Sequential anomaly detection in the presence of noise and limited feedback

This paper describes a methodology for detecting anomalies from sequentially observed and potentially noisy data. The proposed approach consists of two main elements: 1) filtering, or assigning a belief or likelihood to each successive measurement based upon our ability to predict it from previous noisy observations and 2) hedging, or flagging potential anomalies by comparing the current belief against a time-varying and data-adaptive threshold. The threshold is adjusted based on the available feedback from an end user. Our algorithms, which combine universal prediction with recent work on online convex programming, do not require computing posterior distributions given all current observations and involve simple primal-dual parameter updates. At the heart of the proposed approach lie exponential-family models which can be used in a wide variety of contexts and applications, and which yield methods that achieve sublinear per-round regret against both static and slowly varying product distributions with marginals drawn from the same exponential family. Moreover, the regret against static distributions coincides with the minimax value of the corresponding online strongly convex game. We also prove bounds on the number of mistakes made during the hedging step relative to the best offline choice of the threshold with access to all estimated beliefs and feedback signals. We validate the theory on synthetic data drawn from a time-varying distribution over binary vectors of high dimensionality, as well as on the Enron email dataset. © 1963-2012 IEEE.

Resonant transmission near nonrobust periodic slab modes.

We present a precise theoretical explanation and prediction of certain resonant peaks and dips in the electromagnetic transmission coefficient of periodically structured slabs in the presence of nonrobust guided slab modes. We also derive the leading asymptotic behavior of the related phenomenon of resonant enhancement near the guided mode. The theory applies to structures in which losses are negligible and to very general geometries of the unit cell. It is based on boundary-integral representations of the electromagnetic fields. These depend on the frequency and on the Bloch wave vector and provide a complex-analytic connection in these parameters between generalized scattering states and guided slab modes. The perturbation of three coincident zeros-those of the dispersion relation for slab modes, the reflection constant, and the transmission constant-is central to calculating transmission anomalies both for lossless dielectric materials and for perfect metals.

Long-term changes in phytoplankton, zooplankton and salmon related to climate

Recently, large-scale changes in the biogeography of calanoid copepod crustaceans have been detected in the northeastern North Atlantic Ocean and adjacent seas. Strong biogeographical shifts in all copepod assemblages were found with a northward extension of more than ° in latitude of warm-water species associated with a decrease in the number of colder-water species. These changes were attributed to regional increase in sea surface temperature. Here, we have extended these studies to examine long-term changes in phytoplankton, zooplankton and salmon in relation to hydro-meteorological forcing in the northeast Atlantic Ocean and adjacent seas. We found highly significant relationships between (1) long-term changes in all three trophic levels, (2) sea surface temperature in the northeastern Atlantic, (3) Northern Hemisphere temperature and (4) the North Atlantic Oscillation. The similarities detected between plankton, salmon, temperature and hydro-climatic parameters are also seen in their cyclical variability and in a stepwise shift that started after a pronounced increase in Northern Hemisphere Temperature anomalies at the end of the 1970s. All biological variables show a pronounced change which started after circa 1982 for euphausiids (decline), 1984 for the total abundance of small copepods (increase), 1986 for phytoplankton biomass (increase) and Calanus finmarchicus (decrease) and 1988 for salmon (decrease). This cascade of biological events led to an exceptional period, which is identified after 1986 to present and followed another shift in large-scale hydro-climatic variables and sea surface temperature. This regional temperature increase therefore appears to be an important parameter that is at present governing the dynamic equilibrium of northeast Atlantic pelagic ecosystems with possible consequences for biogeochemical processes and fisheries.

A regime shift in the North Sea circa 1988 linked to changes in the North Sea horse mackerel fishery

After 1987, Phytoplankton Colour (a visual estimate of chlorophyll) measured on samples taken by the continuous plankton recorder (CPR) in the North Sea increased substantially, both in level and seasonal extent, compared to earlier years since 1946. Many species of phytoplankton and zooplankton showed marked changes in abundance at about the same time. These events coincided with a large increase in catches of the western stock of the horse mackerel (Trachurus trachurus L.) in the northern North Sea reflecting a northerly expansion of the stock along the shelf edge from the Bay of Biscay to the North Sea after 1987. Using a 3D hydrodynamic model, with input from measured wind parameters, monthly transport of oceanic water into the North Sea has been calculated for the period 1976–1994, integrated for a section from Orkney to Shetland to Norway. A substantial increase in oceanic inflow occurred in the winter months, December to March, from 1988. Higher sea surface temperatures were also measured after 1987 especially in spring and summer months. These biological and physical events may be a response to observed changes in pressure distribution over the North Atlantic. From 1988 onwards, the North Atlantic Oscillation (NAO) index, the pressure difference between Iceland and the Azores, increased to the highest positive level observed in this century. Positive NAO anomalies are associated with stronger and more southerly tracks of the westerly winds and higher temperatures in western Europe. These changing wind distributions may have led to an increase in the northerly advection of water along the western edge of the European shelf and may have assisted the migration of the horse mackerel. This study is possibly a unique demonstration of a correlation between three different trophic levels of a marine ecosystem and hydrographic and atmospheric events at decadal and regional scales. The results emphasise the importance of maintaining into the future long term programmes such as the CPR.

Ocean structure and climate (Eastern North Atlantic): in situ measurement and remote sensing

Structure and climate of the east North Atlantic are appraised within a framework of in situ measurement and altimeter remote sensing from 0 degree - 60 degree N. Long zonal expendable bathythermograph /conductivity-temperature-depth probe sections show repeating internal structure in the North Atlantic Ocean. Drogued buoys and subsurface floats give westward speeds for eddies and wavelike structure. Records from longterm current meter deployments give the periodicity of the repeating structure. Eddy and wave characteristics of period, size or wavelength, westward propagation speed, and mean currents are derived at 20 degree N, 26 degree N, 32.5 degree N, 36 degree N and 48 degree N from in situ measurements in the Atlantic Ocean. It is shown that ocean wave and eddy-like features measured in situ correlate with altimeter structure. Interior ocean wave crests or cold dome-like temperature structures are cyclonic and have negative surface altimeter anomalies; mesoscale internal wave troughs or warm structures are anticyclonic and have positive surface height anomalies. Along the Eastern Boundary, flows and temperature climate are examined in terms of sla and North Atlantic Oscillation (NAO) Index. Longterm changes in ocean climate and circulation are derived from sla data. It is shown that longterm changes from 1992 to 2002 in the North Atlantic Current and the Subtropical Gyre transport determined from sla data correlate with winter NAO Index such that maximum flow conditions occurred in 1995 and 2000. Minimum circulation conditions occurred between 1996-1998. Years of extreme negative winter NAO Index result in enhanced poleward flow along the Eastern Boundary and anomalous winter warming along the West European Continental Slope as was measured in 1990, 1996, 1998 and 2001.

Structure of phytoplankton (Continuous Plankton Recorder and SeaWiFS) and impact of climate in the Northwest Atlantic Shelves

All marine organisms are affected to some extent by the movement and thermal properties of oceanic currents. However phytoplankton, because of its small size is most directly coupled to the physical environment. The intense hydrodynamic activity observed in the Northwest Atlantic Shelves Province makes this region especially intriguing from the point of view of physical-biological interactions. In the present work, remote sensed data of Sea Surface Height (SSH) anomalies, Sea-surface chlorophyll a concentrations (SeaWiFS), and Sea Surface Temperature (SST) are used to complement the Continuous Plankton Recorder (CPR) survey that continuously sampled a route between Norfolk (Virginia, USA; 39° N, 71° W) and Argentia (Newfoundland; 47° N, 54° W) over the period 1995–1998. Over this period, we examined physical structures (i.e. SST and SSH) and climatic forcing associated with space-time phytoplankton structure. Along this route, the phytoplankton structures were mainly impacted by the changes in surface flow along the Scotian Shelf rather than significantly influenced by the mesoscale features of the Gulf Stream. These changes in water mass circulation caused a drop in temperature and salinity along the Scotian Shelf that induced changes in phytoplankton and zooplankton abundance.

Aspects Of Gametogenesis In The Marine Mussel Mytilus-Edulis-L

Marine bivalve molluscs have in recent years attracted considerable attention for a variety of reasons, not least of which is their importance as a source of food for man. Much of this research has concentrated on studies of reproduction; Mytilus viridis (India: Nagabhushanam & Mane, 1975), M. edulis aoteanus and Aulacomya maoriana (New Zealand: Kennedy, 1977);Choromytilus meridionalis and Aulacomya ater (South Africa: Berry, 1978); Mytilus (= Perna) perna (Brazil: Lunetta, 1969); M. edulis planulatus (Australia: Wilson & Hodgkin, 1967); Mytilus californianus and M. edulis (U.S.A.: Hines, 1979); Mytilus galloprovincialis (France: Lubet, 1959) and M. edulis (U.K.: Chipperfield, 1953; Seed, 1975; Bayne et al. 1978). A review of the literature revealed that in the majority of studies cytology was used as a descriptive tool for the ‘staging’ (Chipperfield, 1953; Lubet, 1957; Seed, 1975, 1976) of the developing gametes and certain anomalies were apparent with regard to the nomenclature of the connective tissue matrix of the mantle lobes.

Particulate optical scattering coefficients along an Atlantic Meridional Transect

The particulate optical backscattering coefficient (bbp) is a fundamental optical property that allows monitoring of marine suspended particles both in situ and from space. Backscattering measurements in the open ocean are still scarce, however, especially in oligotrophic regions. Consequently, uncertainties remain in bbp parameterizations as well as in satellite estimates of bbp. In an effort to reduce these uncertainties, we present and analyze a dataset collected in surface waters during the 19th Atlantic Meridional Transect. Results show that the relationship between particulate beam-attenuation coefficient (cp) and chlorophyll-a concentration was consistent with published bio-optical models. In contrast, the particulate backscattering per unit of chlorophyll-a and per unit of cp were higher than in previous studies employing the same sampling methodology. These anomalies could be due to a bias smaller than the current uncertainties in bbp. If that was the case, then the AMT19 dataset would confirm that bbp:cp is remarkably constant over the surface open ocean. A second-order decoupling between bbp and cp was, however, evident in the spectral slopes of these coefficients, as well as during diel cycles. Overall, these results emphasize the current difficulties in obtaining accurate bbp measurements in the oligotrophic ocean and suggest that, to first order, bbp and cp are coupled in the surface open ocean, but they are also affected by other geographical and temporal variations.

Variation in ocean colour may help predict cod and haddock recruitment

Characteristics of the spring and fall phytoplankton blooms in spawning areas on the Scotian Shelf, Canada, were estimated from remote sensing data. These blooms, along with anomalies in the North Atlantic Oscillation, were used to explain variation in the recruitment of 4 populations of cod and haddock. We tested the effects of the timing of the bloom using the chlorophyll a (chl a) signal, the maximum amount of chl a, the timing of the diatom bloom, and the maximum relative dominance of diatoms on the recruitment (to Age 1) of cod and haddock on the Scotian Shelf. Models were run separately for the effects of the spring and fall blooms. Only 3 of 10 models tested (0-lag) explained significant (80 to 92%) variation in recruitment. However, the performance of these models was not consistent across populations or species, suggesting that generalities about how spring and fall phytoplankton blooms affect recruitment cannot yet be made. The differences among models suggest that fish larvae are probably adapted locally to food production and thus indirectly to the characteristics of the phytoplankton bloom, which in turn are influenced by regional (meso-scale) oceanographic conditions.

Unanticipated biological changes and global warming

Evidence of global warming is now unequivocal, and studies suggest that it has started to influence natural systems of the planet, including the oceans. However, in the marine environment, it is well-known that species and ecosystems can also be influenced by natural sources of large-scale hydro-climatological variability. The North Atlantic Oscillation (NAO) was negatively correlated with the mean abundance of one of the subarctic key species Calanus finmarchicus in the North Sea. This correlation was thought to have broken down in 1996, however, the timing has never been tested statistically. The present study revisits this unanticipated change and reveals that the correlation did not break down in 1996 as originally proposed but earlier, at the time of an abrupt ecosystem shift in the North Sea in the 1980s. Furthermore, the analyses demonstrate that the correlation between the NAO and C. finmarchicus abundance is modulated by the thermal regime of the North Sea, which in turn covaries positively with global temperature anomalies. This study thereby provides evidence that global climate change is likely to alter some empirical relationships found in the past between species abundance or the ecosystem state and large-scale natural sources of hydro-climatological variability. A theory is proposed to explain how this might happen. These unanticipated changes, also called ‘surprises’ in climatic research, are a direct consequence of the complexity of both climatic and biological systems. In this period of rapid climate change, it is therefore hazardous to integrate meteo-oceanic indices such as the NAO in models used in the management of living resources, as it has been sometimes attempted in the past.

Diet of auklet chicks in the Aleutian Islands, Alaska: similarity among islands, interspecies overlap, and relationships to ocean climate

Seabirds are effective samplers of the marine environment, and can be used to measure resource partitioning among species and sites via food loads destined for chicks. We examined the composition, overlap, and relationships to changing climate and oceanography of 3,216 food loads from Least, Crested, and Whiskered Auklets (Aethia pusilla, A. cristatella, A. pygmaea) breeding in Alaska during 1994–2006. Meals comprised calanoid copepods (Neocalanus spp.) and euphausiids (Thysanoessa spp.) that reflect secondary marine productivity, with no difference among Buldir, Kiska, and Kasatochi islands across 585 km of the Aleutian Islands. Meals were very similar among species (mean Least–Crested Auklet overlap C = 0.68; Least–Whiskered Auklet overlap C = 0.96) and among sites, indicating limited partitioning of prey resources for auklets feeding chicks. The biomass of copepods and euphausiids in Least and Crested Auklet food loads was related negatively to the summer (June–July–August) North Pacific Gyre Oscillation, while in Whiskered Auklet food loads, this was negatively related to the winter (December–January–February) Pacific Decadal Oscillation, both of which track basin-wide sea-surface temperature (SST) anomalies. We found a significant quadratic relationship between the biomass of calanoid copepods in Least Auklet food loads at all three study sites and summer (June–July) SST, with maximal copepod biomass between 3–6°C (r 2 = 0.71). Outside this temperature range, zooplankton becomes less available to auklets through delayed development. Overall, our results suggest that auklets are able to buffer climate-mediated bottom-up forcing of demographic parameters like productivity, as the composition of chick meals has remained constant over the course of our study.

Synchronous response of marine plankton ecosystems to climate in the Northeast Atlantic and the North Sea

Over the last few decades, global warming has accelerated both the rate and magnitude of changes observed in many functional units of the Earth System. In this context, plankton are sentinel organisms because they are sensitive to subtle levels of changes in temperature and might help in identifying the current effects of climate change on pelagic ecosystems. In this paper, we performed a comparative approach in two regions of the North Atlantic (i.e. the Northeast Atlantic and the North Sea) to explore the relationships between changes in marine plankton, the regional physico-chemical environment and large-scale hydro-climatic forcing using four key indices: the North Atlantic Oscillation (NAO), the Atlantic Multidecadal Oscillation (AMO), the East Atlantic (EA) pattern and Northern Hemisphere Temperature (NHT) anomalies. Our analyses suggest that long-term changes in the states of the two ecosystems were synchronous and correlated to the same large-scale hydro-climatic variables: NHT anomalies, the AMO and to a lesser extent the EA pattern. No significant correlation was found between long-term ecosystem modifications and the state of the NAO. Our results suggest that the effect of climate on these ecosystems has mainly occurred in both regions through the modulation of the thermal regime.

Recent Arctic Climate Change and Its Remote Forcing of Northwest Atlantic Shelf Ecosystems

During recent decades, historically unprecedented changes have been observed in the Arctic as climate warming has increased precipitation, river discharge, and glacial as well as sea-ice melting. Additionally, shifts in the Arctic's atmospheric pressure field have altered surface winds, ocean circulation, and freshwater storage in the Beaufort Gyre. These processes have resulted in variable patterns of freshwater export from the Arctic Ocean, including the emergence of great salinity anomalies propagating throughout the North Atlantic. Here, we link these variable patterns of freshwater export from the Arctic Ocean to the regime shifts observed in Northwest Atlantic shelf ecosystems. Specifically, we hypothesize that the corresponding salinity anomalies, both negative and positive, alter the timing and extent of water-column stratification, thereby impacting the production and seasonal cycles of phytoplankton, zooplankton, and higher-trophic-level consumers. Should this hypothesis hold up to critical evaluation, it has the potential to fundamentally alter our current understanding of the processes forcing the dynamics of Northwest Atlantic shelf ecosystems.

Distribution and multi-annual abundance trends of the copepod Temora longicornis in the US Northeast shelf ecosystem

The average spatial distribution and annual abundance cycle are described for the copepod Temora longicornis from samples collected on broadscale surveys (1977-2006) and along continuous plankton recorder transects (1961-2006) of the US Northeast continental shelf ecosystem. After its annual low in winter, T. longicornis abundance begins to increase in coastal waters with the northern progression of spring conditions. Annual maximum shelf concentrations were found in the more southern inshore waters of the region during the summer months. Abundance throughout most of the ecosystem increased sharply in the early 1990s and remained high through 2001. During this period, the copepod became more numerous and widespread in offshore shelf waters. Abundance declined to approximately average levels in 2002 for the remainder of the time series, but its extended offshore range remained intact. Correlation analysis found that the copepods interannual abundance variability had a significant negative relationship with surface salinity anomalies throughout the ecosystem, with higher correlations found in the northernmost subareas. Temora longicornis abundance in the ecosystem's southernmost subarea (Middle Atlantic Bight) did not increase in the 1990s and was found to be negatively correlated to surface temperature, indicating that continued global warming could adversely impact the copepods annual abundance cycle in this region.

Effects on zooplankton of a warmer ocean: Recent evidence from the Northeast Pacific

The consequences for pelagic communities of warming trends in mid and high latitude ocean regions could be substantial, but their magnitude and trajectory are not yet known. Environmental changes predicted by climate models (and beginning to be confirmed by observations) include warming and freshening of the upper ocean and reduction in the extent and duration of ice cover. One way to evaluate response scenarios is by comparing how "similar" zooplankton communities have differed among years and/or locations with differing temperature. The subarctic Pacific is a strong candidate for such comparisons, because the same mix of zooplankton species dominates over a wide range of temperature climatologies, and observations have spanned substantial temperature variability at interannual-to-decadal time scales. In this paper, we review and extend copepod abundance and phenology time series from net tow and Continuous Plankton Recorder surveys in the subarctic Northeast Pacific. The two strongest responses we have observed are latitudinal shifts in centers of abundance of many species (poleward under warm conditions), and changes in the life cycle timing of Neocalanus plumchrus in both oceanic and coastal regions (earlier by several weeks in warm years and at warmer locations). These zooplankton data, plus indices of higher trophic level responses such as reproduction, growth and survival of pelagic fish and seabirds, are all moderately-to-strongly intercorrelated (vertical bar r vertical bar = 0.25-0.8) with indices of local and basin-scale temperature anomalies. A principal components analysis of the normalized anomaly time series from 1979 to 2004 shows that a single "warm-and-low-productivity" vs. "cool-and-high-productivity" component axis accounts for over half of the variance/covariance. Prior to 1990, the scores for this component were negative ("cool" and "productive") or near zero except positive in the El Nino years 1983 and 1987. The scores were strongly and increasingly positive ("warm" and "low productivity") from 1992 to 1998; negative from 1999 to 2002; and again increasingly positive from 2003-present. We suggest that, in strongly seasonal environments, anomalously high temperature may provide misleading environmental cues that contribute to timing mismatch between life history events and the more-nearly-fixed seasonality of insolation, stratification, and food supply. Crown Copyright (c) 2007 Published by Elsevier Ltd. All rights reserved.