Jun Matsuoka

Seasonal variation of stable isotopic compositions recorded in a laminated tufa, SW Japan

Abstract High-resolution stable isotopic analyses of an annually laminated tufa from Shirokawa, SW Japan, show cyclic variations that correspond to mm-scale summer (dense) and winter (porous) laminae. Both δ 18 O and δ 13 C values are high in winter and low in summer, and correlate well with each other ( r =0.88). Because the δ 18 O content of stream water has remained relatively stable over time, the δ 18 O curve reflects seasonal changes in water temperature. Cyclic variation in δ 13 C suggests that it is controlled by seasonal changes in the δ 13 C of dissolved inorganic carbon in groundwater, which is probably caused by groundwater degassing. This degassing effect is normally strongest during winter when subsurface air is warmer (less dense) than the atmosphere. This decreases P CO 2 in cave air, and the associated degassing of 13 C-depleted CO 2 from the groundwater increases the δ 13 C value of dissolved inorganic carbon in winter. Calcite was deposited in isotopic equilibrium with stream water. Extreme climatic events, such as drought, can be recorded as anomalies in both the δ 18 O and δ 13 C curves. This study demonstrates that annually laminated tufa can be used for high-resolution analysis of terrestrial paleoclimate.

Origin of annual laminations in tufa deposits, southwest Japan

Abstract Laminated tufas and a tufa-depositing stream in SW Japan (Shirokawa, Ehime Prefecture) were studied monthly over a 3-yr period. A series of samples from the tufa clearly reveals the pattern of annual laminations. The annual layering pattern was primarily controlled by changes in the rate of calcite precipitation, as calculated from water chemistry. The concentration of dissolved CaCO 3 , which correlates with the precipitation rate, was high in summer–autumn and low in winter–spring, owing to changes in the partial pressure of CO 2 in underground air. Regular seasonal changes in underground PCO 2 probably resulted from two temperature-dependent processes, the diffusion of soil CO 2 and the ventilation of underground air. These changes, in addition to water temperature changes, altered the precipitation rate, which has a clear seasonal pattern, especially in the lower stream. The seasonal precipitation rate was high in summer–autumn and low in winter–spring, which is consistent with the seasonal lamination pattern seen in the tufas. The textures of collected samples show that the laminations consist of densely calcified summer–autumn (June–October) laminae and lightly calcified winter–spring (November–May) laminae. We infer that the increased precipitation rate stimulated thick calcite encrustation on cyanobacterial filaments to produce the dense textures. This interpretation is supported by the lowered organic/inorganic carbon-production ratio in summer–autumn. Seasonal variations in cyanobacterial assemblages are present, but do not reflect the seasonal lamination pattern. Because the relevant processes are temperature dependent, the seasonal lamination pattern at Shirokawa is thought to generally apply to other laminated tufas deposited in temperate climates. However, a reversed pattern can result from local and climatic circumstances. Dense laminae were deposited in winter near the source of the spring at Shirokawa, because calcite precipitation was high owing to low underground PCO 2 in winter. Reversed patterns reported from northwestern Europe were probably influenced by seasonal rainfall, which is reflected in hydrological conditions.

High-resolution records of rainfall events from clay bands in tufa

Individual rainfall events are recorded as thin clay bands in annually laminated, freshwater-carbonate tufa. Runoff from rainfall events entrains clay that then adheres to the tufa surface. A specimen collected from southwest Japan exhibits >100 clay bands formed during a 15 yr period from 1988 to 2002. The chronology of the clay bands was determined by using the temperature-dependent cycle of oxygen isotope ratios in the carbonate, and the clay content was quantified by electron-probe microanalysis Si X-ray strength. The obtained Si X-ray strength curve and actual rainfall record follow a nearly identical pattern. The clay content of each band correlates with rainfall amount. Rainfall of >50 mm/week produced sufficient amounts of suspended cave clay to produce clay bands in the tufa. Because the clay bands survive postdepositional alteration, this new method can be applied to ancient tufa deposits and has great potential as a high-resolution rainfall record for use in climate studies.