Vegetation

Baksh-Comeau, Yasmin S., Maharaj, Shobha S., Adams, Dennis, Harris, Stephen A., Filer, Denis L. & Hawthorne, William D., 2016, An annotated checklist of the vascular plants of Trinidad and Tobago with analysis of vegetation types and botanical ‘ hotspots’, Phytotaxa 250 (1), pp. 448-450 : 448-450

Vegetation and forest classifications

The extensive botanical explorations undertaken on the islands as described above laid the foundation for vegetation and forest classification in Trinidad and Tobago. Starting with Crüger, his Outline of the flora of Trinidad (1858) was the first description of the vegetation of Trinidad, alongside lists of the families and genera he observed. It was the sole source of information on the vegetation of Trinidad and Tobago for almost 80 years until R.C. Marshall, Conservator of Forests, laid the foundations for a sustainable timber industry in Trinidad and Tobago ( Marshall 1925, 1926). He classified forest based on 800 km of transects and distinguished classes based on habitat (20 years of rainfall data, topography and soil), structure and floristic composition and named them according to dominant timber species ( Marshall 1934b). Marshall noted the presence of ecotones, or transitional belts, between different forest types when distinguishing between zonation and succession in these forests. His work was superseded by Beard’s publication, The natural vegetation of Trinidad ( Beard 1946), designed to form the basis of a land use policy for the island in 1946.

Combining a 14-year enumeration survey of the forests (1927–1941), Marshall’s earlier classification, and aerial photographs of the colony taken in 1938, Beard produced 1:50000 vegetation maps. Beard’s vegetation classification was based on trees> 10 cm DBH (diameter at breast height) and emphasised the dominance of selected valuable species. Beard postulated that for the most part, ‘moisture relations’, then topography and soils, determined forest structure and distribution. Emphasis was placed on the idea of forest formation as a stable entity related to climate. Beard defined a ‘formation’ as a set of floristic associations united by common structure and life-form, and produced a key distinguishing the main formations based on life-form, phenology, number of strata, and tree root and buttress features. He defined ‘association’ as the largest possible group that had consistent dominants of either the same or closely allied species. With the collection of physiognomy data and forest profile diagrams, Beard proposed four climatic and two edaphic (swamps and marshes) formations within which he identified a total of 21 associations. A simplified version, highlighting the formations, is mapped in Figure 2 .

More recently, Nelson (2004) claimed Beard’s vegetation classes had not proved useful for predicting or summarising the distribution of plant species. He proposed a classification of Trinidad and Tobago forests based on a different hierarchy, from eco-region down to landscape level, following Holdridge’s model (1947) of how vegetation is controlled by climate. Nelson (2004) criticised Beard’s system because: it was based on forest and only on large trees; it sought to classify forest in terms of climax conditions; and the original survey data had been lost, making it difficult to evaluate subsequent changes based on new records. Nevertheless, Beard’s associations together with Nelson’s own tree plot data ( Fig. 2 ) were still used as the basis for his modelling of forest community composition.

In spite of reservations such as these, Beard’s system of vegetation classification still remains in use for forest management. Before the current study, the Forestry Division was still mostly focused on managing larger trees. A large part of its research effort was a network of 174 Permanent Sample Plots (PSPs) for monitoring dynamics of trees, in a limited selection of these forest types in Trinidad, but not in Tobago ( Fig. 2 ). The species data from these plots have been tidied up, by the former Director , Forest Research and Inventory Management ( FRIM), Forestry Division , Seepersad Ramnarine , to harmonise nomenclature and these distribution records are included in the checklist. Forest Officers’ involvement in the field survey, with training in the RBS methods, was intended to help the Forestry Division shift from a forest extraction paradigm to one of forest conservation .

The Forestry Division in 1980 attempted to update Beard’s vegetation map as a 1:150000 map ( Thelen & Faizool 1980). It highlighted areas deforested for plantations and farming.

The loss of natural vegetation as indicated in these maps, including conversion of reserved forest areas to exotic plantations, highlighted the need for greater attention to conservation in the management of forest in Trinidad and Tobago. We emphasise the need for more conservation-related floristic assessment, which is at the heart of this checklist.

The first attempt to classify the flora of Trinidad and Tobago into IUCN categories was based solely on a review of the literature and TRIN specimens by Adams and Baksh (1981 –1982). They recommended intensive field study to produce a more accurate assessment of population sizes and distributions, particularly of endangered species. Subsequently, Baksh-Comeau (1999) summarised the ferns and fern allies of Trinidad and Tobago into IUCN categories using the results from a review of the literature, TRIN specimens and field sampling. This present study expands these first attempts, and aims to assess the conservation status of all vascular flora of Trinidad and Tobago, by applying IUCN categories and the RBS Star rating system ( Hawthorne 2012, IUCN 2001).

Also included in the checklist are data from other studies that have focused on specific vegetation types of Trinidad and Tobago :

• Chacachacare Island and Five Islands ( Adams & Baksh-Comeau 2005, Chaboo 1989 –1990, Darlington

1967);

• Bush Bush forest and the Nariva Swamp ( Ramcharan 1980, Aitken 1973);

8 • Phytotaxa 250 (1) © 2016 Magnolia Press

BAKSH-COMEAU ET AL.

• • • •

Savannas of Trinidad ( Comeau 1989 –1990, Schwab 1988);

Vegetation surrounding mud volcanoes in southern Trinidad ( Comeau 1993 –1994); Dry forest of Little Tobago ( Oatham & Boodram 2006);

Riparian vegetation of Trinidad ( Boodram 2009).

The records of species in these publications have been vouchered in TRIN and the distributional data are included in the checklists, with a few extra unvouchered species records added if mentioned only in the publications themselves. By contrast, distributional records from the baseline survey of the Matura National Park

CHECKLIST TO VASCULAR PLANTS OF TRINIDAD AND TOBAGO

Phytotaxa 250 (1) © 2016 Magnolia Press • 9 Environmentally Sensitive Area, where all trees and lianas ≥ 10 cm DBH were sampled within 24 quarter hectare PSPs (van den Eynden et al. 2007), are cited in the checklist even though many of the specimens were unavailable for verification; they had been checked by van den Eynden et al. with reference to TRIN material, and extended the ranges of many species to the previously under-sampled ‘Matura-Salybia’ Minor area. These plots are mapped in Figure 2 .

Understanding plant community variation is an important component of biodiversity analysis to complement the conservation assessments summarised below. It was not an aim of this study to reappraise the classification or remap comprehensively the vegetation of Trinidad and Tobago. However, the advantage of RBS data is that they can often reveal patterns in forest composition that correspond to the communities and associations based on indicator species and forest physiognomy, such as those of Beard (1946). Unlike earlier vegetation classifications, the RBS also provides information on non-tree species, so for the first time these can be referenced to vegetation classes in an objective way. The resulting vegetation classification is used in this checklist to define the ecological associations of species.

Figures

FIGURE 2. Map of Trinidad and Tobago showing location of sample plots from which species distributions are included in the checklist; and Beard’s (1946) main forest formations (with his numbered associations merged under the same colour).