The invertebrates of Prosperous Bay Plain, St Helena a survey by Philip and Myrtle Ashmole September – December 2003 Commissioned by the St Helena Government and financed by the Foreign and Commonwealth Office

Download 1.13 Mb.
Size1.13 Mb.
1   2   3   4   5   6   7   8   9   ...   17

2.2. Assessment of habitat types

At the start of the study it was assumed that it would be useful to undertake habitat mapping on a substantial scale. It quickly became apparent, however, that characterisation and mapping of habitats in the more remote parts of PBP was largely impracticable and of doubtful value, since we initially had no knowledge of the features determining the distribution of the animals. We therefore decided to give priority to sampling the fauna in a wide variety of places, in the hope that the results would clarify the key parameters determining distributions. The initial results of this sampling showed the value of this approach, and we continued to devote almost all our time to work on specified sites.

At each of the study sites (see Plates 1 and 3) an assessment was made of several characteristics – aspect, altitude, form, substrate, vegetation and the invertebrate species present. Particular attention was paid to the possibility that habitats liable to destruction by airport construction would be matched by similar habitats outside the affected area.

Towards the end of the work, feeling that we had gained understanding of the area, some gross habitat parameters were mapped in an area in and immediately around the Central Basin, using a GPS instrument, aerial photographs and large-scale maps. In addition, prolonged observations in the basin and elsewhere were made after dark, in order to gain understanding of the desert ecosystem. These led, for example, to discovery of a localised condensation phenomenon that was not apparent by day, but which seems to have some significance for the fauna. The resulting outline habitat map is included as Plate 2b.

    • 2.3 Recording, analysis, mapping and photography

Microsoft Access was used for recording the site information, species information and collection data. Adobe Illustrator was used for habitat mapping, and recording the position of sites and other information.

Mapping was initially hampered by the discovery that normally calibrated GPS instruments used in conjunction with Ordnance Survey terrestrial maps of St Helena led to an error of around 500 m. After some delay the Ashmoles obtained correct calibration data by courtesy of Shelco. These have since been confirmed by the Ordnance Survey and are included in Appendix 2. However, small errors do persist.

The digital camera purchased for the project was used to take photographs of many of the invertebrates, either in the field or in the laboratory, during the work on the island. This camera has been deposited with the Environment Co-ordinator. Additional photographs were taken after the Ashmoles returned to Britain, using a privately purchased camera. This photographic effort made it possible to illustrate nearly all of the important species found during the fieldwork in the Guide to Invertebrates of Prosperous Bay Plain, St Helena that accompanies this report.

3. Background information on Prosperous Bay Plain

    • 3.1 Topography and geology

Prosperous Bay Plain (PBP) is part of the eastern arid area of St Helena, and shares the designation of Crown Wastes with much of the outer part of the island. For the purpose of this study we have considered PBP as extending from the north rim of Dry Gut (north of Bencoolen) in the south to the top of the steep slopes leading down to Prosperous Bay beach in the north, and from just east of Cooks Bridge on Fishers Valley in the west, to the tops of the sea cliffs in the east (see Plates 1 and 3).

Defined in this way, the Plain extends to about 2.25 km2 and comprises one of the largest areas of relatively level ground on the island. It was formed about eight and a half million years ago by massive outpourings of basalt and trachyandesite lava from vents on the northeast flank of the Southwest Volcano (Baker 1968, Weaver 1991). These lava flows flooded out to cover a large area extending from Longwood in the west to Horse Point in the north and Bencoolen in the south. It seems that in the east they did not reach the sea, but were penned in by older rocks. One or more of the largest flows created an enormous lava pond, which solidified to form the base of the relatively level tract of land that has persisted to the present day. Fishers Valley, formed later and draining a section of the high part of the island, runs eastward as far as the plain but then turns north to reach the sea at Prosperous Bay; its major canyon separates the Holdfast Tom and Horse Point area from the modern Prosperous Bay Plain.

Prosperous Bay Plain has probably always been an area of relatively low rainfall, and apart from Fishers Valley there has been no development of major gully drainage patterns. Within the plain, however, there is a Central Basin covering about 0.6 km2 and bounded on three sides by the main plateau of PBP, which is some 20 m higher (Plate 2). In the northwest of the basin there is a barely visible raised rim before the land drops away steeply into the canyon of Fishers Valley. This Central Basin of PBP has special ecological significance (see Section 6.2).

The area that is now the Central Basin was doubtless originally part of the surrounding plateau, filled by the relatively friable rock overlying the massive trachyandesite flow that now forms the floor of the basin. The action of sun and rain, combined with the effects of acid leaching from the rock, causes gradual disintegration of this upper rock layer into grit and dust. Erosive removal of this material is first by gravity down the eroding slope, and then by the wind towards the northwest. Any material that reaches the rim of Fishers Valley, where the underlying lava flow is cut by the stream flowing from the Peaks, is washed away towards Prosperous Bay (Plate 4d).

We believe that this process started as soon as the Fishers Valley stream cut through the underlying lava flow, with erosion of friable material close to the developing canyon. The upper rim of the developing basin then gradually retreated towards the south and east, progressively enlarging the Central Basin and leaving the lava flow that forms its floor covered by only a thin layer of fine material. The developing basin provided only a small catchment for rain, and only in the southwest (in the direction of Woody Ridge) was there significant drainage from higher land into the basin past our Site 5 (Plate 4b). That input of water has probably contributed to ponding in the basin in the past (as in 1787: see Ashmole & Ashmole 2000a: 61) but some of it now escapes northwards into Fishers Valley via a shallow gully just east of the concrete hut on the vehicle track from Cook’s Bridge. Outside the basin, a much larger ravine running up towards it from near Gill Point originates at the break in the southeast rim, but there is a watershed at this point and the gully has never captured the whole drainage of the basin. However, the presence of the break in the rim implies that there has been some run-off here during an earlier stage in development of the basin.

The floor of the Central Basin is so nearly level that erosion gullies within it are minimal; they are only strongly developed close to the rim of the Fishers Valley canyon in the north. A central boss of (presumably) more resistant rock (designated Stone Hill during the present survey) has been left isolated by erosion of deposits to the north, west and south of it (Plate 2). The summit of this hill provides an excellent vantage point for viewing the basin (see panoramic photos on front cover).

The deposits on the floor of the Central Basin consist of the dust and grit produced by weathering of the overlying rock. In places the dust is rather uniform to a depth of more than one metre; it was not investigated below this depth. This material has evidently been deposited by a combination of wind and water. We frequently saw small dust devils swirling across the floor of the Central Basin from east to west. These must gradually transport light fragments of decayed lava from the eroding southern and eastern walls of the basin. Occasional flooding of the area after heavy rain doubtless sorts and redeposits this material to some extent. The coarser white grit to the east and southeast of Stone Hill seems to be derived from the whitish rock forming the plateau to the east in the area of our site 21 (Plate 6).

In some places in the west the dust appears to have been consolidated to form a sedimentary rock, which is now itself subject to weathering (Site 14, Plate 8a). The surrounding plateau also has varied characteristics, depending on the nature of the rock and the extent to which weathering has progressed.

Lucy Caesar (2001) analysed particle size at sites in and around the western part of the Central Basin in terms of the percentage of a sample that passed through a sieve of 2 mm mesh. Caesar’s sampling sites were randomly placed within squares laid out in a grid, and cannot be located precisely. They show, however, that in samples from many parts of the floor of the Central Basin more than 90% by weight of the dried sample passed through a 2 mm mesh, and in some places this figure rose to almost 100%.

In view of the importance of the role of burrowing invertebrates in the PBP ecosystem, which has become more apparent during our work, there is a case for extending this technique to other parts of the Central Basin and elsewhere on PBP, perhaps using 1 mm sieves as well as 2 mm ones. In describing our sites we have used purely subjective descriptors, referring to dusty and gritty substrates to imply, respectively, the prevalence of finer or coarser particles.

The deposits in the floor of the Central Basin contain high concentrations of mineral salts, probably mainly as a result of leaching from the decaying rocks. Maps in Brown (1981) show high sodium concentrations at several sampling points in the basin, but we have not been able to find the Appendix that gives the full data. She also commented (vol. 2: 27) that streams on the island below 550 m a.s.l. tend to become increasing saline, with a noticeable concentration of highly saline water and soils in the eastern part of the island, and suggested that this was associated with the sodium rich trachyandesite flows of the Upper Shield.

Caesar (2001: 34) documented highly saline conditions in parts of the floor of the Central Basin. She found sodium ion concentrations of more than 3000 ppm in the gully near the concrete hut and values around 1000 ppm at several other sites in the western part of the basin. There were, however, large differences between samples only a short distance apart.

High phosphate concentrations are also present in parts of Prosperous Bay Plain. Daly (1927) first suggested that phosphatic deposits there provided evidence of large ancient seabird colonies. Brown (1981: 19) mentioned an analysis showing total P2O5 content of 20-25%. In 1995 we found many fossil bird bones of the large petrel Pterodroma rupinarum and other species in the shallow gully in the northwest part of the Central Basin, and revisited the previously known deposits in Dry Gut that are dominated by the same species. We also found bones, mainly of the storm petrel Pelagodroma marina, above our 2003 Site 5, which drains into the Central Basin. Old guano deposits and bird bones can still be found in crevices in outcrops of hard rock on PBP, for instance at our Site 4, where we found storm petrel bones in association with remains of the Giant Earwig (see Dermaptera, below).

In 1995 we collected samples of whitish powder from the ground surface in many parts of the island, in an attempt to identify phosphatic deposits possibly derived from seabird guano (Ashmole & Ashmole 2000a: 102-3). We analysed the samples by X-ray fluorescence spectroscopy, with the help of Godfrey Fitton of Edinburgh University, and the results throw some light on the situation in PBP.

Using a basalt standard with about 1% by weight P2O5 as a baseline, our sample from 2003 Site 4 had a value indicating around 39% phosphate. That was in a sheltered crevice, but a sample on the open ridge near our 2003 Site 17 (under the proposed airport terminal) gave the high value of about 19%, strongly suggesting that this ridge was once occupied by a major bird colony. Furthermore, a sample from near the centre of the proposed runway at 2003 Site 21 indicated 5%, and one from the large excavations in white deposits in the same area gave 7%. High levels of phosphate were also found further south, with a value around 7% north of the Gill Point waterfall, 12% near Gill Point and 14% in deposits in Dry Gut.

Intriguingly, however, the sample of white powder from excavations at the watershed at the head of the gully that breaks the southeast rim of the Central Basin (north of our Site 17 and east of Site 12) showed a phosphate level close to the control: it is probably mainly gypsum. We do not have samples from the floor of the Central Basin, but phosphates concentrations around 3% were obtained from two samples further west, one in the gully between 2003 Sites 5 and 15 where bird bones were collected, and the other north of the Fishers Valley canyon, midway between 2003 Site 11 (east of Government Garage) and the ruin of Bradleys near Cooks Bridge.

It will be clear from the preceding discussion that not only is there complex variation in landform and substrate over PBP as a whole, but also that there is significant variation in particle size and chemical composition in the substrate of the Central Basin, which doubtless determine which plants can flourish in the various parts of the area.

1   2   3   4   5   6   7   8   9   ...   17

The database is protected by copyright © 2020
send message

    Main page