- Email: [email protected]
The “seven-coloured earth” of Chamarel, Mauritius
Journal of African Earth Sciences 57 (2010) 169–173
Contents lists available at ScienceDirect
Journal of African Earth Sciences journal homepage: www.elsevier.com/locate/jafrearsci
The ‘‘seven-coloured earth” of Chamarel, Mauritius H.C. Sheth a,*, C.P. Johnson b, C.D. Ollier c a
Department of Earth Sciences, Indian Institute of Technology Bombay (IITB), Powai, Mumbai 400 076, India C-DAC School of Advanced Computing (C-SAC), Quatre Bornes, Mauritius c School of Earth and Geographical Sciences, The University of Western Australia, Nedlands, WA 6009, Australia b
a r t i c l e
i n f o
Article history: Received 11 March 2009 Received in revised form 21 July 2009 Accepted 23 July 2009 Available online 6 August 2009 Keywords: Mauritius Chamarel Basalt Weathering Rills Coloured earth
a b s t r a c t The ‘‘seven-coloured earth” of Chamarel is a geological curiosity and a major tourist attraction of Mauritius. This is a small (7500 m2) area of strikingly bare landscape showing well-developed rills and various shades of red, brown, grey, and purple. Curiously, it is located within a large, dense forest. Prevalent misconceptions are that the landscape formed due to a volcanic eruption, or from volcanic ash. Whereas the bedrock is undoubtedly an old volcanic rock (basalt), the colours are due to weathering of the basalt and the formation of secondary iron oxides and hydroxides in it, and the rilling is a result of deforestation and sheet erosion, i.e., human modification of the landscape. Such features, inadequately described in the literature so far, also occur in Papua New Guinea, and may be common in tropical, high-rainfall regions with volcanic bedrock. Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction The ‘‘seven-coloured earth” of Chamarel is a geological curiosity and one of the major tourist attractions in Mauritius, a small African island country in the Indian Ocean (Fig. 1) that has a growing tourism industry. The seven-coloured earth is a small (7500 m2) site in the Black River district of southwestern Mauritius. Though in the heart of a large, dense forest, it is strikingly devoid of any vegetation. It is a landscape with severe rill erosion and surface colouration in various shades of red, brown, grey, and purple that gives it its name (Fig. 2a–d). The site is on private land and a fee (75 Mauritian rupees per person in June 2008) is charged for entry to the seven-coloured earth as well as the beautiful Chamarel waterfall on the same premises. Searching ‘‘Chamarel” in Google, or in the scientific literature (e.g., Elsevier’s ScienceDirect) produced no results of a technical nature. We present the geological and geomorphological background to the Chamarel seven-coloured earth, so that this, and similar features known elsewhere but insufficiently described, are better understood. 2. Some misconceptions Widespread misconceptions exist about the seven-coloured earth. Links to volcanoes are indicated by enthusiastic tourists. A * Corresponding author. Tel.: +91 22 25767264. E-mail address: [email protected] (H.C. Sheth). 1464-343X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jafrearsci.2009.07.009
tour guide we met at the site told us very confidently that the different colours (and the puzzling total absence of vegetation we pointed out to him) were ‘‘due to a volcanic eruption”. A particular traveller’s website that we hit with Google (by searching ‘‘Chamarel”) says that the different colours are developed due to different lava flows cooling at different temperatures. These conceptions are quite mistaken. The island of Mauritius is indeed volcanic, but volcanism ended long ago. The shield stage of volcanism, which built up most of the island itself, is 8 million years old, and the youngest volcanic activity occurred 30,000 years ago (McDougall and Chamalaun, 1969; Baxter, 1972, 1975, 1976; Sheth et al., 2003; Paul et al., 2007). The bedrock at Chamarel is volcanic rock (basalt) that has been considerably weathered. The basalt lithology cannot be ascertained at the site itself, as this site is fenced off, denying access to tourists. However, 1.3 km away the spectacular Chamarel Falls plunge a hundred metres or so over two cliff-forming basalt lava flows, the upper one of which shows well-developed columnar jointing (Fig. 3a). Saddul (1995) considers the seven-coloured earth as formed on altered volcanic tuffs. He writes that the tuffs are as much as 18–20 m thick and overlie basalt lavas with a sharp contact. He speculates that the volcanic tuffs are derived from the Bassin Blanc volcano nearby (Fig. 1), which we have not visited. His view is echoed in The Mauritius Telephone Directory ‘‘tourist information”, which states: ‘‘These slivers of colours are believed to be the result of the erosion of the volcanic ash”. This sentence confuses erosion, meaning removal of rock material, with weathering or alteration, which happens to the rock in place. The same Directory refers to
170
H.C. Sheth et al. / Journal of African Earth Sciences 57 (2010) 169–173
Fig. 1. (a) Map of the Indian Ocean region showing the locations of the three Mascarene Islands (Réunion, Mauritius, and Rodrigues) and major physical and tectonic features. Based on Sheth et al. (2003). (b) Simplified geological map of Mauritius, showing the presently recognized areal distribution (Paul et al., 2007) of the Older, Intermediate, and Younger Series lava flows (as defined by Baxter, 1972, 1975, 1976), their K–Ar ages in millions of years (McDougall and Chamalaun, 1969), and the locations of the geological features described in this study. Small islets offshore Mauritius are not shown.
Fig. 2. Photographs showing the Chamarel seven-coloured earth landscape. (a) Dendritic drainage pattern, with rounded interfluves and V-shaped valleys. (b, c, d) The various colour shades. (c) A minor watershed crossing the area, and drainage going off to both right and left, illustrating the ’normal’ dendritic drainage pattern of the rills. It also shows the occasional bits of gravel (apparently white) along the drainage lines. (d) This shows the general steepness of the ground. Note the great contrast between the bare area and the forest beyond in (a) through (d).
H.C. Sheth et al. / Journal of African Earth Sciences 57 (2010) 169–173
171
Fig. 3. (a) The Chamarel Falls plunging over an arcuate cliff, with the two lava flows marked. Note the forested landscape in the background. (b) A forested hill in the background in distinct contrast with the bare rilled landscape in the foreground. (c) Saprock right at the surface, with jointing, and the rills containing bedload. (b, c, d) Abundant rubble on the interfluves.
the Grand Bassin (Ganga Talao, Fig. 1) as a volcanic crater lake. This is only 12.5 km from the Chamarel seven-coloured earth. Another volcanic crater (Trou aux Cerfs) is approximately 20 km from the Chamarel site. We have visited both. At Grand Bassin, the thick vegetation and extensive construction related to its Hindu shrines effectively hide all rock outcrops, and we are unable to say whether it is a volcanic crater or not. Trou aux Cerfs is certainly volcanic. Saddul (1995) considers it a monogenetic volcanic cone with a central vent, formed during the late stages of volcanism on the island, 700,000–600,000 years before the present. It is certainly well preserved, though thick vegetation masks rock outcrops here as well. We believe that it may be a monogenetic cinder or tuff cone, like the many such cones that cap and postdate oceanic shield volcanoes such as those of Hawaii or Easter Island. Trou aux Cerf also resembles, in its morphology and size (a few hundred metres), pit craters on Kilauea and other Hawaiian basaltic shield volcanoes. Pit craters are circular or elliptical, and though called ‘‘craters”, they do not erupt lava or ash, but are formed by dislodgement of the solidified roofs of large lava tubes and their collapse into the magma beneath (e.g., Macdonald et al., 1983; Sheth, 2006). Overall, however, we consider Saddul’s (1995) proposal, that the seven-coloured earth is developed from altered volcanic tuffs, to be untenable. The sharp contact between ‘‘tuffs” and ‘‘lavas” which he mentions may actually be a weathering front, described later, and the material exposed at Chamarel does not have the characteristics of tuff but of weathered basalt. 3. The seven-coloured earth: site description The Chamarel seven-coloured earth is a landscape with sheet erosion and rill formation. The landforms are typical small-scale fluvial landforms, with gently convex slopes, and rounded interfluves separated by sharply defined watercourses with broad Vshaped cross sections (Fig. 2a–d). The streams have tributaries in
the normal dendritic way, i.e., the tributaries join the main stream at acute angles. Despite some partial superficial resemblance the landscape has no relation to dunes, and is not really badlands, in which the rills tend to be much deeper and the interfluves sharp and angular. Some of the little valleys have a ‘‘bedload”, just a few grains wide, of fragments of saprock. The bare surfaces expose saprolite, the highly weathered remains of basalt, in place, which shows the famous colours. The total absence of vegetation (apart from one or two small plants) at the site (Figs. 2a–d and 3b–d) is most unusual, given the tropical climate of Mauritius with warm, dry winters and hot, humid summers, and an abundance of rain. In such a climate, vegetation grows surprisingly rapidly on weathered basalt, as the authors’ combined observations in basaltic terrains such as the Deccan plateau (India), Africa, and elsewhere can confirm (e.g., Ollier, 1984; Ollier and Powar, 1985; Ollier and Pain, 1996; Ollier and Sheth, 2008). The barrenness of this small area in a dense forest immediately strikes one as artificial. It gives the impression that the jungle has been cleared manually at this spot, and that the owners of this property periodically pluck out any new vegetation that appears. There is not much soil cover at the site. The upper saprolite is completely structureless. The saprock (basalt rock intermediate in depth and degree of weathering between saprolite above and fresh bedrock below) is right at the surface in places, and shows the joints in the original basalt intact (Fig. 3c). This can be called structured saprolite. Such saprock or saprolite is common on basalts everywhere (e.g., the Deccan highlands, India; Ollier and Sheth, 2008). We suppose (as we cannot reach the outcrops) that the rubblylooking material at the surface (Fig. 3c and d) is of harder saprock. 4. The colours It so happened that there was a drizzle immediately after we reached the site, and we particularly observed if this affected the
172
H.C. Sheth et al. / Journal of African Earth Sciences 57 (2010) 169–173
Fig. 4. (a, b) Brilliantly and variously coloured saprolite profile in a section at the top of the Deccan plateau in the Satara region (between Sadavaghapur and Borgevadi villages, see Ollier and Sheth, 2008), Western Ghats, India.
surface colours. Apart from the darkening due to moistening, it did not, inspiring confidence that the colours are natural. But how do these colours originate? Largely due to the chemical weathering of the basalt and the formation of iron oxides and hydroxides that impart multiple colours to the weathered rock. The weathering product of basalt is clay. We think the surface material is a mix of clay minerals and iron oxides and hydroxides. A signboard at the site gives the following scientific explanation: ‘‘This natural phenomenon is due to decomposed basalt gullies. The hot and humid climate helps in the decomposition of the basalt into clay. As a result of total hydrolysis (chemical breakdown of minerals by water), the soluble elements such as silicic acid and cations are washed, leaving a large composition of iron and aluminium which constitute a ferralitic soil. The iron sesquioxides (Fe2O3) have a red and anthracite colour, whereas the aluminium sesquioxides (Al2O3) have a blue or purplish colour.” This appears more or less correct except that it is not clear what is meant by anthracite, which is mature, shiny, pitch-black coal. Perhaps the reference is to the grey colour of the earth in patches, but iron oxides are red or brown, not grey. In basalt weathering smectites usually form first, and kaolinites later (both are types of clay minerals). But it is not the clay mineral itself that gives the colour; it is the iron-oxide films at microscopic level. It takes a remarkably small amount of iron oxide to give a red colour to the saprolite or soil. Usually the more the hydroxides the yellower the material. We note that ferralitic soils are a major group of soils in the French soil classification, and also in the UNESCO soil classification (Eswaren et al., 2003). Finally, the range of colours at Chamarel is not really exceptional. We can present many examples of brilliantly coloured weathered basalt outcrops, or duricrust-saprolite-basalt profiles, from areas such as the Deccan, India (e.g., Ollier and Sheth, 2008). A very striking example is presented in Fig. 4a and b. 5. The origin of Chamarel seven-coloured earth We believe that the Chamarel seven-coloured earth is partly natural (basalt weathering) and partly artificial (total absence of vegetation in the middle of a forest). How does this landscape originate? Huggett (2003, p. 182) notes that much current gullying apparently results from human modification of the land surface leading to disequilibrium in the hillslope system. We think the area would originally have been forested like the surroundings. Human clearing and burning reduced it to bare land. A hypothesis we propose is that this particular spot was probably cleared to grow crops of
Fig. 5. Red, rilled ground in a village in Papua New Guinea. Photo courtesy Dr. Colin F. Pain.
some sort, but a severe storm came at the wrong time, removed the topsoil by sheetwash erosion, and cut rills. Water can flow over a surface as a thin sheet, and is capable of carrying suspended particles. But water cannot flow very far before it becomes channelized. Rills go from the time sheet flow turns into stream flow in distinct linear channels, until some arbitrary time when we call the result a gully, or a stream. Note that Huggett (2003) defines ‘‘rills” as tiny hillside channels a few centimetres wide, deep cut by ephemeral rivulets. An arbitrary upper size limit for rills given by him is less than one third of a metre wide and two-thirds of a metre deep. He would call any fluvial hillside channel larger than that a ‘‘gully”, and still larger incised stream beds ‘‘arroyos”. We think that Huggett’s boundary between rills and gullies is on the lower side, and that the channels at Chamarel are better described as rills than gullies. In any case, once the topsoil at this site was removed, by sheetwash, the land was no longer suitable for the proposed crops. Safford (1997) writes that Mauritius was originally completely covered by wet or dry evergreen forest and scrub, and palm savanna. Habitat destruction following human colonization in 1638 resulted in the reduction of native vegetation cover on the mainland. Humans sent the native bird dodo into extinction by 1681 (Ash, 2006). Subsequently the colonists have modified habitats on Mauritius at an alarming rate. In the 19th century, large areas of dense Mauritius forests were cleared and converted wholesale into sugarcane plantations. A huge wave of mass migration of labourers from India to work in these fields took place in the second half of the 19th century. By this time, habitat modification
H.C. Sheth et al. / Journal of African Earth Sciences 57 (2010) 169–173
on Mauritius had reached almost every corner of the island. In fact, Safford (1997) estimates that only 5% of the original native vegetation cover remained on Mauritius by 1993. Although it is clear that human impact has much to do with the origin of the bare and rilled landscape of Chamarel, it is probably maintained in part because erosion has exposed the saprolite, and active and continued erosion by rain and runoff provide very little opportunity for vegetation to reestablish itself. We think this topography is what one could get in many parts of Mauritius (or any other area of deeply weathered basalt) if the vegetation were removed and the rain and runoff were allowed to work on a bare surface. Intense runoff is probably all that is required, but humans can help (!) by preventing new vegetation growth. Google Earth images of the area show red-coloured bare areas near the Chamarel site, and these may be forerunners of the seven-coloured earth. This kind of bare, rilled landscape may also be expected to occur elsewhere where climate and geomorphic conditions are similar. In fact, the Chamarel landscape does have counterparts elsewhere. Fig. 5 shows what is evidently the same feature, in Papua New Guinea. These small bare areas with rounded rills, formed on weathered volcanic rock, are either in villages or in old village sites. They too have their origins in human activity, and because of topsoil stripping and saprolite exposure, and continued erosion, they persist in the landscape for long periods.
6. Summary The Chamarel seven-coloured earth of Mauritius is a phenomenon of weathering of basalt associated with rill development due to torrential rain and topsoil removal, along with periodic elimination of vegetation. The rilling itself is a result of human modification of the landscape, namely deforestation. In that sense, the Chamarel seven-coloured earth is partly natural and partly artificial. We hope that this article will trigger many more studies of the geology, geomorphology and the natural environment of Mauritius. Today the tourism industry in Mauritius is much more lucrative than the sugar industry, and we also note that all over the world there is an increase in eco-tourism and geo-tourism (e.g., Eder, 2008). It is important that the scientific basis of tourist sites should be correctly explained, to enhance the interest of the tourists, and also to help in developing an ecofriendly management system. We offer this contribution as an example of improving
173
the perception of an important tourist attraction and geological curiosity. Chamarel-type geomorphic features have really not been described in any detail before, perhaps because they are always of small areal extent, and seemingly unimportant. Observations suggest, however, that such features may be quite common in tropical regions with high rainfall and volcanic bedrock. Acknowledgements We appreciate the constructive journal reviews by Colin Pain and James Terry, and the editorial handling of P. Eriksson. References Ash, R., 2006. Whitaker’s World of Facts. Penguin Books. 320 p. Baxter, A.N., 1972. Magmatic evolution of Mauritius, western Indian Ocean. Ph.D. Thesis, University of Edinburgh, 177 p. Baxter, A.N., 1975. Petrology of the older series lavas from Mauritius Indian Ocean. Geological Society of America Bulletin 86, 1449–1458. Baxter, A.N., 1976. Geochemistry and petrogenesis of primitive alkali basalt from Mauritius Indian Ocean. Geological Society of America Bulletin 87, 1028–1034. Eder, W., 2008. Geoparks – promotion of earth sciences through geoheritage conservation, education and tourism. Journal of Geological Society of India 72, 149–154. Eswaren, H., Rice, T.J., Ahrens, R., Stewart, B.A., 2003. Soil Classification: A Global Desk Reference. CRC Press, Boca Raton. Huggett, R.J., 2003. Fundamentals of Geomorphology. Routledge. 386 p. Macdonald, G.A., Abbott, A.T., Peterson, F.L., 1983. Volcanoes in the Sea: The Geology of Hawaii, 2nd ed. University of Hawaii Press, Honolulu. 517 p. McDougall, I., Chamalaun, F.G., 1969. Isotopic dating and geomagnetic polarity studies on volcanic rocks from Mauritius, Indian Ocean. Geological Society of America Bulletin 80, 1419–1442. Ollier, C.D., 1984. Weathering, 2nd ed. Longman, Harlow, Sussex. Ollier, C.D., Pain, C.F., 1996. Regolith, Soils and Landforms. Wiley, Chichester. Ollier, C.D., Powar, K.B., 1985. The Western Ghats and the morphotectonics of peninsular India. Zeitschrift für Geomorphologie Supplement N. F. 54, 57–69. Ollier, C.D., Sheth, H.C., 2008. The high Deccan duricrusts of India, and their significance for the ‘laterite’ issue. Journal of Earth System Science 117, 1–15. Paul, D., Kamenetsky, V.S., Hofmann, A.W., Stracke, A., 2007. Compositional diversity among primitive lavas of Mauritius, Indian Ocean: implications for mantle sources. Journal of Volcanology and Geothermal Research 164, 76– 94. Saddul, P., 1995. Mauritius: A Geomorphological Analysis. Mahatma Gandhi Institute Press, Moka, Mauritius. 354 p. Safford, R.J., 1997. A survey of the occurrence of native vegetation remnants on Mauritius in 1993. Biological Conservation 80, 181–188. Sheth, H.C., 2006. The emplacement of pahoehoe lavas on Kilauea and in the Deccan Traps. Journal of Earth System Science 115, 615–629. Sheth, H.C., Mahoney, J.J., Baxter, A.N., 2003. Geochemistry of lavas from Mauritius, Indian Ocean: Mantle sources and petrogenesis. International Geology Review 45, 780–797.
Contents lists available at ScienceDirect
Journal of African Earth Sciences journal homepage: www.elsevier.com/locate/jafrearsci
The ‘‘seven-coloured earth” of Chamarel, Mauritius H.C. Sheth a,*, C.P. Johnson b, C.D. Ollier c a
Department of Earth Sciences, Indian Institute of Technology Bombay (IITB), Powai, Mumbai 400 076, India C-DAC School of Advanced Computing (C-SAC), Quatre Bornes, Mauritius c School of Earth and Geographical Sciences, The University of Western Australia, Nedlands, WA 6009, Australia b
a r t i c l e
i n f o
Article history: Received 11 March 2009 Received in revised form 21 July 2009 Accepted 23 July 2009 Available online 6 August 2009 Keywords: Mauritius Chamarel Basalt Weathering Rills Coloured earth
a b s t r a c t The ‘‘seven-coloured earth” of Chamarel is a geological curiosity and a major tourist attraction of Mauritius. This is a small (7500 m2) area of strikingly bare landscape showing well-developed rills and various shades of red, brown, grey, and purple. Curiously, it is located within a large, dense forest. Prevalent misconceptions are that the landscape formed due to a volcanic eruption, or from volcanic ash. Whereas the bedrock is undoubtedly an old volcanic rock (basalt), the colours are due to weathering of the basalt and the formation of secondary iron oxides and hydroxides in it, and the rilling is a result of deforestation and sheet erosion, i.e., human modification of the landscape. Such features, inadequately described in the literature so far, also occur in Papua New Guinea, and may be common in tropical, high-rainfall regions with volcanic bedrock. Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction The ‘‘seven-coloured earth” of Chamarel is a geological curiosity and one of the major tourist attractions in Mauritius, a small African island country in the Indian Ocean (Fig. 1) that has a growing tourism industry. The seven-coloured earth is a small (7500 m2) site in the Black River district of southwestern Mauritius. Though in the heart of a large, dense forest, it is strikingly devoid of any vegetation. It is a landscape with severe rill erosion and surface colouration in various shades of red, brown, grey, and purple that gives it its name (Fig. 2a–d). The site is on private land and a fee (75 Mauritian rupees per person in June 2008) is charged for entry to the seven-coloured earth as well as the beautiful Chamarel waterfall on the same premises. Searching ‘‘Chamarel” in Google, or in the scientific literature (e.g., Elsevier’s ScienceDirect) produced no results of a technical nature. We present the geological and geomorphological background to the Chamarel seven-coloured earth, so that this, and similar features known elsewhere but insufficiently described, are better understood. 2. Some misconceptions Widespread misconceptions exist about the seven-coloured earth. Links to volcanoes are indicated by enthusiastic tourists. A * Corresponding author. Tel.: +91 22 25767264. E-mail address: [email protected] (H.C. Sheth). 1464-343X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jafrearsci.2009.07.009
tour guide we met at the site told us very confidently that the different colours (and the puzzling total absence of vegetation we pointed out to him) were ‘‘due to a volcanic eruption”. A particular traveller’s website that we hit with Google (by searching ‘‘Chamarel”) says that the different colours are developed due to different lava flows cooling at different temperatures. These conceptions are quite mistaken. The island of Mauritius is indeed volcanic, but volcanism ended long ago. The shield stage of volcanism, which built up most of the island itself, is 8 million years old, and the youngest volcanic activity occurred 30,000 years ago (McDougall and Chamalaun, 1969; Baxter, 1972, 1975, 1976; Sheth et al., 2003; Paul et al., 2007). The bedrock at Chamarel is volcanic rock (basalt) that has been considerably weathered. The basalt lithology cannot be ascertained at the site itself, as this site is fenced off, denying access to tourists. However, 1.3 km away the spectacular Chamarel Falls plunge a hundred metres or so over two cliff-forming basalt lava flows, the upper one of which shows well-developed columnar jointing (Fig. 3a). Saddul (1995) considers the seven-coloured earth as formed on altered volcanic tuffs. He writes that the tuffs are as much as 18–20 m thick and overlie basalt lavas with a sharp contact. He speculates that the volcanic tuffs are derived from the Bassin Blanc volcano nearby (Fig. 1), which we have not visited. His view is echoed in The Mauritius Telephone Directory ‘‘tourist information”, which states: ‘‘These slivers of colours are believed to be the result of the erosion of the volcanic ash”. This sentence confuses erosion, meaning removal of rock material, with weathering or alteration, which happens to the rock in place. The same Directory refers to
170
H.C. Sheth et al. / Journal of African Earth Sciences 57 (2010) 169–173
Fig. 1. (a) Map of the Indian Ocean region showing the locations of the three Mascarene Islands (Réunion, Mauritius, and Rodrigues) and major physical and tectonic features. Based on Sheth et al. (2003). (b) Simplified geological map of Mauritius, showing the presently recognized areal distribution (Paul et al., 2007) of the Older, Intermediate, and Younger Series lava flows (as defined by Baxter, 1972, 1975, 1976), their K–Ar ages in millions of years (McDougall and Chamalaun, 1969), and the locations of the geological features described in this study. Small islets offshore Mauritius are not shown.
Fig. 2. Photographs showing the Chamarel seven-coloured earth landscape. (a) Dendritic drainage pattern, with rounded interfluves and V-shaped valleys. (b, c, d) The various colour shades. (c) A minor watershed crossing the area, and drainage going off to both right and left, illustrating the ’normal’ dendritic drainage pattern of the rills. It also shows the occasional bits of gravel (apparently white) along the drainage lines. (d) This shows the general steepness of the ground. Note the great contrast between the bare area and the forest beyond in (a) through (d).
H.C. Sheth et al. / Journal of African Earth Sciences 57 (2010) 169–173
171
Fig. 3. (a) The Chamarel Falls plunging over an arcuate cliff, with the two lava flows marked. Note the forested landscape in the background. (b) A forested hill in the background in distinct contrast with the bare rilled landscape in the foreground. (c) Saprock right at the surface, with jointing, and the rills containing bedload. (b, c, d) Abundant rubble on the interfluves.
the Grand Bassin (Ganga Talao, Fig. 1) as a volcanic crater lake. This is only 12.5 km from the Chamarel seven-coloured earth. Another volcanic crater (Trou aux Cerfs) is approximately 20 km from the Chamarel site. We have visited both. At Grand Bassin, the thick vegetation and extensive construction related to its Hindu shrines effectively hide all rock outcrops, and we are unable to say whether it is a volcanic crater or not. Trou aux Cerfs is certainly volcanic. Saddul (1995) considers it a monogenetic volcanic cone with a central vent, formed during the late stages of volcanism on the island, 700,000–600,000 years before the present. It is certainly well preserved, though thick vegetation masks rock outcrops here as well. We believe that it may be a monogenetic cinder or tuff cone, like the many such cones that cap and postdate oceanic shield volcanoes such as those of Hawaii or Easter Island. Trou aux Cerf also resembles, in its morphology and size (a few hundred metres), pit craters on Kilauea and other Hawaiian basaltic shield volcanoes. Pit craters are circular or elliptical, and though called ‘‘craters”, they do not erupt lava or ash, but are formed by dislodgement of the solidified roofs of large lava tubes and their collapse into the magma beneath (e.g., Macdonald et al., 1983; Sheth, 2006). Overall, however, we consider Saddul’s (1995) proposal, that the seven-coloured earth is developed from altered volcanic tuffs, to be untenable. The sharp contact between ‘‘tuffs” and ‘‘lavas” which he mentions may actually be a weathering front, described later, and the material exposed at Chamarel does not have the characteristics of tuff but of weathered basalt. 3. The seven-coloured earth: site description The Chamarel seven-coloured earth is a landscape with sheet erosion and rill formation. The landforms are typical small-scale fluvial landforms, with gently convex slopes, and rounded interfluves separated by sharply defined watercourses with broad Vshaped cross sections (Fig. 2a–d). The streams have tributaries in
the normal dendritic way, i.e., the tributaries join the main stream at acute angles. Despite some partial superficial resemblance the landscape has no relation to dunes, and is not really badlands, in which the rills tend to be much deeper and the interfluves sharp and angular. Some of the little valleys have a ‘‘bedload”, just a few grains wide, of fragments of saprock. The bare surfaces expose saprolite, the highly weathered remains of basalt, in place, which shows the famous colours. The total absence of vegetation (apart from one or two small plants) at the site (Figs. 2a–d and 3b–d) is most unusual, given the tropical climate of Mauritius with warm, dry winters and hot, humid summers, and an abundance of rain. In such a climate, vegetation grows surprisingly rapidly on weathered basalt, as the authors’ combined observations in basaltic terrains such as the Deccan plateau (India), Africa, and elsewhere can confirm (e.g., Ollier, 1984; Ollier and Powar, 1985; Ollier and Pain, 1996; Ollier and Sheth, 2008). The barrenness of this small area in a dense forest immediately strikes one as artificial. It gives the impression that the jungle has been cleared manually at this spot, and that the owners of this property periodically pluck out any new vegetation that appears. There is not much soil cover at the site. The upper saprolite is completely structureless. The saprock (basalt rock intermediate in depth and degree of weathering between saprolite above and fresh bedrock below) is right at the surface in places, and shows the joints in the original basalt intact (Fig. 3c). This can be called structured saprolite. Such saprock or saprolite is common on basalts everywhere (e.g., the Deccan highlands, India; Ollier and Sheth, 2008). We suppose (as we cannot reach the outcrops) that the rubblylooking material at the surface (Fig. 3c and d) is of harder saprock. 4. The colours It so happened that there was a drizzle immediately after we reached the site, and we particularly observed if this affected the
172
H.C. Sheth et al. / Journal of African Earth Sciences 57 (2010) 169–173
Fig. 4. (a, b) Brilliantly and variously coloured saprolite profile in a section at the top of the Deccan plateau in the Satara region (between Sadavaghapur and Borgevadi villages, see Ollier and Sheth, 2008), Western Ghats, India.
surface colours. Apart from the darkening due to moistening, it did not, inspiring confidence that the colours are natural. But how do these colours originate? Largely due to the chemical weathering of the basalt and the formation of iron oxides and hydroxides that impart multiple colours to the weathered rock. The weathering product of basalt is clay. We think the surface material is a mix of clay minerals and iron oxides and hydroxides. A signboard at the site gives the following scientific explanation: ‘‘This natural phenomenon is due to decomposed basalt gullies. The hot and humid climate helps in the decomposition of the basalt into clay. As a result of total hydrolysis (chemical breakdown of minerals by water), the soluble elements such as silicic acid and cations are washed, leaving a large composition of iron and aluminium which constitute a ferralitic soil. The iron sesquioxides (Fe2O3) have a red and anthracite colour, whereas the aluminium sesquioxides (Al2O3) have a blue or purplish colour.” This appears more or less correct except that it is not clear what is meant by anthracite, which is mature, shiny, pitch-black coal. Perhaps the reference is to the grey colour of the earth in patches, but iron oxides are red or brown, not grey. In basalt weathering smectites usually form first, and kaolinites later (both are types of clay minerals). But it is not the clay mineral itself that gives the colour; it is the iron-oxide films at microscopic level. It takes a remarkably small amount of iron oxide to give a red colour to the saprolite or soil. Usually the more the hydroxides the yellower the material. We note that ferralitic soils are a major group of soils in the French soil classification, and also in the UNESCO soil classification (Eswaren et al., 2003). Finally, the range of colours at Chamarel is not really exceptional. We can present many examples of brilliantly coloured weathered basalt outcrops, or duricrust-saprolite-basalt profiles, from areas such as the Deccan, India (e.g., Ollier and Sheth, 2008). A very striking example is presented in Fig. 4a and b. 5. The origin of Chamarel seven-coloured earth We believe that the Chamarel seven-coloured earth is partly natural (basalt weathering) and partly artificial (total absence of vegetation in the middle of a forest). How does this landscape originate? Huggett (2003, p. 182) notes that much current gullying apparently results from human modification of the land surface leading to disequilibrium in the hillslope system. We think the area would originally have been forested like the surroundings. Human clearing and burning reduced it to bare land. A hypothesis we propose is that this particular spot was probably cleared to grow crops of
Fig. 5. Red, rilled ground in a village in Papua New Guinea. Photo courtesy Dr. Colin F. Pain.
some sort, but a severe storm came at the wrong time, removed the topsoil by sheetwash erosion, and cut rills. Water can flow over a surface as a thin sheet, and is capable of carrying suspended particles. But water cannot flow very far before it becomes channelized. Rills go from the time sheet flow turns into stream flow in distinct linear channels, until some arbitrary time when we call the result a gully, or a stream. Note that Huggett (2003) defines ‘‘rills” as tiny hillside channels a few centimetres wide, deep cut by ephemeral rivulets. An arbitrary upper size limit for rills given by him is less than one third of a metre wide and two-thirds of a metre deep. He would call any fluvial hillside channel larger than that a ‘‘gully”, and still larger incised stream beds ‘‘arroyos”. We think that Huggett’s boundary between rills and gullies is on the lower side, and that the channels at Chamarel are better described as rills than gullies. In any case, once the topsoil at this site was removed, by sheetwash, the land was no longer suitable for the proposed crops. Safford (1997) writes that Mauritius was originally completely covered by wet or dry evergreen forest and scrub, and palm savanna. Habitat destruction following human colonization in 1638 resulted in the reduction of native vegetation cover on the mainland. Humans sent the native bird dodo into extinction by 1681 (Ash, 2006). Subsequently the colonists have modified habitats on Mauritius at an alarming rate. In the 19th century, large areas of dense Mauritius forests were cleared and converted wholesale into sugarcane plantations. A huge wave of mass migration of labourers from India to work in these fields took place in the second half of the 19th century. By this time, habitat modification
H.C. Sheth et al. / Journal of African Earth Sciences 57 (2010) 169–173
on Mauritius had reached almost every corner of the island. In fact, Safford (1997) estimates that only 5% of the original native vegetation cover remained on Mauritius by 1993. Although it is clear that human impact has much to do with the origin of the bare and rilled landscape of Chamarel, it is probably maintained in part because erosion has exposed the saprolite, and active and continued erosion by rain and runoff provide very little opportunity for vegetation to reestablish itself. We think this topography is what one could get in many parts of Mauritius (or any other area of deeply weathered basalt) if the vegetation were removed and the rain and runoff were allowed to work on a bare surface. Intense runoff is probably all that is required, but humans can help (!) by preventing new vegetation growth. Google Earth images of the area show red-coloured bare areas near the Chamarel site, and these may be forerunners of the seven-coloured earth. This kind of bare, rilled landscape may also be expected to occur elsewhere where climate and geomorphic conditions are similar. In fact, the Chamarel landscape does have counterparts elsewhere. Fig. 5 shows what is evidently the same feature, in Papua New Guinea. These small bare areas with rounded rills, formed on weathered volcanic rock, are either in villages or in old village sites. They too have their origins in human activity, and because of topsoil stripping and saprolite exposure, and continued erosion, they persist in the landscape for long periods.
6. Summary The Chamarel seven-coloured earth of Mauritius is a phenomenon of weathering of basalt associated with rill development due to torrential rain and topsoil removal, along with periodic elimination of vegetation. The rilling itself is a result of human modification of the landscape, namely deforestation. In that sense, the Chamarel seven-coloured earth is partly natural and partly artificial. We hope that this article will trigger many more studies of the geology, geomorphology and the natural environment of Mauritius. Today the tourism industry in Mauritius is much more lucrative than the sugar industry, and we also note that all over the world there is an increase in eco-tourism and geo-tourism (e.g., Eder, 2008). It is important that the scientific basis of tourist sites should be correctly explained, to enhance the interest of the tourists, and also to help in developing an ecofriendly management system. We offer this contribution as an example of improving
173
the perception of an important tourist attraction and geological curiosity. Chamarel-type geomorphic features have really not been described in any detail before, perhaps because they are always of small areal extent, and seemingly unimportant. Observations suggest, however, that such features may be quite common in tropical regions with high rainfall and volcanic bedrock. Acknowledgements We appreciate the constructive journal reviews by Colin Pain and James Terry, and the editorial handling of P. Eriksson. References Ash, R., 2006. Whitaker’s World of Facts. Penguin Books. 320 p. Baxter, A.N., 1972. Magmatic evolution of Mauritius, western Indian Ocean. Ph.D. Thesis, University of Edinburgh, 177 p. Baxter, A.N., 1975. Petrology of the older series lavas from Mauritius Indian Ocean. Geological Society of America Bulletin 86, 1449–1458. Baxter, A.N., 1976. Geochemistry and petrogenesis of primitive alkali basalt from Mauritius Indian Ocean. Geological Society of America Bulletin 87, 1028–1034. Eder, W., 2008. Geoparks – promotion of earth sciences through geoheritage conservation, education and tourism. Journal of Geological Society of India 72, 149–154. Eswaren, H., Rice, T.J., Ahrens, R., Stewart, B.A., 2003. Soil Classification: A Global Desk Reference. CRC Press, Boca Raton. Huggett, R.J., 2003. Fundamentals of Geomorphology. Routledge. 386 p. Macdonald, G.A., Abbott, A.T., Peterson, F.L., 1983. Volcanoes in the Sea: The Geology of Hawaii, 2nd ed. University of Hawaii Press, Honolulu. 517 p. McDougall, I., Chamalaun, F.G., 1969. Isotopic dating and geomagnetic polarity studies on volcanic rocks from Mauritius, Indian Ocean. Geological Society of America Bulletin 80, 1419–1442. Ollier, C.D., 1984. Weathering, 2nd ed. Longman, Harlow, Sussex. Ollier, C.D., Pain, C.F., 1996. Regolith, Soils and Landforms. Wiley, Chichester. Ollier, C.D., Powar, K.B., 1985. The Western Ghats and the morphotectonics of peninsular India. Zeitschrift für Geomorphologie Supplement N. F. 54, 57–69. Ollier, C.D., Sheth, H.C., 2008. The high Deccan duricrusts of India, and their significance for the ‘laterite’ issue. Journal of Earth System Science 117, 1–15. Paul, D., Kamenetsky, V.S., Hofmann, A.W., Stracke, A., 2007. Compositional diversity among primitive lavas of Mauritius, Indian Ocean: implications for mantle sources. Journal of Volcanology and Geothermal Research 164, 76– 94. Saddul, P., 1995. Mauritius: A Geomorphological Analysis. Mahatma Gandhi Institute Press, Moka, Mauritius. 354 p. Safford, R.J., 1997. A survey of the occurrence of native vegetation remnants on Mauritius in 1993. Biological Conservation 80, 181–188. Sheth, H.C., 2006. The emplacement of pahoehoe lavas on Kilauea and in the Deccan Traps. Journal of Earth System Science 115, 615–629. Sheth, H.C., Mahoney, J.J., Baxter, A.N., 2003. Geochemistry of lavas from Mauritius, Indian Ocean: Mantle sources and petrogenesis. International Geology Review 45, 780–797.