Palaeoclimatic interpretation of high-resolution oxygen isotope profiles derived from annually laminated speleothems from Southern Oman

Countries located in the Indian monsoon domain are densely populated with their economy depending to a large extent on rain-fed agriculture. Major departures from normal monsoonal rainfall cause serious famines and migration and politic instability. To better predict catastrophic monsoon failures, the monsoon dynamics and its linkages to other climate phenomena, such as the El Niño-Southern Oscillation (ENSO) in the Pacific Ocean and Eurasian snow cover, must be well understood. Our current knowledge of the Indian monsoon dynamics remains limited, mainly due to the lack of high-resolution and long-term monsoon records. The longest instrumental data of monsoon rainfall variability cover the last 150 years (Parthasarathy et al., 1994) and are, thus, too short to fully reveal monsoon variability on century-long time scales. On longer time scales only a few high-resolution geographically widely distributed paleomonsoon records, based on ice cores (Thompson et al., 1997), tree-rings (Bräunig, 1994) and marine sediments (von Rad et al., 1999; Agnihotri et al., 2002; Anderson et al., 2002), are currently available. Even more acute is the lack of information on the Arabian Peninsula and in Northern Africa. A source for information of Indian Ocean monsoon variability is annually laminated speleothems, such as stalagmites and flowstones, forming in caves in Southern Oman where still a monsoon-type climate exists (Burns et al., 2002; Fleitmann et al., 2003). To date, most speleothem-based palaeoclimate reconstructions rely either upon oxygen (δ¹⁸O) isotopic measurements of speleothem calcite or upon thickness of annual growth bands. The importance of both proxies as palaeoclimate variables is described hereinafter. When speleothems are deposited under conditions of isotopic equilibrium, δ¹⁸O of speleothem calcite reflect either variations in δ¹⁸O of the seepage water forming the speleothem and/or variations in cave air temperature. In the first case, the oxygen isotopic composition of the seepage water from which the speleothems are formed reflects the isotopic composition of mean annual rainfall (Yonge et al., 1985). In the second case, the temperature-dependent fractionation of δ¹⁸O between water and calcite (−0.24‰ per 1°C; O’Neil et al., 1969) can be used to reconstruct cave air temperatures (e.g. Lauritzen and Lundberg, 1999), which relates in many caves to mean annual surface air temperature (Wigley and Brown, 1976). However, in most caves the temperature-controlled variations in δ¹⁸O are obscured by changes in δ¹⁸O of the seepage water and surface precipitation, respectively. Generally, changes in δ¹⁸O of rainfall result from a variety of factors, including: (1) changes in δ¹⁸O of the oceanic source region (important on glacial–interglacial time scales), (2) changes in moisture sources or storm tracks, (3) changes in the proportion of rainfall (e.g. winter/summer rainfall), (4) air temperature, (5) amount of rainfall, and (6) evaporation. Particularly evaporation can alter the original oxygen isotopic composition of rainfall and seepage water, respectively (e.g. Bar-Matthews et al., 1996) when the cave is located in an arid or semi-arid regions, such as Southern Oman. To rule out the importance of evaporation processes, the oxygen isotopic composition of water along the pathway from precipitation to speleothem calcite must be traced. The knowledge of the principal controls on the oxygen isotopic composition of present-day stalagmites is essential before δ¹⁸O values of fossil stalagmites can be interpreted correctly in terms of palaeoclimate variability.

The thickness of annual growth bands in stalagmites is also a frequently used speleothem climate proxy, because band thickness is controlled by the drip rate, which relates to surface precipitation (Baker et al., 1993; Genty and Quinif, 1996; Holmgren et al., 1999; Qin et al., 1999). Qin et al. (1999) suggested that the thickness of annual growth bands is a proxy for the amount of drip water and surface precipitation respectively when the cave has a simple hydrological connection to the surface (thin overburden bedrock thickness). Furthermore, the stalagmite should have a columnar shape indicating that calcite precipitated at the top and no stalactite as a drip source.

Before speleothem-based time series can be used for palaeoclimate studies the following fundamental questions must be answered:

• 1.
  Are the stalagmites deposited in isotopic equilibrium with the cave waters?
• 2.
  Is the oxygen isotopic composition of speleothem calcite mainly controlled by the oxygen isotopic composition of rainfall or by air temperature?
• 3.
  Do multiple proxies, in this case δ¹⁸O and annual band thickness, relate to the same climate variable?
• 4.
  Are oxygen isotope profiles of individual stalagmites reproducible?
• 5.
  Do the time series reflect local or regional climate variability?

In this paper, we present high-resolution stable isotope profiles derived from three contemporaneously deposited annually laminated stalagmites from Kahf Defore located in Southern Oman. In addition, we have sampled precipitation, cave drip waters and actively growing stalagmites to study the active water-carbonate system and to test for isotopic equilibrium. By comparing δ¹⁸O and thickness of annual bands of contemporaneously deposited stalagmites we test whether both proxies relate to the same climatic variable. To date speleothem studies combining both proxies at high-resolution are extremely rare. Finally, we compare our records with meteorological observations form the surrounding areas to validate and to put our palaeoclimate reconstructions into a broader context.