To fully understand the physical geography of St. Helena, a knowledge of the island's geological structure, lithology and history is essential. This page augments a concise description of these elements with a consideration of St. Helena in the wider context of ocean-floor spreading and plate tectonics. A description of the island's physiography and comments on the economic geology are also included.
Physiography
St. Helena is the deeply eroded summit of a composite volcano rising from the sea floor at a depth of 4,224m. The sub-aerial portion is approximately 122 sq. km. (47 sq. miles) in area - about half the size of the Isle of Wight. The highest point, Diana's Peak, is 823m above sea level. The small size of the visible island belies the true dimensions of this massive volcano; it is a major feature on the earth's surface. The base of the volcanic pile measures some 130km in diameter and the volume of the cone is estimated to be twenty times that of Mt. Etna - the largest European volcano. St. Helena is truly remote and, together with Ascension Island (1,127km distant), comprises the only land in 3 per cent of the earth's surface area.
The island is approximately 16 km long, 10 km at its widest and elongated in a NE/SW direction. A high central ridge occupying the major axis, dominates the topography. Radiating from the ridge, deep gorge-like valleys are incised to depths of up to 300 m. The valleys, with precipitous sides and narrow valley floors are spectacular sights; Thompson's, Fisher's, Sharks, Lemon and Deep valleys are among the more impressive.
Despite their tremendous dimensions, many valleys are dry in all but the wettest of years and only James Valley, host to the island's capital, possesses a permanent stream. Few areas of level ground have been preserved. Deadwood and Longwood Plains in the interior, are the only sizeable areas of pasture and cultivation and the latter supports the largest settlement outside Jamestown.
The arid Prosperous Bay Plain in the north east, is particularly interesting, but the most striking physiographic unit on the island is undoubtedly Sandy Bay Valley, an enormous bowl-like depression, extending back from Sandy Bay to the mountains of the central ridge.
The grandeur of the interior scenery is complemented by a series of stupendous 300 m cliffs along the island's coastline. These forbidding precipices, combined with the unpredictable Atlantic swell, restrict practical landing sites to those few places where valleys breach the cliffs at sea level.
The northern lee shores are relatively calm for much of the year, but are subject to the potentially destructive effects of the 'rollers' - at their most ferocious in February and March.
Thought to originate from storms in the North Atlantic, the rollers can reach many metres in height, and have been known to destroy vessels lying off Jamestown. The southern 'weather' coastline is subject to a permanent running swell making landing or even sailing close inshore a hazardous undertaking. The problem is compounded over the whole coastline by a proliferation of menacing sunken rocks. These difficulties are unfortunate since access to many field sites would be eased considerably if a landing could be made in the appropriate place, reducing the time taken in overland journeys.
The physiography dominates many aspects of human life; agriculture, settlement, communications and perceptions - changes in relative relief are of such magnitude as to distort, temporarily, the perceptions of scale and distance by observers accustomed to gentler landscapes. Geological structure and lithology - the raw materials of geomorphology - are of particular importance in understanding the formation of this rugged and sometimes oppressive landscape.
Structure & Lithology
The island was formed by the coalescence of two broad shield volcanoes with centres of activity located in the north east - in the Flagstaff Hill/Knotty Ridge area - and in the south-west at Sandy Bay. A third and more recent minor centre was located in the east of the island. The Sandy Bay activity occurred later than the activity in the north east and is thought to account for aproximately five-sixths of the volume of the sub-aerial rocks (Daly 1928). Eruptions occurred predominantly along fissures, and both major centres were transgressed by dyke swarms.
Deep erosion of both shields has enabled the stratigraphy to be determined with some precision. Baker (1968) provides the most comprehensive description. The valley sides and cliffs reveal a succession of lava flows with interbedded ash and pyroclastic deposits; the exposed rocks are of the alkali olivine basalt-trachyte-phonolite assemblage the basalts account for by far the greatest volume of the exposed rocks.
Basalt is a basic igneous rock occurring as lavas and dykes widespread on the continents and ocean floors. It is fine-grained, crystalline, dark and heavy, and lacks all signs of stratification. On eruption it has a low viscosity and often forms extensive floods mantling the landscape. The active volcanoes of Iceland and Hawaii are good examples of the type of activity which St. Helena exhibited throughout its active life. The valleys incised in both shields reveal a history of many tens of eruptions, each identifiable by a lava flow ranging in thickness from 1 to 30 metres, commonly with an overlying ash layer of varying thickness and colour. The ash layers are responsible for much structural instability in the cliffs and valley sides since they are easily eroded by rain, wind and waves and compressed by the weight of overlying rocks.
The basalts exposed on St. Helena have undergone much alteration, and are deep-weathered and friable; the clay mineral halloysite is a common alteration product. Spheroidal weathering forms, typical in basalts, are a common sight, particularly in the valleys fringing Deadwood Plain (Netley Gut, Bilberry Field Gut). These valleys are currently affected by accelerated erosion. Fresh exposures, best seen in cliff faces, reveal the basalts to be typically block-jointed. Both vesicular and amygdaloidal forms abound, the latter being infilled with zeolites and calcites. No good examples of the surface forms of the lavas were observed. Their obliteration is in contrast to the much younger flows observed on Ascension Island where both aa (blocky) and pahoehoe (ropey) forms are preserved. Marine volcanic lava forms are rare on St. Helena. Baker (1968) records a small outcrop of marine breccias at 300m above sea level in Sandy Bay Valley and suggests selective uplift was responsible for their elevation. Pillow lavas, which form as a result of submarine extrusion, are not recorded from St. Helena.
Intrusions and extrusions of more viscous trachytes and phonolites formed plugs, domes and, more rarely, dykes. In the field, these intermediate rocks could be distinguished from the ubiquitous basalts by differing qualities of colour, texture and mineralogy, as well as their field relations. Trachytes and trachy-andesites are generally lighter in colour and have a porous texture. Phonolite - so called because of its sonorous qualities when struck - has a distinctive grey colour and often exhibits a columnar structure, only poorly developed, if at all, in the basalts. The Asses Ears and Castle Rock are particularly fine examples of columnar phonolitic intrusions.
The friable nature of much of the exposed rock on St. Helena has important implications for fieldwork. Most field sites of interest are situated well away from roads and lie outside the vegetated heart of the island. Distances to such sites are not great; yet the magnitude of the relief and the treacherous nature of the rocks combine to render travelling by foot arduous and time-consuming.
Many sites, particularly those at the coast, are reached by barely discernible paths (euphemistically termed 'roads' by the islanders) which follow the contours along valley sides or even across cliff faces. Such paths are covered with a veneer of fine ash and weathered basalt and though reasonably safe when dry, become effectively impassable when wet. The use of a guide is essential for certain parts of the island and travelling alone is inadvisable.
The North Eastern Volcano
The eroded remnants of the north eastern volcano provide some of the most dramatic scenery on the island. The pastures of Deadwood Plain extend to the lower slopes of the 700m Flagstaff Hill, thought to be the remains of a parasitic cone. Flagstaff Hill itself is connected by a treacherous knife-edge ridge to the Barn - an epithet applied to a capping of relatively resistant younger lavas on the weaker rocks of the basal complex. Spectacularly cliffed on its seaward side, the Barn towers above the deeply dissected pyroclasts and weak flows of Turk's Cap Valley on its southern margins.
The oldest rocks consist of greatly altered breccias exposed between dyke swarms. Indeed, the bulk of this shield is composed of flows and pyroclastics transgressed by dykes. Some of the swarms are particularly intense - the knife-edge ridge connecting Flagstaff Hill with the Barn is composed of approximately 200 dykes with no intervening tuffs or other material - testimony to the strong tensional conditions that must have been present in the cone. Dyke orientation falls into two categories. One swarm strikes NE/SW - parallel to the dominant dyke system of the south-western volcano - prompting Daly (1928) to suggest that the basal complexes of the two shields are part of the same formation.The second swarm strikes N. 10-20 degrees W. and is best seen in the knife edge ridge mentioned above.
The South Western Volcano
The structure of the south-western volcano is more complex than the north eastern; nonetheless the stratigraphy has been determined from exposures of the deeply eroded shield, particularly in the arid Sandy Bay Valley region. The basal complex is similar to that in the north east, consisting of bedded pyroclastic deposits derived from local cinder-cone concentrations interstratified with lavas and transgressed by dykes trending NE/SW. Above the basal complex, extensive lava flows (ranging in thickness from 1 - 30 m) dominate to form the main shield which overlaps the western and southern flanks of the north eastern volcano. Sandy Bay Valley is of particular interest and will be examined in more detail.
The valley is a remarkable feature; a huge amphitheatre, approximately 5kmin diameter, it extends back to the central ridge defining its northern rim. The more resistant basalts of the main shield have been removed in this area to expose the weaker strata of the basal complex - readily excavated to produce a pattern of deep valleys converging on the main streams forming the axis of the amphitheatre. The area is characterised by steep, barely vegetated scree slopes which are severely gullied. The intricate patterns formed by the myriad dykes are visible in many parts of the valley.
Phonolitic intrusions, more resistant to erosion, have been isolated to form upstanding volcanic necks such as Lot's Wife rock. This, and other peculiarly shaped examples, (Lot, the Riding Stones, Negro's Head and the Asses Ears) represent the last stages of volcanic activity and are not thought to have caused much surface disturbance upon intrusion. According to Daly (1928), both Lot's Wife Rock and Castle Rock form local swellings on the same feeder dyke, whereas Lot is situated on a second dyke and the phonolitic neck forming Speery Island lies on a third.
Darwin (1844) thought this amphitheatre represented the eroded remnants of a huge explosion crater or caldera, breached along its seaward side. Daly (1928) discounted the caldera hypothesis, pointing to the lack of fragmented material on the flanks of the volcano and the existence of other, smaller scale, erosional amphitheatres (Powell's Valley, the Devil's Punch Bowl).
While there are serious objections to the caldera hypothesis, the alternative erosion hypothesis calls for the removal of enormous amounts of material following the cessation of volcanic activity. The intrusion of the phonolites has been dated as the last important volcanic event on St. Helena, they are not thought to have broken the surface. Their present form gives an indication as to the amount of material removed subsequently. To accept the erosion hypothesis, palaeo-erosion processes operating at rates far in excess of the present ones must be invoked.
Age
Rock samples from St. Helena have been dated radiometrically using the Potassium-Argon (K/Ar) method. Abdel-Monern and Gast (1967) made determinations indicating an age of 10-13 million years for the basaits and trachybasalts and 8-8.5 million years for the later intrusions. Baker, Gale and Simons (1967) dated stratigraphically controlled samples to produce the age sequence . Their results show that the north east centre was active from 14.6 million years ago, whereas the Sandy Bay centre was not active until about 11.3 million years ago.
All activity apparently ceased about 6.8 million years ago following the intrusion of the phonolites. Baker (1968) suggested that the late intrusives represented the products of a highly differentiated magma chamber at a high level in the volcanic pile; differentiation occupying a considerable period of time prior to extrusion. The sub-aerial life of the volcano is estimated to be a minimum of 7.5 million years, with the main centre active over 4 million years. It is thought that the late flows and minor flank activity occupied a period of time comparable to that taken to build the bulk of the main shield (Baker et al. 1967).
Summary
Both centres are characterised by thick basal complexes of weak basalt flows, tuffs and agglomerates overlain by more massive flows interbedded with thin tuffs. Streams have cut through the upper flows, generally sharply cliffed,into the weaker basalts beneath. Eruptions from the third minor centre in the last stages of volcanic activity resulted in extensive flooding of both shields in the east and north east.
St Helena & Plate Tectonics
Over the last six decades, advances in geophysics, oceanography and marine geology have enabled earth scientists to build a global geological model. Plate tectonics theory has come to dominate current thinking since it accounts for more features of the earth's geology than other models. Essentially it proposes that the crust is divided into a number of rigid 'plates', whose boundaries are defined by three types of margins-
- Constructional: where new material is added to the edge of the plate. On land, constructional margins are given surface expression by the great continental rift valleys, such as the East African Rift Valley, in the oceans, constructional margins are marked by the line of the great oceanic ridges, such as the mid-Atlantic rise.
- Destructive: where plate material is subducted into the earth's mantle and effectively destroyed. When two continental plates collide, the crust of one plate is forced over the crust of the other to create the great mountain ranges. Thus, the Himalayan mountains were formed as the plate bearing the Indian sub-continent collided with the China plate. Destructive margins in the oceans are marked by the ocean trenches, such as the Marianas Trench in the Pacific.
- Transform Fault: where two plates slide laterally against each other. The San Andreas Fault, marking the continental boundary between the North American and Pacific plates, is a notable example.
The mechanism for plate movement remains unproven. Current theories postulate the existence of huge thermal currents in the earth's mantle. Constructional margins occur at the site of upwelling of these currents, whereas destructive margins occur as the currents cool and sink back into the mantle. The friction caused by plate movement releases seismic energy in the form of earthquake and volcanic activity, and the boundaries of the major plates, coincide broadly with zones of seismic activity.
Plate tectonics has evolved into a complex body of theory and the reader is referred to Wilson (1972), Tarling and Tarling (1971) and Wyllie (1976) for an introduction to the subject. Press and Siever (1974) is a good example of a general geological text written within the framework of plate tectonics. One of the major components of plate tectonics is the sea-floor spreading hypothesis which attempts to account for the formation of the ocean basins.
The hypothesis holds that molten material is added continually to the ocean floors along the line of the mid-oceanic ridges, causing lateral migration or 'spreading' of oceanic crust from either side of the ridge. Thus a transect across an ocean basin should show the age of rocks increasing with distance from the axis of spreading. Further, the age of volcanic islands should also increase with distance from the ridge. St. Helena lies 960km west of the mid-Atlantic ridge; if sea-floor spreading occurs then the island, in common with other Atlantic islands, must have formed on the ridge and since been gradually pushed away.
By relating island ages to distance from the ridge, some idea of the rate of movement can be calculated. Wilson (1963) compared the ages of the oldest exposed rocks on the Atlantic islands with distance from the ridge, and was indeed able to demonstrate a linear relationship For this sort of analysis the age determination is crucial; how accurately do the ages of rocks exposed at the surface today represent the age of formation of the volcano? Wilson assumed that St. Helena formed on the ridge 20 million years ago, indicating a rate of movement of 3.5cm per year to cover the distance to its present location.
The ages Wilson (1963) used in his analysis were given as minima, therefore the rates he calculated are maxima. Later studies comparing the magnetic anomaly patterns of the ocean floor with the magnetic polarity history inferred from dating volcanic rocks (Vine 1966) suggested the considerably lower rate of approximately 2.3cm per year. This figure indicates that St. Helena formed at the ridge between 45-50 million years ago. Magnetic anomaly studies of the ocean floors proved crucial in elevating sea-floor spreading to the status of accepted theory. Abdel-Monernand Gast (1967) suggest that the time represented by the sub-aerial terrestrially-extruded portion of the volcano may be as much as 30 million years, and claim that:
it is rather apparent that the volcanic history covered by the volcanic island covers a time span much longer than this region of the ocean floor spent near the vicinity of the ridge crest for the ocean floor spreading hypothesis.
Baker, Gale and Simons (1967) note the difficulties involved in extrapolating back to the formation of the volcano, since the rocks recording this 1:at the base of the volcanic pile (the sub-aerial portion represents only per cent of the total volume of the cone) It is therefore impossible to determine the rate of submarine extrusion without submarine dredging and deep-drilling - a point also recognised by Abdel-Monern and Gast (1967]. Baker et al claimed that field evidence points to a decline in the rate of extrusion:
the age/volume relationships demonstrate a slowing down in activity throughout the formation of the main volcano.
It may be that the bulk of the cone was built up relatively quickly and the declining rate of activity was linked to increasing distance from the ridge; though such a relationship is difficult to substantiate. On the basis of their evidence, Baker et al (1967) concluded that:
the radiometric ages for St. Helena support in a general way the concept of ocean-floor spreading. It is clear that the formation of St Helena took place more than 14 million years ago, which is certainly much older than the age of volcanism on the North Atlantic Ridge and does in fact accord quite well with the linear correlation between age and distance from the ridge given by Wilson.
If this conclusion is accepted, then St. Helena formed at the mid-Atlantic ridge up to 50 million years ago. A magma source was maintained for approximately 37-40 million years as the volcano moved away from the ridge. The last major volcanic event occurred approximately 7 million years ago with the intrusion of the phonolites, probably the product of a highly differentiated magma chamber.
Economic Geology
The island's geology provides little in the way of mineral resources for its inhabitants - unfortunate for an island economy dependent upon aid from the UK. Hirst (1951) surveyed the economic geology, concentrating particularly on manganese and phosphate deposits. Despite fairly widespread occurrence of these minerals, there is insufficient tonnage for commercial extraction. At present, the only quarrying involves extraction of hard-core from basalt layers on Donkey Plain (which is facing problems owing to the presence of halloysite) and the extraction of calcareous sand from the deposits above Potato Bay.
