有孔蟲殼體微量元素比與穩定同位素組成：南海之校正與古海洋應用

• Main Authors: 黃國芳, Huang, Kuo-Fang
• Other Authors: 地球科學系碩博士班, 游鎮烽, You, Chen-Feng
• Format: Thesis
• Language: English
• Published: 2007
• Subjects: Trace element ratio; South China Sea; Isotope; 微量元素比; 南海; 同位素; Pacific; Sunda; Planktonic foraminifera
• Online Access: http://ir.lib.ncku.edu.tw/handle/987654321/20504; http://ir.lib.ncku.edu.tw/bitstream/987654321/20504/1/

None Long-term climatic and environmental reconstruction from marine sediments relies heavily on the reliable geochemical proxies from foraminiferal shells Therefore the aim of this dissertation is to improve our understanding and confidence in planktonic foraminiferal proxies as indicators of seawater physical (e g sea surface temperature SST) and chemical (e g composition and pH value) properties Toward this goal natural laboratory calibration works of foraminiferal shell chemistry have been carried out by using specimens collected from sediment traps and core-top sediments in the South China Sea (SCS) First of all high-precision and high-accurate measurements on trace element ratios boron and strontium isotopic compositions in planktonic foraminifera were investigated and have been successfully developed for carbonate samples (CHAPTER 2) This technique would allow us to obtain more reliable trace elements and isotopic compositions in foraminiferal shell calcites and provides a powerful tool to study the paleoceanographic reconstructions in the SCS Applying this analytical technique in situ Mg/Ca-SST calibration equation for the three abundant planktonic species in the SCS can be described as Mg/Ca = 0 32 exp (0 090*T) using time-series sediment trap and intense dissolution artifact has also been taken into account for obtaining more accurate SST records (CHAPTER 3) In addition a dissolution-corrected equation was developed using trap data from different depths These temperature equations are Mg/Ca= (0 38-0 02* water depth (km)) exp (0 090*T) and Mg/Ca= 0 30 exp (0 090*T) for fall-winter and spring-summer season respectively These equations can be applied for accurate reconstruction of mixed layer and thermocline temperatures in sediment cores Changes in the local SST and freshwater budget over the last 22 kyrs have been reconstructed from the tropical SCS (CHAPTER 4) Through the comparison of the east-west (zonal) SST gradient between western and eastern Pacific a persistent ENSO-like pattern can be seen during the Last Glacial Maximum (LGM) This is contrasted with mid-Holocene cooling suggest a La Niña-like pattern with enhanced SST gradients and strengthened trade winds Low salinity in the northern SCS during the LGM probably reflects an increase in freshwater inputs from several emerged rivers across the Sunda Shelf during glacial shelf exposure Alternatively an influence by the closure of the southern straits would cut off the inflow of high saline waters from the Indo-Pacific into the SCS cannot be excluded For the thermal structure of the upper water column Mg/Ca-based temperature variability reconstructed by three different planktonic species shows a strong mixing during the LGM and becomes more stratified during the Holocene Combining with shell chemistry the covariation of Mg/Ca and Ba/Ca over the last 220 kyrs indicates a variation in the continental input associated with the glacial-interglacial changes Results from core-top sediments in the SCS demonstrate that B/Ca ratios in the three planktonic foraminiferal species are strongly affected by seawater temperature and pH (CHAPTER 5) Species-specific foraminiferal B/Ca ratios from core-top sediments are in equilibrium with ambient seawater [B(OH)4-/HCO3-] in response to their habitat depths in the modern water column indicating shell B/Ca can be used as a reliable paleo-pH proxy by applying the dissolution-corrected Mg/Ca-derived SST and KD of B/Ca The pH and pCO2 reconstructions at Site 1145 indicate that the variability of surface-ocean pCO2 shows a more variability than atmospheric CO2 records and the ocean-atmosphere CO2 flux has substantially changed with time in the SCS The pCO2 (or pH) in glacial surface waters was approximately 150 ppmv lower (or 0 3 pH units higher) than the Holocene This amplitude in surface-water pCO2 is significantly larger than changes in atmospheric pCO2 recorded in the ice cores (~ 90 ppmv) This observation indicates that this area was a strong CO2 sink to the atmosphere at the LGM and presumably reflects a significant increase in regional primary productivity in the north SCS during the LGM The B isotopic analyses presented here have been successfully developed and extend its potential applications for various natural samples including seawaters pore waters and biogenic carbonates (CHAPTER 6) For the natural seawaters 11B shows a rather homogeneous distribution in the open ocean (11B= 39 6±0 2‰) but slightly negative 11B value can be determined in the coastal ocean (11B= 38 6±0 3‰) This can be attributed to the influence of riverine inputs (averaged 11B= +10‰ [Lemarchand et al 2000]) with substantially lighter 11B relative to seawater B and 11B in pore fluids collected from the southern Okinawa Trough and South China Sea show a complicated distribution pattern with depths indicating distinct geochemical processes or admixtures of fluids with various source end-members of B were involved in different geological settings Because of the large variability of B (200-2000 M) and 11B (+32-+51‰) in pore fluids the approach using foraminiferal 11B to estimate pH into long-term and short-term timescale needs to be interpreted cautiously due to potential influences of diagenetic processes that reacted with the surrounding sediment pore waters The result also indicates that the recovery yields of B extraction by microsublimation is close to 100% and shows that a precise determination of B isotopic ratios is possible for coral skeletons and foraminiferal shells This advance would offer an opportunity to evaluate the empirical calibration of 11B/pH and further refine the relationship between surface water pCO2 and atmospheric CO2 level in short- and long-term timescales The first detailed observations of Sr isotopic compositions (Sr ICs) in surface and vertical seawater profiles in the SCS and around Taiwan were investigated through the high-precision Sr isotopic measurement (CHAPTER 7) Combining results of Sr ICs T S and δ18O we have found that Sr ICs are distributed rather complicated and in-homogeneously both in vertical (Δ87Sr variation >40 ppm) and surface (>50 ppm) distributions after a heavy flood event In the Kao-ping river-sea system the surface Sr ICs were affected by mixing of three end-members water masses: two episodic continental runoffs with radiogenic Sr ICs and a modified SCSSW The two radiogenic sources were most likely resulted from: (1) unusual inputs of radiogenic Sr due to typhoon disturbance in the upper stream (i e top soils or ambient rocks) (2) intense water/sediment interaction in seawater column in response to heavy rainfalls or (3) normal river discharge plume into the coastal zone Vertical Δ87Sr profiles along the Kao-ping canyon show large isotopic variations in the upper most 200 m possibly were affected by continental runoffs (e g Kao-ping River: Δ87Sr~ 5500 ppm) and local seawater masses Below 200 m Sr ICs gradually become radiogenic while decreasing distance toward river mouth and reflect relatively high contribution of terrestrial inputs Sr isotopic composition in planktonic foraminifera shows a species dependency but is independent of pre-cleaning method and is in equilibrium with ambient seawater Sr isotopes The results presented here strongly suggest that the use of foraminiferal Sr isotopes for reconstructing seawater 87Sr/86Sr ratios with time A small but significant variation of about 40 ppm in the seawater 87Sr/86Sr can be found in the SCS and seems to follow a cycle to the previously reported 100-kyr cycle This periodicity can be linked to one of the prominent cycles in the Earth’s orbital parameters which are known to modulate the patterns of solar insolation and climate On this short timescale these changes are most likely be controlled by variations in the riverine Sr input