Quantifying Upwelling and Freshening in Nearshore Tropical American Environments Using Stable Isotopes in Modern Gastropods

To identify and quantify upwelling and freshwater influences in contrasting tropical ecosystems, we performed stable isotope analyses (δ13C and δ18O) on 13 serially-sampled modern Conus shells collected from coastal waters in the southwestern Caribbean (SWC, non-upwelling) and gulfs of Chiriquí (nonupwelling) and Panama (upwelling) in the tropical eastern Pacific (TEP). Most shells reveal seasonal variations in temperature and/or seawater δ18O in their δ18O profiles. Unusually high or low seasonal δ18O values measure the intensity of seasonal upwelling or freshwater input, respectively. To quantify upwelling and freshening signals, baseline δ18O values free of seasonal upwelling and freshening have been calculated from average temperatures during rainy (nonupwelling) seasons and average salinities during dry (upwelling) seasons. Baseline-normalized δ18O profiles reveal little or no upwelling in the SWC and Gulf of Chiriquí, and strong upwelling in the Gulf of Panama, as well as strong freshwater input in most areas. Dry-season δ18O values for Gulf of Panama Conus can exceed the baseline by as much as 2‰, equivalent to seawater temperatures approximately 9 °C lower than normal. In contrast, rainy-season δ18O values can be as low as 1.8‰ below the baseline, equivalent to salinities approximately 7 units lower than dry-season values. We use shell δ18O range (Δ18O) and δ18O-δ13C (O-C) correlation to further identify upwelling and freshening environments and thus nutrient source and status. Eutrophic environments like the Gulf of Panama are characterized by high Δ18O (2.3‰–3.2‰) and little O-C correlation. In contrast, the oligotrophic environments of the SWC lead to low (0.4‰) to moderate (1.6‰) Δ18O and non-significant to positive O-C correlation. When applied to fossil shells, these methods can characterize the nutrient status of ancient ecosystems. The formation of the Central American isthmus isolated the Pacific and Caribbean during the late Neogene and Pleistocene (Coates and Obando 1996, Jackson and O’Dea 2013), resulting in major oceanographic and biological changes in tropical America (Fuglister 1960, Glynn 1972, Keigwin 1978, D’Croz et al. 1991, O’Dea et al. 2007). Understanding the relationship between this environmental perturbation and the evolutionary and ecological origins of the modern tropical American marine fauna requires detailed environmental data from coastal waters during the formation of the isthmus (Schmidt 2007, Farris et al. 2011), which are currently lacking. Modern coastal waters around the isthmus show extreme hydrologic variability. This is principally caused by the presence of strong seasonal wind-jet driven upwelling in the tropical eastern Pacific (TEP) and the general absence of upwelling in the southwest Caribbean (SWC) (D’Croz and Robertson 1997, D’Croz and O’Dea 2007). Additionally, within each ocean there are substantial environmental differences, particularly in the origins of nutrient-rich waters critical for maintaining productivity. BULLETIN OF MARINE SCIENCE. VOL 00, NO 0. 0000 2 In the present study, we explore novel approaches to distinguish and quantify seasonal upwelling and freshwater input preserved in the oxygen and carbon isotopic profiles of serially sampled gastropod shells from modern tropical American coastal environments. The aim is to provide high-resolution data to be used as reference points to help interpret the significance of isotopic variations in fossil mollusk shells. In doing so, the data will contribute to characterizing the changing environmental conditions associated with the uplift of the Isthmus of Panama. Modern Tropical American Hydrology.—Trade winds blowing across the isthmus from late December to late March/April form wind jets where land elevation is <500 m (O’Dea et al. 2012). These wind jets push Pacific coastal waters offshore resulting in strong seasonal upwelling that brings cold, nutrient-rich water to the surface and greatly increases primary productivity (D’Croz and O’Dea 2007). The TEP is also affected by El Niño events, although the influence of El Niño on upwelling in the TEP remains unclear (O’Dea and Jackson 2002, Agujetas and Mitchelson-Jacob 2008, Okamura et al. 2013). In contrast, SWC waters generally do not experience seasonal upwelling, although there are three areas in the northern coast of Colombia and Venezuela that do experience Ekman-driven upwelling (Reuer et al. 2003). Consequently, the most striking difference across the isthmus is that upwelling in the TEP brings nutrients that drive a fervent planktonic productive system, whereas in the SWC productivity is shifted to the benthos because dissolved nutrient levels are considerably lower and principally supplied by terrigenous runoff (D’Croz et al. 2005, D’Croz and O’Dea 2007). The net movement of evaporative water by the trade winds from the Caribbean into the Pacific also results in the Caribbean being more saline (Fuglister 1960, Glynn 1972, Keigwin 1978, D’Croz et al. 1991). In addition to inter-ocean differences, TEP and SWC environments are internally variable. In the TEP, the variation of upwelling is correlated with the altitude of the isthmus (O’Dea et al. 2012). During the boreal winter, the trade winds are built up by the high pressure in the SWC and Gulf of Mexico. Only where the isthmus is sufficiently low in elevation are trade winds able to push surface waters away from the Central American coast, where they are replaced by upwelled deep water (D’Croz and O’Dea 2007, 2009). This wind-driven upwelling occurs in the Gulf of Panama but not in the Gulf of Chiriquí, where the cordillera is much higher. In the SWC, in addition to the three Ekman-driven upwelling regions, the seasonal surface water run-off pattern also varies along the Caribbean coast resulting from the seasonal movement of the Intertropical Convergence Zone (ITCZ, Lachniet and Patterson 2002, 2006). Stable Isotope Profiles as Records of Environmental Conditions.— Oxygen isotopes in serially sampled mollusks have been used as a proxy for seasonal variation in sea surface temperature (SST) for decades (e.g., Lowenstam and Epstein 1954, Killingley 1981, Krantz 1990, Jones and Allmon 1995, Kobashi and Grossman 2003, Surge and Walker 2006). Carbon isotopes, though influenced by vital effects (e.g., Lorrain et al. 2004, Gillikin et al. 2006), still provide an environmental record of the δ13C of ambient dissolved inorganic carbon (DIC; Mook 1971, Fritz and Poplawski 1974, Gentry et al. 2008, McConnaughy and Gillikin 2008, Beirne et al. 2012). The coupling of oxygen and carbon isotopic profiles can be used to detect seasonal upwelling (Killingley and Berger 1979) and freshwater input (Mook 1971). In the absence of other influences, shell accreted during times of upwelling should have higher δ18O and lower δ13C. This is because upwelled waters are cool, 18O-enriched, TAO ET AL.: QUANTIFYING UPWELLING AND FRESHENING IN NEARSHORE TROPICAL ENVIRONMENTS 3 and 13C-depleted, resulting in a negative δ18O-δ13C (O-C) correlation (Killingley and Berger 1979, Kroopnick 1980). This relationship, however, is not always simple as the δ13C minima and δ18O maxima are not always synchronous (Killingley and Berger 1979, Robbins 2010). Conversely, carbonate deposited during times of increased freshwater is typically depleted in both δ18O and δ13C, a reflection of the 18O-depletion of meteoric waters and 13C depletion of soil CO2, resulting in a positive O-C correlation in the isotopic profiles of mollusks (Mook 1971, Surge et al. 2003). Previous studies of modern mollusks from the TEP and SWC have successfully used oxygen isotopes to distinguish upwelling and non-upwelling environments (Geary et al. 1992, Bemis and Geary 1996). Geary et al. (1992) observed a significantly higher δ18O range for a strombid gastropod from the Gulf of Panama (4.5‰) vs one from a Caribbean non-upwelling area (1.5‰). Bemis and Geary (1996) obtained similar results for venerid bivalves. The present study builds upon the work by Bemis and Geary (1996), but rather than analyzing slow-growing bivalves, we analyze fast-growing gastropods to provide high-resolution isotopic profiles. The data are used for the first time to quantify upwelling and freshening, both critical factors in the delivery of nutrients. These data will then be used to develop a method for quantifying upwelling and freshwater influence in ancient tropical environments. Samples and Study Area.—To help isotopically characterize the environmental variability in isthmian waters, we analyzed 13 modern gastropod shells, some collected live, from a variety of environmental settings. These include three non-upwelling areas in the Caribbean (Bocas del Toro, San Blas Archipelago, Golfo de los Mosquitos), and upwelling (Gulf of Panama) and non-upwelling (Gulf of Chiriquí) areas on the Pacific coast of Panama (Table 1, Fig. 1). Specimens were Table 1. Taxonomy, location information, collection date, and dimensions of Conus shells used in this study. Up = upwelling, SL = shell length, SW = shell width, WL = whorl length. Sample ID Species Locale Collection date Lat (°N) Lon (°W) Up Depth (m) SL (mm) SW (mm) WL (mm) Sediment type1