The Newport line off Oregon - Studies in the North East Pacific

The Newport Hydrographic (NH) Line along 44.65°N off central Oregon was sampled seasonally during two epochs: 1961-1971 through the TENOC program and 1997-2003 through the GLOBEC Northeast Pacific Long Term Observations Program (LTOP); some observations are available for 2004 and 2005. During TENOC, the line extended 305 km offshore to 128°W, with stations 18 km apart over the continental shelf and 36 km offshore. During LTOP, the line was shorter (to 126°W) with closer station spacing over the continental shelf (9 km apart) and slope (18 km apart). LTOP cruises included biochemical sampling and underway current measurements. During both TENOC and LTOP, the seasonal cycle is very strong (accounting for >50% of the variance in surface layer properties), with rapid transitions in spring and fall. The summer regime is subject to coastal upwelling driven by southward winds, equatorward surface currents, and advection of low-salinity waters from the Columbia River. The winter regime off Newport is subject to coastal downwelling and poleward surface currents driven by northeastward winds. Comparison between TENOC and LTOP summer regimes shows the near-surface layer (0-100 m) at most locations is significantly warmer and fresher during LTOP than TENOC, and steric heights over the continental margin are significantly higher. Comparison of LTOP and TENOC winters shows that average differences at most locations were not statistically significant, but that the variance of steric height and shelf-break temperatures was significantly higher during LTOP than TENOC. Interannual variability of climate indices is also stronger during LTOP, which included a rare Subarctic invasion in 2002 as well as the strong 1997-1998 El Niño. During both TENOC and LTOP, interannual variability of steric height is closely related to the El Niño/La Niña cycle. Nutrient concentrations and nitrate-to-phosphate ratios of upwelling-source waters vary inversely with halocline temperature. Both reflect alongshore advection by coastal currents: southward currents bring cool, nitrate-rich waters in summer (especially during the Subarctic invasion), and northward currents bring relatively warm, nitrate-poor waters to the NH line in winter (especially during El Niño). Seasonal and interannual variations in the nutrient level of upwelling-source water are reflected in time series of vertically-integrated chlorophyll over the LTOP survey region (about 150 km by 300 km). Seasonal variations in chlorophyll and currents are congruent with seasonal variations in copepod biomass and diversity. We were not successful in establishing a clear connection between chlorophyll levels and interannual variations in copepod biomass or diversity, nor in explaining the large decrease in the survival rate of coho salmon between TENOC (6%) and LTOP (3%).     

Effects of furadan in the brown mussel Perna perna and in the mangrove oyster Crassostrea rhizophorae

Furadan is a carbamate pesticide used widely to combat agricultural pests. However little information is available about the toxicity of furadan in aquatic macroinvertebrates. The in vivo effects of furadan were evaluated in mussels, Perna perna, and oysters, Crassostrea rhizophorae. Glutathione S-transferase (GST), catalase (CAT) and cholinesterase (ChE) activities were measured in the gills of both species exposed to furadan (100 microg/l) for 96 h. No changes were observed in GST activity in the exposed groups. CAT activity was higher (9%) in the oysters exposed to furadan. ChE activity was inhibited by 64 and 35%, respectively, in C. rhizophorae and P. perna exposed to furadan, suggesting that the former is more susceptible to the toxic effects of furadan.     

Mechanisms of imposex induction in the mud snail, Ilyanassa obsoleta: TBT as a neurotoxin and aromatase inhibitor

The occurrence of imposex, imposition of male sex characteristics on female snails, has been extensively documented throughout the world. Tributyltin (TBT) and other organotins have been causally linked to imposex induction at levels as low as 2 ng/l. There are several proposed mechanisms of action. First, TBT has been shown to be neurotoxic and to accumulate in snail ganglia. Peptide hormones control sexual differentiation in gastropods, and one hypothesis is that TBT acts as a neurotoxin to abnormally release the peptide hormone Penis Morphogenic Factor (PMF). However, PMF has not been characterized to date. The neuropeptide APGWamide significantly induces imposex in the mud snail, Ilyanassa obsoleta, at 10⁻¹⁶ moles sub-cutaneous (SQ) injection over 2 weeks, and could be the PMF in this species. A second hypothesis is that TBT inhibits aromatase activity leading to increased testosterone levels and decreased estradiol. In vitro studies with snail digestive gland microsomes showed that TBT-dosed snails not exhibiting imposex had a 52% reduction in aromatase activity. Although the role of vertebrate sex steroids is not known in gastropods, it is possible that the combination of changes in peptide and steroid hormones may lead to imposex induction at extremely low doses of TBT.     

New production in the subarctic Pacific Ocean: Net changes in nitrate concentrations, rates of nitrate assimilation and accumulation of particulate nitrogen

Changes in water column nitrate and particulate nitrogen (PN) concentrations and rates of nitrate assimilation at 50°N 145°W were measured over a four-month interval for 1984, 1987 and 1988. Rates of nitrate depletion in the upper 80m of the water column averaged 12mg N m⁻²d⁻¹, but most of the net depletion occurred during May when rates were high (75mg N m⁻²d⁻¹) compared to later in the year. Particulate nitrogen (collected on GF/F filters) increased 2- to 3-fold during the month of May and accounted for 30–60% of the net nitrate depletion for May. Mean rates of PN accumulation for the 4-month intervals were 2.4mg N m⁻²d⁻¹ and accounted for about 20% of the net nitrate depletion. Rates of nitrate assimilation (measured in incubation bottles with ¹⁵N) averaged 45.0±4.5mg N m⁻²d⁻¹ (mean±SD), and appeared to decrease between May and September. A good correspondence between in situ and incubation estimates of nitrate assimilation was found for the 4-month comparison, but not for the month of May when net changes in nitrate concentrations were greatest. Vertical and horizontal inputs of nitrate are about the same order of magnitude as biological removal, thus the high inout of nitrate into the euphotic zone contributed to the continuously high nitrate concentrations in this region. Seasonal changes in nitrate and PN were significant and need to be considered in comparisons of new and export production.     

Evaluation of the effects of alteration and leaching on Sm–Nd and Lu–Hf systematics in submarine mafic rocks

Nd and Hf isotope systematics of oceanic basaltic rocks are often assumed to be largely immune to the effects of hydrothermal alteration. We have tested this assumption by comparing Nd and Hf isotope data for acid-leached Cretaceous oceanic basalts from Gorgona and DSDP Leg 15 with unleached data on the same rocks. Hf isotope values and Lu/Hf ratios are relatively unaffected by leaching, but ¹⁴³Nd/¹⁴⁴Nd values of leached samples are significantly higher than those of unleached fractions of the same sample in most cases. Furthermore, the Sm/Nd ratios of the majority of leached samples are 10–40% greater than those of unleached samples. X-ray diffraction studies indicate that selective removal of secondary minerals, such as smectite, during the acid leaching process is responsible for the fractionation of Sm/Nd ratios. These results have implications for interpretation of the Hf–Nd isotope systematics of ancient submarine rocks (older than ~ 50 Ma), as (1) the age-corrected ¹⁴³Nd/¹⁴⁴Nd ratio may not be representative of the primary magmatic signature and (2) the uncertainty of the age-corrected ε_Nd value may exceed the assumed analytical precision.     

Toponym matching through deep neural networks

Toponym matching, i.e. pairing strings that represent the same real-world location, is a fundamental problemfor several practical applications. The current state-of-the-art relies on string similarity metrics, either specifically developed for matching place names or integrated within methods that combine multiple metrics. However, these methods all rely on common sub-strings in order to establish similarity, and they do not effectively capture the character replacements involved in toponym changes due to transliterations or to changes in language and culture over time. In this article, we present a novel matching approach, leveraging a deep neural network to classify pairs of toponyms as either matching or nonmatching. The proposed network architecture uses recurrent nodes to build representations from the sequences of bytes that correspond to the strings that are to be matched. These representations are then combined and passed to feed-forward nodes, finally leading to a classification decision. We present the results of a wide-ranging evaluation on the performance of the proposed method, using a large dataset collected from the GeoNames gazetteer. These results show that the proposed method can significantly outperform individual similarity metrics from previous studies, as well as previous methods based on supervised machine learning for combining multiple metrics.