Non-climatic growth of the saline Lake Beseka, Main Ethiopian Rift

https://doi.org/10.1016/j.jaridenv.2008.09.015Get rights and content

Abstract

When using rift lakes as proxies for palaeoclimate it is essential to know if water level changes are of geological or climatic origin, and whether a reinterpretation of palaeo-lake levels derived from lithostratigraphic sequences is required. The saline, endorheic Lake Beseka is located in the tectonically active Main Ethiopian Rift. Despite the aridity of the rift valley the lake's surface area quadrupled from 11.1 km2 in 1973 to 39.5 km2 in 2002. We quantify the lake growth by means of a detailed bathymetric model and high-resolution satellite time series. We analyse the potential climatic, anthropogenic, and tectonic agents of Lake Beseka's growth. Multitemporal remote sensing data and meteorological records were compared with in-situ measurements of hydrochemical parameters and water depth to address the reason for the lake level rise. Our results suggest that Lake Beseka's expansion originates from an increased discharge of the hot springs. The combined analysis of satellite data and in-situ measurements proved to be a valid tool for the quantification of lake level changes and can help to detect the causes of these variations.

Introduction

Past lake levels are used as proxy for palaeoclimate in numerous studies (e.g., Kiage and Liu, 2006, Stager et al., 2005, Verschuren et al., 2000). For a meaningful interpretation, signals derived from bio- and lithostratigraphic data need to be corrected for influences other than climate variations. Our study aims to determine these factors taking Lake Beseka as an example for a rift lake in a tectonically active environment. The endorheic Lake Beseka, in the centre of the semi-arid rift valley of the northern Main Ethiopian Rift (nMER, cf. Fig. 1), has rapidly expanded over recent decades (e.g. Williams et al., 1981).

East Africa is a region of extremely complex meteorological and climatological phenomena and coupling-mechanisms, possibly one of the most complicated of the continent. The topography of the rift exerts a strong influence on microclimate, drainage systems, and local ecosystems (e.g. Seleshi and Zanke, 2004, Gissila et al., 2004). The lakes occupying the rift valley are either dominated by highland runoff and riverine delivery or by groundwater and local runoff (Brandt, 1982). In arid areas like the one exemplified in this study, a high fraction of lakes will be endorheic and thus more susceptible to slight perturbations of the water balance. According to Mercier et al. (2002) height changes of a lake's water column may be caused by

  • (i)

    Changes in surface pressure;

  • (ii)

    Wind-driven events (i.e. seiches) and tides;

  • (iii)

    Fluctuations in the volume of this column due to an alternating temperature or composition;

  • (iv)

    Modifications of the hydrologic cycle (i.e. recharge, outflow, evaporation).

Items (iii), (iv) will ultimately result in lake water volume changes.

The first three factors may be relevant for larger rift lakes, but cannot explain a water level increase of more than 4 m within less than three decades (Ayenew, 2004) of the comparatively small Lake Beseka. Cause (iv) can be directly triggered by non-climatic factors, for example tectonically induced alteration of flow patterns, changes in the heat budget due to geothermal effects, and artificial regulation. Anthropogenic influence on rift lake levels can also be exerted indirectly through water withdrawal (Ayenew, 2004) or an increased sediment input as a result of inadequate agricultural practices (Geremew, 2000; WWDSE, 2001). Lakes also fluctuate in accordance to climatic variations (Street and Grove, 1976; Chernet, 1982; Chalié and Gasse, 2002), especially when they are terminal (Ayenew, 2004). All these potential contributors to lake level change must be quantified in order to understand the hydrological setting of a drainage area.

Changes in the level of continental lakes can be assessed by measuring the absolute or relative elevations of the water level by continuous gauge records. However, in many settings it might be difficult to use standard hydrological survey data. A historical data record might not exist or not be accessible. Archives of satellite data covering up to 35 years can provide an inexpensive alternative or addition to incomplete ground measurements. RADAR altimeters were shown to yield reasonably accurate measurements of variations in water surface elevation (e.g., Mercier et al., 2002). However, they are only applicable for lakes situated along the satellite altimeter's ground track. In addition, the lakes need to be wider than 500 m and need to have a surface area of more than 100 km2 (Crétaux and Birkett, 2006). In this study we quantify the expansion of Lake Beseka over the past decades with multitemporal optical remote sensing data, a bathymetric model, and lake level records. Moreover, we apply hydrochemical analyses and temperature measurements to contribute to the study of Beseka's rapid lake level change.

Section snippets

Study area

The East African Rift System (EARS) cuts through the Ethiopian highlands (Fig. 1). One of its subsystems, the nMER, is among the few sites worldwide featuring a transition from continental to oceanic rifting (Ebinger and Casey, 2001). Present-day volcanic and seismic activity and faulting (Keir et al., 2006, Keranen et al., 2004, Kurz et al., 2007) are phenomena linked to this geologic process. Located within an active tectono-magmatic segment (e.g. Kurz et al., 2007), Lake Beseka's basin is

Quantification of the growth of Lake Beseka

Data from Landsat (180-054, 168-054: 1973, 1986, 1989, 2000, 2001, 2002) and ASTER (2000, 2001) were investigated in order to describe temporal variations in the surface area of Lake Beseka. To avoid obscuration by clouds, these data were acquired in the dry season. Due to a similar spectral signature of the littoral and recent basalt flows on the northern shore, the precise shoreline of Lake Beseka is difficult to map. Hence, two methods were applied on the satellite data to measure the

Results

The time series of satellite data documents that Lake Beseka's surface area quadrupled from 11.1 km2 in January 1973 to 39.5 km2 in February 2002 (Fig. 5). Aerial photographs from 1957 and 1964 show a lake area of about 3 km2 (Tessema, 1998). Putting aside seasonal fluctuations the lake has been steadily rising at an average rate of approximately 18 cm/yr (Fig. 6). In total the water level increased by about 8 m from 1976 to reach 955 m asl in September 2004.

The striking similarity between the shapes

Discussion and conclusions

We used several methods to examine the growth of Lake Beseka and establish the contribution of various potential water sources. The influence of neotectonic activity has rarely been taken into account. Previous approaches did consider the basin morphology only to a very limited extent (Tessema, 1998, MWR, 1998, Engida and Russom, 2004). In this study, we present the most detailed model of the lake floor yet considered. In combination with measurements from optical satellite data and digital

Acknowledgements

The realisation of this project would never have been possible without the logistical support of the staff of the Department of Biology (Addis Ababa University) and the kind assistance of K. Adem, T. Ayenew, H. Hagos, Z. Shiferaw and M. Umer (all Department of Geology, AAU). T. Eshete, M. Adem and L. Gurmu (Metehara Sugar Plantation) helped us with detailed information on the sugar plantations near Beseka. M. Buchwitz (TU Bergakademie Freiberg) supported us with detailed geologic knowledge of

References (49)

  • F. Mercier et al.

    Interannual lake level fluctuations (1993–1999) in Africa from Topex/Poseidon: connections with ocean-atmosphere interactions over the Indian Ocean

    Global and Planetary Changes

    (2002)
  • M.H. Trauth et al.

    East African climate change and orbital forcing during the last 175 kyr BP

    Earth and Planetary Science Letters

    (2003)
  • F.M. Williams et al.

    Tensional fissures and crustal extension rates in the northern part of the Main Ethiopian Rift

    Journal of African Earth Sciences

    (2004)
  • Ayenew, T., 1998. The Hydrogeological System of the Lake District Basin. Central Main Ethiopian Rift. PhD thesis, Free...
  • T. Ayenew

    Environmental isotope-based integrated hydrogeological study of some Ethiopian rift lakes

    Journal of Radioanalytical and Nuclear Chemistry

    (2003)
  • T. Ayenew

    Environmental implications of changes in the levels of lakes in the Ethiopian Rift since 1970

    Regional Environmental Change

    (2004)
  • R. Bilham et al.

    Secular and tidal strain across the Main Ethiopian Rift

    Geophysical Research Letters

    (1999)
  • Brandt, S.A., 1982. A Late Quaternary Cultural/Environmental Sequence from Lake Beseka, Southern Afar, Ethiopia. PhD...
  • M. Buchwitz
  • Chernet, T., 1982. Hydrogeology of the Lakes Region, Ethiopia. Memoir No. 7, Ministry of Mines and Energy, Addis Ababa,...
  • C.J. Ebinger et al.

    Continental breakup in magmatic provinces: an Ethiopian example

    Geology

    (2001)
  • Z.A. Engida et al.

    Determination for the cause of rising water levels in Lake Beseka and design of remedial measure

    Ethiopian Engineers Association Journal

    (2004)
  • Eshete, T., 2004. Chief of agriculture at the Metehara Sugar Factory, personal...
  • Geremew, Z., 2000. Engineering Geological Investigation and Lake Level Changes in the Awassa basin. M.Sc. thesis,...
  • Cited by (43)

    • Origin and migration of fluoride in the area of the Aluto Volcanic Complex (Main Ethiopian Rift)

      2022, Applied Geochemistry
      Citation Excerpt :

      This correlation was already demonstrated in groundwater of the East African Rift system, which often has elevated temperatures due to volcanic activity (e.g. Ayenew 2008). The volcanic and tectonic activity caused by rift propagation also results in a complex fault system allowing fluid migration along permeable fractures (Goerner et al., 2009; Hutchison et al., 2015; Jolie et al., 2019; Jolie, 2019), which could enhance fluoride circulation. Associated geothermal activity affects the groundwater flows and the water chemistry of the Main Ethiopian Rift (MER; Kebede et al., 2010).

    • Estimation of groundwater contribution to Lake Basaka in different hydrologic years using conceptual netgroundwater flux model

      2020, Journal of Hydrology: Regional Studies
      Citation Excerpt :

      The GW inflow is from the south-western shore of the lake where there are a number of hot springs, which are currently submerged. The direction of GW inflow is also reported by others (e.g. Ayenew, 2007b; Belay, 2009; Georner et al., 2009; Dinka, 2010; Dinka et al., 2014; Dinka, 2017c). The groundwater outflow from the lake during heavy rainy period is towards north-east, east and south-east directions, but mostly towards north-east direction.

    • Use of MODIS satellite images for detailed lake morphometry: Application to basins with large water level fluctuations

      2016, International Journal of Applied Earth Observation and Geoinformation
      Citation Excerpt :

      On the other hand, during the ascending period, low vegetation that grew on the exposed sediment may enhance the contrast with open water in near-infrared, thus increasing the accuracy. An effect similar to the water trapped in ponds has been explained by the inconsistent water level due to strong winds (Goerner et al., 2009; Zolá and Bengtsson, 2007). Another reason for the differences between the two versions of the DDMs could be due to the continuous natural sedimentation and scouring processes, influenced by the management policy of Lake Kerkini and the seasonal changes of upstream soil erosion (Alexandridis et al., 2015b; Panagos et al., 2011).

    View all citing articles on Scopus
    View full text