Last update: August 15, 2011 10:31:41 AM E-mail Print





P.W. Roux

Grootfontein Agricultural Development Institute, Private Bag X529,

Middelburg CP 5900


In 1983 several large circular areas were observed on a black and white satellite image of South Africa. Since the introduction of a daily satellite weather map on South African television the existence of these circular areas were conclusively established as being definite fixed features. Fig. 1 shows the location of the three most commonly observed circular areas. An interesting aspect is that they appear to be more or less of the same size. The most prominent one is situated in the Northcape Province and covers an area of approximately 125 000 km2. A number of smaller circular areas also appear to exist (Fig.l); these, however, have to be positively verified as they are but very vague features.



On a NOAA A VHRR satellite image of the western Karoo a great dark semicircular feature can be seen, bordering on the southern boundary of Great Bushmanland and covering a very large area of arid karoo-bush vegetation and most of the so-called Great Floors such as Verneuk Pan. This dark semicircular feature appears to have some connection with the North-Cape circular area as it forms part of its northern boundary.

The circular areas cover different geological formations and appear only to have some vague connection with the geological structures as mapped (Visser,1984) and some loose connection with topographic features. The North-West Province circle appears to have a direct connection with the Olifantshoek Sequence (Matsap) and Kuruman Hills which apparently exist as short arcs on the boundary of the circular area.

The southern Namibian circular area appears to have some relationship with the southern arc of geological structures which apparently form part of it. However, to the north there are no visual indications of such a relationship. An important aspect is that there exists another more or less equally sized circular area to the north. The important aspect here is that this circular area overlaps the southern circle in the north. This possibly indicates that circular areas could be of different age. Overall it appears that a direct relationship between circular areas and the geology, as mapped (Visser et al., 1984), is small.

A primary interest of this article centres on the influence that circular areas have on the cloud cover and the condensation of clouds. It is very clear from the weather satellite images that the circular areas (Fig. 1) have a definite influence on the cloud cover. It can very frequently be seen (on the TV -weather satellite image) that the cloud cover either shows signs of dissipating on reaching a circular area, sometimes apparently more or less bypassing the circular area, or that a circular area might show through the clouds as a somewhat transparent circular area. More frequently, however, a circular area shows up as a dark area more or less surrounded by clouds. Observations have also indicated that the circular areas have no effect on high altitude clouds, such as cirrus, or heavy cloud covers. Further, the North-Cape circle appears to be generally more "active" during summer, and the North-West Province and Namibian circles more in winter. However, this phenomenon might be purely incidental or related to the seasonal incidence of clouds.

The reader should try to locate these circular areas on the daily satellite weather image and observe the effect they have on clouds.

Judged from cloud behaviour it is most probable that temperature differences exist between the circular areas and the surrounding land. On the satellite TV-image (June 1994) it was very clearly seen after a night with heavy frost, that the North-West circular area showed up dark against a white background owing to greater radiation from the circular area than from the surrounding land. These circular areas are thus apparently "hotspots" that affect condensation of moisture and also atmospheric pressure and wind. It is also logical to accept, seeing that the cloud cover is affected, that the circular areas would most likely have a lowered rainfall.

An interesting aspect is the possibility that a smaller cloudless area could form in the centre of a circular area, surrounded by clouds. This could relate to geologic formations, or, more likely, that the central area of a hot spot is warmer. The existence of concentric hotspots cannot be ignored.

Two large circular areas also appear to be present in Zaire and Mozambique. There is also a vague indication of the existence of a very large circular area in the Indian Ocean off the south-eastern coast of South Africa. The occurrence of these areas, however, need to be verified. On two other occasions, months apart, it was noticed that on the same spot in central Africa, moisture had condensed to form a perfect ring of clouds. This phenomenon was interpreted as to indicate that the cloud rings had condensed above the perimeter of a circular area as a result of a difference in temperature between the boundary of a circular area and the surrounding land. The diameter of this particular ring of clouds was estimated to have been about 100 km.

These circular areas could be of considerable significance, not only from a climatic point of view, but also, to a considerable extent in regard to the structure of the deeper regions under the earth's crust. They probably indicate  the existence of great and specific magmatic bodies in the upper mantel.

To explain the occurrence of the circular areas at this stage, could at best purely be on a hypothetical basis. It was initially thought that these circular areas had been caused by the impact of large asteroids or similar bodies, or  even gigantic volcanoes that became levelled through erosion and were buried later under magmatic outflows. In a review (Melosh,1994) of a book by Spudis, it appears that lunar and planetary rings occur widespread and have most likely been caused by meteoroid impacts. The possibility that circular hotspots could have been similarly caused, cannot be overlooked.

Impact craters were largely ruled out as a cause of a circular area because such an area was also located on a section of a satellite image of the Patagonian desert; it is unlikely that bodies from outer space would always strike the earth at a vertical angle so as to cause a neat circular feature. In this respect the Vredefort dome is a good example of a slanting impact. Furthermore, impact craters leave definite surface characteristics and are, as such, usually identifiable. A volcanic origin of the circular areas is more likely but can be largely ruled out as this would certainly have been reflected more definitely in the known geological structures, unless such volcanic action predates the present structures. The most likely cause of these hotspots, especially in view of the overlapping of one circular area by another, is that they are of magmatic origin.

Hotspots are known to exist on earth and are described as igneous provinces by Coffin and Eldholm (1993). These hot spots vary in area, are highly irregular in shape and size, usually several km thick and often of great length. Some of the largest of these are the Deccan and Siberian Traps, the Chagos-Laccadive Ridge, the Kerguelen Plateau and the Ontong Java Plateau. These igneous provinces exist under the sea bed as well as on land.

According to Coffin and Eldholm the igneous provinces were formed by magmatic plumes arising from a relatively narrow molten zone between the outer core of the earth and its lower mantle. When a plume reaches the base of the upper mantle it decompresses and partially melts, producing a locus of volcanic activity. Some of the magma may produce, by eruption, an extensive flood of lava creating an igneous province. Coffin and Eldholm presumed that the rising magmatic plumes are spherical in shape and that this shape is the most efficient form for the transport of magmatic material and to allow a plume to rise. Similarly, the circular hotspots may have originated as gigantic bodies of magma evaginating (boiling off) from the molten zone. These bodies break away from the main source as a result of a lower density, due to expansion from the great heat, and consequently gradually drift upwards in the form of gigantic molten blobs. Probably such dislodgments did not take place simultaneously as circular areas can overlap, as is also evidenced by the (apparent) occurrence of smaller circular areas on, or within, the boundary of a hotspot (Fig. 1). Whether these smaller hot spots occur above, below, or embedded in a larger hotspot, is purely a matter of conjecture. The ascending magmatic bodies apparently had become arrested and spread out under the lithosphere ceiling forming the circular hot spots. If this be the case, it cannot be ruled out that some of these hotspots may still even be in a process of cooling. It is also quite possible that the intrusion and expansion of these magmatic bodies could have caused cracks and weak areas in the lithosphere through which magma intruded or extruded and possibly also influenced the geology. Under such cataclysmic occurrences lateral and vertical pressure would be exerted on the containing crustal structures causing faulting, slip- pages, folding and upliftments even at some considerable distance from the area of activity.

A very important aspect concerning the igneous provinces of Coffin and Eldholm, is that significant changes in the global environment could have been triggered. If this be the case, it can be concluded that the circular hotspots could also have influenced the environment (especially temperature) to a major or minor degree. Their greatest effect would have been during and soon after their formation. The primary effect would most certainly have been the creation of deserts and consequently the course of the evolution of the biosphere.

The pertinent questions that now arise are:

Research on these circular anomalies could possibly open up new perspectives on geology, climate and vegetation. The North-Cape hotspot appears to be the most likely area for research.



COFFIN, M.F. & ELDHOLM, 0.,1993. Large igneous provinces. Scientific American, 269, 4 : 26-33.

MELOSH, H.J. 1994. Missing rings. Nature, 368, March, p. 24. Book review of Spudis, P.D.; The geology of multi-ring impact basins: The moon and other planets. Cambridge University Press.

VISSER, D.J.L.(Compiler), 1984. Geological map of the Republic of South Mrica, Transkei, Boputhatswana, Venda, Ciskei and the kingdoms of Lesotho and Swaziland. Dept. of Mineral and Energy Affairs, Republic of South Africa.



Karoo Agric, Vol. 6, No 1, 1994 (3-5)