found that the annual production of leaf litter correlated positively with annual rainfall and that the mean basal area increment per tree correlated positively with total rainfall during the previous two years. Soil water availability also controls the timing of growth within seasonal forests. For the evergreen species, Exostema caribaeum, in dry forest in Puerto Rico (mean annual rainfall 929 mm), Lugo et al. (1978) found maximum daily rates of photosynthesis were five times higher during the wet season compared to the dry season. In addition, marked fluctuations in tree girth in parallel with monthly rainfall were described by Swaine et al. (1990) for very dry tropical forest in Ghana (c. 750 mm yr–1), where seasonal variation in girth was about ten times greater than the underlying annual increment in Millettia thonningii. Direct measurement of wood production in seasonal forest from examination of the pattern of cambial activity, by measuring either the width of the band of differentiating xylem which stains for cellulose (Amobi 1973), or the width of unlignified xylem (Lowe 1968), has also demonstrated the importance of increased soil water availability for the initiation of growth in seasonal forests (Borchert 1999). Soil water availability also influences the timing and amount of growth in forest with substantially higher mean annual rainfall than the previous examples. For instance, in a semi-deciduous forest in Venezuela (mean annual rainfall 1700 mm), using tree ring chronologies of seven species, Worbes (1999) reported positive correlations between annual rainfall and mean annual growth rates, and at three sites in seasonal forest in Panama, Devall et al. (1995) found that annual rainfall correlated with variation in tree ring width for three species. In contrast to these correlative studies, field experiments that directly test whether water supply limits growth in a particular forest are rare. However, an irrigation experiment over five dry seasons in semi-evergreen forest in Panama (mean annual rainfall 2600 mm) showed that ameliorating the seasonal drought could increase fine-root production and leaf longevity, and influence phenology. Dry season community-level fine-root production, measured using in-growth cylinders with root-free soil, was five times greater in irrigated plots than in control plots

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

24

T. R. Baker et al.

(Cavelier et al. 1999), and dry season leaf fall was delayed for two out of nine tree species (Wright & Cornejo 1990). In addition, for three shrub species, the proportion of leaves retained, of those that were newly flushed at the start of the experiment, was approximately 20% higher over the first two years in the irrigated plots (Mulkey et al. 1993). However, this experiment also demonstrated that not all aspects of growth, and not all species, are limited by seasonal water shortage in this forest. Irrigation did not affect community level leaf-litter production (Cavelier et al. 1999), stem diameter growth of five shrub species (Wright 1991), or the timing of leaf fall for seven out of nine study species (Wright & Cornejo 1990). There is mixed evidence on whether water supply substantially limits growth in forests growing in aseasonal climates, where annual variation in soil water availability is minimal. At La Selva in Costa Rica (mean annual rainfall 3859 mm), Hazlett (1987) found that the lowest rates of girth change for two tree species, Carapa guianensis and Goelthalsia meiantha, occurred during the driest period of the year, and at a welldrained site; Breitsprecher & Bethel (1990) found that cessation of growth for five out of eight species was also associated with the drier part of the year. However, in the same forest, Clark & Clark (1994) found that variation in growth rate failed to correlate with variation in rainfall over eight years, for both adult trees and juveniles. In another forest in Costa Rica (mean annual rainfall 3970 mm), a negative correlation was found between rainfall and shoot growth rate of a liana species, Passiflora pittieri (Longino 1986). All the studies described above concentrate on single sites; forest growth in the tropics has rarely been compared along rainfall gradients. However, Harrington et al. (1995) examined the growth of Acacia koa along an altitudinal gradient in Hawaii, where annual rainfall ranged from 850–1800 mm. These authors found that increment in aboveground woody biomass increased as phyllode δ13C values became more negative, but was not correlated with phyllode nutrient concentrations, suggesting that water rather than nutrient availability limited growth. Soil water availability also varies at small spatial scales, differing between topographic positions in both evergreen and semi-deciduous tropical forests (Becker et al. 1988; Daws et al. 2002; Green & Newbery 2002; Baker et al. 2002, 2003) but the influence of these differences on tree growth has rarely been studied. However, topographic variation in soil water availability does influence plant water relations and patterns of cambial activity in seasonal forests. Becker et al. (1988) demonstrated that during the dry season in a semi-evergreen forest in Panama, pre-dawn water potentials of the shrub Psychotria horizontalis and of saplings of the

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

tree Trichilia tuberculata, were half as low in slope compared to plateau positions. In addition, in seasonally dry forest in Costa Rica, species such as Astronium graveolens, which occur in moist localities, are able to maintain cambial activity into the dry season and are leafless only for a short period, whereas girth increment ceases during the dry season in deciduous species in drier sites (Borchert 1999). In both examples, greater water availability in lower topographic positions improves conditions for tree growth.

Irradiance Irradiance declines exponentially with decreasing height in tropical forests (Yoda 1978) and increased canopy illumination has a clear, positive effect on tree growth. For example, in moist evergreen forest in Panama, saplings of ten species showed greater height growth in high-light environments (King 1994) and at La Selva, Costa Rica, positive correlations were found between growth rate and crown illumination category for eight species in size classes up to 30 cm diameter (Clark & Clark 1992). For large trees, many studies have used Dawkins’ (1958) classification of tree crown position in the canopy to show effects on tree growth (e.g. Korsgaard 1986; Silva et al. 1995; Alder & Silva 2000). For example, in individuals of Carapa nicaraguensis, 30–40 cm in diameter in swamp forest in Costa Rica, Webb (1999) showed that median diameter increments were significantly higher in higher crown-score classes. Also, in a study of 15 species in tropical forest in Puerto Rico (mean annual rainfall range 2500–4500 mm), Parresol (1995) demonstrated that maximum growth rates were five times higher in dominant trees than in trees in the suppressed crownclass categories. Variation in irradiance also influences temporal patterns of growth, and may be particularly important in aseasonal forests, where water availability is not limiting. For instance, in the absence of any correlation of growth with rainfall at La Selva, Costa Rica (mean annual rainfall 3659 mm), Clark & Clark (1994) suggested that inter-annual variation in irradiance determined the consistent long-term growth patterns of adult and juvenile trees. In addition, studies of phenology have indicated that trees may time the production of new leaves during the months of the highest irradiance, if water supply is not limiting. Wright & van Schaik (1994) found that in two weakly seasonal forests, at La Selva, in Costa Rica and at the Ducke Reserve in central Brazil, the number of species that centred leaf production in the three-month period of highest irradiance was significantly greater than expected by chance. Also, they found that in four seasonal forests, two thirds of deep-rooted species flushed

Variation in tropical forest growth rates

during the dry season, when irradiance was highest. Finally, results from a biogeochemical model of forest productivity also suggested that light may limit growth in aseasonal forests, as a decline in modelled estimates of net primary productivity at the wettest sites along a transect in Brazilian Amazonia was attributed to reduced irradiance (Potter et al. 1998).

Nutrients Most tropical forests grow on relatively nutrient-poor soils. The effect of soil fertility on growth has therefore been a focus of much experimental work, and fertilisation experiments on adult trees in natural forests provide some evidence that nutrient supply does limit growth rates. In montane forest in Venezuela, Tanner et al. (1992) demonstrated that trunk growth of all species approximately doubled in plots fertilized with N and P, and in montane forest in Colombia, N and P fertilization also significantly increased growth for two out of three species (Cavelier et al. 2000). On soils of volcanic origin in montane forest in Hawaii, Vitousek et al. (1993) demonstrated that N limited growth on the two youngest sites (<30 and 200 years old), while P was limiting on older soils (2000 years). However, few studies have directly examined nutrient limitation in lowland tropical forests. In one study in Kalimantan (mean annual rainfall c. 3600 mm), N and P fertilisation over four years had no effect on trunk growth, but did increase litterfall (Mirmanto et al. 1999). The few field studies comparing patterns of soil nutrient availability with variation in adult tree growth in natural forest have failed to find significant correlations. Clark et al. (1998), working in forest in an aseasonal climate in Costa Rica with adults of nine canopy or emergent tree species, found no difference in growth rate between two well-drained soil-types that differed in available P concentration by a factor of two. In addition, Ashton & Hall (1992) found no correlation between the proportional diameter growth (1965–1985) of large trees (dbh >30 cm), and soil nutrient status across 13 plots in high-rainfall forests in Borneo. However, neither of these studies compared growth rates across the large differences in soil fertility that are found along regional gradients of rainfall, and differences in site fertility on this scale do influence plantation performance. In stands of teak (Tectona grandis) more than 10 years old in West Africa, for example, total N in the topsoil (0–10 cm) and rooting depth were the most important factors determining variation in growth (Drechsel & Zech 1994). Also, for plantations of Terminalia ivorensis across seven forest reserves in Ghana (mean annual rainfall 1280–1650 mm), soil total N and C concentrations, cation exchange capacity and exchangeable Ca and Mg concen-

25

trations were significantly negatively correlated with the degree of die-back, a condition that leads to wilting, chlorotic leaves, poor growth and high mortality (Agyeman & Safo 1997). Plantation studies have also indicated that variation in nutrient supply may be important over topographic gradients. Variation in growth of teak over a catena in the rain forest zone of Liberia (mean annual rainfall 2553 mm) was related most strongly to variation in soil pH and percentage base saturation (Zech & Drechsel 1991). Experimental studies of seedling growth also demonstrate that regional and topographic gradients of soil fertility can cause differences in growth of individual species. Veenendaal et al. (1996a) studied seedling growth in a range of irradiances in well-watered pots of soils from semi-deciduous (pH >6.1) and evergreen forest (pH <4.9) in Ghana, and found that two out of 15 species grew faster on soil from semi-deciduous forest, and one species grew faster on soil from evergreen forest. The greater growth of species on generally more fertile soils seems likely to be caused by nutrient availability: foliar concentrations of P, N, Ca and Mg were generally higher in seedlings grown on soil from semi-deciduous forest (Veenendaal et al. 1996a). In contrast, the greater growth of one species on generally less fertile soils was tentatively attributed to a specific requirement for a trace element, such as Cu or Zn (Veenendaal et al. 1996a). In another study, in a comparison of growth in seedlings transplanted into soils collected along a topographic gradient in aseasonal forest in Sri Lanka, five out of eight species of Shorea showed greater dry mass after two years when grown on valley soil as opposed to ridge soil, and seven species grew more on valley soil than on midslope soil (Gunatilleke et al. 1996). Valley soils had a higher pH than both mid-slope and ridge soils, and higher concentrations of total N & Mg, and the differences in growth were attributed to topographic variation in soil nutrient availability.

Interactions between resources Previously, we have considered individual resources as if they act independently. However, variation in irradiance, nutrients and water availability can have complex interacting effects on tree growth. For example, the response of plant growth to nutrient availability increases at higher flux densities of irradiance, because faster growth imposes an increased demand for nutrients (Peace & Grubb 1982). For tropical trees, this positive interaction between irradiance and nutrient supply has been shown in studies of pot-grown seedlings (Thompson et al. 1992; Lehto & Grace 1994; Veenendaal et al. 1996a; Huante et al. 1998a,b). The demonstration of similar interactions in fertilisa-

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

26

T. R. Baker et al.

tion experiments in the field appears to depend on the inherent fertility of the soil and the magnitude of difference in irradiance between gap and understorey treatments. For example, Dalling & Tanner (1995) found a response to fertilisation of seedlings of three species transplanted onto nutrient-poor landslides in Jamaica under high irradiance, but no effect of fertilisation in the understorey. However, Fetcher et al. (1996) found no difference in the effect of fertilisation in a comparison of gap centre and gap edge sites on a landslide in Puerto Rico. In contrast, Denslow et al. (1990) found no effect of fertilisation in either gap or understorey conditions for seven species of shrubs on more fertile soils at La Selva, Costa Rica. Other interactions, between light and water supply, or water and nutrient supply have received even less attention, even for pot-grown seedlings, and there have been very few field-based studies. However, Fisher et al. (1991) showed a positive interaction between water supply and irradiance on height growth and leaf area production of Virola surinamensis seedlings in an irrigation experiment in semi-evergreen forest in Panama. An interaction between nutrient supply and water availability may be important in determining differences in the spatial pattern of tree growth observed in different years. Baker et al. (2003) studied diameter growth of trees > 20 cm diameter of two species over two years in semi-deciduous and evergreen forest in Ghana. During the first year, when there was similar, low rainfall in both sites, growth did not differ between forests. However, during the second year, when soil water availability was generally higher at both sites, growth was greater in the semi-deciduous than in the evergreen forest. This difference between the two sites, apparent only when there was sufficient rainfall, was attributed to the higher soil nutrient availability in semi-deciduous forest. In summary, within tropical rain forests, it is evident that water availability is an important influence on tree growth rates, particularly in seasonal forests, where it determines both the inter- and intra-annual patterns of growth. However, it is less clear whether the substantial spatial variation in soil water availability between different tropical forests, or the smaller fluctuations in soil water availability within aseasonal forests, determine important differences in tree growth. Variation in irradiance is the primary factor that limits plant growth within forest stands, and possibly over long timescales for aseasonal forest as a whole. Variation in soil fertility appears to be important for determining variation in the growth rates of some species, but may have a smaller role in controlling overall stand-level patterns of growth. However, truly broad-scale comparisons from a large number of sites are currently lacking.

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

Tree growth and functional groups We now consider variation in tree growth from the perspective of differences between functional groups. A functional group can be defined as a suite of species that share similar species-specific patterns (i.e. evolved traits, not phenotypic) of resource use, a similar response to disturbance, or a similar range of growth, mortality and recruitment rates (Gitay & Noble 1997). By recognising groups of species that share suites of covarying characteristics we may be able to scale-up from the well-studied responses of few species, to model community- or stand-level responses to perturbations such as climate change or tree harvesting (Vanclay 1994; Condit et al. 1996). The main purpose of recognising aggregations of traits in life histories is thus to provide a semi-quantitative foundation for these practical objectives, while in practice most ecologists emphasise continua rather than discontinuities in the spectrum of life histories of coexisting species. Turner (2001) identified regeneration class and maximum height as the primary factors that differentiate tropical tree species, and below we review briefly the evidence for trends in growth rate and response to resource availability among functional groups defined in these ways. Finally, we propose a new way of defining a functional group, based on differences among species in their distribution at large spatial scales in respect of soils and climate, and illustrate differences among groups defined in this way for trees growing at one site in semi-deciduous forest in Ghana.

Groups defined by regeneration strategy The major axis of differentiation in ecological characteristics used to separate species in tropical forests has been related to light requirements for regeneration, because light is the primary limiting factor for growth rate within a forest stand, and because it is possible to allocate species on the basis of field observation alone. Species that require the high light environment of a gap for seed germination and/or establishment (‘pioneers’), are separated from species that are able to establish under canopy shade (‘non-pioneers’) (Swaine & Whitmore 1988; Whitmore 1989). These two categories mark sections on what is now agreed to be a continuum of responses to the range of light environments found on the forest floor (Swaine & Whitmore 1988; Alvarez-Buylla & Martinez-Ramos 1992). Many studies have examined growth responses of tropical tree seedlings with differing regeneration strategies to variation in irradiance (e.g. Thompson et al. 1992; Lehto & Grace 1994; Veenendaal et al. 1996a). This work has shown that under high-light conditions, pioneers have higher growth rates than non-pioneers.

Variation in tropical forest growth rates

For example, in a comparison of Ghanaian tree seedlings, three pioneer species (Terminalia ivorensis, Milicia excelsa and Albizia zygia), all had higher relative growth rates than a group of five non-pioneer shade bearers (Strombosia glaucescens, Cynometra ananta, Guarea cedrata, Celtis mildbraedii and Chrysophyllum pruniforme) at 16% of ambient irradiance (Veenendaal et al. 1996a). At lower light levels, whether pioneers have higher growth rates than non-pioneers appears to depend on exactly how the comparison is performed. A compilation of the results of seedling growth studies of 194 species grown under contrasting light regimes found that pioneers and light demanding non-pioneer species had higher relative growth rates than more shade-tolerant species, at up to 5% irradiance (Veneklaas & Poorter 1998). However, 5% irradiance is greater than the compensation point of all tropical trees so far tested, and when lower irradiances are included in experiments (e.g. Agyeman et al. 1999) pioneers are found to grow more slowly than shade-bearers, and typically show negative growth at 2% irradiance. The timescale of the experiment also influences the interspecific differences that are observed (Sack & Grubb 2001). When measured over short periods in low irradiance, seedlings of light demanding species may have higher growth rates than more shade-tolerant species, due to the higher growth relative growth rates associated with smaller seed sizes. However, over longer timescales, shade-tolerant species are expected to out-perform light demanders (Sack & Grubb 2001). Amongst adult trees in the field, pioneer species are usually found to grow faster than more shade-tolerant species, presumably because they both have higher intrinsic growth rates at a given irradiance, and because they are found in high-light sites. Swaine (1994) reported growth rates of pioneer trees in semi-deciduous forest in Ghana more than double those of any more shade-tolerant group, and in semi-evergreen forest in Panama, Condit et al. (1996) found that species ‘colonizing index’, defined as the proportion of recruits found in light gaps, was positively correlated with species growth rates in the 10–20 cm diameter class. However, adult tree growth rate is not always simply related to regeneration strategy. For example, Milton et al. (1994), working in semi-evergreen forest in Panama, compared the growth rate of two groups of species, defined on the basis of whether recruitment was significantly higher, or did not differ, between low and high canopy sites, based on previous work by Welden et al. (1991). Over 13.6 years, for trees >19.1 cm diameter, growth rates were significantly higher in species whose regeneration was not related to low canopy sites. This complexity is at least partly caused by the considerable variation in growth rate that is found in non-pioneer species beyond the seedling stage

27

due to changes in species light requirements during ontogeny (Clark & Clark 1992, 1999). For example, many species that are shade-tolerant as seedlings require a large increase in canopy illumination as saplings to reach the canopy (Jones 1956; Clark & Clark 1992; Hawthorne 1995). In addition, others have noted small-stature species whose seedling establishment requirements match those of pioneers, but which later in life tolerate deep shade, and behave as shade-tolerant understorey trees, such as the ‘cryptic pioneers’ of Hawthorne (1995, p. 16). An analogous ontogenetic shift in light requirement is displayed by the small-seeded canopy tree Alseis blackiana on Barro Colorado Island, Panama, which establishes only in canopy gaps but persists for several years in the shade after canopy closure (Dalling et al. 2001). In addition to higher growth rates at high irradiance, pioneer species also have greater photosynthetic plasticity and show greater growth responses to increased irradiance than shade-tolerant species. A comparison of the range of maximum photosynthetic rates, under high and low light, for species subjectively described as early- and late-successional (16 and 24 species, respectively), showed that early-successional species have a greater range in Amax values than latesuccessional species, because of significantly higher values under high light (Strauss-Debenedetti & Bazzaz 1996; Thomas & Bazzaz 1999). For adult trees in the field, Welden et al. (1991) working in a semi-evergreen forest in Panama, noted that increased growth in low compared to high canopy sites was a particular feature of six pioneer species. Nutrient addition experiments with seedlings show similar patterns to those with irradiance, with a greater response in pioneer and early-successional species. Huante et al. (1995) demonstrated that three early-successional species from a deciduous forest in Mexico showed a greater proportional increase in biomass (3.7–124 fold increase) between low and high (0 and 41 ppm) P treatments, than four late-successional species (1.2–2.4 fold increase). In addition, Raaimakers & Lambers (1996) showed that the biomass of a pioneer species from Guyana, Tapirira obtusa doubled over six months at high (>250 mg P per plant) compared to low (<10 mg P per plant) P-supply, whereas growth of the non-pioneer species Lecythis corrugata was not affected. In natural forest on a landslide in Puerto Rico, Fetcher et al. (1996) found that two pioneer species showed a greater response in growth to N and P fertilisation than two shade-tolerant species. Also, in lowland forest in Puerto Rico, saplings of the pioneer Cecropia schreberiana were three to five times more numerous in fertilized compared to control plots after four years (Walker et al. 1996).

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

28

T. R. Baker et al.

Groups based on maximum height A second axis of differentiation amongst tree species in tropical rain forest, related to the vertical gradient in irradiance, is caused by differences in maximum height. Species with low maximum heights typically have low growth rates. For example, in semi-deciduous forest in Ghana, Swaine (1994) found that mean annual diameter-increment of large shade-tolerant species was double that of small-stature species, and in Sungei Menyala Forest Reserve in Malaysia, Manokaran & Kochummen (1987) found that mean annual diameter increments of understorey species were less than 2 mm yr–1, compared to increments greater than 3 mm yr–1 for emergent species. Growth rates are also less variable in species with low maximum heights. At La Selva, Costa Rica, Lieberman & Lieberman (1987) demonstrated that total variation in growth rate is approximately five times greater in shade-tolerant canopy species than in understorey species. Distinguishing whether these patterns are caused by a correlation between tree diameter and absolute measures of diameter growth, lower levels of crown illumination for smaller trees, or because small trees have lower inherent growth rates under similar levels of irradiance, requires comparisons that control for tree diameter and the light environment. In Pasoh Forest Reserve, Malaysia, using asymptotic height derived from height/diameter relationships as an estimator of maximum species height, Thomas (1996) found a linear correlation (r2 = 0.56, P < 0.001) between maximum height and mean annual growth rate of adult trees for 38 species. In addition, the same positive correlation (r2 = 0.20, P < 0.01) was found using the growth rates of saplings (1–2 cm diameter) rather than adult trees, suggesting that the relationship is not confounded by variation in tree diameter. Also, in a comparison of photosynthetic characteristics of similar-sized saplings, under similar light conditions, of species with different asymptotic heights, smaller-statured species were found to have an inherently lower leaf level photosynthetic capacity (Amax), compared to species with greater maximum height, within the same genus (Thomas & Bazzaz 1999). Both inherently lower diameter growth rates and lower levels of crown illumination therefore characterise species with low maximum heights. However, it is important to note that all of the above studies refer predominately to shade-tolerant tree species. These species establish in the understorey and therefore experience increasing light levels as they increase in size. This pattern will inevitably lead to increased growth rates in larger individuals, and, on an evolutionary timescale, favour increased capacity for growth in tall-

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

statured species (Thomas & Bazzaz 1999). In contrast, pioneer species establish under high light conditions, so there will not be a positive correlation between light availability and plant size. Therefore, similar ecological or evolutionary arguments for faster growth in species with greater maximum height are unlikely to apply to pioneers (Thomas & Bazzaz 1999).

Groups defined by species’ associations with specific environmental conditions: a test from the forest zone of Ghana A third axis of differentiation of tropical forest species may relate to the associations between species’ distributions and climatic and edaphic factors. Species’ distribution in respect of edaphic factors has been demonstrated to be a useful predictor of growth rates in studies of temperate plants, where species from different habitats have been grown under the same environmental conditions. The generalisation to emerge has been that species from nutrient-poor habitats typically have lower maximum growth rates than species from nutrient-rich sites when they are grown under the same experimental conditions (Grime & Hunt 1975; Grime 1979; Chapin 1980; Chapin et al. 1986), although this is not universally applicable (Grubb 1998). Tropical forest tree species have often been shown to have distinctive distributions across largescale edaphic and climatic gradients (e.g. Gartlan et al. 1986; Newbery et al. 1986; Baillie et al. 1987; Tuomisto et al. 1995; Swaine 1996). However, the relevance of these associations to variation in growth rate within one site has not previously been examined. Here, we test this idea using the extensive ecological knowledge of Ghanaian forest species, and inventory data from a semi-deciduous forest. Firstly, we developed a classification of tree species incorporating both their regeneration requirements and their association with different parts of a major resource gradient. Within Ghanaian tropical forests, regional scale patterns of species distribution have been comprehensively described across a gradient of increasing rainfall and decreasing soil fertility towards the south-west of the forest zone (1200–2300 mm yr–1, soil pH 7.0–3.8; Hall & Swaine 1976, 1981). The first axis of a multivariate ordination of compositional data from 155 closed-canopy forest plots was shown to quantify species position along this gradient, with decreasing axis one scores in the wettest, least fertile forest sites (Hall & Swaine 1981). To categorise species’ environmental preferences, the median score (25) of all species from 23, 1-ha plots in a range of Ghanaian forest types was used as an arbitrary threshold value (Baker 2000). Species with axis one scores above this value were classified as dry forest special-

Variation in tropical forest growth rates

Fig. 1. Distribution of diameter increments within Ghanaian semi-deciduous forest, for six functional groups, defined by regeneration strategy and distribution pattern with respect to climatic and edaphic conditions. (a) Shade-bearers, (b) non-pioneer light-demanders, and (c) pioneers. There is a significant difference in median growth rate between dry- and wet-forest pioneer species. Box size represents the range between the first and third quartiles, and whiskers extend 1.5 times this range. Outliers are shown as asterisks; mean (open circle) and median (shaded circle) are also indicated.

29

ists, and species with scores below this value were classified as wet forest specialists. This categorization represents a coarse division of species responses to a regional scale environmental gradient, and does not consider local (e.g. topographic) influences of water and nutrient availability. However, the environmental factors that determine species’ distributions at different scales are likely to be similar. For example, in Ghana, the abundance of Celtis mildbraedii declines as the landscape becomes wetter, both at local and regional scales (Swaine & Hall 1986; Hawthorne 1995). Therefore, we are confident that our classification adequately, if crudely, represents species’ environmental preferences. Species were also grouped according to their regeneration requirements (‘pioneer’, ‘non-pioneer light-demander’ and ‘non-pioneer shade-bearer’), following Hawthorne (1995). Therefore, overall, species were grouped into six categories, each including species with a range of maximum heights. Inventory data were obtained from permanent sample plots located in Bobiri Forest Reserve, in Ghana (1°15–23′W, 6°40–42′N), which is classified as Moist Semi-deciduous Forest (Hall & Swaine 1981). Mean annual rainfall (1961–1993) for Kumasi, 15 km west of the forest is 1183 mm. The main dry season occurs from December to March (Swaine et al. 1997), with a less severe drier period from July to September (Veenendaal et al. 1996b). Most of the reserve contains soils developed over upper and lower Birrimian phyllite (Adu 1974), with gently undulating topography. The soils consist of red, silty clay loams in summit areas and become paler in colour lower downslope. Valley soils are grey, sandy loams and clays (Adu 1974). Seven 1-ha plots that had not been damaged by fire or degraded by logging were selected in a range of topographic positions. They were established by the Ghana Forestry Department in July/August 1990 and re-enumerated in July/August 1996. The girth of each tree >20 cm dbh was measured to the nearest mm and each measured tree was scored for the degree of crown illumination (‘crown score’) on a five point scale (Dawkins 1958). The painted point of measurement was at a height of 1.3 m, or 50 cm above the top of any buttress. Annual diameter growth of surviving trees was compared between wet- and dry- forest species for each of the three regeneration categories using Kruskal-Wallis tests. For the three comparisons, the only significant difference in growth rate was found between wet- and dry-forest pioneer species (Fig. 1). Dry-forest pioneer species had significantly higher median growth rates than wet-forest pioneers (H = 6.25, P < 0.012, n = 92, dry-forest pioneers, and 27, wet-forest pioneers). The

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

30

T. R. Baker et al.

Fig. 2. Annual diameter increment as a function of tree diameter for wet and dry forest pioneers, within semi-deciduous forest in Ghana.

difference in growth rate between these groups does not appear to be caused by differences in tree size or crown illumination. Wet- and dry-forest pioneers comprise trees of a similar diameter range, and there is no significant difference in the median diameter between the two groups (Fig. 2; H = 0.52, P > 0.05). In addition, the proportion of trees with high levels of crown illumination are similar: 30% of dry forest pioneers and 29% of wet forest pioneers were classified as Dawkins’ crown score 4 or 5. The lower growth rates of wet-forest compared to dry-forest pioneers may reflect an inability of the wet forest species, that typically occur on less fertile soils, to respond to the higher nutrient status of the soils of the semi-deciduous forest site. This contrast is supported by seedling experiments on two pioneer species that have different distribution patterns, Triplochiton scleroxylon and Lophira alata (Table 2). Both species are strongly light demanding, and attain large sizes, greater than 90 cm dbh (Hawthorne 1995). However, they show strongly contrasting patterns of seedling growth on low and high fertility soils. Under well-watered conditions, T. scleroxylon showed significantly higher growth on soil from semi-deciduous compared to evergreen forest sites, whereas L. alata grew fastest on the less fertile, evergreen forest soil (Swaine et al. 1997; Veenendaal et al. 1996a). With the results here, these patterns suggest that there may be variation in the maximum growth rate of pioneer species in Ghanaian forest, related to their association with particular edaphic and climatic conditions. In summary, in terms of growth rate, the classifications of functional groups proposed by Turner (2001) differentiates groups of similar species. Classifications

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

based both on regeneration strategy and maximum size, generally divide species that have low growth rates and show little response to increasing resources, from more responsive, faster-growing species. However, in both cases, exceptions occur. The growth rate of adult trees of non-pioneer species may be modified by variation in light demand during ontogeny, and maximum size may not be associated with any systematic variation in growth rate for pioneer species. Classifications of species based on their association with particular edaphic and climatic conditions may be useful for categorizing important variation in the ecology of pioneer species in tropical forests.

Functional gradients in forest composition So far, we have focussed on two rather different perspectives on tropical tree growth, and separately considered variation in growth rates with resource availability, and between functional groups. However, both factors will be important for understanding spatial variation in stand-level growth rates, if there are differences in the relative abundance of functional groups between forests. There is increasing evidence of such functional gradients at a range of scales within tropical forests. Studies of gradients in the functional composition of tropical forests have concentrated on variation in the relative abundance of pioneer and non-pioneers, or variation in stand-level mean values for traits such as wood density, that are strongly correlated with species’ light demand (Whitmore 1998). For example, in Ghana, pioneer species are both most abundant and

Variation in tropical forest growth rates

31

Table 2. Dry- and wet-forest pioneers in Ghanaian tropical forest. Nomenclature and classification of pioneers follows Hawthorne (1996); association with dry or wet forest based on a multivariate ordination of species presence or absence in 155 plots across the forest zone of Ghana (Hall & Swaine 1981). Dry forest pioneers ––––––––––––––––––––––––––––––––––––––––––––––––– Species Family

Wet forest pioneers –––––––––––––––––––––––––––––––––––––––––––– Species Family

Baphia pubescens Canarium schweinfurthii Ceiba pentandra Celtis adolfi-friderici Cleistopholis patens Draceana arborea Ficus exasperata Holoptelea grandis Lannea welwitschii Milicia excelsa Morus mesozygia Newbouldia laevis Rauvolfia vomitoria Bombax buonopozense Ricinodendron heudelotii Stereospermum acuminatissimum Terminalia superba Tetrapleura tetraptera Tetrorchidium didymostemon Triplochiton scleroxylon Zanthoxylum leprieurii

Cola caricifolia Daniellia ogea Daniellia thurifera Hannoa klaineana Lophira alata Petersianthus macrocarpus Bombax brevicuspe Zanthoxylum gillettii

Fabaceae Burseraceae Bombacaceae Ulmaceae Annonaceae Agavaceae Moraceae Ulmaceae Anacardiaceae Moraceae Moraceae Bignoniaceae Apocynaceae Bombacaceae Euphorbiaceae Bignoniaceae Combretaceae Fabaceae Euphorbiaceae Sterculiaceae Rutaceae

diverse in semi-deciduous forests, where rainfall is lower and soils are more fertile than in evergreen forest (Hawthorne 1996). In addition, mean stand level wood density and seed size are lowest in southern compared to central Guyana, indicating that southern forests have a greater relative abundance and diversity of more light-demanding species (ter Steege & Hammond 2001). Also, in a comparison of 59 plots across Amazonia, mean stand level wood density was 12% higher in eastern, compared to north-western Amazon forests (Baker et al., in press). Variation in past and current rates of disturbance are probably the most important factor determining the variation in the composition of tropical forests with respect to light demand. For both Ghana and Guyana, higher proportions of light-demanding taxa are associated with areas that have a history of higher rates of anthropogenic disturbance (Fairhead & Leach 1998; ter Steege & Hammond 2001). Past human activity has also been suggested as an important factor determining the pioneer-rich flora of Barro Colorado Island in Panama (Sheil & Burslem 2003). Across Amazonia, it seems unlikely that there are important systematic differences in past human activity that have generated the observed variation in stand-level wood density (Baker et al., in press). Here, higher abiotic disturbance rates in western compared to eastern Amazon forests may explain the pattern. However, it is also

Sterculiaceae Fabaceae Fabaceae Simaroubaceae Ochnaceae Lecythidaceae Bombacaceae Rutaceae

likely that soil nutrient availability influences the functional composition of tropical forest, as the growth rates of pioneer species are particularly responsive to higher soil nutrient concentrations. The forests of both Ghana and Guyana that have a higher abundance of light demanding taxa, occur on more fertile soils (Hawthorne 1996; ter Steege & Hammond 2001). As this pattern coincides with variation in the intensity of past human disturbance, it is difficult to distinguish the relative importance of the two factors. These functional gradients in tropical forest composition have important implications for spatial variation in forest growth and dynamics, as light demand is generally positively correlated with species’ growth and mortality rates (Whitmore 1998). For example, at a small scale, Tim Whitmore’s long-term study of forest dynamics on Kolombangara in the Solomon Islands has yielded evidence of spatial variation in stand-level turnover rates that correlates with variation in functional group composition (Burslem & Whitmore 1999, in press). In addition, within a network of plots in aseasonal climates across Sarawak, a guild of pioneer species increased in abundance at a high-nutrient site, but were rare elsewhere, resulting in positive correlations between soil nutrient concentrations and diameter growth of recruits (Ashton & Hall 1992). At a larger scale, the lower wood density of western compared to eastern Amazon forest is associ-

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

32

T. R. Baker et al.

ated with higher tree turnover rates (Phillips et al. 1994, in press) and higher stand-level wood production, calculated on both a volume and mass basis (Y. Malhi et al., unpubl. results). Finally, in a transcontinental comparison of the dynamics of an evergreen and a semi-deciduous forest, a guild of pioneer species present only in the semi-deciduous forest was suggested to determine overall plot differences in growth and mortality (Condit et al. 1999). The more seasonal forest had higher overall growth and mortality rates than a plot lacking the guild of fast-growing pioneer species. In summary, these studies demonstrate that spatial variation in the proportion of light-demanding taxa, occurs at range of scales in tropical forests. Variation in disturbance rates appears to be the main factor that determines these patterns, although soil nutrient availability may also be important. The combined effects of functional group composition and resource availability will therefore determine spatial variation in standlevel growth.

Future directions In this review, we have used a broad definition of ‘growth rate’ incorporating both changes in size and carbon content of plants, plant parts and populations. Size is important, both ecologically as an important determinant of competitive success, and for silviculture as timber yields are typically determined on the basis of volume. In contrast, growth rates calculated in terms of changes in carbon content are more closely related to variation in rates of photosynthesis and respiration and hence resource availability, and are important for calculating stand-level carbon balance. The two quantities are related through variation in allocation strategies, which vary both along resource gradients and between functional groups. In terms of aboveground growth, inclusion of wood density in calculations of the growth rate of adult trees is essential for understanding how different life history strategies compare in terms of carbon uptake (e.g. Enquist et al. 1999). In addition, more work is required on how trends in belowground allocation correspond to variation in aboveground productivity. For example, fineroot densities in the topsoil are higher in tropical forests on highly infertile soils (Coomes & Grubb 2000), but across Amazonia it is not known to what extent these kind of belowground patterns may offset spatial variation in aboveground productivity (Y. Malhi et al., unpubl. results). Comparative studies of the ecology and dynamics of a large number of forests are required to understand the combined effects of functional groups and resource

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

availability on spatial variation in forest growth rates. These studies will need to account for variation in water and nutrient availability as well as variation in light. In addition, quantifying the variation in life-history strategies using appropriate traits, such as wood density and maximum size, will allow tests of whether effects of resource availability are independent of variation in species composition. The ideas of Tim Whitmore in defining and exploring functional differences between tropical forest trees will therefore remain central to future research into this fundamental aspect of tropical rain forest ecology. Acknowledgements. For advice and assistance during fieldwork in Ghana we thank Dr. V.K. Agyeman (FORIG, Kumasi) and Dr. T.K. Orgle (Forest Management Support Centre, Kumasi). We also thank Kofi Affum-Baffoe, Yaw Atuahene and Raymond Votere (Inventory Unit, Forest Management Support Centre, Kumasi) for compiling the forest inventory data used in this study. Peter Grubb, Ian Turner and Oliver Phillips provided very helpful comments on a previous version of this manuscript. This work was largely funded by a University of Aberdeen Faculty studentship to TRB, who also acknowledges current financial support from the Max-Planck-Institut für Biogeochemie, Jena, Germany.

References Adu SV (1974) Soil Map of the Ashanti Region 1:250,000. Soil Research Institute, Kumasi. Agyeman VK & Safo EY (1997) Mineral nutrition and dieback in Terminalia ivorensis A. Chev. in Ghana. Journal of Tropical Ecology 13: 317–335. Agyeman VK, Swaine MD & Thompson J (1999) Responses of tropical forest tree seedlings to irradiance and the derivation of a light response index. Journal of Ecology 87: 815–827. Alder D & Silva JNM (2000) An empirical cohort model for management of terra firme forests in the Brazilian Amazon. Forest Ecology and Management 130: 141–157. Alvarez-Buylla ER & Martinez-Ramos M (1992) Demography and allometry of Cecropia obtusifolia, a neotropical pioneer tree – an evaluation of the climax-pioneer paradigm for tropical rain forests. Journal of Ecology 80: 275–290. Amobi CC (1973) Periodicity of wood formation in some trees of lowland rainforest in Nigeria. Annals of Botany 37: 211–218. Anonymous (1976) Solomon Islands Timbers. Timber Booklet 1. Major Species. Solomon Islands Forestry Division, Honiara. Anonymous (1979) Solomon Islands Timbers. Timber Booklet 2. Minor Species. Solomon Islands Forestry Division, Honiara. Ashton PS & Hall P (1992) Comparisons of structure among mixed dipterocarp forests of north-western Borneo. Journal of Ecology 80: 459–481.

Variation in tropical forest growth rates Baillie IC, Ashton, PS, Court MN, Anderson JAR, Fitzpatrick EA & Tinsley J (1987) Site characteristics and the distribution of tree species in Mixed Dipterocarp Forest on Tertiary sediments in central Sarawak, Malaysia. Journal of Tropical Ecology 3: 201–220. Baker TR (2000) Spatial and temporal patterns of growth in Ghanaian tropical rain forest. PhD thesis, University of Aberdeen. Baker TR, Affum Baffoe K, Burslem DFRP & Swaine MD (2002) Phenological differences in tree water use and the timing of tropical forest inventories: conclusions from patterns of dry season diameter change. Forest Ecology and Management 171: 261–274. Baker TR, Burslem DFRP & Swaine MD (2003) Associations between tree growth, soil fertility and water availability at local and regional scales in Ghanaian tropical rain forest. Journal of Tropical Ecology 19: 109–125. Baker TR, Malhi Y, Phillips OL, Almeida S, Arroyo L, Di Fiore A, Killeen TJ, Laurance SG, Laurance WF, Lewis SL, Lloyd J, Monteagudo A, Neill DA, Patiño S, Pitman NCA, Silva JNM & Martínez RV (in press) Variation in wood density determines spatial patterns in Amazonian forest biomass. Global Change Biology. Becker P, Rabenold PE, Idol JR & Smith AP (1988) Water potential gradients for gaps and slopes in a Panamanian tropical moist forest’s dry season. Journal of Tropical Ecology 4: 173–184. Borchert R (1999) Climatic periodicity, phenology and cambium activity in tropical dry forest trees. IAWA Journal 20: 239–247. Breitsprecher A & Bethel JS (1990) Stem growth periodicity of trees in a tropical wet forest of Costa Rica. Ecology 71: 1156–1164. Brokaw NVL (1985) Gap-phase regeneration in a tropical forest. Ecology 66: 682–687. Brokaw NVL & Busing RT (2000) Niche versus chance and tree diversity in forest gaps. Trends in Ecology and Evolution 15: 183–188. Brown ND & Jennings S (1998) Gap-size differentiation by tropical rainforest trees: a testable hypothesis or a brokendown bandwagon? Dynamics of Tropical Communities (eds. DM Newbery, HHT Prins & N Brown), pp. 79–94. Blackwell Science, Oxford. Budowski G (1965) Distribution of tropical American rain forest species in the light of successional processes. Turrialba 15: 40–42. Bullock SH (1997) Effects of seasonal rainfall on radial growth in two tropical tree species. International Journal of Biometeorology 41: 13–16. Burslem DFRP & Swaine MD (2002) Forest dynamics and regeneration. Foundations of Tropical Forest Biology: Classic Papers with Commentaries (eds. RL Chazdon & TC Whitmore), pp. 577–583. Chicago University Press, Chicago. Burslem DFRP & Whitmore TC (1999) Species diversity, susceptibility to disturbance and tree population dynamics in tropical rain forest. Journal of Vegetation Science 10: 767–776. Burslem DFRP & Whitmore TC (in press) Functional diversity and response to disturbance in lowland tropical rain

33

forest on Kolombangara, Solomon Islands. Long term changes in composition and diversity as a result of natural and man made disturbances: case studies from the Guyana Shield, Africa, Borneo and Melanesia (ed. H ter Steege). Tropenbos, The Netherlands. Cavelier J, Tanner E & Santamaria J (2000) Effect of water, temperature and fertilizers on soil nitrogen net transformations and tree growth in an elfin cloud forest of Colombia. Journal of Tropical Ecology 16: 83–99. Cavelier J, Wright SJ & Santamaria J (1999) Effects of irrigation on litterfall fine root biomass and production in a semideciduous lowland forest in Panama. Plant and Soil 211: 207–213. Chapin FS (1980) The mineral nutrition of wild plants. Annual Review of Ecology and Systematics 11: 233–260. Chapin FS, Vitousek PM & van Cleve K (1986) The nature of nutrient limitation in plant communities. The American Naturalist 127: 48–58. Clark DA & Clark DB (1992) Life history diversity of canopy and emergent trees in a neotropical rainforest. Ecological Monographs 62: 315–344. Clark DA & Clark DB (1994) Climate-induced annual variation in canopy tree growth in a Costa Rican tropical rain forest. Journal of Ecology 82: 865–872. Clark DA & Clark DB (1999) Assessing the growth of tropical rain forest trees: issues for forest modelling and management. Ecological Applications 9: 981–999. Clark DB, Clark DA & Read JM (1998) Edaphic variation and the mesoscale distribution of tree species in a neotropical rain forest. Journal of Ecology 86: 101–112. Condit R, Hubbell SP & Foster RB (1996) Assessing the response of plant functional types to climatic change in tropical forests. Journal of Vegetation Science 7: 405–416. Condit R, Ashton PS, Manokaran N, LaFrankie JV, Hubbell SP & Foster RB (1999) Dynamics of the forest communities at Pasoh and Barro Colorado: comparing two 50–ha plots. Philosophical Transactions of the Royal Society of London (Series B) 354: 1739–1748. Coomes DA & Grubb PJ (2000) Impacts of root competition in forests and woodlands: a theoretical framework and review of experiments. Ecological Monographs 70: 171–207. Dalling JW & Tanner EVJ (1995) An experimental study of regeneration on landslides in montane rain forest in Jamaica. Journal of Ecology 83: 55–64. Dalling JW, Winter K, Nason JD, Hubbell SP, Murawski DA & Hamrick JL (2001) The unusual life history of Alseis blackiana: a shade-persistent pioneer tree? Ecology 82: 933–945. Dawkins HC (1958) The management of tropical high forest with special reference to Uganda. Imperial Forestry Institute Paper 34. Daws MI, Mullins CE, Burslem DFRP, Paton SR & Dalling JW (2002) Topographic position affects the water regime in a semideciduous tropical forest in Panama. Plant and Soil 238: 79–90. Denslow JS (1987) Tropical rainforest gaps and tree species diversity. Annual Review of Ecology and Systematics 18: 431–451.

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

34

T. R. Baker et al.

Denslow JS, Schultz JC & Vitousek PM (1990) Growth responses of tropical shrubs to treefall gap environments. Ecology 71: 165–179. Devall MS, Parresol BR & Wright SJ (1995) Dendroecological analysis of Cordia alliodora, Pseudobombax septenatum and Annona spraguei in central Panama. IAWA Journal 16: 411–424. Drechsel P & Zech W (1994) DRIS evaluation of teak (Tectona grandis L.f.) mineral nutrition and effects of nutrition and site quality on teak growth in West Africa. Forest Ecology and Management 70: 121–133. Enquist B, West GB, Charnov EL & Brown JH (1999) Allometric scaling of production and life-history variation in vascular plants. Nature 401: 907–911. Fairhead J & Leach M (1998) Reframing Deforestation: Global Analyses and Local Realities: Studies in West Africa. Routledge, London. Fetcher N, Haines BL, Cordero RA, Lodge DJ, Walker LR, Fernández DS & Lawrence WT (1996) Responses of tropical plants to nutrients and light on a landslide in Puerto Rico. Journal of Ecology 84: 331–341. Fisher BL, Howe HF & Wright SJ (1991) Survival and growth of Virola surianamensis yearlings: water augmentation in gap and understorey. Oecologia 86: 292–297. Gartlan JS, Newbery DM, Thomas DW & Waterman PG (1986) The influence of topography and soil phosphorus on the vegetation of Korup Forest Reserve, Cameroun. Vegetatio 65: 131–148. Gitay H & Noble IR (1997) What are functional types and how should we seek them? Plant Functional Types: their Relevance to Ecosystem Properties and Global Change. (eds. TM Smith, HH Shugart & FI Woodward), pp. 3–20. Cambridge University Press, Cambridge. Green JM & Newbery DM (2002) Drought avoidance and the effect of local topography on trees in the understorey of Bornean lowland rain forest. Plant Ecology 164: 1–18. Grime JP (1979) Plant Strategies and Vegetation Processes. Wiley, Chichester. Grime JP & Hunt R (1975) Relative growth-rate: its range and adaptive significance in a local flora. Journal of Ecology 63: 393–422. Grubb PJ (1996) Rainforest dynamics: the need for new paradigms. Tropical Rainforest Research – Current Issues (eds. DS Edwards, SC Choy & WE Booth), pp. 215–233. Kluwer Academic Publishers. Grubb PJ (1998) A reassessment of the strategies of plants which cope with shortages of resources. Perspectives in Plant Ecology, Evolution and Systematics 1: 3–31. Gunatilleke CVS, Perera GAD, Ashton PMS, Ashton PS & Gunatilleke IAUN (1996) Seedling growth of Shorea section Doona (Dipterocarpaceae) in soils from topographically different sites of Sinharaja rain forest in Sri Lanka. The Ecology of Tropical Forest Tree Seedlings (ed. MD Swaine), pp. 245–266. UNESCO, Paris. Hall JB & Swaine MD (1976) Classification and ecology of closed-canopy forest in Ghana. Journal of Ecology 64: 913–952. Hall JB & Swaine MD (1981) Distribution and Ecology of Vascular Pants in a Tropical Rain Forest. Dr W Junk Publishers, The Hague.

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

Harrington RA, Fownes JH, Meinzer FC & Scowcroft PG (1995) Forest growth along a rainfall gradient in Hawaii: Acacia koa stand structure, productivity, foliar nutrients, and water- and nutrient-use efficiencies. Oecologia 102: 276–283. Hawthorne WD (1995) Ecological profiles of Ghanaian forest trees. Tropical Forestry Papers 29: 1–345. Hawthorne WD (1996) Holes and sums of parts in Ghanaian forest: regeneration, scale and sustainable use. Proceedings of the Royal Society of Edinburgh 104: 75–176. Hazlett DL (1987) Seasonal cambial activity for Pentaclethra, Goelthalsia, and Carapa trees in a Costa Rican lowland forest. Biotropica 19: 357–360. Holdridge LR (1967) Life Zone Ecology. Tropical Science Center, San José. Huante P, Rincón E & Chapin FS (1995) Responses to phosphorus of contrasting successional tree-seedling species from the tropical deciduous forest of Mexico. Functional Ecology 9: 760–766. Huante P, Rincón E & Chapin FS (1998a) Effect of changing light availability on nutrient foraging in tropical deciduous tree-seedlings. Oikos 82: 449–458. Huante P, Rincón E & Chapin FS (1998b) Foraging for nutrients responses to changes in light and competition in tropical deciduous tree seedlings. Oecologia 117: 209–216. Jones EW (1956) Ecological studies on the rain forest of southern Nigeria IV. The plateau forest of the Okomu Forest Reserve Part II: The reproduction of the forest. Journal of Ecology 44: 83–117. King DA (1994) Influence of light level on the growth and morphology of saplings in a Panamanian forest. American Journal of Botany 81: 948–957. Korsgaard S (1986) An analysis of the potential timber production under conservation management in the tropical rain forest of South-east Asia. PhD thesis, Royal Danish Veterinary and Agricultural University, Copenhagen. Lehto T & Grace J (1994) Carbon balance of tropical tree seedlings: a comparison of two species. New Phytologist 127: 455–463. Lieberman D & Lieberman M (1987) Forest tree growth and dynamics at La Selva, Costa Rica (1969–1982). Journal of Tropical Ecology 3: 347–358. Longino JT (1986) A negative correlation between growth and rainfall in a tropical liana. Biotropica 18: 195–200. Lowe RG (1968) Periodicity of a tropical rain forest tree Triplochiton scleroxylon K. Schum. Commonwealth Forestry Review 47: 150–159. Lugo AE, Gonzalez-Liboy JA, Cintrón B & Dugger K (1978) Structure, productivity and transpiration of a subtropical dry forest in Puerto Rico. Biotropica 10: 278–291. Manokaran N & Kochummen KM (1987) Recruitment, growth and mortality of tree species in a lowland dipterocarp forest in Peninsular Malaysia. Journal of Tropical Ecology, 3: 315–330. Milton K, Laca EA & Demment MW (1994) Successional patterns of mortality and growth of large trees in a Panamanian lowland forest. Journal of Ecology 82: 79–87. Mirmanto E, Proctor J, Green J, Nagy L & Suriantata (1999) Effects of nitrogen and phosphorus fertilization in

Variation in tropical forest growth rates a lowland evergreen rainforest. Philosophical Transactions of the Royal Society of London (Series B) 354: 1825–1829. Mulkey SS, Wright SJ & Smith AP (1993) Comparative physiology and demography of three neotropical forest shrubs: alternative shade-adaptive character syndromes. Oecologia 96: 526–536. Newbery DM, Gartlan JS, McKey DB & Waterman PG (1986) The influence of drainage and soil phosphorus on the vegetaion of Douala-Edea Forest Reserve Cameroun. Vegetatio 65: 149–162. Parresol BR (1995) Basal area growth for 15 tropical tree species in Puerto Rico. Forest Ecology and Management 73: 211–219. Peace WHJ & Grubb PJ (1982) Interactions of light and mineral nutrient supply in the growth of Impatiens parviflora. New Phytologist 90: 127–150. Phillips OL, Hall P, Gentry AH, Sawyer SA & Vasquez R (1994) Dynamics and species richness of tropical rain forests. Proceedings of the National Academy of Sciences, USA 91: 2895–2809. Phillips OL, Baker TR, Arroyo L, Higuchi N, Killeen T, Laurance WF, Lewis SL, Lloyd J, Malhi Y, Monteagudo A, Neill D, Núñez Vargas P, Silva JNM, Terborgh J, Vásquez Martínez R, Alexiades M, Almeida S, Brown S, Chave J, Comiskey JA, Czimczik CI, Di Fiore A, Erwin T, Kuebler C, Laurance SG, Nascimento HEM, Palacios W, Patiño S, Pitman N, Olivier J, Quesada CA, Saldias M, Torres Lezama A & Vinceti B (in press) Pattern and process in Amazonian tree turnover, 1976–2001. Philosophical Transactions of the Royal Society of London, Series B. Potter CS, Davidson EA, Klooster SA, Nepstad DC, De Negreiros GH & Brooks V (1998) Regional application of an ecosystem production model for studies of biogeochemistry in Brazilian Amazonia. Global Change Biology 4: 315–333. Raaimakers D & Lambers H (1996) Response to phosphorus supply of tropical tree seedlings: a comparison between a pioneer species, Tapirira obtusa, and a climax species, Lecythis corrugata. New Phytologist 132: 97–102. Richards PW (1996) The Tropical Rain Forest. Cambridge University Press, Cambridge. Sack L & Grubb PJ (2001) Why do species of woody seedlings change rank in relative growth rate between low and high irradiance? Functional Ecology 15: 145–154. Sheil D & Burslem DFRP (2003) Disturbing hypotheses in tropical forests. Trends in Ecology and Evolution 18: 18–26. Silva JNM, de Carvalho JOP, Lopes JCA, de Almeida BF, Costa DHM, de Oliveira LC, Vanclay JK & Skovsgaard JP (1995) Growth and yield of a tropical rain forest in the Brazilian Amazon 13 years after logging. Forest Ecology and Management 71: 267–274. Strauss-Debenedetti S & Bazzaz FA (1996) Photosynthetic characteristics of tropical trees along successional gradients. Tropical Forest Plant Ecophysiology (ed. AP Smith), pp. 162–186. Chapman and Hall, New York. Swaine MD (1994) Long-term studies of tropical forest dynamics. Long-term Experiments in Agricultural and Eco-

35

logical Sciences (ed. AE Johnston), pp. 305–320. CAB International, Wallingford. Swaine MD (1996) Rainfall and soil fertility as factors limiting forest species distributions in Ghana. Journal of Ecology 84: 419–428. Swaine MD, Agyeman VK, Kyereh B, Orgle TK, Thompson J & Veenendaal EM (1997) Ecology of Forest Trees in Ghana. ODA Forestry Series 7. University of Aberdeen, Aberdeen. Swaine MD & Hall JB (1986) Forest structure and dynamics. Plant Ecology in West Africa (ed. GW Lawson), pp. 47–95. Wiley, Chichester. Swaine MD, Lieberman D & Hall JB (1990) Structure and dynamics of a tropical dry forest in Ghana. Vegetatio 88: 31–51. Swaine MD & Whitmore TC (1988) On the definition of ecological species groups in tropical rain forests. Vegetatio 75: 81–86. Tanner EVJ, Kapos V & Franco W (1992) Nitrogen and phosphorus fertilization effects on Venezuelan montane forest trunk growth and litterfall. Ecology 73: 78–86. Ter Steege H & Hammond DS (2001) Character convergence, diversity, and disturbance in tropical rain forest in Guyana. Ecology 82: 3197–3212. Ter Steege H, Sabatier D, Castellanos H, van Andel T, Duivenvoorden J, de Oliveira A, Ek R, Lilwah R, Maas P & Mori S (2000) An analysis of the floristic composition and diversity of Amazonian forests including those of the Guiana Shield. Journal of Tropical Ecology 16: 801–828. Thomas SC (1996) Asymptotic height as a predictor of growth and allometric characteristics in Malaysian rain forest trees. American Journal of Botany 83: 556–566. Thomas SC & Bazzaz FA (1999) Asymptotic height as a predictor of photosynthetic characteristics in Malaysian forest trees. Ecology 80: 1607–1622. Thompson WA, Kriedemann PE & Craig IE (1992) Photosynthetic response to light and nutrients in sun-tolerant and shade-tolerant rainforest trees. I. Growth, leaf anatomy and nutrient content. Australian Journal of Plant Physiology 19: 1–18. Tuomisto H, Ruokolainen K, Kalliola R, Linna A, Danjoy W & Rodriguez Z (1995) Dissecting Amazonian biodiversity. Science 269: 63–66. Turner IM (2001) The Ecology of Trees in the Tropical Rain Forest. Cambridge University Press, Cambridge. Vanclay JK (1994) Modelling Forest Growth and Yield. CAB International, Wallingford. Veenendaal EM, Swaine MD, Lecha RT, Walsh MF, Abebrese IK & Owusu-Afriyie K (1996a) Responses of West African forest tree seedlings to irradiance and soil fertility. Functional Ecology 10: 501–511. Veenendaal EM, Swaine MD, Blay D, Yelifari NB & Mullins CE (1996b) Seasonal and long-term soil water regime in West African tropical forest. Journal of Vegetation Science 7: 473–482. Veneklaas EJ & Poorter L (1998) Growth and carbon partitioning of tropical tree seedlings in contrasting light environments. Inherent Variation in Plant Growth: Physiological Mechanisms and Ecological Consequences (ed. MMI Van Vuuren), pp. 337–361. Backhuys Publishers, Leiden.

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

36

T. R. Baker et al.

Vitousek PM, Walker LR, Whiteaker LD & Matson PA (1993) Nutrient limitations to plant growth during primary succession in Hawaii Volcanoes National Park. Biogeochemistry 23: 197–215. Walker LR, Zimmerman JK, Lodge DJ & Guzmán-Grajales S (1996) An altitudinal comparison of growth and species composition in hurricane-damaged forests in Puerto Rico. Journal of Ecology 84: 877–889. Walter H (1979) Vegetation of the Earth and Ecological Systems of the Geo-Biosphere. Springer, New York. Webb EL (1999) Growth ecology of Carapa nicaraguensis Aublet. (Meliaceae): implications for natural forest management. Biotropica 31: 102–110. Welden CW, Hewett SW, Hubbell SP & Foster RB (1991) Sapling survival, growth, and recruitment: relationship to canopy height in a neotropical forest. Ecology 72: 35–50. Whigham DF, Zugasty Towle P, Cabrera Cano E, O’Neill J & Ley E (1990) The effect of annual variation in precipitation on growth and litter production in a tropical dry forest in the Yucatan of Mexico. Tropical Ecology 31: 23–34. White F (1983) The Vegetation of Africa: A Descriptive Memoir to Accompany the UNESCO/ AETFAT/ UNSO Vegetation Map of Africa. UNESCO, Paris. Whitmore TC (1974) Change with Time and the Role of Cyclones in Tropical Rain Forest on Kolombangara, Solomon Islands. Commonwealth Forestry Institute Paper 46, Oxford.

Perspectives in Plant Ecology, Evolution and Systematics (2003) 6, 21–36

Whitmore TC (1975) Tropical Rain Forests of the Far East. First Edition. Clarendon Press, Oxford. Whitmore TC (1984) Tropical Rain Forests of the Far East. Second Edition. Clarendon Press, Oxford. Whitmore TC (1989) Canopy gaps and the two major groups of forest trees. Ecology 70: 536–538. Whitmore TC (1998) An Introduction to Tropical Rain Forests. Oxford University Press, Oxford. Woodward FI (1987) Climate and Plant Distribution. Cambridge University Press, Cambridge. Worbes M (1999) Annual growth rings, rainfall-dependent growth and long-term growth patterns of tropical trees form the Caparo Forest Reserve in Venezuela. Journal of Ecology 87: 391–403. Wright SJ (1991) Seasonal drought and the phenology of understorey shrubs in a tropical moist forest. Ecology 72: 1643–1657. Wright SJ & Cornejo FH (1990) Seasonal drought and leaf fall in a tropical forest. Ecology 71: 1165–1175. Wright SJ & van Schaik CP (1994) Light and the phenology of tropical trees. The American Naturalist 143: 192–199. Yoda K (1978) Three-dimensional distribution of light intensity in a tropical rain forest of West Malaysia. Malaysian Nature Journal 30: 161–177. Zech W & Drechsel P (1991) Relationships between growth, mineral nutrition and site factors of teak (Tectona grandis) plantations in the rainforest zone of Liberia. Forest Ecology and Management 41: 221–235.