An experimental test of the effects of sedimentation on Lake Tanganyika rocky
shore gastropod survivorship
Student: Alieth Mutatina
Mentor: Ellinor Michel
Introduction
Sediment erosion from land and transport and deposition on littoral habitats has been identified as a form of
pollution threatening the biodiversity of Lake Tanganyika (Cohen et al., 1993, 1996; Alin et al. 1999;
Donoghue & Irvine, 2003; McIntrye et al. submitted). This anthropogenic sediment can alter the native
aquatic habitat by blocking light when it is in suspension, affecting water chemistry and changing the
substrate where it is deposited. Effects are most severe in rocky habitats, which in their pristine states have
specialized endemic faunas and floras that are tightly co-evolved to clear water physical surroundings.
Allochthonous sediment in suspension decreases light quantity and quality, reducing algal growth and
potentially changing algal composition. When sediments settle out of the water onto the rock surface, the
particles may interfere directly with algal growth and also survivorship of the fauna that lives on the algae.
Sediment may coat the rock surfaces interfering with grazing and digestion for a very diverse fauna of
herbivores (e.g. see Gilbert, this volume). For the truly benthic animals such as gastropods, shrimp, and
ostracods, sediment also interferes with movement and many more physiological processes such as
respiration and reproduction. Finally, if the sediment blanket on the habitat becomes great enough, it may
decrease the whole available habitat by filling cracks between rocks and successively covering the entire
rocky habitat. Various stages of this process have been documented throughout Lake Tanganyika.
Anthropogenic sedimentation on a large scale is evident along the Burundi shoreline and is correlated with
significantly depauperate diversity (Cohen et al., 1993). Areas of sedimentation, in some cases severe,
although generally patchy, have been documented around Kigoma. This has allowed local comparisons of
field patterns through a number of Nyanza research projects (France & McIntyre, 1998; Sekendende, 1999;
Menone, 2000; Mulongaibalu 2000; Rivers, 2001; Solomon, 2001; Sapp, 2002; Faloon, 2002; Gilber,
2003). Sedimentation has been found to correlate with altered species assemblages, changing population
sizes, decreases in predation and parasites and smaller adult sizes in fish and gastropods. However, direct
testing of the effects of sediment on the animals themselves has been lacking.
I set up a series of laboratory experiments to test the effects of sediment on gastropods from the rocky areas
around Kigoma. My work was modeled on experiments by Donoghue & Irvine, 2003 that subjected
Reymondia horei to inundation with sediment of various size classes in replicated experimental containers
and record the time until death. This was essentially an LD50 experiment. The Donoghue & Irvine (2003)
experiment had two major flaws that I attempted to redress in this study. First, they used a gastropod
species that does not occur on rock surfaces, rather it is generally lives on vertical surfaces and under
overhangs or in very deep water, thus is not generally exposed to sedimentation of any kind in its natural
environment (Powers, 2002). Their experiment covered the snails with a heavy sediment load, thus their
results might not be relevant to the real life, where Reymondia are unlikely to experience sedimentation at
all. I used snail species that naturally occur on upward-facing rock surfaces and do directly experience
sedimentation when it occurs in their natural habitat. Second, Donoghue & Irvine (2003) used sediment that
was not very close to the naturally deposited sediment on three criteria: 1) they use terrestrial sediment
(dirt), 2) they separated it into three separate, single grain-size distributions and 3) they sterilized all the
experimental sediment, removing organics. We could see no reason for these deviations from natural
conditions except to create artifacts. I used sediment that had been deposited in the lake and retained a
natural range of grain sizes. Sediment with a range of grain sizes is likely to have more natural interstitial
space. As I was interested in comparing my results to the Donoghue & Irvine (2003) results, I used both
sterilized and native sediment.
Material and methods
To determine the effects of deposited sediments on rock dwelling endemic snails, I conducted three
different experiments:
Experiment 1
This experiment tested six species and two sediment particle sizes plus a control, replicated eight times for
each treatment for a total of 24 experimental units. I used 2 liter round chambers (approx. 20cm diameter,
10 cm height), each with 1500ml of lake water and a small, flattish cobble from the lake of essentially
similar sizes of 10cm diameter. The cobbles came from 3-5m depth at Kalalangabo 3, a pristine area with
high snail diversity and also the source of most of the tested snails (see map in Berrett et al, this volume).
Two sediment particle size treatments from lake were used in this experiment, with grain size determined
by sieiving representative samples from the bulk lot. These were:
Fine sediment which ranged from 500µm to 2mm collected from approximately 5m water depth inside
Nondwa Point
Very fine sediment which ranged from 63µm to 250µm collected from the east of Hilltop point, near
Kigoma prison. The sediments were not sterilized, thus had a natural complement of organic matter and
lake bacteria.
Each replicate was filled with a constant amount of sediments forming a layer of 1.5cm thick on the
container bottom and covering, but not hiding, the cobble. No sediments were added to control replicates.
Six species of rock-dwelling gastropods were used; these were Lavigeria coronata, L. grandis, L. nassa,
Lavigeria new species W, Vinundu guillemei and Reymondia horei. Taxa follow names in Michel et al.
(2003, in press) and West et al. (2003). Two individuals from each species were placed in each jar after
addition of sediments and water. All treatments were aerated using diaphragm pumps, air stones. Fresh lake
water was replaced every second day. Behavior was noted and the number of live individuals of each
species in each replicate was recorded daily to produce survivorship curves.
Experiment 2
This experiment used a wider range of different grain sizes and sources of sediments and only one species
of snail, Reymondia horei, in an attempt to test results from Donoghue & Irvine (2003). The sediments
were: fine sediments from lake; very fine sediments from lake, fine sediments from land; very fine
sediments from land and mixed fraction of fine and very fine sediments from land. Fine sediments ranged
from 500µm to 2mm and very fine sediments ranged from 63µm to 250µm that burned in muffle furnace at
550oC for one hour to remove organic matter. There were eight replicates per each treatment, and each
replicate containted a small rock and filled with 2cm of sediments and 500ml of lake water in a 1 liter
container (beaker-like cut “MajiPoa” bottle). No sediments were added to control replicates.
Eight individuals of herbivorous rock-dwelling gastropod Reymondia horei were placed in each jar after
settling of sediments and water. Treatments were not aerated but fresh sediments and fresh lake water were
replaced for every second day. The numbers of live individuals in each replicate were recorded daily.
Experiment 3
This experiment used the range of sediments as in experiment 2, but with a selection of species similar to
experiment 1. This was run at the end of the program to test whether the results that were evident from
experiment 1 were due to the species and lack of sterilized sediments or the experimental set up from
experiment 2. Five species of rock-dwelling gastropods were used, these were Lavigeria coronata, L.
grandis, L. nassa. Reymondia horei and Vinundu guillemei. There were eight replicates per each sediment
treatments and each replicate filled with 2cm thickness of sediments and 500ml of lake water. No
sediments were added to control replicates. One individual from each species was placed in each jar after
addition of sediments and water. Treatments were not aerated but fresh sediments and fresh lake water were
replaced every second day. The number of surviving individuals (for each species) in each replicate was
recorded daily.
Results and Discussion
Experiment 1 had continuous suvivorship for almost all individuals of all species near the end of the
research period – 18 days! Only at the end did snails begin to die. This indicates that natural lake
sediments, if well aerated and not sterilized, are not immediately deadly to rock dwelling snails. It
challenges the generality of results from Donoghue & Irvine (2003). These results could be consistent
with previous field results showing decreased sizes of individuals at sedimented sites, as the snails were
likely not to be thriving under sedimented conditions and could be growing more slowly.
Experiment 2 had essentially immediate death of all test individuals in 4 days (Figure 1). This indicates
that sterilized sediments are immediately deadly to rock dwelling snails (if not aerated). It also suggests
that Donoghue & Irvine (2003) should not have sterilized their sediments, as sterile sediments are not
found in the natural world. Control snails lived somewhat longer, but also all died within 8 days. Thus
aeration is critical for snail survival.
Experiment 3 had rapid death of all treatment gastropods with no recoverable pattern by sediment grain
size. However, some species appeared to be more robust, as was also shown in survivorship of control
groups (Figure 2).
Conclusions
•
•
•
•
•
Sediment has different effects based on size fractions (as shown by Donoghue & Irvine, 2003).
Sterilizing sediments is not an appropriate methodology
Aeration changes the effects of sedimentation – lots of aeration allows snails to survive sedimentation
Snails from different species respond differently to sediments
Sedimentation into the lake affects benthic communities. It can lead to the extinction of benthic
species, especially rock-dwelling populations, as it changes the substrate, light penetration and alters
the algae.
Future research
I suggest that this work should essentially be redone with tests using more aeration on multiple sediment
types. It will be very useful to compare results of species survivorship with field results of snail
distributions such as Benthic Survey Team, Nyanza 2003. Finally the critical tests will include exploring
physiological mechanisms of snail death: suffocation by sediment, interference with digestion, growth,
reproduction.
Acknowledgements
I thank the Benthic Team and Katie Wagner for collections of snails, sediments and rocks, Ellinor Michel,
Andy Cohen and Pete McIntyre for comments on methodology and direction, Kamina Chorokora, and
Nathan Stewart for help with analysis and writing and Mr. Cobra for making my first solo research
snorkel such a thrilling experience!
References
Alin, S.R., A.S. Cohen, R. Bills, M.M. Gashagaza, E. Michel, J.-J. Tiercelin, K. Martens, P. Coveliers, S.K.
Mboko, K. West, M. Soreghan, S. Kimbadi, and G. Ntakimazi. (1999) Effects of Landscape
Disturbance on Animal Communities in Lake Tanganyika, East Africa. Conservation Biology
13(5): 1017-1033.
Cohen, A.S., L. Kaufman, and R. Ogutu-Ohwayo. (1996) Anthropogenic Threats, Impacts and
Conservation Strategies in the African Great Lakes: A Review. In: Johnson, T.C., and E.O. Odada,
Eds. The Limnology, Climatology and Paleoclimatology of the East African Lakes. Gordon and
Breach, Newark, NJ. 575-614.
Cohen, A.S., R. Bills, C.Z. Cocquyt, and A.G. Caljon. (1993) Impact of Sediment Pollution on Biodiversity
in Lake Tanganyika. Conservation Biology 7(3): 667-677.
Coulter, G.W. (1991) Lake Tanganyika and Its Life. New York: Oxford University Press
Donoghue & Irvine (2003) Effects of sediment particle size composition on survivorship of benthic
invertebrates from Lake Tanganyika, Africa. Archiv fuer Hydrobiologie.
Faloon, K. (2002) Trematode parasite prevalence in Lavigeria: The effects of sedimentation, depth, size
and species. The Nyanza Project 2002 Report.
France, K. and McIntyre, P. (1998) Gastropod Community Ecology on the Rocky Eastern Shore of Lake
Tanganyika. The Nyanza Project 1998 Annual Report. Univ. Arizona.
Gilbert, D. (2003) Impact of Sediment Pollution on the Littoral Zone of Lake Tanganyika: A Case Study of
Two Cichlid Fish, Petrochromis polyodon and Tropheus brichardi. The Nyanza Project 2003
Annual Report. Univ. Arizona.
Menone, C. (2000) Snail diversity and abundance at sediment disturbed and undisturbed sites around
Kigoma, Tanzania. The Nyanza Project 2000 Annual Report. Univ. Arizona.
Michel, E., Todd, T., Cleary, F.R., Kingma, I., Cohen, A., and Genner, M. (2003, In Press) Scales of
endemism: Challenges for conservation and incentives for evolutionary studies in a gastropod
species flock from Lake Tanganyika. Jourrnal of Conchology Special publication 3: 1-18.
Mulongaibalu, M. (2000) Effet des sediments exogènes sur la communauté littoral benthique au Lac
Tanganyika: Cas de Petrochromis spp et Tropheus spp (Cichlidae). The Nyanza Project 2000
Annual Report. Univ. Arizona.
Powers, L. (2002). Substrate preference and microhabitat use among gastropods of Lake Tanganyika. The
Nyanza Project 2002 Report.
Rivers, A.R. (2001) The Effect of Sediment Deposition on Gastropod Fecal Content. The Nyanza Project
2001 Annual Report. Univ. Arizona.
Sapp, J. (2002) The Dirt on Fish: Effects of Anthropogenic Sedimentation on Fish Behavior and Nutrition.
The Nyanza Project 2002 Annual Report. Univ. Arizona.
Sekandende, B. (2000). Patchiness in gastropod abundance and diversity at Jakobsen’s Beach: A depth
survey using SCUBA. Nyanza Project 2000 annual report, Univ. Arizona.
Solomon, C. (2001) Trophic ecology of the benthic littoral of Lake Tanganyika: The role of grazers in
determining algal biomass and productivity. . The Nyanza Project 2001 Annual Report. Univ.
Arizona.
West, K., E. Michel, J. Todd, D. Brown, J. Clabough. (2003) The Gastropods of Lake Tanganyika:
Diagnostic key, classification and notes on the fauna. (International Association of Theoretical and
Appliced LimnologOccasional Publication no. 2. 132pp.
Sediment Impact on Reymondia
Not aerated lake & terrestrial seds., 8 reps e. tmt.
8
7
6
5
4
3
2
1
0
0
1
2
3
4
5
6
7
8
Days
Figure 1. Results from Experiment 2
Ave. Number of Surviving
Gastropods spp.
Figure 2: Survivorship of control group in experiment 3.
1
0.8
L. Coronata
0.6
L. Grandis
0.4
L. Nassa
0.2
Raymondia
Vinundu
0
0
1
2
3
4
Days
Control Group, Experiment 3
5
6