This report describes a field visit to the Caribbean island of Bonaire aiming at collection of field data on the arid vegetation, which consists largely of dry thorn scrub, cactus scrub and dry tropical forest. Vegetation releves have been amde at 23 sites, among which eight sites that have been surveyed in 1999. Besides 50 short field descriptions have been made as a basis for a land-use map. The report describes how the more than 1000 releves of the Dutch Caribbean islands, which are stored in the vegetation database CACTUS, may be analysed to provide insight in natural processes like vegetation succession and effects of land use, climate change and other factors. Besides, the report contains an advice for the representative Ministry of Agriculture, Nature and Food Quality on the island about the (better) conservation of sites on the limestone terraces with very well developed woodland.
Vegetation map of St.Maarten. Source unknown.
Caribbean dry forests are among the most endangered tropical ecosystems on earth. Several studies exist on their floristic composition and their recovery after natural or man-made disturbances, but little is known on the small Dutch Caribbean islands. In this study, we present quantitative data on plant species richness and abundance on St. Eustatius, one of the smallest islands of the Lesser Antilles. We collected and identified trees, shrubs, lianas and herbs in 11 plots of 25 x 25 m in different vegetation types. We compared their floristic composition and structure to vegetation surveys from roughly the same locations in the 1990s and 1950s. We found substantial differences among our 11 plots: vegetation types varied from evergreen forests to deciduous shrubland and open woodland. The number of tree species ≥10≥10 cm DBH ranged between one and 17, and their density between three and 82 per plot. In spite that all plots were subject to grazing by free roaming cattle, canopy height and floristic diversity have increased in the last decades. Invasive species are present in the open vegetation types, but not under (partly) closed canopy. Comparison with the earlier surveys showed that the decline of agriculture and conservation efforts resulted in the regeneration of dry forests between the 1950s and 2015. This process has also been reported from nearby islands and offers good opportunities for the future conservation of Caribbean dry forests.
A semi-detailed landscape-based vegetation map (scale: 1: 37,500) is presented for the 13 km2 Lesser Antillean steep volcanic island of Saba, Netherlands Caribbean. The map is based on a total of 49 vegetation plots that were sampled in 1999 using a stratified random sampling design and analysed using TWINSPAN cluster analysis. Three hundred and fourteen (314) plant species, representing 56% of the total known flora (565 species), were recorded in the sample plots. The principal lower sections of the island possess a tropical savannah climate whereas the upper slopes reaching a maximum altitude of 870 m can best be characterized as a tropical rainforest climate.
A total of two main and nine different sub-landscape types were distinguished based on geology, geomorphology and nine distinguished vegetation types. In Saba, sharp contrasts in soil, geomorphology and climatic factors are found on a small spatial scale and this meant that compared to the other islands of the Dutch Caribbean there is little mixing and merging of vegetation types at the landscape vegetation level. Consequently, vegetation type translates relatively directly into landscape vegetation units. Aside from important contrasts in vegetation that correspond to what is known about differences in soil and climate, our study also shows that large vegetation changes have taken place on the island since the survey by STOFFERS, five decades earlier. These largely appear to be due to three major forces: a) hurricane impacts; b) natural succession made possible due to diminished agricultural activity and; c) invasive plants and plant pest species.
The most recent hurricane, hurricane Georges, which struck the island one year before this study, clearly caused much damage to the vegetation, especially high on Mount Scenery. As a consequence, the elfin woodland vegetation has virtually disappeared, while remnant sections have been radically altered. Based on studies elsewhere in the region, the elfin woodland can be expected to take very long (if at all) to gradually recover. The impact of various hurricanes in the last 60 years has clearly caused major disturbance of the vegetation throwing it back into earlier stages of succession. The development of the “Tree fern brake” into “Pioneer forest” vegetation must be seen as a positive change where a secondary community had entered a higher stage in the sequence of succession. The virtual disappearance of the formerly prominent secondary shrub communities like Miconia thickets, Piper dilatatum thickets and Leucaena thickets can also be seen as likely evidence of natural successional forces thanks to diminished agriculture and woodcutting. Invasive species was the third major force of change that clearly appears to have been active on Saba in recent decades. The lasting impacts of insect invaders which have decimated formerly prominent Opuntia (cactus) and Tabebuia (tree) populations testify to the impact of invasive species as a major driver of recent vegetation changes on Saba.
Our field data show that most wilderness areas of Saba remain strongly affected by roaming grazing goats even though the contribution of goats to the local island economy is negligible. Goat dung or traces of grazing were recorded in or adjacent to 46% of the sample plots. Grazing by exotic mammals reduces the resilience of natural vegetation types and interferes with natural succession. Highest livestock densities and impacts seem to be in the more vulnerable coastal arid zones along the western and southern sections of the island with poor soil conditions and more open and shrubby vegetation. The development of ‘Dry evergreen woodland’ under similar conditions on the more remote, windy and salt spray-affected, but less-grazed, northern sectors of the island, suggest that those disturbed areas of the southern and western coastal zones should have potential for woodland recovery if and when goat grazing is reduced. Therefore, a key priority for terrestrial conservation in Saba should be to reduce feral grazer densities to allow vegetation recovery and reduce vulnerability to erosion. We suggest the use of pilot demonstration projects for grazer exclusion as a useful way to help build stronger arguments and public support for tackling the roaming goat problem in Saba.
A semi-detailed landscape-based vegetation map (scale: 1: 37,500) based on field data from 1999 has been available as an update of Stoffers’ 1956 map of the Lesser Antillean island of St. Eustatius, Netherlands Caribbean, but up to now was never finalized or published. In this report we complete the documentation of that map to provide new insights into vegetation change over a period of more than 40 years, and a quantitative reference point for future studies on landscape-level vegetation development for the island.
The principal lower sections of the 21 km2 island of St. Eustatius possess a tropical savannah climate according to the Köppen (1931) classification system, and the documented flora of the island amounts to 505 species. Color aerial photographs (1: 8,000) taken in 1991 and field data from 1999 were used to produce the map. A total of 84 vegetation sample plots were analysed using a stratified random sampling design and TWINSPAN cluster analysis.
Four main and 16 sub-landscape types were distinguished based on geology, geomorphology and different mixes and expressions of the component vegetation types. The five principal landscape types are in descending order of importance: H1, H2, M4, M9 and C, and covered some 67 % of the seminatural habitat of the island. H1 and H2 are the Pisonia-Justicia hills and Pisonia-Bothriochloa hills and are limited to The Mountains area. Analysis of the sampling data resulted in the distinction of 13 (semi)natural vegetation types. The three principal vegetation types were, the Pisonia-Justicia type, Pisonia-Ayenia type and Bothriochloa-Bouteloua type which together accounted for 38 % of total (semi)natural vegetation cover. The following well-developed vegetation types of St. Eustatius represent primary climax communities: Types 1, 2, 3, 5 and 7, all found in and around the Quill in the southwestern part of the island. A comparison of the vegetation types in the present study with those of Stoffers (1956) showed that only one vegetation type closely resembles one in Stoffers’ study. Major changes have taken place in certain types of the natural vegetation of the island in the intervening five decades.
The majority of the central sections of the island around Oranjestad the so-named “Cultuurvlakte”, amounting to approximately 25 % of the surface of the island, have suffered intensive disturbance due to past agriculture, livestock husbandry and invasive species and were not mapped. Only a small remnant portion of the semi-natural lowlands vegetation (L1 and L2) was left in the coastal reaches of Billy’s Gut. Nevertheless, this area is heavily affected by grazing and the invasive vine Antigonon leptopus.
A comparison with the 1950s vegetation map by Stoffers shows that the rarest and most valuable elfin woodland vegetation of the rim of the Quill crater had been largely lost and that the areas he described as “Montane thickets” (Type 2) had declined and been degraded. We speculate that these losses may be most directly attributable to the impact of recent hurricanes and/or grazing by introduced livestock. On the lower slopes of the Quill, several areas mapped by Stoffers as farmland had been abandoned and have evidently regenerated into mixed deciduous and evergreen thorny woodlands.
The vegetation of the Mountains area showed some recovery since the 1950s. There were more evergreen bushes, and less Acacia and Leucaena than Stoffers described. The vegetation Stoffers described for the lowlands had more Acacia than we found but the invasive Antigonon has since dramatically increased as a ubiquitous and often dominant species. The former importance of Opuntia prickly pear cacti in disturbed vegetations has dramatically declined since the 1950s. We ascribe this to the likely effect of the invasive parasitic insect Cactoblastis cactorum. In the 1980s and 1990s many Opuntia cacti were seen affected by this insect (G. Lopes, pers. comm.).
Our field data show that all wilderness areas of St. Eustatius remained heavily affected by grazers. This reduces the resilience of natural vegetations and interferes with natural succession by imparting heavy losses to hardwood seedlings and saplings (see e.g. Melendez-Ackerman et al. 2008), by reducing plant biomass (which increases exposure to wind and sun), and by favoring hardy invasive plant species. In Curaçao, large scale reduction in densities of feral grazers in the Christoffelpark since 1993 has led to rapid recovery of several rare plant species and vegetation types. The problem of feral livestock remains severe. Therefore the number one priority for terrestrial conservation in St. Eustatius will be to reduce feral grazer densities and impacts in key wilderness areas.
Parts of the natural areas of both The Mountains and the Quill are protected by law, but goats and other roaming livestock are omnipresent in all habitats and continue to have evident impacts on the vegetation. Aside from generally reducing the resilience of the vegetation to major disturbance, intensive impact in the herbaceous layers likely affects regeneration of rare hardwood species directly through selective predation and indirectly by overall desiccation and increased exposure to the elements. Goats have a broad diet in the region and species eaten include canopy, mid-canopy and understory species (Melendez-Ackerman et al. 2008). Vegetation degradation further also affects the competitive balance towards (unpalatable) invasive species such as Antigonon leptopus, Pedilanthus tithymaloides, and Leucaena leucocephala, which have expanded massively into natural habitat in the last 50 years. Not only is general vegetation degradation suggested as a problem to endangered breeding seabirds, but goats likely also directly trample nesting burrows of seabirds (Collier & Brown 2009).
Consequently, priorities for nature conservation are to:
- Reduce grazer densities in all areas,
- Protect the most sensitive vegetations using total grazer exclusion, and
- Experiment with methods to spur vegetation recovery, including erosion control, propagation of rare and endangered species and reforestation with indigenous species
- Establish permanent plots in areas with the most sensitive vegetations in order to better understand causal factors of short- and longer term changes.
This report is part of the Wageningen University BO research program (BO-11-011.05-004) and has been financed by the Ministry of Economic Affairs, Agriculture and Innovation (EL&I) under project number 4308202004.
Satellite image-based mapping of tropical forests is vital to conservation planning. Standard methods for automated image classification, however, limit classification detail in complex tropical landscapes. In this study, we test an approach to Landsat image interpretation on four islands of the Lesser Antilles, including Grenada and St. Kitts, Nevis and St. Eustatius, testing a more detailed classification than earlier work in the latter three islands. Secondly, we estimate the extents of land cover and protected forest by formation for five islands and ask how land cover has changed over the second half of the 20th century. The image interpretation approach combines image mosaics and ancillary geographic data, classifying the resulting set of raster data with decision tree software. Cloud-free image mosaics for one or two seasons were created by applying regression tree normalization to scene dates that could fill cloudy areas in a base scene. Such mosaics are also known as cloud-filled, cloud-minimized or cloud-cleared imagery, mosaics, or composites. The approach accurately distinguished several classes that more standard methods would confuse; the seamless mosaics aided reference data collection; and the multiseason imagery allowed us to separate drought deciduous forests and woodlands from semi-deciduous ones. Cultivated land areas declined 60 to 100 percent from about 1945 to 2000 on several islands. Meanwhile, forest cover has increased 50 to 950%. This trend will likely continue where sugar cane cultivation has dominated. Like the island of Puerto Rico, most higher-elevation forest formations are protected in formal or informal reserves. Also similarly, lowland forests, which are drier forest types on these islands, are not well represented in reserves. Former cultivated lands in lowland areas could provide lands for new reserves of drier forest types. The land-use history of these islands may provide insight for planners in countries currently considering lowland forest clearing for agriculture.
A preliminary inventory is given of key terrestrial nature values of Bonaire in order to determine their occurrence in relation to areas designated as “nature” and “open landscape”, according to the Spatial development plan of Bonaire. This was based on a literature study and supplemented by expert advice.
In 2010 a spatial development plan was written in order to determine the spatial policy and regulation for the future development of Bonaire. The island was partitioned into areas for different uses such as agriculture and recreation. Two specific designations are “nature” and “open landscape”. The occurrence of nature values within these areas remained unclear. This makes implementation of protective measures based on international treaties and island legislation problematic. An inventory of the occurrence of these values should help facilitate more effective implementation of these protective measures. In the present study key nature values are determined, both in terms of protected species and essential habitat (e.g. caves).
From the literature study it became apparent that data on the occurrence of most of the priority species of flora and fauna, is limited and scattered, especially with respect to “open landscape” and “nature” outside parks. Therefore, only a preliminary inventory is provided showing the general distribution of nature values across the entire island, as linked to various habitat types. An exact distribution of the different nature values was not possible at this time, but extrapolation from areas of known occurrence into other areas of similar habitat type was used to show the occurrence of overlapping distributions of nature values within the designated areas of “nature” and “open landscape”. The number of overlapping distributions of nature values may contribute to setting conservation priorities.
From the results it can be concluded that the areas of “open landscape” and “nature” (outside the national parks) seem to harbour unique and critical nature values. These areas are not actively managed or protected as national parks. The “open landscape” of Bolivia possibly harbours a few rare plant species (unique), an important population of critical key columnar cacti and at least two columnar cactus-pollinating bat species. The “open landscape” of Washikemba/Bakuna harbours key mangrove species that only have another main location at Lac Bay (national park). The “nature” area of Terrace Landscape Middle Bonaire seems to harbour a concentration of unique (e.g. Tillandsia balbisiana) and rare plant species (e.g. Krugiodendron ferreum etc.) and four bat species. The same is the case for Lima (e.g. Sabal palm, Maytenus versluysii and three bat species) while in Southern Bonaire key mangrove species also still occur. Table 1 shows which nature values are found or expected to occur within each “open landscape” and “nature” (outside national parks) area.
It can be concluded that outside the current parks, the main regions that harbour a concentration of key nature values are Terrace Landscape Middle Bonaire/Sta. Barbara, Bolivia, Washikemba/Bakuna and Lima. Terrace Landscape Middle Bonaire is designated as “nature” area, while Washikemba/Bakuna and Bolivia are in part designated as “open landscape”. Lima has both “nature” and “open landscape” designations. Sta. Barbara is designated for other uses, but the present review shows that the occurrence of several significant nature values is likely within this area.
Additionally, based on the preliminary inventory, the combination of apparent concentrations of rare plants, occurrence of critical bat species and the high probability of corridor values show that the areas of Terrace Landscape Middle Bonaire/Sta. Barbara and Lima are important areas concerning conservation and further research. The areas of Bolivia and Washikemba/Bakuna follow closely.
To be able to implement the necessary protective measures within these areas, it is recommended that more extensive research through fieldwork is done, in order to obtain a complete inventory of the different nature values found on Bonaire, not only in the areas of “nature” and “open landscape” but also in areas with other designations. Additionally, it is recommended to assess the list of vulnerable and endangered species (‘Informatieblad beschermde dier- en plantensoorten Bonaire’) as certain species that may be of importance to Bonaire are not included.
When executing a complete and extensive inventory of Bonaire it would be of value to also determine the ecological conditions needed for the different species to survive. Based on the ecological conditions necessary for their life functions, it may be possible to pinpoint those areas of main ecological importance per species. A complete inventory of the nature values on the island can contribute to better management of nature values (e.g. determining the distribution of caves and the distribution, health status and diversity of keystone cacti species for better management of bat populations). It is also recommended to determine areas with high potential for the occurrence of rare or relict species and which areas harbour high corridor values.
For future research it is recommended to execute a complete and extensive inventory of Bonaire, through fieldwork, in order to implement the necessary protective measures to ensure the conservation of these nature values. The present study shows that the areas of WNSP/Brasil, Terrace Landscape Middle Bonaire, Lima and Bolivia may be of priority as these areas seem to harbour a concentration of unique and critical plants.
Present studies shows that key nature values may occur in areas with a different designation than “nature” or “open landscape”. For future research it is recommended not to limit inventory research to the areas of “nature” and “open landscape”, but to include other areas with different designations.
In the present study the nature values chosen were based on the list of vulnerable and endangered species (Informatieblad beschermde dier- en plantensoorten Bonaire). During the study several species were added based on expert knowledge. The list used therefore seems to be limited. For future research it is recommended to assess if there are other nature values that are important to Bonaire that should be included on the list (e.g. Clusia sp, Ammodramus savannarum).
A complete inventory of the nature values on the island can contribute to the better management of nature values. A good example is the management of Bonaire’s bat population. In order to define the priority areas to maintain for the management of the different bats on Bonaire it is essential to obtain a detailed inventory of the different caves that these species use as habitat.
Additionally for the nectar-feeding bats it is crucial to map the occurrence of the different candle cacti on which they feed. The nectar-feeding bats are the critical pollinators of the three candle cacti (Petit, 2001). As already mentioned these cacti are key species on the island as they provide food for several species of animals during the dry season, when many other plant species are non-productive (Petit, 2001). Research on the distribution, health status and diversity of candle cacti on Bonaire is recommended in order to pinpoint priority areas for nectar- feeding bats. The cactus populations are threatened severely by feral livestock (goats, donkeys) which remove the bark of the mature trees, thereby threatening the food supply for frugivores and nectarivores. From our analysis open land areas of Bolivia would seem to possess large cactus populations of vital interest to conservation of endangered bird species on an island-wide scale.
When executing a complete and extensive inventory of Bonaire it would be of value to determine the ecological conditions and various habitats needed for the different species to survive. Based on the ecological conditions necessary for their life functions it may be possible to pinpoint those areas of principal ecological importance per species.
It is necessary to identify those areas with a high potential for the concentration of nature values rare species or relict vegetation species in order to secure the survival of these species and to be able to implement the necessary protective measurements. Such areas for instance are the open land and nature sections of Lima, Terrace Landscape Middle Bonaire (nature) and Bolivia (open). For future research it is recommended to determine those areas with high corridor values for the implementation of ecological corridors and buffer zones on Bonaire.
A semi-detailed landscape-based vegetation map (scale 1:50,000) is presented for the southern Caribbean arid island of Bonaire (mean annual precipitation is 463mm). Color aerial photographs (1:8,000) taken in 1995 and 1996 were used to produce the map. A total of 302 vegetation sample plots were analyzed using a stratified random sampling design and twinspan cluster analysis.
A total of 18 vegetation types, and 32 (sub-)landscape types were distinguished. The three principal vegetation types, Casearia tremula-Prosopis juliflora type (Type 17), Croton flavens-Haematoxylon brasiletto type (Type 14) and Prosopis juliflora- Opuntia wentiana type (Type 16), account for 40% of total vegetation cover. The four principal landscape types also cover 40% of the island and are: D3 (Prosopis- Casearia Landscape), TH1(Haematoxylon-Croton Higher Terrace), D2 (Haematoxylon- Casearia Landscape) and TM7 (Acacia-Croton Middle Terrace). The vegetation on the volcanic Washikemba Formation is more uniform than that on the limestone forma- tions. Most of the vegetation types can be categorized as secondary. This is consid- ered mainly to be the result of the impact of introduced grazing mammals (principally goats and donkeys) and woodcutting in the past. Six vegetation types are considered of relatively high natural value. Three of these (Types 1, 9, and 10) are comparable to Stoffers’ less degraded communities. The other three have been selected based on cri- teria such as structural complexity, diversity and number and rarity of rare species.
A comparison with a vegetation map from the 1950’s shows that three types of areas can be distinguished: areas in which the vegetation has remained more or less the same, areas in which the vegetation shows improvement and areas that show broadscale de- terioration of the vegetation. The largest area that shows deterioration is the southern part of Bonaire. The northern part of the Washington-Slagbaai National Park is the largest area with improvement. The findings are discussed in relation to the Nature Management Plan for Bonaire and conservation recommendations are made.