Volume 1 Issue 9 - October 19, 2007
Tracing Fresh Water Plume after Typhoon by High Precision Sr Isotopic Composition in the Kao-Ping Estuary, Southern Taiwan
Kuo-Fang Huang and Chen-Feng You*

Department of Earth Sciences, National Cheng Kung University; Also at Earth Dynamic System Research Center, National Cheng Kung University
Email: cfy20@mail.ncku.edu.tw
Geophysical Research Letter 34, doi:10.1029/2006GL028253
In tropical estuarine, plumes of freshened water derived from runoffs or episodic floods (e.g., Typhoon) influence greatly the physical, chemical and biological conditions of these systems, and which may also have profound effects on the biogeochemical cycles of elements in coastal oceans. In addition to the riverine transport, submarine groundwater discharge provides up to 40% river fluxes into the oceans. Tracing fresh water plumes, therefore, are critical issues for a better understanding of nutrient and chemical supplies in the coastal regions. Tracers including T, S, trace elements, stable isotopes, and nutrients are useful tools for water masses identification. However, some tracers were affected significantly by biological uptake or particle scavenged. Oxygen and neodymium isotope applications were limited due to analytical complications in pre-concentration and purification, or lacks of superior sensitivity. Sr isotopic compositions (ICs) have been widely used in global climate change and marine carbonate studies. These applications assumed strontium behaves conservatively because of its long residence time compared with accepted ocean mixing model. Homogeneous Sr ICs were reported in the Hudson Bay, Pacific Ocean, Atlantic Ocean and Artic Ocean, as well as in marine carbonates. Recent more vigorous studies, however, have found considerable differences in regional Sr flux and Sr ICs do occur, especially in marginal seas.

Figure 1: Distribution of Sr isotopic compositions in surface seawater collected from the KPE.
Distributions of water masses at estuary are complicated due to intense regional mixing and hydro-morphological restriction. The spatial distributions of surface waters at the Kao-Ping Estuary (KPE) are rather complicated due to regional intense mixing of various sources, which vary significantly with seasonal changes. Here we apply high precision 87Sr/86Sr measurement (2σ= ±3 ppm) in seawater, together with dissolved Ba and Mn, to trace freshened plumes after Typhoon Toraji. The surface patterns of Sr ICs in the KPE are rather heterogeneous (Fig. 1), and small, but distinguishable Sr isotopic variations were detected in the KPE, Δ87Sr varied up to 55 ppm (12.7-64.9 ppm). In the following discussion, we focus to shed lights on circulation pattern and water masses near the KPE in a wet season to assess potential application of Sr ICs for studying freshwater migration. In wet season, continental runoff plays the most important role in nutrient supply and hydrology in the coastal ocean because approximately 90% of annual runoffs occur in May to October. Conversely, the Sr eolian inputs would be significantly furnished in dry season. The estimated dust Sr in KPE is rather small compared with river fluxes in wet season. Therefore, the Sr isotopic variation in the KPE should be related to enhanced river inputs delivered from the KPR after flood event. The freshwater plumes in the KPE generated by Toraji typhoon resulted in ~1-2‰ lower salinity compared with water samples collected in the summer and winter 2002, notably indicating large amounts of fresh water remained at the surface ocean.

The surface ocean circulation near the KPE is strongly influenced by prevailing winds in response to local topography and seasonal reversing monsoon. For example, in summer, the cyclonic gyre in the northern SCS transports the modified SCSSW to mix subsequently with local masses around the KPE. This process might also affect seawater Sr ICs significantly. The Sr ICs distribution showed a rather complicated picture and no straightforward correlation with distance from the KPE and may partly due to complexity in circulation structure, local wind forcing and the KC intrusion.

Of special interesting is that the most radiogenic Δ87Sr occupied in the southern part, ~30 km away from the KPE and formed an eddy-like pattern. Since chemical or biologic processes could not fractionate Sr ICs significantly, physical disturbances associated with typhoon are likely causes to explain these observations. Three possible scenarios that resulted in heavy Sr ICs after the flood event are: (1) intense seawater/sediment interaction caused by episodically increased suspended particles after large storms; (2) compositional changes in solid phases under episodically environmental disturbances; (3) inputs of more radiogenic Sr due to source region change in stream discharge during flood events. The latter two scenarios are in good agreement with Sr ICs in the adjacent river, Erjen River, where riverine Sr isotopes are positively correlated with discharge rates. In the Δ87Sr- salinity diagram (Fig. 2a), the characteristics of Sr ICs in regional sources are identified clearly by three end-members mixing trends. The Sr ICs and hydrographic characteristics are (1) EM-1: the most radiogenic Sr ICs and relatively low salinity, (2) EM-2: enriched Δ87Sr and the lowest salinity, and (3) EM-3: depleted in Sr ICs together with high salinity. Although limited in database, those results can be fitted with two highly correlated linear trends, allowing us to define possible Sr ICs in each end-member component. The water masses mixing processes are likely to be more complicated, but the present data do warrant value discussion on source mixing among the three dominant end-members near the KPE coastal zone. The saline water, EM-3, is possibly modified from the prime SCSSW by inflowing KC through the Luzon Strait. The less radiogenic Sr is possibly fingerprints of volcanic materials transported from Luzon Arc. Oxygen isotopes also supported this viewpoint and fall in a mixing trend between typical SCS water and typical KC water.

Figure 2: (a) Δ87Sr vs. salinity, (b) Δ87Sr vs. dissolved Ba, and (c) Δ87Sr vs. dissolved Mn in surface layer of the KPE after the Typhoon Toraji. These observations provide strong evidences for three end-members mixing between coastal waters and freshwater near this coastal zone.
The EM-1 and EM-2 end-members share common characteristics of low salinity and heavy Sr ICs. Apparent end-member for EM-2 and EM-1 can be estimated by extrapolating the correlation plot to zero salinity (Fig. 2a), Δ87Sr= 337 and 1339 ppm, respectively. These values are, however, much lower than that of the KPR river end-member, indicating strong removal of radiogenic Sr occurred at the freshwater-seawater boundary. Formation of Fe-Mn oxyhydroxides can scavenge Sr from dissolved loads, and tends to decrease the differences in 87Sr/86Sr ratio. Further seaward, ion exchanges between coastal water and sediment or dissolution of Fe-Mn particulates taking place and Sr redistribution may have occurred. The EM-2 component shows distinct chemical features of low salinity and median Δ87Sr. Dissolved Ba and Mn in these samples display large enrichment relative to EM3 (Fig. 2b), possibly reflecting a more reductive condition in the coastal zone after typhoon. Dramatic changes in physicochemical and hydrodynamic conditions after episodic floods were reported in the sub-tropical Brunswick Estuary, Australia. They observed increases in concentrations of nutrients and suspended sediments, as well as decreased dissolved oxygen in water columns during one week post-flood. In addition, dissolution of Fe-Mn particulates via bacterial reduction or photo-assisted reduction of particulate Mn (IV) by organic substances in estuary will release adsorbed-Ba and Mn back to dissolved pool. Similar processes modify also the Δ87Sr significantly. For EM-1, additional sources of heavy Sr ICs could only be derived from terrigenous particles carried by flood plume. The dissolved Ba and Mn results agree well with this viewpoint, indicating Ba desorption via ion-exchange with less amounts of Mn.

Based on these geochemical constraints, two unique sources with heavy Δ87Sr were identified and subsequently mixed with the South China Sea Surface Water. The most radiogenic Sr source can be attributed to intensive water-sediment interaction generated by typhoon. River discharge related redox variations cause the second radiogenic end-member.
< previousnext >