ENSO Project
Recent ENSO and paleo-ENSO of the last 1000 years in Lake Tanganyika Project (contact address: P.-D. Plisnier, Royal Museum for Central Africa, Geology & Mineralogy Department, Leuvensesteenweg 13 , B-3080 Tervuren, Belgium, tel. (32)-2-7695 405, fax: (32)-2-7695 432, e-mail: pierre-denis.plisnier@arcadis.be) :

•   Background: Lake Tanganyika is very sensitive to climate variability. Its sediments might prove to be real archives of paleoclimatic variability. Several environmental parameters have been correlated with ENSO (El Niño Southern Oscillation) in this area (Plisnier, Serneels & Lambin, in print for Global Ecology and Biogeography). Laminated sediments are known to exist in Lake Tanganyika. The diatom community is sensitive to climatic change (Degens et al., 1971; Haberyan & Hecky, 1987);
•   Execution: Royal Museum of Central Africa (MRAC), Université Catholique de Louvain (UCL) and Universiteit Gent (RUG), Belgium;
•   Duration: 4 years, 1997-2000;
•   Funding: Belgian Federal Government;
•   Activities: a multidisciplinary scientific programme including remote sensing and teleconnection as well as field sampling for sedimentology and planktonic analysis of the sediments of Lake Tanganyika. A large climatological databank has already been established. A coring expedition, together with a Swiss research team, was executed in November 1998. Many sediment cores were sampled, and several CTD measurements were taken between Mpulungu (Zambia) and Kipili (Tanzania) and up to greater depths (down to 1300 m) in the Tanzanian waters close to the Congolese border across Kipili.
•   Results: part of their findings were published in the following article: "Impact of ENSO on East African ecosystems: a multivariate analysis based on climate and remote sensing data" by Plisnier P.D., Serneels S. & E.F. Lambin, 2000, Global Ecology & Biogeography (2000) 9: 481-97.    ENSO is an important driver  of inter-annual variations in climate and ecosystem productivity in tropical regions. The findings of the study confirm the ENSO impact on the climatic and ecological variability in East Africa. However, it is also shown that the pattern of teleconnections is much more complex than generally assumed, both in terms of spatial distribution and impact on different ecosystem variables. Not all climate and land surface variables are teleconnected to ENSO in the same way, which leads to a complex impact of ENSO on the ecosystem.  Moreover, the ENSO impact is highly differentiated in space, as the direction, magnitude and timing of this impact are controlled by the local climate system, the presence of large lakes, proximity to the coast and, possibly, local topography and land cover.     
•   Final summary:
Summary of the  ENSO project

Lake Tanganyika is one of the oldest lakes in the world. Its present limnological cycle shows that it is very sensible to climate variability (Coulter, 1991; Plisnier et al., 1999). The winds and the air temperature fluctuations influence the mixing of the lake, allowing access of nutrient-rich deeper water toward the surface where biotic production takes place. When organisms die, they accumulate into layers at the bottom of the lake, where some of them may be identified and related to the lake conditions prevailing at the time of their deposit.

Global signals of oceanic/atmospheric origin affect the local climate in East Africa and probably the hydrodynamic and the biotic environment of Lake Tanganyika. Its sediments could thus archive paleoclimatic variability partially related to world climatic variability such as ENSO (El Niño/Southern Oscillation). ENSO is a major climatic/oceanographic event characterised by the apparition every 2-10 years of a warm current in the Pacific Ocean and numerous climatic anomalies around the Pacific Ocean and in many others areas of the world.  The influence of El Niño stretches across the Indian Ocean to southern Africa where it brings droughts and to equatorial Africa where it brings flooding (WMO-UNDP, 1993; IPCC, 1990).

The objective of the project was to check the possible influence of climate variability on Lake Tanganyika at two scales of time: the recent period (+/- last 50 years) and the last 1000 years. For the recent period, the project studied the teleconnections (correlation with remote climatic or proxy data) between ENSO and various data (oceanic, climatic, remote sensing, hydrological, fisheries catches...) in the Lake Tanganyika area.  For the last 1000 years, paleo-signals were studied in the sediments of the lake. The finely laminated sediments were studied particularly using diatom composition and thin sections analysis to detect variability of organisms as possibly related to limnological and climatic variability.

For the study of the recent period, available climate data in the Lake Tanganyika area (Tanzania, Burundi, Zambia and R.D.Congo), were processed using various statistical methods for the 1981-94 period at different time lags. The results of teleconnections with climate data were compared with remote sensing data from NOAA/AVHRR (National Oceanic and Atmospheric Administration / Advanced Very-High Resolution Radiometer) using the same procedures.

Significant ENSO teleconnections were found with average air temperature, maximum and minimum air temperature, humidity, rainfall, winds, pressure and radiation. The strongest teleconnections were found between monthly air temperature anomalies with the sea surface temperature anomalies in the west equatorial Pacific Ocean. A time lag of 4-6 months generally gave the strongest correlation (e.g. in the range of R=0.6). ENSO events were characterised by average air temperature increase (+0.26°C) while extreme air temperature could reach +/- 0.8°C as observed during a strong El Niño event for the recent period. During ENSO events, winds decreased but air pressure and radiation increased. This seems to impact the mixing of the lake. Catches per unit of effort of the main pelagic fishes were partially correlated with ENSO for the last 30 years in two stations of Lake Tanganyika. Hypotheses of changes in hydrodynamic and upwelling intensity have been presented to explain this. More winds and lower temperature seem favourable for clupeids fishes (and possibly phytoplankton and zooplankton) while Lates stappersi catches are lower maybe because of lower transparency unfavourable to this visual predator.

One of the major findings of this project is that several climate and land surface attributes show a partial correlation with the fluctuation of the Pacific sea surface temperature (SST) index. These different variables are not all teleconnected to ENSO in the same way, which leads to a complex impact of ENSO on the ecosystem. The SST anomalies in the Pacific Ocean induce changes in climate variables. These ENSO-driven changes in climate induce changes in vegetation activity, as measured by changes in several remote-sensing measured variables (a vegetation index, the surface temperature etc…). The exact response of the vegetation to the ENSO-induced climate variations depends on the land cover type.

The project showed that the ENSO impact is highly differentiated in space in the investigated East African area. Many previous studies have postulated a single, region-wide impact of ENSO. It appears that the direction, magnitude and timing of this impact is controlled by the climate system at a regional scale and at a more local scale - e.g. as influenced by the presence of large lakes, local topography or proximity to the coast. Surface attributes, as determined by geology, soil and vegetation might also influence the magnitude and the time lag of the ENSO impact. Hence different zones are recognised in the East African area studied, each subject to different combinations of ENSO induced climate variations.

ENSO is however not the only source of inter-annual variability in climate conditions in the region. The changes of the variables studied, not taken into account by ENSO, could be related to the variability of neighbouring oceans as well (Saji et al., 1999). Beside ENSO, a warming was observed in the recent decades in the air temperature at Lake Tanganyika (>0.7-0.9°C). This was apparently linked to a water temperature increase and a higher stability of the lake. Decreased winds and changes in fish catches were observed during the same period for the clupeids fishes and Lates stappersi.  Those observations suggest that the lake is sensitive to other climate variability such as the recent global temperature increase besides ENSO.

In order to study the sediments of Lake Tanganyika, an expedition was organised in collaboration with the EAWAG-team from Switzerland. During this expedition, 16 cores were retrieved in the southern basin of the lake, between 330 and 1200m water depth. All cores were taken in deep water where permanent anoxic conditions exist and where bioturbation is minimal.

Thin section analysis was performed on parts of the sediments of the two cores taken on the Kalemie Ridge in 424 and 428m water depth and of one core taken in the southern basin in 1200m water depth. Both laminated as well as homogeneous layers are present in all cores. The laminated layers result from the settling of alternating organic sedimentation and terrigenous sediments. Homogeneous parts in the sediment cores probably correspond to periods with a steady settling of a continuous rain of debris.

The results of the thin section analyses support the hypothesis that the laminated sediment, present in the sediments of Lake Tanganyika, reflects a varying biogenic production. More intense winds prevailing apparently in La Niña years could possibly have favoured this increased production. Because the laminations were not continuous in the entire core and are difficult to count reliably, we do not expect them to provide a comprehensive annual reconstruction of the ENSO-signal. However, it seems possible to produce a reconstruction of the primary productivity related to the strongest ENSO events. An accurate chronology for the sedimentary records is necessary to test this hypothesis.

Sediment material for dating was taken from a core recovered near Kasaba in 500m water depth and from a core taken near Kipili in 1200m water depth. The results for the Kasaba core are the only one available at the moment. 14C AMS dating indicates that the sediments between 10.5 and 11.5cm are from the historical period (830 (±40) BP), corresponding to a sedimentation rate of about 0.134mm/y during recent periods. The sediments between 48 and 52cm depth are much older (8060 (±60) BP), which would correspond with a very low sedimentation rate of 0.06mm/y. It is known that sedimentation rates in Lake Tanganyika have not been constant and vary considerably in the sedimentary record (Coulter, 1991). Mean reported sedimentation rates in Lake Tanganyika vary between 0.4 and 0.6mm/y (Haberyan and Hecky, 1987; Tiercelin et al., 1988; Coulter, 1991). Extremely low sedimentation rates (±0.05-0.25mm/y) were only reported from the Kalemie Ridge, a bathymetric high that separates the northern and southern basin (Cohen and Palacios-Fest, 1999). The extremely low sedimentation rate, calculated for deeper parts of core taken near Kasaba and the absence of a layer rich in Aulacoseira (Melosira) species, a layer reported earlier for sediment cores from the southern basin of Lake Tanganyika (Tiercelin et al., 1988), may suggest that part of the sediment is missing, possibly because of sediment-sliding events.

High-resolution diatom analyses were performed on the upper parts of both of the above-mentioned cores. The upper 11cm of the Kasaba core consist of dark finely laminated sediments; the laminations in the upper 18cm of the Kipili core are mostly diffuse. These upper layers were analysed at a vertical resolution of 100µm and 500µm, respectively. Deeper parts of the Kasaba core were also studied at a lower resolution.   

Although 230 diatom taxa were observed in the two cores studied, only a few taxa were relatively abundant. In the upper finely laminated sediments “long” Nitzschia's and Gomphonema clevei were relatively the most important diatoms, followed by Nitzschia lancettula, N. vanoyei, N. frustulum, N. fonticola and Gomphonitzschia spp. The high species diversity was due to the numerous benthic taxa; although individual abundancies were generally low, these benthic taxa dominated the fossil assemblage in certain layers.

The diatom record revealed a cyclicity, with a provisionally estimated period of about 200 years, in the Kasaba core. This might be related to a 200 years cycle in the sun activity but this needs to be confirmed by data from longer or more time series. Among the diatoms, “long” Nitzschia spp. most clearly demonstrate this cyclicity. They are pennate, planktonic species which can replace Cyclostephanos spp. (centricate planktonic diatoms) in the plankton of East African lakes when Si/P ratios decrease (Haberyan and Hecky, 1987).

Benthic diatoms are potentially good indicators of lake level fluctuations, indicating the presence of nearby littoral zones (Barker, 2001). In the Lake Tanganyika cores the significant changes in the proportion of benthic diatoms may likewise indicate lake level changes. In addition, our data suggest that the interpretation of the benthic diatom assemblages might be more complicated. A notable change in the benthic assemblage was the dominance of Gomphonema clevei in the lower part of the sediment section studied for the Kasaba core, while in the upper layers other species were abundant. Although this pattern may reflect changes in the littoral environments in Lake Tanganyika, more needs to be known about the ecology of key taxa. Similarly, further investigation of the ecology of planktonic and facultative planktonic key-taxa, e.g. “long” Nitzschia's, Nitzschia lancettula, N. vanoyei, N. frustulum, N. fonticola and Gomphonitzschia spp., is needed to permit a detailed interpretation of the L. Tanganyika sediments.

The present data and work in progress illustrate the unique potential of these cores. It seems feasible to infer records of historical changes in paleo-ENSO intensity, as well as longer-term variation in regional and global climatic conditions.

The recent ENSO signal confirms thus the hypothesis that ENSO may impact significantly the climate in the Lake Tanganyika area and the mixing conditions of the lake in the actual period. It is recommended that a multidisciplinary database using recently collected information in various fields (climate-hydrodynamics-phytoplankton-fisheries) at regular sampling intervals, could provide the required information to interpret the various sediments signals and decrypt the coded information stored in the sediments.

Bibliography

Cocquyt, C and D. Schram, 2000. Diatom assemblages in surface sediments along the eastern coast of Lake Tanganyika. Hydobiologia 436: 59-71.
Cocquyt, C. , 1999. Seasonal variations in periphytic diatom communities in the northern basin of Lake Tanganyika. Systematics and geography of Plants 69: 265-273.
Cocquyt, C. ,1999.  Diatoms from a hot spring in Lake Tanganyika. Nova Hedwigia 68: 425-439.
Cocquyt, C., 1999. Seasonal dynamics of diatoms in the littoral zone of Lake Tanganyika, Northern Basin. Arch. Hydrobiol. series: Algological Studies 92: 73-85.
Cocquyt, C., 2000. Biogeography and Species Diversity of Diatoms in the Northern Basin of Lake Tanganyika.  Advances in Ecological Research, 31, 2000, 125-150.
Cocquyt, C., 2001. Notes on Cymatopleura calcarata Hustedt (Bacillariophyceae), an endemic diatom from Lake Tanganyika. Festschrift für H. Lange-Berlalot: in press.
Cocquyt,C., 2000. A light and microscopic investigation of Surirella brevicostata O. Müller (Bacillariophyceae), an endemic Tropical African diatom. Systematics and Geography of Plants 70: 245-254.
Francus P. & Karabanov E., 2000. A computer-assisted thin-section study of lake Baikal sediments: a tool for understanding sedimentary processes and deciphering their climatic signal. International Journal of Earth Sciences, Geologische Rundschau, 89, 2, 260-267.
Francus P., 1999. Using image analysis to estimate quantitatively some microstructural parameters of detrital sediments. Geologica Belgica, 2/3-4, 173-180. (published in 2000).
Francus P., 2001. Quantification of bioturbation in hemipelagic sediments via thin-sections image analysis. Journal of Sedimentary Research, 71, 3, 501-507.
Plisnier P.D. Climate and Limnology changes in Lake Tanganyika. Verhandlungen Internationale Vereinigung für theoretische und angewandte Limnologie. In press.
Plisnier P.-D. Limnological profiles and their variability in Lake Tanganyika. IDEAL, Proceedings of the second Internationl symposium on East African Lakes at Lake Malawi, Makakola 01/2000. In press.
Plisnier P.-D., 1997. Lake Tanganyika Research seeks evidence of the El Niño-Southern Oscillation during the past 1000 years. Bulletin of the International Decade for the East African Lakes (IDEAL), Spring issue: 4.
Plisnier P.-D., 1998. Lake Tanganyika: Recent climate changes and  teleconnections with ENSO. Proceedings of the International Conference “Tropical Climatology, Meteorology and Hydrology” (Brussels, 22-24 May 1996). G. Demarée, J. Alexandre and M. De Dapper (ed.):Royal Academy of Overseas Sciences and Royal Meteorological Institute of Belgium: 228-250
Plisnier P.-D., 1998. Limnologische Untersuchungen am Tanganjikasee. DATZ, 09/1998: 594-600.
Plisnier P.-D., D. Chitamwebwa, L. Mwape, K. Tshibangu, V. Langenberg and E. Coenen, 1999. Limnological annual cycle inferred from physical-chemical fluctuations at three stations of Lake Tanganyika.  Hydrobiologia 407: 47-61.
Plisnier P.-D., S. Serneels and E.F. Lambin, 2000. Impact of ENSO on East African ecosystems: a multivariate analysis based on climate and remote sensing data. Global Ecology and Biogeography 9 (6): 481-497.
Plisnier P.-D. and E.C. Coenen, 2001. Pulsed and dampened annual limnological fluctuations in Lake Tanganyika. In: The Great Lakes of the World (GLOW): Food-web, health and integrity.  Munawar M. and Hecky R. (Eds). Ecovision World Monograph Series, Leiden, The Netherlands: 81-94.
Sarvala J., K. Salonen, M. Järvinen, E. Aro, T. Huttula, P. Kotilainen, H. Kurki, V. Langenberg, P. Mannini, A. Peltonen, P.-D. Plisnier, I. Vuorinen, H. Mölsä and O. V. Lindqvist, 1999. Trophic structure of Lake Tanganyika: carbon flows in the pelagic food web. Hydrobiologia 407: 155-179.