Ocean Sciences

COMPASS (Combined OCE MPO ATM Seminar Series) FRIDAY


(11:00 in the Auditorium)


Feb 10 --- 15-min seminars:

Lisa Bucci (MPO)

"An Airborne Doppler Wind Lidar in Tropical Cyclones"

For the first time, the NOAA P3 Orion flew a Doppler Wind Lidar (DWL) in 3 tropical cyclones (TCs) during the 2015 and 2016 hurricane seasons. Instruments typically flown on the Hurricane Hunter Aircraft include a Tail Doppler Radar (TDR) that provides the most comprehensive coverage of the wind field in TCs. TDR has the ability to measure winds in areas of precipitation but outside of these regions and below approximately 1 km, it either cannot make observations or the data is less reliable. This study shows that the DWL is able to retrieve measurements in these areas making it a complementary observation set to data from the TDR. A field experiment was designed for the 2016 hurricane season to target the regions the TDR cannot make measurements. Validation of DWL observations is in progress. Comparisons to other in situ wind measurements, like from the GPS dropwindsondes, show varying results depending on the scanning pattern on the DWL.

Valeria Donets (ATM)

"Shear Lines Facilitate Transport of Reactive Trace Gases from East Asia into the Deep Tropics"

Rapid industrialization in Eastern Asia has in recent decades led to increases in anthropogenic pollutants in the regional atmosphere. This pollutant burden is transported from local and regional sources into the global atmosphere and has been shown to be a significant source of intercontinental and cross-border anthropogenic pollution. Such transport is greatly facilitated during the winter phase of the East Asian monsoon, when cold air-masses originating over regions of Siberia are carried across East Asia into the Philippine and South China Seas, and further towards lower latitudes, sweeping off low-level pollution in the process. Over the ocean water heating from the ocean surface reduces thermal gradients across the frontal boundary, leaving behind a shear line. North of the shear line, the formerly cold air mass still contains the diluted pollutants picked up as the air crossed the source regions in East Asia.

Here we present measurements of chemically reactive trace gases collected during the CONvective TRansport of Active Species in the Tropics (CONTRAST) mission, an airborne research study conducted during January-February, 2014, over the tropical West Pacific Ocean. This presentation describes shear line facilitated transport of pollution into the tropics during boreal winter and its effects on chemical composition of the troposphere in the deep tropics.

Feb 17 (VISITOR) --- Dr. Hezi Gildor (Weizmann Institute of Science, Israel)

"Ocean Mixing: From Horizontal Barriers to Mixing to Vertical Density Currents"

Feb 24 --- 15-min seminars:

Rachel Zelinsky (MPO)

"Convective Characteristics of the ITCZ and MJO over the Indian Ocean"

The MJO and the ITCZ are both distinctive forms of large-scale convective organization over the Indian Ocean.  One of the surprising outcomes of the DYNAMO field campaign was that an interaction was observed between the MJO and ITCZ for the first time. In this study, cloud and rainfall statistics are calculated to determine if the interaction between the ITCZ and MJO initiation is just a transition of convection in space or if a convective regime change occurs. A method is developed to objectively identify the ITCZ over the Indian Ocean using TRMM 3B42 daily rain rate data. The MJO is identified through a local precipitation tracking method and using the RMM index. The cloud characteristics are calculated for all identified ITCZ and MJO events from 2002-2013 using TRMM precipitation feature dataset, establishing the climatology of organized convection over the Indian Ocean. This climatology is described in terms of stratiform and convective precipitation, precipitation feature size, and other statistics of cloud and precipitation features. Preliminary results indicate the transition of convection from the ITCZ to the MJO does not involve a dramatic regime change in the convective characteristics. Instead, the absolute number of precipitation features change depending on the phase of the MJO but all convection has the same underlying distribution over the Indian Ocean.

Alessandro Cresci (OCE)

"Magnetic Orientation of Early Life Stages of Eels"

The Earth magnetic field (EMF) is a vector quantity composed by horizontal and vertical components. Its intensity and inclination change in latitude and longitude. Therefore, it is widely used in the animal kingdom for migrations. In the marine environment, organisms such as lobster, sea turtles, and salmon also use the EMF to navigate. However, whether fish have such magnetic sense at early life history stages is still unknown. My dissertation will focus on how post-larval eels (Anguilla spp.) use the EMF for orientation. Eels are catadromous; they inhabit freshwater stream and migrate to the Sargasso Sea to spawn. Their larvae drift for months up to the continental shelf, where they metamorphose into glass eels, which then recruit in the estuaries. How glass eels find the streams has yet to be discovered. It has been assumed that glass eels use the magnetic field for compass orientation. To test this hypothesis, we collected glass eels in Austevoll (Norway) and observed their orienting behavior in a circular arena, both drifting in situ under natural environmental conditions and placed in a magnetic laboratory where the EMF was rotated. Most (98%) of the glass eels tested in situ exhibited a preferred bearing at the individual level. At the population level, eels had a common orientation during the ebb. In the laboratory, in presence of the manipulated magnetic field only, 71% of the same eels significantly oriented as individuals, and showed the same common orientation during the ebb tide. These results demonstrate that glass eels possess a magnetic compass linked to a circatidal rhythm. Numerical modeling will be used to understand the role of the orientation behavior during their recruitment phase of the extraordinary eel migration. Preliminary simulations using the CMS coupled with HYCOM-Global 1/12o demonstrate the possible role of compass orientation in glass eel.

Mar 03 --- 15-min MPO seminars

Diane Palko

"Understanding Coastal Sea Level and its Relationship to Ocean Circulation"

The accurate prediction of trends at a regional scale is essential for adapting to and preparing for the effects of climate change. The prediction of local sea level trends is particularly critical. Sea level rise is one of the most often cited consequences of climate change and is presently impacting cities along the east coast of the United States. Currently, typical general circulation models (GCMs) have 1.0° resolution in the ocean, and are unable to resolve features, like the Gulf Stream, which may have large effects on regional sea level. The main goal of this research is to utilize high-resolution (HR) (0.1° resolution in the ocean) coupled model runs of CCSM4 to analyze regional sea surface height (SSH) trends. Our first step is to characterize the natural variability of SSH along the east coast of the United States in HR runs of CCSM4 with fixed forcing. The variability of coastal SSH from these runs is then compared to reanalysis and unforced, low-resolution (LR, 1.0°) runs of CCSM4. The connection of coastal SSH to basin wide climate patterns is visualized by plotting the correlation between SSH at coastal locations and SSH, SST, and surface KE at all other locations in the north Atlantic. By characterizing the natural variability of SSH in models and observations we strive to understand what processes influence SSH along the east coast of the United States and to improve the prediction of sea level rise. The next step in our research is to utilize HR hosing experiments to understand the regional SSH response to anthropogenic forcing.

Steven Simon "Applying a Statistical Pattern Matching Algorithm to Correct for Spatial Biases in GCM Precipitation Climatologies: Applications to South Florida Water Resource Management"

Under the scenario of greenhouse gas-forced anthropogenic global warming, Coupled Global Climate Models (GCMs) commonly project a tendency for wet regions to become wetter and dry regions to become drier. Such projections pose an assortment of problems for water resource managers and planners, since future regional rainfall patterns may fall outside the historical range of variability on which water resource districts are managed. In South Florida, the federal government is currently investing several billion dollars to fund efforts to restore the Everglades ecosystem. However, these efforts are currently not guided by credible projections of future regional rainfall patterns. Due to their large regional biases, future rainfall projections simulated by GCMs have proven to be ineffective for direct use by South Florida water resource management. Recent research has attributed this intermodel spread in regional precipitation projections to differences in the GCMs’ climatological distributions of precipitation. Operating under this framework, this project posits that correcting for spatial biases in GCM precipitation climatologies, derived from CMIP5 historical runs, by using merged rain gauge-satellite observations from the GPCP monthly dataset will yield more consistent and accurate regional projections of precipitation change. While GCMs succeed in simulating recognizable precipitation patterns, they are often spatially displaced due to regional biases in model climatology. To address these regional spatial biases, this project presents a new approach for statistically downscaling GCM projections, which involves utilizing a “spatial pattern matching algorithm” that explicitly adjusts for regional displacements in GCM precipitation climatologies using GPCP observations and remaps them accordingly. This algorithm is evaluated under an array of parameter permutations using GCM precipitation climatologies. Preliminary results reveal that this algorithm provides a framework to remap the aforementioned regional spatial displacements in model precipitation climatologies, January and July in this case, so as to reduce regional biases. Future work will address and evaluate the robustness of this algorithm across 17 CMIP5 models and their associated precipitation climatologies and remappings.

Brian Matilla

"Extreme Precipitation Events over Tropical Asia: A Case Study over Hainan Island"

Cases of extreme precipitation accumulations are identified using rainfall observations from the Tropical Rainfall Measuring Mission (TRMM) 3B42 Version 7 and reanalysis of total precipitation from the Modern-Era Retrospective-analysis for Research and Applications (MERRA) over tropical Asia. We classify extreme precipitation events as rainfall accumulations exceeding the 99th percentile. Because of their differing spatial scales, both datasets were combined, re-gridded, and remapped to a common 2º x 2º mesh such that a scatterplot of the two datasets can be generated. Data from 1998 to 2013 over a latitude-longitude box 10-30º N, 90-110º E is considered. Comparisons between the two datasets within the specified latitude-longitude box show that MERRA under-analyzes precipitation compared to TRMM 3B42 observations, consistent with previous studies of similar nature.

A case was selected using TRMM 3B42 observations described in Mapes (2011): a weak tropical cyclone (Tropical Storm 28W; 2000; Western Pacific) that meandered over the South China Sea. Rainfall over Hainan island was estimated to be over 200 mm over a 36-hour period from 13 October 0900Z to 14 October 2100Z. Dynamics and thermodynamic parameters pertaining to 28W, including select variables from MERRA and the European Reanalysis (ERA)-Interim products, will be presented to detail the meteorological story of this case.

Our next step involves studying factors that most significantly contribute to this extreme precipitation event. We plan to employ the Weather Research and Forecast (WRF) model to allow for modifications of boundary conditions and then simulating under these modifications.

Mar 10 --- 15-min seminars: 

Mingming Shao (AMP)

The Variability of Fluxes Observed near Sub-Mesoscale Fronts in the Gulf of Mexico"

Submesoscale fronts (SF), O(1–10 km), spontaneously emerge during the evolution of mesoscale eddies, and they can be easily detected by their surface expression. These SF may be strongly coupled with the air-sea interaction near the frontal boundary, which has been seldom studied and may effect the air-sea exchange coefficients. The reliable estimates of these coefficients are essential for climate studies, model development and quantify the net air-sea gas fluxes over the global ocean. To address these issue, direct measurements of momentum, heat and water vapor flux were collected on the R/V Walton Smith during the LAgrangian Submesoscale Experiment (LASER) in the Northern Gulf of Mexico in January-February 2016. Twin flux towers were mounted on each of the prows of the R/V Walton Smith catamaran, and for each tower eddy covariance measurements of the fluxes were made at 3 vertical levels. Surface elevation was simultaneously observed from a bow-mounted array of five Ultrasonic Distance Meters (UDM). Several individual periods of strong frontal action have been identified. Primary analysis result showed an atmospheric horizontal gradients of 0.5° to 1° appeared near the SF in stable condition. Wind intensity decrease from warm side to cold side when synoptic wind blow from warm side. A sudden increase of wind stress appeared in the down wind direction, which indicate the existence of internal boundary layer. Surface latent heat fluxes always increase from cold side to warm side due to the increase wind speed and temperature. All these features are in agreement with former work focus on mesoscale fronts. But under unstable conditions, the variability of momentum and heat fluxes show distinct features.

Marcia Trillo (OCE)

"Insights from the Gulf Science Data on the Nature of the Macondo Blowout"

During the Deepwater Horizon accident in 2010, oil spilled into the Gulf of Mexico and approximately 770,000 gallons of chemical dispersants were injected directly at the Macondo wellhead attempting to keep the oil submerged. Despite conducting thorough research in the past seven years, a detailed empirical analysis as the oil was spewing from the wellhead and after the well capping, has not yet been completely undertaken. Recently, BP made available a unique Gulf Science dataset of oil samples collected from May 2010 to July 2012. In this research, we examine this fully comprehensive dataset to determine the temporal and spatial distribution of the chemical signature of Macondo oil up to December 2010. Oil samples are classified monthly in two hydrocarbon categories (i.e., C5-C12 and C13+) and the distance from the wellhead is also computed for these classifications in order to better analyze the different hydrocarbon partitioning in the water column. We also split the hydrocarbon chemical data depending on the variable dispersant application periods to provide an exhaustive sub-sea dispersant injection (SSDI) analysis. Our spatial data results confirm the presence of the so called “deep-plume” (i.e. a dominant intrusion layer centered around 1,100 m) and show a clear vertical partition of heavier hydrocarbons higher in the water column, which is partly explained by the physical properties of the hydrocarbon chain length. SSDI analysis suggests that deep surfactant injections reveal no clear evidence of a decrease in surface hydrocarbon concentrations, and questions the effects of sub-sea dispersants in deep-sea oil spills. Despite inhomogeneous sampling, the comprehensive Gulf Science data sheds some light on understanding the nature of deep blowouts and required responses.

John Lodise (MPO)

"Surface Velocity Fields in the Gulf of Mexico Obtained Through LAVA Blending of Altimetry and LASER Drifter Data"

Thus far, the surface currents of the Gulf of Mexico have been vastly understudied and much of the variability of these currents on the order of 1 day to a few weeks is yet to be documented. Through the use of a large data set containing trajectories from over 1000 surface drifters from the LAgrangian Submesoscale Experiment, LASER, that took place in January-February of 2016, this study’s main focus is to map the surface currents over the spatial and temporal scales of the submesoscale, that are missing from other observational platforms. Using the LAgrangian Variational Analysis (LAVA) approach to blend altimetry data with available drifter data, we aim to resolve large-scale to mesoscale flows calculated from altimetry data, which have temporal and spatial scales on the order of weeks and 100km, respectively, as well as smaller scale flows using data retrieved from drifters, which intrinsically measure the combined effects of smaller scale processes including Stoke’s drift, convection, Langmuir turbulence, and submesoscale flows. In addition, vertical shear observations in the first 60cm of the ocean are made possible through the use of both drogued and undrouged surface drifters during LASER. Drogued drifters span from the surface down to 60cm, where the undrogued drifters only span the top 5cm of the ocean. The velocities derived from these drifters are also compared to the 10m wind output data from the UWIN-CM model, which has been validated with observational wind data over the duration of the experiment. This assimilation of data leads to more accurate measurements of velocity fields while still maintaining a wide range of scales, which is vital in any upper ocean transport problem.

Mar 17 --- Dr. Ray Bell

"North Atlantic Storm Driving of Extreme Wave Heights in the North Sea"

The relationship between storms and extreme ocean waves in the North sea is assessed using a long-period wave dataset and storms identified in the Interim ECMWF Re-Analysis (ERA-Interim). An ensemble sensitivity analysis is used to provide information on the spatial and temporal forcing from mean sea-level pressure and surface wind associated with extreme ocean wave height responses. Extreme ocean waves in the central North Sea arise due to either the winds in the cold conveyor belt (northerly-wind events) or winds in the warm conveyor belt (southerly-wind events) of extratropical cyclones. The largest wave heights are associated with northerly-wind events which tend to have stronger wind speeds and occur as the cold conveyor belt wraps rearwards round the cyclone to the cold side of the warm front. The northerly-wind events also provide a larger fetch to the central North Sea to aid wave growth. Southerly-wind events are associated with the warm conveyor belts of intense extratropical storms developing in the left upper-tropospheric jet exit region. There is predictability in the extreme ocean wave events up to two days before the event associated with a strengthening of a high pressure system to the west (northerly-wind events) and south-west (southerly-wind events) of the British Isles. This acts to increase the pressure gradient over the British Isles and therefore drive stronger wind speeds in the central North sea.

Mar 24 --- Xing Lu (1-hour MPO student semimar)

"South Pacific Tropical Water: Variability and Downstream Impacts"

South Pacific Tropical Water (SPTW) is characterized by a vertical salinity maximum 35.6-36.5 psu located at 8°S-25°S, 160°W-110°W and lying in the upper thermocline between 24.0 and 25.0 σθ. SPTW propagates westward and equatorward after being formed by subduction from late winter to early spring. As temperature (T) and salinity (S) of SPTW have increased due to climate change, there is interest in its variability and downstream impacts on the tropical Pacific. SPTW subduction variability and downstream impacts of T/S anomalies are examined using data. Subduction rates of SPTW correlate with Southern Oscillation Index (SOI) interannually from 2005 to 2013. Contributions from different components of SPTW subduction rates in interannual and decadal timescales are discussed. The South Equatorial Current transports SPTW westward and exhibits two three-year cycles, with each cycle containing an El Niño-Southern Oscillation (ENSO) event. Rossby waves play an essential role in this variability and consequently the westward extension of the SPTW. Non-spiciness T/S anomalies of SPTW are found for the first time using Argo data for the period 2005 to 2014. The T/S anomalies are induced by ENSO, and anomalies within the strongest propagation area are correlated with SOI. Temperature anomalies are having a stronger influence on density than the salinity anomalies. The downstream vertical exchange of water between surface and subsurface is intensified during warm/salty anomaly years and slowed down during cold/fresh anomaly years. SPTW anomalies appear to affect the downstream vertical stability and the surface characteristics. Output from a model will be used to further confirm the results and to study coupled air-sea processes.

Mar 31 --- Shitao Wang (1-hour MPO student seminar)

"The Application of Uncertainty Quantification Techniques and Information Theory to Oil Spill and Ocean Forecasting"

The reliability of the oil fate model depends critically on the quality of its input data, such as the initial release flow rate, droplet size distribution and oceanic state variables. Quantifying uncertainties in these upstream input data will enhance the forecasting abilities of the oil fate model. The main objective of my research is to explore the usefulness of using uncertainty quantification techniques and information theory in the ensemble oceanic and oil plume simulations. In this talk, I will mainly focus on the ensemble ocean forecasting.

Polynomial Chaos (PC) methods have recently been used to quantify uncertainties in the circulation forecast of the Gulf of Mexico due to initial conditions and wind forcing uncertainties. The input uncertainty was specified by decomposing the variability in those fields using Empirical Orthogonal Functions(EOF); an ensemble of simulations was then launched to explore the ensuing parameter space.
Three common questions of ensemble simulations that we wish to address here concerning the “realism" of the uncertainty analysis, the computational and information trade-off in choosing different uncertain inputs, and the exploration of the statistical information conveyed by the PC approach. The surface and subsurface model data comparison shows that the observational data falls into the envelope of the ensemble and the modal decomposition delivers "realistic" perturbations. The quantification of model output variability suggests that adding additional sources of uncertainty is more useful than increasing the variance of uncertainty. For the center of Gulf of Mexico, two initial condition EOF modes are enough to capture the ensemble variability while additional wind forcing EOF modes are needed to capture the ensemble variability in the coastal zone. The ensemble statistics are then explored using the PC approach and the newly developed contour boxplot method.

Apr 07 --- 15-min seminars:

Romain Chaput (OCE)

"Can Fish Larvae Stay in Group from Hatching to Settlement?"

On the light of recent discoveries on fish larvae behavior some theories have been formulated concerning the possibility and advantages they would have to stay in group during their pelagic life. Patchy distribution of fish larvae can be found in retention areas like sub-mesoscale eddies, but the distinction between passive aggregation and active group cohesion has not been yet studied due to observation limitation. Larvae of Chromis atripectoralis have been observed to orient cardinally only within a group, confirming the theory that group interaction enhances abilities of the larvae.

The group behavior could be common among different taxa of fish. Group dynamic, cohesion and structure definition should be adapted to fish larvae. When and where larvae can form groups are fundamental questions to understand this behavior. Larvae must be within detection distance in order to maintain group cohesion. This condition seems to be met only right after hatching for demersal spawning species, or later during the dispersal phase when larvae aggregate and stay in retention areas. The first case involves early stage larvae with small size and low swimming abilities, both characteristics having consequences regarding the efficiency of their forward movements. Larvae are living a Reynold’s regime where viscosity create an important drag to every movement, and group cohesion could also be limited by microscale turbulences that tend to disperse the larvae.
By reviewing available literature, we quantified the potential for coral reef fish larvae of various taxa to maintain cohesion at different stage of the ontogeny. We adapted equations to find values of critical turbulences associated with different swimming behavior and highlighted the gaps of knowledge about parameters necessary to further describe the swimming behavior of fish larvae in natural conditions.

Molly Martin (MAC)

"Decadal Variability in Oxygen and AOU in the South Pacific"

Effects of climate change, warming and stratification, suggest impacts on oceanic oxygen concentrations. Oxygen concentrations are affected by physical circulation and biological processes. Earlier studies have shown decreasing oxygen in oxygen minimum zones (OMZs) over the past few decades, and increased ventilation of the South Pacific subtropical gyre during the 1990s. Data from a GEOTRACES 2013 zonal section along about 12S are used to investigate variability on decadal time scales in the South Pacific, and are compared to mid-1990s WOCE and 2000s CLIVAR data. Spatial differences in oxygen and apparent oxygen utilization (AOU) are analyzed over the density range of lower thermocline to intermediate levels 26.5-27.2 sigma theta. In the OMZ, oxygen decreases from 9.14 ± 3 μmol/kg during the 1990s to 3.31 ± 3 μmol/kg in 2013. The OMZ AOU increases from 267.79 ± 3 μmol/kg during the 1990s to 272.48 ± 3 μmol/kg in 2013. The subtropical gyre shows the opposite trend with AOU decreasing by ~5 ± 3 μmol/kg over the same time period. To determine if biological activity is a source of the changes in oxygen and AOU the O2* tracer is examined. The O2* tracer uses the O2 and PO4 stoichiometric ratio to create a conservative tracer of O2 addition or removal. The difference between O2* and the solubility curve reflects the biological component. In the OMZ, O2* remained constant over two decades, while in the subtropical gyre, O2* decreased. These calculations suggest the cause of the OMZ oxygen decrease is primarily related to physical processes, while in the subtropical gyre changes in biological activity are likely playing a relatively larger role over the two decades.

Juan Pinales (AMP)

"Developing a Marine Oil Spill Detection System with C-Band Synthetic Aperture Radar Using Machine Learning"

Previous studies have demonstrated the ability to detect and classify marine hydrocarbon films with spaceborne synthetic aperture radar (SAR) imagery. The dampening effect of hydrocarbon discharges on surface capillary-gravity waves renders the ocean surface "radar dark" compared with the standard wind-roughened ocean surfaces. The research presented here describes the development of an algorithm that detects marine oil spills utilizing single-polarization C-band SAR data acquired during the Deepwater Horizon oil spill. Radar-borne, as well as environmental data variables serve as inputs for machine learning algorithms like artificial neural networks (ANNs) and support vector machines (SVMs). Shapefiles produced by an experienced human analyst served as targets (reference) during the training portion of the investigation. Previous results of this investigation show that a single-step classifier using machine-learning methods is not suitable for robust classification due to susceptibility to oil spill look-alikes and radiometric errors inherent to SAR images. Previous results also indicate that both artificial ANNs and SVMs provide efficient and comparable results for oil spill candidate detection. As a result of these findings, a 3-step ANN hierarchical classifier approach consisting of dark spot detection, feature extraction, and classification of the candidates into oil spills or look-alikes will be considered going forward. A training dataset deemed representative of surface conditions was obtained from 20 images, while a set of 17 testing images were used to determine algorithm effectiveness. Performance metrics based on well-known tools like the confusion matrix and receiver operating characteristics (ROC) curves were calculated based on the spot detection results. Results for the dark spot detection portion show average performance metrics above 70% for relevant metrics (sensitivity, area under the curve, and accuracy) obtained from the testing images. The best performing models used the radar-borne variables and a wind-speed model.

Apr 14 --- 15-min MPO seminars:

Greg Koman "Transport and Variation of the East Reykjanes Ridge Current"

 The principal features of the subpolar gyre circulation in the Iceland Basin include the northward flowing North Atlantic Current (NAC) in the eastern part of the basin and a southwestward flow along the eastern flank of the Reykjanes Ridge - the East Reykjanes Ridge Current (ERRC).  The ERRC is effectively a western boundary current that recirculates a portion of the NAC as well as Labrador Sea water within the Iceland Basin.  Previous estimates of ~2 Sv flow in the upper 400m of the ERRC are documented, but full-depth estimates of its transport are sparse and neither its annual mean transport nor seasonal cycle is accurately known.  Since 2014, the OSNAP project has maintained the first continuous Eulerian array across the North Atlantic Subpolar Gyre, from Labrador to Greenland, and from Greenland across the northern North Atlantic to Scotland. In the Iceland Basin, continuous measurements of the ERRC have been maintained through ADCPs, current meters and dynamic height moorings at multiple mooring sites near 58°N.  Together with satellite altimetry, the mean transport and synoptic variability of the ERRC are studied for the period from July 2014 to July 2016.  Results suggest that the ERRC is highly variable with a mean southwestward flow of ~7 Sv using 1200m ADCP reference velocities and ~4 Sv using satellite altimetry with extreme variations of up to 22 Sv southwestward and 11 Sv northwestward. 

Wei Zhang "Estimates of Decadal Climate Predictability Using the Nonlinear Local Lyapunov Exponent"

Abstract: Over the past several decades, great progress has been made in seasonal climate prediction, which is partly driven by better understanding of the limits and mechanisms of seasonal predictability (e.g. ENSO). However, little is known about decadal predictability and decadal climate prediction remains a challenging and debatable question. In this study, the recently developed Nonlinear Local Lyapunov Exponent method is applied to investigate the limit of decadal climate predictability. Here we discuss results from three sets of Sea Surface Temperature (SST) model data (one set of CCSM4 control run output and two sets of higher resolution output) and SST observational estimates. Both models and observations show that the limit of decadal predictability is relatively larger in North Atlantic, North Pacific and Southern Ocean and the lowest values lie in the tropical central-eastern Pacific Ocean. Also, the distributions of the limit of predictability are consistent with those of the intensity and persistence on decadal timescales, which is estimated by the ratio of 6-year filtered variance and total variance of SST data. The limit of predictability based on observational data is generally larger than that of model datasets. Comparisons within model datasets indicate that with an increased resolution, no significant difference is detected.

Jeremy Klavans "The Role of Ocean Dynamics in North Atlantic Climate Variability"

Variability in the North Atlantic basin is a major feature of the global climate; oscillations in this region can influence weather across North America and Eurasia. Traditionally, multi-decadal variations in the Atlantic have been viewed as a consequence of changes in ocean dynamics. However, internal atmospheric noise (as described by the North Atlantic Oscillation) is the primary mode of variability in the region. Recent work suggests that atmospheric variability can drive multi-decadal sea-surface temperature variability, either directly or as moderated by the overturning circulation (Clement et al. 2015; Delworth et al. 2017). We will evaluate the role of the NAO and ocean dynamics in multi-decadal variability using a hierarchy of climate models, extending from a general circulation model (GCM) with fixed sea-surface temperatures to a fully-coupled atmosphere-ocean GCM.

Apr 21 (Special location: MSC343) --- Mike Rudko (1-hour MPO seminar)

"Anomalous Structures of Oceanic Turbulence: Dynamics, Energetics, Transport"

Oceanic mesoscale turbulent flows on a beta-plane contain vortices and zonally-elongated patterns. This study addresses the stability of and transport by baroclinic vortices and the phenomenology and sensitivity analysis of zonally-elongated large-scale transients (ZELTs).

For the first part of this study, we adapt a wave-mean flow formalism and examine interactions between the axisymmetric flow (“vortex") and residuals (“waves"). Unlike baroclinically unstable vortices on the f-plane, such vortices on the beta-plane can be also vulnerable to barotropic instability as revealed by the globally integrated energy-balance analysis. The spatial structure of energy fluxes shows the energy leakage inside the vortex core when its breakdown occurs. Mixing by stable and unstable vortical flows is quantified through the computation of Finite-Time Lyapunov Exponent (FTLE) maps. Depending on the strength of wave radiation, the upper-layer FTLE maps of stable vortices show either an annulus or volute ring of vigorous mixing inside the vortex interior. This ring region is disrupted when the vortex becomes unstable. Both stable and unstable vortices show the wavy patterns of FTLE in the near- and far-fields. Despite the fact that the initial vortex resides in the top layer only, significant FTLE patterns are observed in the deep layer at later times. Lagrangian analysis of the vortex-induced change of large-scale tracer gradient demonstrates significant effects of vortex instability in the top layer and the importance of the wave-like anomalies in the bottom layer.

In the second part of the study, we use Empirical Orthogonal Functions (EOF) decomposition that allows to separate ZELTs from the background turbulent flow as several leading EOF modes. The leading Extended EOF reveals that ZELTs propagate westward with the speed of
1.15 cm s−1. The decrease in the planetary vorticity gradient and increase in the bottom drag coefficient each leads to flattening of the variance spectrum, isotropization of the leading EOF and fast decay of the autocovariance function of its corresponding Principal Component. The variability of the leading EOF anisotropy due to simultaneous variations in and bottom drag coefficient is also reported.