Home | Texas A&M College of Veterinary Medicine ...



INTRODUCTION

Palms and the Tropical Landscape

Tropical forests are biodiversity hotspots and contain the most diverse plant communities on the planet (Givinish 1999). Neotropical rainforests are the most extensive of all global tropical forests; around half of the global total, 4 x 106 km2 in area, and one-sixth of the total broad-leaf forest of the world (Whitmore 1998). Tropical forest habitats can support a host of different tree species even in small areas; up to 283 species per hectare (Phillips et al. 1994).

Palms are among the dominant vascular plants in many tropical forests, and are important components of the forest structure in Amazonian forests (Scariot 1999). Palms (Arecacea family) belong to the class of monocotyledons; they form a distinctive crown of feathery or fan-shaped leaves which are always alternate, often very large and pinnately incised to a varying degree (Bruggeman 1962). There number of palm species is estimated at 2500 – 3500 in approximately 210 – 236 genera (Jones 1995). Montufar & Pintaud (2006) have estimated the number of native palms in Western Amazonia to be about 121 species and 33 genera (roughly two-thirds of the Amazon palms and one-fifth of the New World palms).

The richness and abundance of palm taxa, as well as their occurrence in all strata of the forest, and their importance as a vital food source for wildlife make this family a primary target for conservation (Scariot 1999). The sedentary habitat and patchy distributions of plants make them susceptible to habitat destruction, which may cause changes in taxa composition and population sizes (Schemske et al. 1994). And despite the importance of plants, particularly palms, in the forest structure and function, their response to habitat fragmentation has been studied less than animals (Laurence & Bierregaard 1997).

There are 165 individual species registered on the 2006 IUCN Red list, 9 of those are critically endangered, 14 individuals are endangered and 67 species are registered as vulnerable. Almost a third of the species on the list are data deficient (45 individuals), where there is inadequate information to make an assessment on the extinction risk of an individual species. Only one species, Amorphophallus preussii (endemic to Ecuador), has a recorded population trend; which is declining (Darbyshire 2004), all other individuals have not had their population trends studied. The major threat for all of the individuals is ongoing human induced habitat loss/degradation (ICUN 2006), whilst other major threats include large-scale wood plantations, livestock and crop agriculture, fires, infrastructure growth and in rare occasions; natural disasters such as volcano eruptions (Benavides & Pitman 2003).

Factors Influencing Species Abundance

Land Use in Tropical Habitats

Much of the Amazon Basin is a mosaic of forest types produced by erosion/deposition cycles of the major rivers (Clarke et al. 1995) and variation in terre firme parent material and geochemistry (Guillaumet 1987). Another source of small to medium scale spatial heterogeneity within tropical forests is the historical or current impact of human activity, from agriculture, silviculture and selective harvesting, even within stands considered to be old growth (Clark et al. 1995; Bush & Colinvaux 1994).

Agriculture is the biggest land use of tropical forests and serves the main purpose for which rain forests are cleared (Whitmore 1998). Conversion of tropical rainforests to pasture for cattle husbandry is particularly widespread in the neotropics. Shifting agriculture, commonly known as ‘swidden’, is a sustainable low-input form of cultivation which can continue indefinitely on the infertile soils underlying most tropical rainforests, provided the carrying capacity of the land is not exceeded. Dufour (1990), claims that under some circumstances, shifting agriculture based on long fallow periods can be an ecologically and an economically sustainable practice in tropical forests. Not all shifting agriculture is practised in a sustainable manner, in some circumstances farmers fell and burn the forest and grow crops on the released nutrients for several years in succession, continuing until coppicing potential and the soil seed bank are exhausted, invasive species take hold and soil nutrients are depleted. They then move onto a new patch of virgin forest. In western Amazonia, peasants from the Andean plateau are moving into the forest with no previous experience of forest agriculture, after growing a few crops they sell-out to a pastoralist who raises cattle, creating a wave of cultivation and poor pasture sweeping western from the foothills of the Andes (Whitmore 1998).

On a more commercial scale, tropical forests are frequently used as a source of timber and non-timber products, and are an important and widespread forest land use in much of the tropics (Collins et al 1991). Clear felling results in greater levels of habitat loss and degradation, and it is believed that many palms do not regenerate in open areas and are therefore threatened in areas with extensive deforestation (Pedersen, 1994; Moraes et al., 1995), whereas selective logging; the periodical extraction of commercially valuable trees from forests (Johns 1988), has a much less detrimental effect on the forest ecosystem (Grieser Johns 1997; Asner et al 2004).

The rate at which habitats are disturbed influences the species assemblages and diversity within a plant community. At very low rates of disturbance, such as infrequent clear-felling, competitive exclusion would occur and species richness would be low, species richness would be greatest at moderate levels of disturbance, because dominance is prevented, and the pool of potential colonists is relatively large (Armesto & Pickett 1985).

Habitat Associations

The distribution of individuals within a population of plants is rarely random across a landscape (Harms et al. 2001). Tropical trees and shrubs often display distributional biases with respect to environmental variables such as soil type and water availability, across spatial scales of several ha to many km2.

The association of species with physical habitat variables generates some of the most obvious patterns in the distribution and abundance of organisms, and its study has a long history (Cowles 1899; Whittaker 1956). On a smaller scale, microhabitat heterogeneity (referring to environmental conditions that vary at scales less than 10m2, e.g. treefall gaps or local topographic variations (Svenning 1999)), has been suggested to play an extensive role in determining species distributions (Grubb 1977; Ricklefs 1977). Whereas several authors (Molofsky & Augspurger 1992; Nicotra et al. 1999) have shown heterogeneity in environmental conditions that clearly influence seed and seedling survival, such as light, litter and soil moisture, have been shown to occur at spatial scales of less than 1m. Webb & Peart (2000) suggest that the most influential hypothesis for habitat partitioning in rain forest trees relies on not on any underlying heterogeneity in the physical environment, but on the endogenous, local heterogeneity created by canopy openings. A tropical landscape scattered with small indigenous communities practising traditional agricultural methods would create a mosaic of habitat types, and increasing heterogeneity on a landscape and microhabitat scale.

Only one group of tropical forest trees, the pioneer or early-successional species, has been clearly and consistently associated with a specific habitat type; disturbed areas such as large canopy openings (Whitmore 1989). Among non-pioneers species, little evidence has been compiled to link any differences in individual-level response to environmental conditions to differences in distribution along environmental gradients (Baraloto & Goldberg 2004). This shows a critical gap in current knowledge, as non-pioneer species form a large part of the tropical forest ecosystem, and in the case of many species such as palms, are fundamental keystone species, components for the present and future of the tropical forests globally. Consequently, it is important to identify any strong habitat associations resulting from human influenced habitat loss/degradation and their effects on the abundance and diversity of palms.

Niche Width & Occupation

Several authors (Grubb 1977; Denslow 1987) believe that the exceptional diversity within a tropical forest is maintained through niche differentiation with respect to resources. The niche is a multidimensional description of a species’ resource needs, habitat requirements and environmental tolerances (Hutchinson 1957). Where the niche structure of complex plant communities has been investigated in some detail, niche differences have been found in germination behaviour, root depth, temperature thresholds, grazing tolerance, phenology and many other factors (Crawley 1997).

The niche width of a species refers to the area which a species could physically inhabit; the niche width often differs from the area that a species actually inhabits, or its realized niche width. The niche concept has rarely been used in plant ecology due to the difficulty in defining a species’ niche breadth, in zoological terms it is fairly straightforward to define an individual’s niche in relation to diet and range size, yet it is much more difficult to apportion resources such as light, water or nitrogen between species. Within plant communities, a species with a ‘broad niche’ will be found growing abundantly over a wide range of niche conditions, and are known as generalists, and species with a ‘narrow niche’ , a specialist, will only be found under a restricted range of conditions. Specialist species with a narrow realized niche width will eventually become extinct from the landscape due to competition from more abundant species with a broader niche width. Crawley (1997) suggests that in order to maintain diversity, species-rich plant communities may be composed of: (i) species with narrower niches; (ii) species with more broadly overlapping niches; (iii) habitats providing ‘longer’ niche axes; or (iv) a combination of these.

The ability for species to coexistence rather than out-compete in plant communities is driven by niche specialisation along gradients or resource availability (Grubb 1977; Crawley 1990). The niche diversification hypothesis created by Connell (1978) is an important equilibrium hypothesis stating that species coexist by occupying different niches. There are many authors that discredit such as simple hypothesis, and a much more complex version proposes that coexistence is a result of habitat or microhabitat specialisation (Denslow 1987; Gentry 1988; Welden et al. 1991; Clark et al. 1993) and that much of the tropical plant diversity therefore depends on habitat and microhabitat heterogeneity.

Aims & Objectives

The aim of this study is to investigate possible species associations between palms and balsas across a range of habitat types, displaying different degrees of anthropogenic alteration and supporting a range of microhabitat variables. Specifically the questions addressed are (i) What is the palm density and abundance in the study area? (ii) Are there any habitat associations between palm species in natural and anthropogenically altered tropical forest? (iii) What are the species associations between balsa trees and palm species? The objectives derived from these aims are (i) Obtain the number of palms and balsas present in the study area and calculate species density and abundance through plot sampling. (ii) Identify differences in habitat type and land use and record microhabitat variables across the study site. (iii) Identify differences in species abundance and assemblages between natural and anthropogenically altered habitats.

METHODS

Study Site

The study took place during July and August 2007, conducted in the native community of La Torre, within the buffer zone of the Tambopata Nature Reserve, 27km southwest of Puerto Maldonado in south-eastern Peru (12°49’S, 69°17’W). The area is at the moist tropical/subtropical vegetation boundary (Brightsmith 2004). Rainfall is on average 2810 mm per year (Pearson & Derr 1986). The dry season starts in April and ends in October (Pearson & Derr 1986). The surrounding area is made up of a mix of floodplain and terra firme forests (Brightsmith 2004). There are no large-scale deforestation or agricultural areas in the La Torre community, although along the river edge and close to tourist and community residence there are small cleared sections for subsistence agriculture (less than 3 hectares) and homegardens (R. Harris, pers. Obs.). The dominant habitat is secondary floodplain forest, with fragments of understorey Guadua bamboo (Lloyd 2004) and remnants of primary forest. A seasonal palm swamp (12°42’S, 69°20’W), which has not been subjected to anthropogenic disturbances or land use change, located approximately 5km downstream of the La Torre community was also sampled.

[pic]

Figure 1. Location of study site within South-eastern Peru.

[pic]

Figure 2. Location of transects within study site. 1= 1250m, 2 = 1750m, 3 = 600m, 4 = 250m, 5 = 250m, 6 = 700m (Palm swamp), 7 = 1500m. Total length surveyed = 6.3km

Habitat Sampling

A total of 136 survey plots were sampled laid out along transects following existing pathways, Figure 2 shows the location and length of the transects. A 100m2 plot was sampled every 50m along the transect, a pilot study using 25m intervals between plots was initially trailed but the method was deemed too time-consuming for the limited study period. Plots were sampled at alternate sides of the transect to include more habitat types and ensure a more random sampling strategy. In total an area of 1.36ha was surveyed. Each plot was categorized at the time of the survey according to habitat type or past/present land use. The habitat categories were:

Regenerated secondary forest with no visible anthropogenic alteration (SEC). Regrowth of secondary forest after large scale disturbances both natural and anthropogenic develops rapidly in tropical habitats, but species richness often accumulates quite slowly. Even regrowth over a century old does not contain all the species present in primary forest. (Corlett & Turner 1997).

Garden, either abandoned or receiving lowest level of management such as occasional fruit harvesting (GARD). Home gardens usually contain small numbers of useful plant species grown such as trees for fruit and firewood, specialised vegetables and medicinal plants. Small populations of animals such as poultry and pigs are sometimes raised in homegardens (Kellman & Tackaberry 1997).

Residential (RES); either tourist residence or member of the La Torre Community, including the land immediately surrounding the buildings of a distance of less than 10m and not including land managed for agriculture purposes (managed land within the vicinity of residential property classed as home gardens).

Agricultural land (AGRI); land recently (within last 20 years) >1ha used for the growth of crop species cleared completely of the original forest species present and may or may not of been subjected to burning. Agricultural land includes small-scale shifting agriculture and larger monoculture plantations.

Dominant bamboo understorey within secondary forest (BAM). Bamboo grows in monotypic stands, which is unusual in tropical plant communities. The structure of bamboo stands differs from other habitats in that the relatively thin-stemmed plants are densely packed and have a thick subcanopy of similarly shaped leaves (Kratter 1997).

Periodic palm swamp (SWA). Soils tend to be richer in plant nutrients due to seasonal flooding (Whitmore 1998). Yet flooded forests are usually less floristically diverse than non-flooded forests and contain more specialised species (Kellman & Tackaberry 1997).

Habitat measurements were taken at each plot, the canopy openness was taken from the centre of the plot with an acetate sheet, the sheet had a 10cm x 10cm grid split into 100 cm squares, the number of squares which the sky was visible through when held directly above at arms length, equalled the percentage of canopy openness.

The five tallest trees within the survey plot were selected and their heights estimated by sight alone, the estimation of tree heights was based on previous knowledge of vegetation sampling in the same habitat, using an automatic range -finder and clinometer to record tree heights. The tree architecture of the five tallest trees was taken to give an indication of the recent history of the forest (Torquebiau 1986; Jones et al. 1995). The categories are:

BA – ‘branching above’, if the first or major branch was above half of the tree height, indicative of tree growth under a closed canopy (undisturbed primary forest) or;

BB – if it was ‘branching below’ half of the tree height, suggesting growth under an open canopy due to disturbance or treefall

BASB – ‘branching above scars below’, where the first or major branch is above half height of the tree but there is evidence of dropped branches below half the tree height, caused by tree growth under a regenerating, closing canopy

BBVG – ‘branching below vegetated growth’, the branches are growing vertically from close to its base, suggesting tree growth in a heavily altered habitat with frequent tree-cutting.

Palm and Balsa sampling

Previous field work within the same study region and a pilot study revealed that Balsa (Ochroma pyramidale) was present along with five palm (Araceae) species.

[pic]

Figure 3. Ochroma Pyramidale

[pic]

Figure 4. The stilt roots of a Socratea exorrhiza palm.

[pic]

Figure 5. Attalea butyracea

[pic]

Figure 6. The underside and trunk of Astrocaryum murumuru

[pic]

Figure 7. Euterpe edulis

[pic]

Figure 8. Root structure of Iriartea deltoidea

The number of Ochroma pyramidale individuals with a diameter at breast height (d.b.h) ≥300mm were recorded, whereas with the palm species, the diameter of the palm was recorded at 500mm above ground level rather than the d.b.h, this would include the diameter of the root stilts in Iriartea and Socratea rather than the main stem, as the diameter of stilt roots can be used as an indicator to the age of an individual (A. Lee, pers. Comm.). The palm species were identified to species level and all palm and Ochroma species surveyed with a diameter ≥ 300mm had their heights recorded and whether the tree was fruiting, flowering or neither.

RESULTS

Species Abundance

There were 195 individuals recorded over all 7 transects at an overall abundance of 144 individuals per ha, 171 of those were palm species, 24 were Ochroma pyramidale. The most abundant species was Iriartea deltoidea, recorded at an abundance of 53 individuals per ha (Table 1 shows all abundances), the least abundant species was Socratea exorrhiza, recorded at 4.4 individuals per ha, Euterpe edulis also had a very low abundance, 6.6 individuals per ha. Transect 1 was the only transect to have all species present, whereas in transect 4 and 5, only one species was recorded, both these transects had the lowest abundance, 33.3 individuals per ha for both. Transect 6, located in the palm swamp, had the highest abundance of palms recorded, 214.3 per ha, but no Ochroma pyramidale individuals. No species was recorded on every transect, although Attalea butyracea occurred most frequently, on 6 out of the 7 transects, Socratea exorrhiza and Euterpe edulis occurred the least frequently, on only 3 out of the 7 transects.

Table 2 shows the abundance of individuals based on habitat type at each plot. Secondary forest was the only habitat with all species present, agricultural habitats supported the lowest number of species, only Ochroma pyramidale and Attalea butyracea were present there. Agricultural habitats also had the lowest abundance, 20 individuals per ha, the highest abundance within a habitat was within the palm swamp, 214 individuals per ha. The palm swamp also had the highest abundances of 3 palm species over all habitats; Iriartea deltoidea, Euterpe edulis & Socratea exorrhiza (129, 36 & 29 individuals per ha respectively). Garden habitats had the lowest abundance for palm species (119 individuals per ha), but the highest abundance for Ochroma pyramidale (81 individuals per ha).

Table 1. Abundance of individuals by transect. Key to habitats; SEC – secondary forest, GARD – garden, RES – residential including homegardens, AGRI – agricultural land, BAM – dominant bamboo understorey, SWA – seasonal palm swamp

| | |Transect |  |  |  |  |

|  |

|Ochroma pyramidale |9.7 (3) |50 (18) |16.7 (2) |0 |0 |0 |

|Number of plots |86 |16 |6 |5 |8 |14 |

|Area surveyed (Ha) |0.86 |0.16 |0.06 |0.05 |0.08 |0.14 |

| | | | | | | |

|Abundance (individuals per ha plus number recorded in parentheses) |

|Ochroma pyramidale |5.8 (5) |81.2 (13) |50 (3) |20 (1) |25 (2) |0 |

|Astrocaryum murumuru |26.7 (23) |12.5 (2) |16.6 (1) |0 |12.5 (1) |0 |

|Iriartea deltoidea |52.3 (45) |0 |16.6 (1) |0 |62.5 (5) |128.5 (18) |

|Attalea butyracea |27.9 (24) |6.2 (1) |0 |20 (1) |12.5 (1) |21 (3) |

|Euterpe edulis |3.4 (3) |0 |0 |0 |0 |35.7 (5) |

|Socratea exorrhiza |2.3 (2) |0 |0 |0 |0 |28.5 (4) |

| | | | | | | |

|All palm species |112.7 (97) |18.7 (3) |33.3 (2) |20 (1) |87.5 (7) |214.2 (30) |

A non-parametric test was carried out to investigate differences in species abundances between habitats. Ochroma pyramidale & Iriartea deltoidea were significant to the P 0.05 |P > 0.05 |P > 0.05 |

|Iriartea deltoidea | | | |+ + 2.9% |+ + 2.2% |

| | | | |+ - 28.6% |+ - 29.4% |

|  |  |  |  |P > 0.05 |P > 0.05 |

|Euterpe edulis | | | | |+ + 0% |

| | | | | |+ - 8.1% |

|  |  |  |  |  |P > 0.05 |

There were only two significant correlations between species (Table 6), Ochroma pyramidale and Iriartea deltoidea not once occurred in the same plot (X2 = 6.74, P < 0.05), and Euterpe edulis and Attalea butyracea only occurred in the same plot on 2.9% of the time (X2 = 9.17, P ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download