OS33B-1330
Effects of Climate Change on Biogeochemistry in Lake Superior: A Numerical Modeling Study
Physical and biogeochemical dynamics in mid-latitude lakes are tightly coupled due to strong seasonal forcing. Regional climate change has influenced and will continue to influence the physical dynamics in temperate lakes such as Lake Superior. A negative trend in average areal ice coverage on Lake Superior due to regional climate change exists. Recently, an unexpectedly strong positive trend in summer surface water temperatures (twice the rate of regional atmospheric temperature increase) was shown to exist. A positive trend in wind speeds over Lake Superior, possibly connected to increasing water temperatures, was also shown. Studies using numerical modeling of the impacts of changing physical dynamics in Lake Superior due to regional climate change on biogeochemical dynamics within the lake are presented. The Regional Ocean Modeling System (ROMS), a three-dimensional high resolution hydrodynamic model, is used. A dynamic-thermodynamic sea ice model is coupled to the physical model. An ecological- biogeochemical model based on the Fasham et al. (1990) NPZD model, but with multiple plankton size classes and biological limitation by phosphorus, is used. We report the results of our sensitivity experiments examining the effects of variability in winter ice coverage and increasing surface water temperatures and wind speeds on seasonal biogeochemical dynamics in Lake Superior.
OS33B-1331
A High Resolution Study of Black Sea Circulation and Hypothetical Oil Spills
A 1/24 deg resolution adaptation of the DieCAST ocean model simulates a realistically intense Rim Current and ubiquitous mesoscale coastal anticyclonic eddies that result from anticyclonic vorticity generation by laterally differential bottom drag forces that are amplified near Black Sea coastal headlands. Climatological and synoptic surface forcings are compared. The effects of vertical momentum transfer by known (by Synop region fishermen, as reported by Ballard National Geographic article) big amplitude internal waves are parameterized by big vertical viscosity. Sensitivity to vertical viscosity is shown. Results of simulated hypothetical oil spills are shown. A simple method to nowcast/forecast the Black Sea currents is described and early results are shown.
OS33B-1332
Circulation Patterns in the Ancient Epicontinental Seas
We describe our numerical investigation of circulation patterns that arise when the oceans encroach over broad continental areas, as was common during much of the Paleozoic and Mesozoic portion of geologic history. For this effort we have used a code that solves the shallow water equations on a rotating sphere, one developed at the National Center for Atmospheric Research. The code allows for arbitrary surface topography. Initial calculations using Chris Scotese's (http://www.scotese.com) continental configurations during this geological interval give time-varying circulation patterns of flow in these shallow seas. The flow patterns appear to be able to account for some noteworthy features of the sedimentary record during this period.
OS33B-1333
Surface Ocean Circulation and Dynamics in the Philippine Archipelago Region
The dynamics of the seasonal surface circulation in the Philippine Archipelago (117E-128E, 0N-14N) are investigated using a high-resolution Regional Ocean Modeling System (ROMS) for the period of 2004 – March 2008. Three experiments are performed to estimate the relative importance of local, remote and tidal forcing. On annual mean, circulation in the Sulu Sea is dominated by the inflow from the South China Sea at the Mindoro and Balabac Straits, outflow into the Sulawesi Sea at the Sibutu Passage, and cyclonic circulation in the Sulu basin. A strong jet forms in the northeast Sulu Sea when currents from the Mindoro and Tablas Straits converge. Within the Archipelago, strong westward currents in the Bohol Sea carry the western Pacific (WP) Water from the Surigao Strait into the Sulu Sea via the Dipolog Strait. In the Sibuyan Sea, currents flow westward, which carry part of the WP water from the San Bernardino Strait into the Sulu Sea via the Tablas Strait. The surface circulations exhibit strong seasonal variations or reversals from winter to summer. The variations of volume transports above 40m at the Mindoro Strait, Balabac Strait and Sibutu Passage of the Sulu Sea and at the Verde Island Passage are dominated by remote forcing, although local forcing can be large sometimes. At the Tablas and Dipolog Straits, variations of transports result from both remote and local forcing. The cyclonic (anticyclonic) gyres during winter (summer) in the Sulu Sea and seasonally reversing currents in the Archipelago result mainly from local forcing. Nonlinear rectification of tides acts to reduce the mean transports at the Surigao, San Bernardino and Dipolog Straits, and it also affects the Sulu Sea circulation perhaps by tidal mixing and shifting the locations of jets and eddies.
OS33B-1334
The Origin of the Tsushima Warm Current in a High Resolution Model
Using a high resolution global ocean circulation model results, the present study investigates the origin of the Tsushima Warm Current and related East China Sea Circulation. The simulated Tsushima Warm Current is weaker than the observations by about 30 %, but the persistence of the Taiwan-Tsushima Current System shows that the Taiwan Warm Current is the main source of the Tsushima Warm Current. The high resolution model results allow us to distinguish the Kuroshio intrusion north of Taiwan and west of Kyushu from the Taiwan-Tsushima Current System. West of Kyushu the onshore intrusion of the Kuroshio is strong between September and February, and north of Taiwan between June and November. The annual mean strength of the intrusion is 0.32 Sv west of Kyushu, and 0.22 Sv north of Taiwan. Since the simulated Tsushima Warm Current is weaker than the observation while that of the Taiwan Current is comparable to the observations, the strength of the intrusion is weaker than the reality. In addition, a linear relation is found between the transport of the Tsushima Warm Current and the sea level difference between the Korea/Tsushima Strait and the Tsugaru/Soya Straits, and we can conclude that the sea level difference is the main driving force of the current.
OS33B-1335
A Lagrangian Model Analysis of Arctic Water Mass Transformations and Exports.
Many recent studies indicate that a change in the volume, heat or freshwater export from the Arctic Ocean to the North Atlantic could strongly affect the deep convection regions and thus the global thermohaline circulation. However, the origins of the exported water, mostly along both sides of Greenland through Davis Strait and Fram Strait, are still largely unknown, as strong incertitude remains about the dynamics in the Arctic Ocean and the water mass transformations that occur in this basin. An original approach is presented here to investigate these issues. A quantitative Lagrangian method is applied to Eulerian fields of a global high resolution model (around 12~km grid size in the Arctic). First, the simulated Arctic is validated against available observations. Then, the Lagrangian method allows to establish a quantitative circulation scheme. We quantify the relative contributions of the different branches of circulation to the export to the North Atlantic, as well as the related timescales and water mass transformations. The role of the Barents Sea in the modification of the different branches of Atlantic Water entering the Arctic is specially emphasized.
OS33B-1336
How does western Pacific intraseasonal variability affect the prediction of ENSO?
Intraseasonal wind variability in the western Pacific has been found to have substantial influences on the evolution of El Nino Southern Oscillation (ENSO) in the tropical Pacific. In this study, we analyze the impacts of intraseasonal wind anomalies in the tropical Western Pacific (WPac) on the prediction of ENSO based on the real-time forecast during 2005-2008 from the National Centers for Environmental Prediction (NCEP) Climate Forecast System (CFS). We will first diagnose the overall relationship between WPac intraseasonal wind anomalies and subsequent subsurface and surface ocean temperatures in the Pacific. We will then focus on the analysis of the forecast of the decay of the 2006-2007 El Nino and the evolution of the 2007- 2008 La Nina. Our analysis shows that sea surface temperatures in the eastern Pacific respond to western pacific wind anomalies with a lag of three months. The Madden-Julian Oscillation (MJO) during Dec 2006 and Jan 2007 contributed to the demise of the 2006/2007 weak El Nino. Forecast of 2007-2008 La Nina strongly depends on initial phase of the MJO: forecast from initial easterly anomalies in WPac produced too cold Nino34 SSTs, while forecast from initial westerly anomalies in WPac produced warmer Nino34 SSTs.
OS33B-1337
Eddy permitting simulation of the global ocean model COCO4.3 driven by the CORE inter- annual forcing
We are developing ocean component of the CCSR/NIES/FRCGC climate model to conduct high-resolution global warming simulations under IPCC scenarios. This presentation focuses on the performance and the behavior and role of eddies in the global ocean. The Ocean model is CCSR Ocean Component Model (COCO) version 4.3, which is a z-coordinate, free-surface primitive equation ocean model with multi-category sea ice model. The geographical North Pole is moved to 40W, 77N on Greenland and the geographical South Pole is moved to 40E, 77S. The computational domain covers global ocean, with zonal grid spacing of 0.28125 degree and meridional grid spacing of 0.1875 degree. There are 50 vertical levels excluding the bottom boundary layer, and 7 of which are within the sigma-coordinate (~42m). The model employs the momentum advection algorithm of Ishizaki and Motoi (1991), which is a pseudo-enstrophy preserving scheme with a consideration for up-/down-sloping advection. The model's tracer advection is based on the second-order moment (SOM) advection scheme of Prather, M. J. (1986). The vertical mixing of momentum and tracers is represented by a harmonic form. The coefficients are calculated by the parameterization of Noh and Kim (1999), but the formulation is slightly modified (see K1-developers, 2004). As background diffusivity, a minimum value is set for each level, suggested by Tsujino et al. (2000). The Smagorinsky's (1963) biharmonic viscosity is applied for the lateral momentum mixing, and its coefficient is dependent on the grid width and the strain rate, and its controlled by a single non- dimensional parameter whose values is taken to be 2.5. The constant coefficient biharmonic diffusion is applied with the coefficient value of 1.0E9 m4/s. The model is driven by the inter-annual forcing data set adopted by common ocean-ice reference experiments (CORE). The results are reported by focusing on heat transport in strong eddy activity regions, such as the Kuroshio-Oyashio Extension, equatorial oceans, and the western tropical Pacific. These output data are also used for the marine ecosystem modeling project conducted by Yamanaka ea al.
OS33B-1338
Eddy Permitting Simulations of Biogeochemical Cycles in the Global Ocean
A 3D ecosystem-biogeochemical model simulation for the global domain is performed in order to investigate variability of oceanic ecosystem on time scales of years to decades. The model has a horizontal resolution of 1/4 times 1/6 degrees and 51 vertical levels, covering the entire domain of the world ocean. The ecosystem- biogeochemical part of the model is based on NEMURO (North Pacific Ecosystem Model Used for Regional Oceanography), and is coupled with CCSR Ocean Component Model (COCO) version 4.3 by an offline technique. The physical part of the model is driven by the inter-annual forcing by common ocean-ice reference experiments (CORE) data from 1958 to 2004, and reasonably simulates inter-annual to decadal variabilities of ocean conditions related to biogeochemical cycles. These properties of the physical model with its eddying filed enable us to reproduce the realistic distributions of nutrients and plankton productions. Comparisons with historical station data show that the model also reasonably simulates the observed variabilities of ecosystem on time scales of years to decades. In particular, the model captures the transitions of biogeochemical cycles associated with regime shifts.
OS33B-1339
Modeled MOC Variability: Its Probability Distribution and Relation to Observations
A set of IPCC AR4 climate models and a set of higher resolution ocean models are used to examine the distribution of the strength of the Meridional Overturning Circulation (MOC). Observations of the MOC are used to quantify the representation of the distribution and to explore the probability of observed measurements occurring in the modeled simulations. Monthly values of the MOC strength from between 25 degrees North and 30 degrees North are determined from a set of 20th century simulations as well as simulations under a variety of CO2 emission scenarios. Probability density functions are created from both the absolute values and the trends in the strengths for each simulation. Biases, as related to observations, of the distributions are examined and the distributions are analyzed to whether or not they represent probable estimates of the observed MOC strength. High frequency estimates of strengths from models are used to compare to observed measurements from the RAPID array to further understand the sampling and distribution of the MOC variability in models.
OS33B-1340
Impact of high-frequency atmospheric forcing on oceanic mixed layer variability
The effect of high-frequency variability in the surface boundary conditions on the upper ocean physics is an important issue in numerical ocean modeling. This study focuses on the response of the upper ocean to the atmospheric forcing at a variable time frequency (ranging from hourly to monthly). NCEP's GFS (Global Forecast System) surface fluxes averaged over different time intervals will be used to force ocean models based on HYCOM in global coarse (0.7 degree) and basin-wide fine (1/12 degree) resolution configurations. The experiments will also include sensitivity studies with various vertical mixed layer parameterization in vogue with global and basin-wide ocean models. This work builds on similar studies involving mixed layer models in the recent past (Halliwell, 2003 ; Kara et al., 2008).
OS33B-1341
A Comparison of the Annual Circulation of the Western South Atlantic Ocean Based on two OGCMs, the Parallel Ocean Model (POP) and the Ocean Circulation and Climate Advanced Modelling Project (OCCAM)
The Western South Atlantic Ocean is a highly complex region that includes different ocean regimes. One important aspect in these regimes is the bifurcation of the South Equatorial Current (SEC), which at the surface and sub-surface levels is related to the origin of the southward flowing Brazil Current (BC) and the northward flowing North Brazil Current/North Brazil Undercurrent System (NBC/NBUC). This study investigates the annual circulation of the Western South Atlantic using: i) POP with a 1/10° horizontal resolution over a period of 10 years and ii) OCCAM with a 1/4° horizontal resolution over a period of 8 years. The variables analyzed are the temperature, the salinity and the associated velocity field. The depth-averaged volume transport show values varying from 8.5 to 10.6 Sv at 30°W meridional section between 5°-15°S, which for the surface and sub-surface levels are associated to the waters coming from the Benguela Current region towards the Brazilian shelf/slope region. At 5°S both models show a consistent 9.2 Sv northward flow, while at 15S POP shows a net southward flow whereas OCCAM shows a net northward flow. This difference in the net depth-averaged flow is a result of the complex vertical structure of the flow, with opposing flows throughout the water column. The zonal velocity field at 30°W presents a deep SEC in the South Atlantic around 30°S with lower intensity while at the bifurcation (10°- 15°S) region it gets stronger and shallow. The NBC/NBUC core, with velocities greater than 0.45 m/s, appears as a clear feature at the 5°S and 10°S zonal sections. South of this region, around 15°S the BC presents southward velocities of up to 0.55 m/s at the first 250 m, associated to the Tropical Water (TW). Below 250 m, the northward flowing NUBB carrying the South Atlantic Central Water (SACW) occurs due to the shift of the SEC bifurcation with depth, which for this level is near 22°S. Both models estimated a long-term annual variability and reproduced the strength and vertical structure of the SEC, BC and NBC/NUBC. This evaluation is an important process to provide confidence in the water masses vertical structure represented by the models, even in the case of OCCAM, which has a lower horizontal resolution. The large-scale patterns presented here can also be important to provide initial and boundary conditions for regional numerical models.
OS33B-1342
Covariance regularization through a Gaussian graphical model and its application to ocean data assimilation
In data assimilation, covariance matrices are introduced to prescribe weights of the initial (background) state, model dynamics, and observation and suitable specification of the covariances is known to be essential for obtaining sensible state estimates. Covariance matrices are specified by sample covariances and modeled covariance structure. Covariance modelling consists of regularization of the sample covariance and constraint of dynamical relationship. Regularization is required for converting the singular sample covariance into non-singular one, removing spurious correlation between variables at distant points, and reducing a required number of parameters that specify the covariances. Regularization of the sample covariance has been carried out in physical (grid) space, spectral space, and wavelet space. We propose a covariance regularization method in inverse space with a Gaussian graphical model. Defining a neighbor of a variable, we assume conditional independence between the variable and those outside of the neighbor. Conditional independence is expressed by specifying zero elements in the inverse covariance matrix. We give an illustrative example using a simple matrix, and show application to sample covariance obtained from sea- surface height (SSH) observations.
OS33B-1343
Problem of Western Boundary Current Separation
By analyzing the results of a realistic ocean general circulation model (OGCM) and conducting a series of idealized OGCM experiments, the problem of western boundary current separation is examined. Our main subjects are the Kuroshio Current System and an idealized western boundary current in a rectangular basin. In the realistic configuration, the Kuroshio Current System is successfully simulated when the horizontal resolution of OGCMs is increased from 1/2 to 1/10. In the high-resolution run, a jet, the model's Kuroshio Extension, and a pair of cyclonic and anticyclonic recirculation gyres (RGs) are found on the northern and southern flanks of the jet in the Sverdrup subtropical gyre. Similar zonal jet and RGs are obtained also in the idealized model with a rectangular basin and a flat bottom with a horizontal resolution of 1/6. The northern RG is generated by the injection of high potential vorticity (PV) created in the viscous sublayer of the western boundary current. Since there is no streamline with such high PV in the Sverdrup interior, the eastward current in the northern RG region has to lose its PV anomaly by viscosity before connecting to the interior. The southern RG is due to the classical Rhines-Young mechanism. The westward-flowing southern branch of this southern RG stabilizes the inertial western boundary current and prevents its separation in the northern half of the Sverdrup subtropical gyre, where the western boundary current is unstable without the stabilizing effect of the southern RG. Therefore, in the equilibrium state, the southern RG should be located just to the north of the center of the Sverdrup subtropical gyre, which is defined as the latitude of the Sverdrup streamfunction maximum. The zonal jet (the Kuroshio Extension) and the northern RG gyre are formed to the north of the southern RG.
OS33B-1344
Tale of two currents: The collision of the Gulf Stream and Deep Western Boundary Current at the Tail of the Grand Banks of Newfoundland
The Tail of the Grand Banks of Newfoundland is a dramatic topographic ridge that protrudes southeastward from the continental shelf in the western North Atlantic. It is a critical location for the North Atlantic circulation because here the major branches of the wind-driven and thermohaline circulations meet, with dramatic changes in the currents' pathways. The Gulf Stream approaches from the west, splitting into four recognized branches at the Tail. A critical branch turns north, crossing the Tail to feed the North Atlantic Current. The Labrador and Deep Western Boundary Currents approach from the north. At the Tail the cold currents bifurcate, branching westward and retroflecting northward, and eventually come into contact with the warm branches offshore. Here we analyze all available post-1948 hydrographic data together with satellite derived sea surface height observations to characterize the meeting of the Gulf Stream Branch, the retroflecting Labrador Current Branch, and the Deep Western Boundary Current over the southern part of the Tail of the Grand Banks. Based on the geometry of the vertical superposition of the oppositely flowing currents, we hypothesize that the path of the DWBC is not determined by the Tail's topography alone, but by the orientation of the Gulf Stream Branch crossing over the Tail and the Deep Western Boundary Current.
OS33B-1345
Hydrographic and Circulation Variability of the East Asian Marginal Seas Inferred from an Ocean Reanalysis
Long-term variability of the hydrographic conditions and circulation in the East Asian Marginal Seas (EAMS) is examined using the Simple Ocean Data Assimilation (SODA) product. Even though the SODA data resolve the overall pattern of circulation in the EAMS, there are differences between the reanalysis product and observations in detail. The volume transport of the Tsushima Current through the Korea Strait, which supplies warm water into the East/Japan Sea (EJS) to influence significantly the hydrographic conditions of the upper layer in the southern region of the EJS, estimated from the SODA data is much less than the observed one. In terms of decadal variability of temperature, there is discrepancy between the SODA data and observations in some regions though the warming trend shown in most regions is similar between them. For example, while decadal variability of temperature at intermediate depths in the southern region of the EJS is reproduced well in the SODA data, it is not realistic in the northwestern region, which might come from the lack of observation data assimilated into the SODA system.
OS33B-1346
Past sea level reconstruction based on tide gauges and spatial patterns from OGCMs with assimilation
Abstract A two-dimension sea level reconstruction is proposed over the period 1950-2003 using tide gauge records at 99 selected sites and 46-year (1960-2005) of 2°x2° gridded dynamic heights from the OPA/NEMO global ocean circulation model with data assimilation. Spatial trend patterns of the reconstructed sea level display high regional variability that reflects both influences of large-scale ocean- atmosphere perturbations such as ENSO or NAO, and of longer-term signal. Tests show that the reconstructed trend spatial patterns strongly depend on the length of the used 2-D OPA/NEMO grids: under ~15-20 years, interannual variability is correctly captured, unlike multidecadal variability, especially in the tropical and north Pacific. EOF analysis of the reconstruction over 1950-2003 using the 46 years of OPA/NEMO displays a leading mode essentially representing the long-term (but regionally variable) signal (evidenced through spatial trends over 1950-2003) and a second mode representing interannual (ENSO- dominated) variability. Reconstruction based on only ~15 years of 2-D sea level data (e.g., from satellite altimetry) only contain the interannual variability signal. The robustness of the results is cross-tested, leaving out successively each one of the 99 tide gauges when reconstructing the sea level signal and then comparing original and reconstructed time series at the left out tide gauge site. In general the reconstruction performs well at most tide gauges, except in the northeast Atlantic, where we diagnose a deficiency of the OPA/NEMO model in this region.
OS33B-1347
Interannual to multidecadal variability of the mean sea level
Several previous studies based on tide gauges records over the 20th century have reported important mean sea level variability at interannual to multi decadal time scales (enhanced in sea level rates). We have avalysed mean sea level rates based on different published sea level reconstructions over the past century and found significant pseudo periodicities in the 4-8-year band, likely linked with ENSO forcing. Other peaks are visible around 10-12 years and 20 years. Longer quasi periodic fluctuations are also observed. Such interannual to multidecadal variability seen in tide gauge-based mean sea level rates is a composite response due to different regional forcing acting at different time scales. Using long, high-quality tide gauge records, we determined the dominant interannual/decadal/multidecadal sea level response periods depending on tide gauge locations, and further investigated the expected sea level response to regional variability of the ocean-atmosphere system (i.e., NAO, PDO, etc.) using runs of the OPA/NEMO general ocean circulation model with data assimilation over the 1960-2005 time span. Using the model data, we further investigated which pseudo periodic fluctuations will dominate the mean sea level curve (or mean sea level rate) based on different selections of tide gauge records. Inferences on the present-day high global mean sea level rate based on satellite altimetry and its interpretation in terms of acceleration or decadal variability are discussed.
OS33B-1348
Interdecadal Variability in Bias-Corrected Ocean Temperature Data
Interdecadal variability of ocean temperature data is presented which has been bias-corrected for XBT instruments. Such a correction has been a hot topic recently, and some preliminary results suggested that interdecadal variability is greatly reduced in world-ocean averages by applying this correction to XBT data in the global temperature record. We present results which show that significant interdecadal variability is still present both regionally and globally. This variability shows up in surface temperature records, such as from I-COADS, and in subsurface data from World Ocean Database 2005. These data sets both contain various sources of measurement bias, but these sources are different. The strength of interdecadal variability on regional, basin-wide and global scales, even after some bias corrections are applied, implies that previous concerns regarding the averaging of interdecadal data, collected by a decadally-varying observing system, still exist.
OS33B-1349
Westward Traveling Speeds of Sea-level Variations in the North Pacific as Revealed by 2D Frequency-wavenumber Spectra From 9 Years of TOPEX Data
The spectral method is refined to better isolate traveling features from standing features, which include everything that does not propagate, such as the sea-level response to the seasonal heating and cooling. Examining the traveling spectra from the equatorial region to latitudes south of the Kuroshio Extension, it often appears that there is a dominant zonal speed such that the spectral energy tends to congregate around a straight line on the spectra representing this dominant speed of westward propagation. This qualitative description has been put on a quantitative basis. One certainly likes to know how much spectral energy is congregated around the line and how narrow is the concentration. Thus a speed range centered on the dominant speed is computed so that more than 50% of the westward propagating power is contained in this speed range. The computation has been carried out for 9 latitude zones centered on the midpoint latitudes with three estimates at each latitude zone representing respectively the entire ocean basin and the east and west basins. The results are discussed in the context of various existing theories.
OS33B-1350
Seasonal and Interannual Variability in the California Current System over the Washington Continental Slope observed with Seagliders
Nearly continuous Seaglider repeat sections over the Washington continental slope since August 2003 reveal seasonal and interannual patterns of circulation, oxygen and phytoplankton distribution. Structure in the upper 1 km of the water column has been resolved to ~5 km along sections completed roughly fortnightly using Seaglider long-range autonomous underwater vehicles. More than 11,000 profiles of temperature, salinity, dissolved oxygen, chlorophyll fluorescence and optical backscatter on over 25,000 km of survey track have been collected in 5 years along a pair of sections normal to the coast. Both seasonal and interannual signals are prominent in the observations. Isopycnal tilts delineating the equatorward California Current offshore and the poleward inshore countercurrent reverse seasonally. Alongshore current fluctuations as well as property surfaces tend to propagate offshore near the speed of a gravest mode long Rossby wave. Annual signals are nearly oppositely phased from the continental shelf edge to the end of the survey track, 220-240 km offshore. Interannual differences in water type and depth averaged transport in the upper 1 km are apparent. Alongshore transport variations are strongly interannual, amounting to a few Sverdrups poleward or equatorward in different years. A particularly large equatorward transport pulse was observed over a several month episode centered on January 2007, where average transports exceeded 10 Sverdrups with a few fortnightly estimates exceeding 20 Sverdrups.
OS33B-1351
Cyclonic Eddies as Observed by Mid-depth RAFOS Floats off Central California
Meso- and submesoscale eddy-like structures with counterclockwise rotation are investigated using
subsurface isobaric RAFOS floats launched off Central California between 150 and 600 dbar from 1992-
2006. For the 26071 days of observed float trajectories, cyclonic (anticyclonic) motion was observed 2716
(4943) days or ~10% (~19%) of the time. While most floats made a single cyclonic and/or
anticyclonic circuit during their mission, only those with at least two consecutive loops were counted in the
above numbers and, following Richardson (1993), are referred to as loopers. Cyclonic looping motion was
usually of much shorter duration than anticyclonic looping. 60% of the cyclonic loopers had only two loops
and only 20% made more than 10 consecutive loops. The two longest incidents of cyclonic looping lasted
for 365 and 405 days and made 78 and 66 loops, respectively. Two thirds of the floats that exhibited
cyclonic looping were first entrained by cyclonic eddies during the period from the end of September to the
beginning of December.
In order to characterize the kinematics of the looping floats, the following kinematic characteristics of the float
rotational motion have been calculated: period of rotation, characteristic swirl velocity, mean radius of the
rotation, angular velocity, relative vorticity, and Rossby number. The evolution of the rotating loopers was
studied using continuous wavelet transformation. This technique was also applied to filter measurement
noise and select the energy-dominant temporal scales. Additional data were used in the analysis to provide
insight into three-dimensional structure of cyclonic eddies in which the loopers were entrained: maps of SSH
anomalies; information from ARGO floats, as well as CTD and ADCP measurements for the time of the
RAFOS deployment. Several case studies of two floats entrained by the same eddy are also presented. The
complex motion of the RAFOS floats moving with the cyclonic eddies is described and dynamical implications
discussed.
Reference: Richardson, P. L. (1993), A census of eddies observed in North Atlantic SOFAR float data, Prog.
Oceanogr., 31(1), 1–50.
http://www.oc.nps.edu/npsRAFOS/
OS33B-1352
On the influence of the Galapagos Islands on the Humboldt Current System
Recent studies have shown that water masses which compose the poleward Peru-Chile Undercurrent (PCUC) and the equatorward Peru Coastal Current are influenced by the mean state and variability of the Equatorial Under Current (EUC) and of the South Equatorial Current. Several modeling studies emphasized that EUC intensity is strongly affected by the topographic barrier formed by the Galapagos Islands (GI) and the spatial resolution of the model, which can reduce the eastward flow and the associated cold bias usually present in the cold tongue in the majority of CGCMs. Consequently, the archipelago appears to be a key element connecting the Equatorial Pacific and the Humboldt Current System. A ROMS (Regional Oceanic Model System) model configuration was designed to address the role of the small scale topography of the GI on the dynamics of the area. The employed methodology was quite simple: 2 climatological simulations (15 years after spin ups) were run, one including the archipelago with a 1/6 ° resolution, the other one excluding all bathymetric features above 3000m depth in the vicinity of GI. The main investigated questions are how the model spatial fine resolution impacts the Eastern Pacific mean state, seasonal variability, meso-scale dynamics in the region and the EUC/PCUC pathways, which are studied using Lagrangian diagnostics. The influence of these islands on the propagation of equatorial and coastally trapped Kelvin waves, and on the reflection of Rossby waves is also diagnosed, because of the role of these waves on the surface and sub-surface water properties off South America coast. The possible rectification of Tropical eastern Pacific mean state through changes in waves propagation features, namely changes in vertical stratification, is also investigated. Results will be confronted to TAO mooring data (95 ° W, just west of GI) and to satellite altimetry in order to validate our modeling procedure.
OS33B-1353
Hydrographic variability patterns in the tropical oceans during 2003-2007
Quality controlled ARGO data have been used to produce a gridded field of global temperature and salinity from the surface down to 2000m depth during 2003-2007. The analysis of this product has shown that the dominant signature of global interannual variability occurs in the upper 500m depth of the tropical oceans. In the near surface layer, interannual fluctuations are mostly associated to the strong signature under ENSO conditions and are low in the Atlantic and Indian Ocean. This is considerably different in the salinity field since there, interannual fluctuations are dominant in all three ocean basins. Previous results have also shown that the vertical maximum of temperature variability is associated with the equatorial and ventilated thermocline. Salinity fluctuations are characterized by two maxima, one is surface confined whereas the second is subsurface intensified. Seasonal to interannual fluctuations of the tropical salinity field play a fundamental role in changing the potential density field - which occurs not only in the surface layer - and thus the analysis of its variability is vital. In addition, bi-annual fluctuations characterized by vertical propagation of salinity anomalies are observed in the upper 300m, predominantly in the tropical Pacific Ocean. Those features are not density compensated as similar patterns are not evident in the temperature field.
OS33B-1354
Ensemble reconstruction of small-scale variability in Atlantic sea surface temperatures from 1870 - 2001
Existing historical records of sea surface temperature extending back to the mid 1800's are a valuable source of information for understanding climate variability at interannual and decadal time-scales. However, the temporal and spatial irregularity of these data make them difficult to use in scientific climate research, where gridded data fields are preferred for both direct analysis and forcing of numerical models of the atmosphere. Infilling methods based on constraining the leading eigenvectors of the global-scale covariance have proven very successful in creating gridded estimates of sea surface temperature. These methods are especially useful for infilling within the vast regions of unobserved ocean that characterize the earliest segments of the data record. Regional variability, on the other hand, is not well represented by these methods. This is especially true in data-poor regions. Here we present a method for augmenting the existing large-scale reconstruction methods with a statistical model of the regional scale variability. Using high quality sea surface temperature data from the last 25 years, including satellite-derived records, we specify a spatially non-stationary covariance model for the regional scale marine surface temperature variability. The use of a non-stationary, non-isotropic correlation function in the covariance model is a novel aspect in this work. With the covariance model estimated from the modern record, historical observations are used to condition posterior distributions on the regional scales back to the mid 1800's It is common in the geosciences for the expected value of the distribution to be offered as the data reconstruction. If uncertainty information is provided, it often takes the form of a point-wise estimate that neglects the covariability inherent in the full distribution. In contrast to this common practice, we generate multiple realizations from the full posterior distribution. These samples will be made available to the climate research community via the internet. The presentation of an ensemble of possible realizations of sea surface temperature is especially important in data-poor regions of the ocean. It is a natural consequence of Bayesian inference that the expected value of the reconstruction in unobserved areas will relax towards the mean of the prior distribution. When considered outside the context of the full covariance information, data users can falsely interpret these places in the data record as less energetic; a proper interpretation, in contrast, would be that there is little constraint on the possible states of the system. The sparse and irregular nature of this historical data thus makes ensemble presentation an important contribution to the research community. For demonstration purposes, this work focuses on a limited domain in the mid-latitude Atlantic. However, the method employed here can be extended to global reconstructions.
OS33B-1355
Changes in the temperature and salinity tendencies of the upper subtropical North Atlantic Ocean.
The Ocean section at 24.5⁰N has a doubtless interest for climate studies because it is the place where the northward ocean transport of heat throughout the Atlantic has its maximum. Data from the Argo network and from five previous oceanographic cruises are used to quantify the temperature and salinity variation at this latitude. Results show that there has been a significant cooling (-0.11°C) in the upper ocean (600-1800 dbar range) during the last seven years (1998-2005), contrary to the continuous warming (0.26°C) observed in the previous forty years (1957-1998). This cooling significantly decreases the maximum upper-ocean warming found in 1998 to only 0.15°C for the period 1957-2005, slightly higher than half of that found during the 1957-1998 period. Salinity shows a similar change in tendency to that found in temperature. The results demonstrate that there are important oscillations in the tendencies of the interior North Atlantic.
OS33B-1356
Circulation and Volume Balance in the Bering Sea: Analysis of the Inverse Solution
The Bering Sea circulation is reconstructed as a variational inverse of the hydrographic (temperature, salinity and velocity) and atmospheric climatologies to obtain a reliable estimate of the Bering Sea volume balance. The optimized transports through the Kamchatka Strait, Near Strait, Amchitka and Amukta passes are -28, 13, 6, and 3.5 Sv respectively. These transports are significantly higher than conventional climatological estimates but agree well with recent transports calculations based on direct velocity measurements. Posterior error analysis and satellite sea surface height observations indicate high interannual and seasonal variability of the transports through the Aleutian passes. It is shown that the changes in the Kamchatka Strait transport are controlled by variability of the Near Strait inflow and Alaska Stream transport. It is demonstrated that the reconstructed climatological SSH distribution can be used as a reference state for the satellite altimetry data in the region. Combination of the obtained reference SSH with satellite altimetry observations results in a realistic reconstruction of the Amukta Pass circulation.
OS33B-1357
Toward the reanalysis of the of the circulation in the Chukchi and East Siberian Seas
Several data assimilation algorithms were combined to develop an efficient data assimilation system in the Arctic Ocean. Conventional four-dimensional variational technique for the oceanic data and nudging data assimilation for the ice data are proposed to build reanalysis of the ice and ocean conditions in the Arctic Ocean for a period of 20 years. In a preliminary effort, the approach was used to reconstruct the circulation in the Chukchi Sea in 1990-1991 and circulation in the East Siberian Sea during the fall of 1994. The reconstructed circulation in the Chukchi Sea is in a good agreement with observations. The obtained velocity, temperature and salinity fields are used to estimate volume, heat, salt transports. The analysis revealed some specific features of the Chukchi Sea circulation. The estimates of the circulation in the East Siberian Sea are used to derive the non-stationary reference sea surface height for the Chukchi and East Siberian seas. The proposed approach also involves the optimization of the sampling strategy through the analysis of dynamically induced correlations. The study of the sampling strategy is performed with respect to robustness of the reconstruction of the circulation with in the vicinity of the Bering Strait with a particular emphasis on the accurate monitoring of the mean Bering Strait transport. Computations reveal four major regions that are highly correlated with the Bering Strait transport. Apart from the regions within the Bering Strait itself, they include the Anadyr Strait and a region 100 km south of the Cape of Prince of Wales. Results of the sensitivity analysis are tested in the framework of twin data experiments with quasi-stationary and oscillatory backgrounds.