engleski

A Comparison of Isotopic and Geochemical Environmental Records in a Modern Laminated Tufa (River Krka, Croatia)

An attempt was made to estimate environmental conditions during the 60 years of precipitation of a subrecent annually laminated tufa incrustation, found in the artificial tunnel diverting river water to the Jaruga hydroelectric power plant on the river Krka (Croatia). Continuous records (1938 – 1998) of the stable isotope composition of carbonate (δ 13C and δ 18O) and Mg, Sr and Ba contents in the laminae were analysed and compared to the measured ambient temperature and river discharge records. The C and O stable isotope compositions and metal concentrations are partially correlated with environmental parameters (temperature, discharge and precipitation). Only a poor negative correlation between δ 13C and δ 18O of carbonate was found, whereas no correlation existed between metal concentrations and isotopic composition of carbonate. Decreasing δ 18O with time is related to the long-term increase in temperature and decreasing discharge, whereas the increase of δ 13C cannot be unambiguously explained at data available to the moment and is most probably governed by several interplaying mechanisms, such as long-term variation in discharge, temperature, vegetation cover and probably microbial communities on the site of precipitation. The oxygen isotope thermometers could only provide a relative trend of temporal temperature changes and largely overestimated the realistic temperature differences during the precipitation period. One reason for this discrepancy is the unknown δ 18O of water at the time of precipitation, although in contrast to ancient tufa formations, it is less likely that significant changes in water isotope composition could occur within only a few decades. A possible cause of changes in mean annual δ 18O of precipitation would be a change in seasonal distribution of rainfall. In fact, following the hydrometeorological data, summer and winter rainfall amounts in the recharge area show the same long-term trends till 1991, whereas later on the seasonal rainfall pattern changed: summer precipitation exhibits an increasing trend, while winter precipitation decreased. This most probably caused a shift in the mean annual δ 18O of the river water toward less negative values after 1991, which increased the estimated temperatures of precipitation in the last decade. However, it cannot explain the exaggerated long-term and interannual variation in precipitation temperature estimates before 1990. There is no theoretical basis for the forms of the commonly applied equations linking the precipitation temperature and oxygen isotope separation between carbonate and water ; as they are simply the best fit of experimental data to a second-order polynomial. Since every tufa precipitating river system is unique with respect to the seasonal and annual water temperature and water δ 18O variations, flow regime etc., these equations obviously cannot be used universally. Moreover, one could assume that for each system, its own temperature equation should be derived, taking into account its unique characteristics. Long-term trends in variation of metal concentrations are synchronous and well correlated with long-term temperature changes before the onset of industrial pollution, which began in 1971. In this period, the estimated temperatures of precipitation derived from δ 18O records and Mg concentration in carbonate match well, while later on the Mg thermometer gave much higher temperature estimates than the isotope thermometer.

laminated tufa; stable isotope; Mg; palaeotemperature

nije evidentirano