seaturtle.org : MTN : ARCHIVES : Sign In

Oceanic beaches are continuously changing as natural forces of wind and water meet land. These changes, which have been taking place for millions of years, are linked to variation in wind, waves, currents and sea level. It is not only natural forces that change beaches, however, human activities do as well, especially stone mining and sand harvesting, construction of buildings and walls close to the sea, and beach tourist activities (Nelson & Dickerson 1988). Mature marine female turtles nest on specific beaches, that is, their natal beach where they were incubated and hatched in the past (Bowen et al. 1989). Consequently, many mature females may fail to nest in suitable areas if their natal nesting habitats have been destroyed. Loss of suitable nesting habitats has been identified as one of the major threats facing the recovery of sea turtle populations in Kenya (UNEP 1998). Encroachment by manmade structures prevents the natural movement of sand, thus resulting in the degradation of beach habitats through factors such as beach erosion.

Much of the Kenya coastal zone has experienced rapid human population growth, spiraling demand for marine resources and unplanned development (UNEP 1998). Despite protection by the Kenya government (Fisheries Act 1989; and the Wildlife Management and Conservation Amendment Act 1989), illegal harvesting of adult nesting female turtles and egg collection are rampant (Olendo 1993).

In view of these anthropogenic activities and natural beach processes it is important to study the impact of beach development, tourism, and beach erosion and accretion on sea turtle nesting habitats. These activities and processes affect the biophysical characteristics of beaches, which may in turn affect the beaches’ suitability for turtle nesting. Previous studies, (e.g., Horrocks & Scott 1991; Kamel & Mrosovsky 2004; Marcus & Maley 1987; Mortimer 1990), have shown that suitability for nesting by turtles depends on the following biophysical characteristics; beach vegetation, offshore approach, beach slope, beach width and beach material. This study therefore determined how biophysical characteristics have changed in different locations at the beaches between 1986 and 2006, and in turn, how these changes have impacted the availability of potential suitable sites for nesting. The study was conducted on about 45 km of shoreline on the western Indian Ocean (Fig. 1) using field surveys and satellite imagery. Ground-truthing, i.e., the identification of features on the beach to aid the interpretation of satellite imagery was conducted from September 2006 to December 2006.

Figure 1. Location of marine turtle nesting beaches along the Mombasa-Kilifi shoreline in Kenya.

The nature of the offshore approach was characterized as the presence or absence of any obstructions such as rocks or strewn boulders from the low water mark to the start of the beach vegetation line shoreward. Presence of such obstructions may prevent marine turtle emergence onto the beach. The beaches were then classified as either having an open offshore approach or an obstructed approach (including partial obstruction). The study included partially obstructed with obstructed approaches so there were only two categories, i.e., clear/open and obstructed.

Beach width was determined as the horizontal distance between the low water mark (the start of the foreshore) and the start of the beach vegetation line. In the absence of vegetation, the distance was measured to the nearest beach structures at the landward edge of the sandy beach. We measured width using a tape measure, and recorded the presence or absence of vegetation on the nesting beaches.

The composition of beach substrate was determined from sand samples taken from all beaches. Three samples were collected from a random area of the beach, at nest depth using a soil auger. These samples were then thoroughly mixed to make a representative sample and a sub-sample of 200 g was taken. Granulometric analysis was then conducted on the sub-samples and a mean particle diameter for each beach was determined. The Wentworth scale of particle diameter (Penthick 1984) was then used to determine the type of beach material at each beach. To determine the slope of the beach, the Wentworth scale for slope (Penthick 1984) was used.

Existing nesting beaches were mapped using coordinates determined by a GPS during ground-truthing. These coordinates were then entered into ArcGIS 9.2 and polygons representing the nesting beaches were created. Existing nesting beaches were identified by the presence of nests observed during the field survey and information from key informants (fishermen and turtle conservation group members).

Potential suitable nesting sites in 1986, 2001 and 2006, were mapped by identifying sites that had 1) sandy beaches and open offshore approaches; 2) slopes greater than 4° and less than 12.5°; and 3) were approximately perpendicularly located 30 m away landwards from the shoreline. This identification was done from a combination of classified Landsat imagery and a GIS-generated slope map. Three separate map layers representing approach, slope and beach material were created. These maps were then combined to distinguish suitable and unsuitable areas.

Contours were digitized from a topographic map and a Digital Elevation Model (DEM) was created. A slope image derived from the DEM was then reclassified to create a Boolean image where a value of 1 was given to areas with slopes between 4° and 12.5°, and a value of 0 was assigned to all other areas. A third criterion map was created for areas 30 m landwards away from the shoreline by first digitizing the shoreline for the entire study area and creating a 30 m buffer. A Boolean image was created with areas 30 m landward away from the shoreline having a value of 1 and those less than 30 m being assigned a value of 0.

An overlay operation (multiply option) was then performed where the three Boolean criteria images were imposed on each other. The resulting image showed areas with suitable nesting beaches with a value of 1 and those beaches that were unsuitable habitats had a value of 0. This was done for both 1986 and 2001.

To determine the extent and pattern of change in the nesting beaches between 1986, 2001 and 2006, an overlay operation (subtraction option) was carried out. The 2001 map of suitable nesting beaches was overlaid on the 1986 map and again for 2006 compared to the 2001 map. The Area module in Idrisi 3.2 was used to calculate the changes in size of suitable nesting beaches.

A combination of GIS-based change detection methods and field surveys was used to link changes in land use and land cover characteristics to the changes in the sizes of nesting beaches. Field surveys were used to obtain complementary information on possible causes of land use and land cover in the study area. A total of 109 nests of three turtle species: green turtle, hawksbill turtle and olive ridley turtle were recorded (Fig. 2). These nests were observed on seven beaches that were being used for nesting during the study.

Figure 2. Distribution of nests along the study beaches.

All beaches used for nesting were found to be composed of sandy surface material. No nesting activity was observed on beaches with muddy or rocky surfaces. Of the 109 nests observed, 59 were found on beaches with medium-grain sand, 26 nests on coarse sand and 24 on fine sand beaches. Of the seven beaches that were used for nesting, 3 had a medium-grain sand surface, 2 had fine sand and 2 had a coarse sand surface (Table 1).

Table 1. Mean sandy soil grain diameter for each beach.

Beach slopes that were used for nesting ranged from 5° to 9°, while beach width varied from 31 m to 60 m (Table 2). Most of the beaches used for nesting had open offshore approaches consisting of sand. A small section at English Point Beach, however, had obstacles mainly consisting of debris washed ashore and concrete blocks from destroyed sea walls. We also observed that a section between English Point and Nyali Beach had a sandy beach but it was not being used for nesting because there was a rocky approach.

Table 2. Beach slope and width and the number of nests on each beach in the study area in 2006.

Nests were mostly found where some vegetation was present. About 80% of the nests were found at the vegetation line while 10% of the nests were found on the open beach. The remaining 10% were found within the vegetation. The most common vegetation present where the eggs were laid included vines (Ipomoea pes-caprae) and Marram grass (Ammophila spp).

Table 3. Size (ha) of nesting beaches suitable for nesting in 1986, 2001 and 2006.

In 1986, 11 beaches exhibited suitable characteristics (slope, beach width, approach and vegetation) for sea turtle nesting habitats. These beaches had a total combined area of 84.6 ha. In 2001, 10 beaches were found to be suitable for nesting, with a total combined area of 144.9 ha. This study, therefore, showed an increase in the area suitable for nesting between 1986 and 2001 by 60.3 ha. The area suitable for nesting in 2006 was only 41.7 ha. This area was spread across 7 beaches (Table 3) indicating a decline of 103.2 ha in suitable beaches for nesting between 2001 and 2006.

Table 4. Land use and land cover changes (ha) between 1986 and 2001.

Eight land cover and land use classes were created from the two Landsat images (Table 4). Although some land classes that were inland, such as plantations, did not have any influence on nesting beaches, other land use changes such as built-up areas and thickets that were close to the beaches were found to have an influence on the suitability of nesting beaches. Increased beach development, especially on the Mombasa shoreline, was determined to be one of the major causes of altered nesting beach area, accounting for nearly 60% of the changes. Development, e.g., construction of sea walls within the Mombasa shoreline where most of the nests were located hindered accessibility by hatchlings to the water. Multiple turtle species have been found to nest on beaches with an open offshore approach. This was also observed for the three species nesting in the study area. Mortimer (1995) and Godley et al. (2001) concluded that the primary reasons green and hawksbill turtles at Ascension Island avoided beaches with rock strewn approaches was because moving over such rocks was not only dangerous but also increased the threat of predation due to obstructed movement. All three turtle species in the study area were found to nest on sandy surface beaches. Kamel et al. (2004) observed olive ridley turtles nesting on muddy shores; however, no nests were observed in this study despite the availability of muddy shoreline habitat at Mtwapa beach. This study did not establish the influence of sand particle size on the three observed species, thus confirming the findings of Mortimer (1990) that beach sand types were less important than slope and offshore configuration of the beach. Most nests in the study were found on beaches with steep slopes. Fish et al. (2005) determined that in Bonaire, Dutch Caribbean Antilles, nesting density increased with beach slope for the greens, hawksbills, loggerheads, olive ridleys and leatherbacks.

Despite having a short beach width, Vipingo and Msumarini beaches still had nesting activity due to their steep gradient. This concurred with the findings of both Horrocks et al. (1991), and Weishampel et al. (2003), who observed that beach width and gradient were crucial for nest placement. In this study, short beaches with a steep gradient had nesting activity because they were safe from being inundated by seawater. Of the nests in this study, 72% were found in areas where vegetation was present. The results, therefore, concurred with those of Bustard (1972) and Mortimer (1990) who found that the presence of vegetation at nesting sites was important. These studies both showed that hatchling emergence success at beaches with loose sand was low as the nests tended to crumble and suffocate some of the hatchlings. This did not happen when sand was held together by vegetation.

This study concludes that there were changes in the area of suitable nesting beaches along the Mombasa-Kilifi coastline between 1986 and 2006. Some of the beaches had completely been abandoned by nesting turtles due to habitat loss and alteration while other beaches had either decreased or increased in size (Table 3). English Point was one of the beaches with notable changes despite exhibiting the highest nesting density. This beach, located on the Mombasa shoreline, is fronted by cliffs and residential areas. By the year 2006, six nesting sites had been abandoned in this location due to encroachment by residential buildings and the construction of sea walls around the sites.

Other beaches, such as Kenyatta, Nyali and Serena were also affected by the encroachment of human activities. Tourist hotels and tourist activities (e.g., beach football and volleyball) were prevalent on these beaches. In Kenyatta and Serena beaches, some tourist hotels have completely encroached upon beaches that had previously been used for nesting through construction of sea walls and the use of beach chairs.

Other activities that affected the nesting beaches included the removal of beach front vegetation and the planting of exotic vegetation, such as the Australian pine (Casuarina equisetifolia). The increased presence of the Australian pine on the Kenya coastline has been attributed to efforts to control beach erosion (UNEP 1998), but it might be detrimental to sea turtle nesting; this has been observed by Marcus et al. (1987) in South Florida where Australian pine roots prevented nesting turtles from building good nests.

The land use and land cover change analysis showed an increase in the size of exclusive expansive high cost residential areas on the Mombasa shoreline where most of the beaches were found to have been affected by structures such as sea walls and buildings. Also, in these areas the vegetation present near the beaches had been cleared, thus leaving the nesting sites exposed. Conversely, beaches along Kilifi’s shoreline such as Jumba Ruins, Msumarini and Vipingo did not have large areas of encroachment. Most of these beaches were bordered by cliffs and thickets, which seemed to deter any encroachment. Nesting beaches in both the Mombasa and Kilifi shorelines that were bordered by cliffs were the least affected by development. The beaches in Kilifi seemed to be primarily negatively affected by erosion caused by wave action.

At Jumba Ruins, the lack of encroachment by human activities was likely due to the fact that most of the beach was fronted by the Jumba Ruins prehistoric site. This site is managed by the National Museums of Kenya, which has protected it from any form of human activities. Mtwapa beach experienced some unique changes during the study period. Despite having suitable nesting habitat, the approach from the sea at Mtwapa consisted mainly of muddy shores with dense mangrove vegetation. The land use and land cover change analysis showed an increase in the area occupied by muddy shores over the study period in this area. Olive ridley turtles did not use this beach even though they are known to use beaches with muddy shores for nesting (Pritchard 1979). This is likely due to the presence of dense mangrove vegetation that hindered access to the beach.

The increase in size of suitable nesting areas especially between 1986 and 2001, and in the English Point region between 2001 and 2006 may be attributed to the natural beach processes of erosion and accretion. The areas with sandy beaches could have varied between 1986 and 2001 in the amount of sand probably because the two satellite images despite being taken during the same season could have been taken when erosion or accretion had occurred. It was difficult during this study to ascertain exactly the time of the year when these two processes actually occurred in 1986 and 2001 due to lack of information. Although the sandy beach areas available in 1986 and 2001 varied due to either accretion or erosion, the processes could have been influenced and accelerated by human activities like beach armoring. At the English Point beach for example, the presence of sea walls and sand bags placed on the beaches indicated that wave erosion was affecting residences and that action was being taken to further prevent erosion.

Acknowledgements. The authors wish to thank the Baobab Trust for allowing us to use its facilities during the field survey. We would also like to thank Jonathan Charo for assistance with data collection.

BOWEN, B.W., A.B. MEYLAN & J.C. AVISE. 1989. An odyssey of the green turtle: Ascension Island revisited. Proceedings of the National Academy of Sciences 86: 573-576.

BUSTARD, H.R. 1972. Australian sea turtles. Collins and Sons. London 220 pp.

FISH, M.R., I.S. COTE, J.A. GILL, A.P. JONES, S. RENSHOFF & A.R WATKINSON. 2005. Predicting the impact of sea level rise on Caribbean sea turtle nesting habitat. Conservation Biology 19: 482-491.

FRAZIER, J. 1975. Maziwi Island. Interim report, mimeo, 2 pp.

GODLEY, B.J., A.C. BRODERICK & G.C. HAYS. 2001. Nesting of green turtles (Chelonia mydas) at Ascension Island, South Atlantic. Biological Conservation 97: 151-158.

HORROCKS, J.A. & N.M. SCOTT. 1991. Nest site location and nest success in the hawksbill turtle, Eretmochelys imbricata, in Barbados, West Indies. Marine Ecology Progress Series 69: 1-8.

KAMEL, S.J. & N. MROSOVSKY. 2004. Nest site selection in leatherback, (Dermochelys coriacea): Individual patterns and their consequences. Animal Behaviour 68: 357-366.

KAMEL, S.J. & N. MROSOVSKY. 2005. Repeatability of nesting preferences in the hawksbill sea turtle, (Eretmochelys imbricata), and their fitness consequences. Animal Behaviour 70: 819-829.

MARCUS, S.J. & C.G. MALEY. 1987. Comparison of sand temperatures between a shaded and unshaded turtle nesting beach in South Florida. Seventh Annual Workshop on Sea Turtle Biology and Conservation. Wekiwa Springs State Park, Florida. U.S. Fish and Wildlife Service, Vero Beach, Florida.

MORTIMER, J.A. 1990. The influence of beach sand characteristics on nesting behavior and clutch survival of green turtles (Chelonia mydas). Copeia 1990: 802-817.

MORTIMER, J.A. 1995. Factors influencing beach selection by nesting sea turtles. In: K.A. Bjorndal (Ed.). Biology and Conservation of Sea Turtles. Smithsonian Institution Press. Washington D.C. pp. 45-51.

NELSON, D. & D.D. DICKERSON. 1988. Effects of beach nourishment on sea turtles In: In: L.S. Tait (Compiler) Proceedings of the Fifth Annual National Conservation on Beach Preservation Technology: New Directions in Beach Management. Florida Shore and Beach Preservation Assessment, Tallahasee, Florida. pp. 285-294.

OKEMWA, G.M, S. NZUKI & E. MUENI. 2004. The status and conservation of sea turtles in Kenya. Marine Turtle Newsletter 105:1-6.

OLENDO, D. 1993. Report on sea turtles and Dugong survey. Kenya Wildlife Service (KWS), Nairobi 22 pp.

PENTHICK, J. 1984. An introduction to coastal geomorphology. Edward Arnold Publishers Limited, London. 260 pp.

PRITCHARD, P.C.H. 1979. Encyclopedia of Turtles. T.F.H. Publications, Inc., Neptune, New Jersey. 895 pp.

UNITED NATIONS ENVIRONMENT PROGRAMME (UNEP). 1998. Overview of land-based sources and activities affecting the marine, coastal and associated fresh-water environment in the East African region. Regional Seas Report and Studies No. 167.

WEISHAMPEL, J.F., D.A. BAGLEY, L.M. EHRHART & B.L. RODENBACK. 2003. Spatiotemporal patterns of annual sea turtle nesting behavior along an East Central Florida beach. Biological Conservation 110: 295-303.