Local Wind Stress Curl Research Articles

Seasonal to interannual variability in density around the Canary Islands and their influence on the Atlantic meridional overturning circulation at 26°N

The meridional interior flow obtained from the RAPID array is determined by horizontal density fluctuations at the eastern and western boundary of 26°N. The physical causes of these density variations are responsible for fluctuations in the Atlantic Meridional Overturning Circulation (AMOC) and through it, the meridional heat transport of the Atlantic. In this modeling study, a high-resolution ocean model is used to investigate the source and origin of the AMOC variability associated with the density fluctuations at the eastern boundary. The AMOC in the model is in good agreement with the RAPID observations and appears to adequately represent the smaller scale features of variability around the Canary Islands. In this paper, we identify a robust relationship between the density structure south of the Canary Islands, the local wind stress curl (WSC) around these islands and the AMOC using an empirical orthogonal functions analysis, wavelet transform, and wavelet coherence. We find that the deep density fluctuations at the eastern boundary of 26°N arise from the pumping effect of the spatial pattern of WSC south of the islands. These deep density fluctuations drive the AMOC both on seasonal and interannual time scales, through their influence on the basinwide tilt of the thermocline. At seasonal time scales, the density fluctuations south of the islands are driven by the WSC and directly influence the AMOC. At interannual time scales, a significant coherence is found between the density fluctuation and the southward Upper Mid-Ocean (UMO) transport although the origin of these density fluctuations is not explained by the direct pumping caused by the WSC.

Interannual forcing mechanisms of California Current transports II: Mesoscale eddies

Mesoscale eddies exert dominant control of cross-shelf exchanges, yet the forcing dynamics underlying their interannual and decadal variability remain uncertain. Using an ensemble of high-resolution ocean model hindcasts of the central and eastern North Pacific from 1950 to 2010 we diagnose the forcing mechanisms of low-frequency eddy variability in the California Current System (CCS).We quantify eddy activity by developing eddy counts based on closed contours of the Okubo-Weiss parameter and find that the spatial and temporal features of model-derived counts largely reproduce the short AVISO observational record.Comparison of model ensemble members allows us to separate the intrinsic and deterministic fractions of eddy variability in the northern CCS (34.5–50°N) and in the southern CCS (28.5–34.5°N). In the North, a large fraction of low-frequency eddy variability (30% anticyclones, 20% cyclones) is deterministic and shared with satellite observations. We develop a diagnostic model based on indices of the large-scale barotropic and baroclinic states of the CCS which recovers this deterministic variance. This model also strongly correlates with local atmospheric forcing. In contrast to the North, Southern CCS eddy counts exhibit very little deterministic variance, and eddy formation closely resembles a red-noise process.This new understanding of the external forcings of eddy variability allows us to better estimate how climate variability and change impact mesoscale transports in the California Current. The skill of our diagnostic model and its close association with local wind stress curl indicate that local atmospheric forcing is the dominant driver of eddy activity on interannual and decadal time scales north of pt. conception (~33°N).

Interannual forcing mechanisms of California Current transports I: Meridional Currents

An ensemble of eddy-resolving ocean model hindcasts integrated from 1950 to 2008 is used to examine and quantify the interannual variability of large-scale (>200km) alongshore equatorward flow in the California Current System (CCS). We also develop a single index of this transport in order to determine what fraction of variance is driven locally, by changes in wind stress curl, and remotely, by the arrival of coastally trapped waves of tropical origin.In agreement with previous studies, coastally trapped waves dominate large-scale interannual CCS sea surface height variability. In contrast, we find that large-scale alongshore currents (v) are driven predominantly by local wind stress curl variability rather than coastally trapped waves. A simple wind-driven diagnostic model of the time-dependent large-scale geostrophic meridional transport captures ~50% (R=0.7) of the total variance. The local wind-stress curl gradient that controls the largest fraction of meridional transport is not independent of the modulations in atmospheric circulation that drive the Pacific Decadal Oscillation (PDO), and a significant fraction of the monthly transport variability in the model ensembles is correlated to the PDO (R=0.4).

Variability of the Denmark Strait overflow: Moored time series from 1996–2011

The Denmark Strait overflow provides about half of the total dense water overflow from the Nordic Seas into the North Atlantic Ocean. The velocity of the overflow has been monitored in the Strait with two moored Acoustic Doppler Current Profilers since 1996 with several interruptions due to mooring losses or instrument failure. So far, overflow transports were only calculated when data from both moorings were available. In this work, we introduce a linear model to fill gaps in the time series when data from only one instrument is available. The mean overflow transport is 3.4 Sv and exhibits a variance of 2.0 Sv2. No significant trend was detected in the time series. The highest variability in the transport is associated with the passage of mesoscale eddies with time scales of 2–10 days (associated with a variance of 1.5 Sv2). Seasonal variability is weak and explains less than 5% of the variance in all time series, which is in contrast to the strong seasonal cycle found in high resolution model simulations. Interannual variability is on the order of 10% of the mean. A relation to atmospheric forcing such as the local wind stress curl, as well as to larger scale phenomena, e.g. the North Atlantic Oscillation, is not detected. Since 2005 data from moored temperature, conductivity and pressure recorders have been available as well, monitoring the hydrographic variability at the bottom of Denmark Strait. In recent years the temperature time series of the Denmark Strait overflow revealed a cooling, while the salinity stayed nearly constant.

Vorticity budget study on the seasonal upper circulation in the northern South China Sea from altimetry data and a numerical model

Based on the EOF analyses of Absolute Dynamic Topography satellite data, it is found that, in summer, the northern South China Sea (SCS) is dominated by an anticyclonic gyre whilst by a cyclonic one in winter. A connected single-layer and two-layer model is employed here to investigate the dynamic mechanism of the circulation in the northern SCS. Numerical experiments show that the nonlinear term, the pressure torque and the planetary vorticity advection play important roles in the circulation of the northern SCS, whilst the contribution by seasonal wind stress curl is local and limited. Only a small part of the Kuroshio water intrudes into the SCS, it then induces a positive vorticity band extending southwestward from the west of the Luzon Strait (LS) and a negative vorticity band along the 200 m isobath of the northern basin. The positive vorticity field induced by the local summer wind stress curl is weaker than that induced in winter in the northern SCS. Besides the Kuroshio intrusion and monsoon, the water transports via the Sunda Shelf and the Sibutu Passage are also important to the circulation in the northern SCS, and the induced vorticity field in summer is almost contrary to that in winter. The strength variations of these three key factors (Kuroshio, monsoon and the water transports via the Sunda Shelf and the Sibutu Passage) determine the seasonal variations of the vorticity and eddy fields in the northern SCS. As for the water exchange via the LS, the Kuroshio intrusion brings about a net inflow into the SCS, and the monsoon has a less effect, whilst the water transports via the Sunda Shelf and the Sibutu Passage are the most important influencing factors, thus, the water exchange of the SCS with the Pacific via the LS changes dramatically from an outflow of the SCS in summer to an inflow into the SCS in winter.

Impact of prolonged La Niña events on the Indian Ocean with a special emphasis on southwest Tropical Indian Ocean SST

This study examines the mechanisms governing the teleconnections associated with the long-lived La Niña variability in the tropical Indian Ocean (TIO) sea surface temperature (SST) anomalies using observational and reanalysis products. Two long-lived La Niña events (1973 to 1976 and 1998 to 2001) are observed in the recent years, one falling before and the other after the mid 1970's climatic shift. The winter (boreal) and spring (November to April) TIO SST is highly influenced by long-lived La Niña forcing. Climatic shift in mid 1970s contributes to the changes in TIO SST pattern during these two long-lived La Niña events. Surface heat flux variations due to long-lived La Niña contribute to the SST changes except in the southwest TIO. The upwelling favorable local surface wind stress curl and upwelling Rossby waves originating from the east are the dominant mechanisms responsible for the La Niña related winter time SST cooling over the southwest TIO. Long-lived La Niña induced surface wind anomalies enhance the fall Wyrtki Jet in the equatorial Indian Ocean resulting large scale anomalous heat transport. Local SST cooling reduces convection and contributes to the low rainfall over southwest TIO and the northern parts of Madagascar Island.

What determines the spatial pattern in summer upwelling trends on the U.S. West Coast?

Analysis of sea surface temperature (SST) from coastal buoys suggests that the summertime over‐shelf water temperature off the U.S. West Coast has been declining during the past 30 years at an average rate of −0.19°C decade−1. This cooling trend manifests itself more strongly off south‐central California than off Oregon and northern California. The variability and trend in the upwelling north of off San Francisco are positively correlated with those of the equatorward wind, indicating a role of offshore Ekman transport in the north. In contrast, Ekman pumping associated with wind stress curls better explains the stronger and statistically more significant cooling trend in the south. While the coast‐wide variability and trend in SST are strongly correlated with those of large‐scale modes of climate variability, they in general fail to explain the southward intensification of the trend in SST and wind stress curl. This result suggests that the local wind stress curl, often topographically forced, may have played a role in the upwelling trend pattern.

Satellite views of Pacific chlorophyll variability: Comparisons to physical variability, local versus nonlocal influences and links to climate indices

Concurrent satellite-measured chlorophyll (CHL), sea surface temperature (SST), sea level anomaly (SLA) and model-derived wind vectors from the 13+ year SeaWiFS period September 1997–December 2010 quantify time and space patterns of phytoplankton variability and its links to physical forcing in the Pacific Ocean. The CHL fields are a metric of biological variability, SST represents vertical mixing and motion, often an indicator of nutrient availability in the upper ocean, SLA is a proxy for pycnocline depths and surface currents while vector winds represent surface forcing by the atmosphere and vertical motions driven by Ekman pumping. Dominant modes of variability are determined using empirical orthogonal functions (EOFs) applied to a nested set of domains for comparison: over the whole basin, over the equatorial corridor, over individual hemispheres at extra-tropical latitudes (>20°) and over eastern boundary current (EBC) upwelling regions. Strong symmetry exists between hemispheres and the EBC regions, both in seasonal and non-seasonal variability. Seasonal variability is strongest at mid latitudes but non-seasonal variability, our primary focus, is strongest along the equatorial corridor. Non-seasonal basin-scale variability is highly correlated with equatorial signals and the strongest signal across all regions in the study period is associated with the 1997–1999 ENSO cycle. Results quantify the magnitude and geographic pattern with which dominant basin-scale signals are expressed in extra-tropical regions and the EBC upwelling areas, stronger in the Humboldt Current than in the California Current. In both EBC regions, wind forcing has weaker connections to non-seasonal CHL variability than SST and SLA, especially at mid and lower latitudes. Satellite-derived dominant physical and biological patterns over the basin and each sub-region are compared to indices that track aspects of climate variability in the Pacific (the MEI, PDO and NPGO). We map and compare the local CHL footprint associated with each index and those of local wind stress curl, showing the dominance in most areas of the MEI and its similarity to the PDO. Principal estimator patterns quantify the linkage between dominant modes of forcing variability (wind, SLA and SST) and CHL response, comparing local interactions within EBC regions with those imposed by equatorial signals and mapping equatorial forcing on extra-tropical CHL variability.

Macro- and micro-nutrient flux to a highly productive submarine bank in the Gulf of Alaska: A model-based analysis of daily and interannual variability

Although the Gulf of Alaska is subjected to intense downwelling through much of the year, during early spring and summer, upwelling due to local wind stress curl can occur over major portions of the shelf, resulting in high production. Satellite observations indicate that shallow banks may have substantially elevated chlorophyll concentrations relative to surrounding waters during much of the summer. We use the Regional Ocean Modeling System (ROMS) and ocean observation data to examine circulation and stratification around Portlock Bank in the Gulf of Alaska, and to explore mechanisms contributing to interannual variability in the supply of iron and nitrate onto the bank in spring and summer. ROMS at 3-km resolution is coupled to a lower trophic level biology model for the Gulf of Alaska; the coupled model is driven by tidal forcing, sub-daily atmospheric forcing, freshwater runoff, and boundary and initial conditions from Simple Ocean Data Assimilation (SODA) products. Hydrographic observations were made as part of six surveys undertaken by the GLOBEC/NEP (Global Ocean Ecosystem Dynamics/Northeast Pacific) program. Modeling results suggest that iron supply to the shallow layer around Portlock Bank is controlled by both advection and vertical diffusion processes, while nitrate supply is dominated by tidally-induced vertical diffusion. Overall, higher chlorophyll concentration in summer around Portlock Bank is attributed to strong vertical mixing, which pumps nutrients onto the bank from the flanks on either side, and from the top of the bank into the euphotic zone. Recirculation attributed to tidal effects increases residence time over the bank, further enhancing potential production, but intense mixing atop the bank in early spring can lead to light limitation of phytoplankton production. We used July 2004 cruise data from Portlock Bank to help verify model results.

Numerical study on the bifurcation of the North Equatorial Current

A 1.5-layer reduced-gravity model forced by wind stress is used to study the bifurcations of the North Equatorial Current (NEC). The authors found that after removing the Ekman drift, the modelled circulations can serve well as a proxy of the SODA circulations on the σθ=25.0 kg m−3 potential density surface based on available long-term reanalysis wind stress data. The modelled results show that the location of the western boundary bifurcation of the NEC depends on both zonal averaged and local zero wind stress curl latitude. The effects of the anomalous wind stress curl added in different areas are also investigated and it is found that they can change the strength of the Mindanao Eddy (ME), and then influence the interior pathway.

Mechanisms controlling the seasonal mixed-layer temperature and salinity of the Indonesian seas

We examine the seasonal mixed-layer temperature (MLT) and salinity (MLS) budgets in the Banda–Arafura Seas region (120–138° E, 8–3° S) using an ECCO ocean-state estimation product. MLT in these seas is relatively high during November–May (austral spring through fall) and relatively low during June–September (austral winter and the period associated with the Asian summer monsoon). Surface heat flux makes the largest contribution to the seasonal MLT tendency, with significant reinforcement by subsurface processes, especially turbulent vertical mixing. Temperature declines (the MLT tendency is negative) in May–August when seasonal insolation is smallest and local winds are strong due to the southeast monsoon, which causes surface heat loss and cooling by vertical processes. In particular, Ekman suction induced by local wind stress curl raises the thermocline in the Arafura Sea, bringing cooler subsurface water closer to the base of the mixed layer where it is subsequently incorporated into the mixed layer through turbulent vertical mixing; this has a cooling effect. The MLT budget also has a small, but non-negligible, semi-annual component since insolation increases and winds weaken during the spring and fall monsoon transitions near the equator. This causes warming via solar heating, reduced surface heat loss, and weakened turbulent mixing compared to austral winter and, to a lesser extent, compared to austral summer. Seasonal MLS is dominated by ocean processes rather than by local freshwater flux. The contributions by horizontal advection and subsurface processes have comparable magnitudes. The results suggest that ocean dynamics play a significant part in determining both seasonal MLT and MLS in the region, such that coupled model studies of the region should use a full ocean model rather than a slab ocean mixed-layer model.

Variations of eddy kinetic energy in the South China Sea

Fifteen years of merged altimetric data were used to acquire the seasonal to interanual variations of eddy kinetic energy (EKE) in the South China Sea (SCS). The results show that climatological mean EKE in the SCS ranges from 50 cm(2)/s(2) to 1,400 cm(2)/s(2), with high values in the regions southeast of Vietnam and southwest of Taiwan Island. The amplitude of the annual harmonic of the EKE is characterized by high values to the southeast of Vietnam where the maximum exceeds 800 cm(2)/s(2). The EKE in the northern SCS reaches its maximum in August-February, while it peaks in September-December in the southern SCS. Besides the seasonal variation, the EKE also shows strong interannual variation, which has a negative (positive) anomaly in boreal winter during El Nino (La Nina) events. The interannual variation of local wind stress curl associated with El Nino-Southern Oscillation events may be the cause of the interannual variation of the EKE in the SCS.

The eddy pair off eastern Vietnam: Interannual variability and impact on thermohaline structure

The main features, interannual variability of the eddy pair (an anti-cyclonic southern eddy and a cyclonic northern eddy) off eastern Vietnam and its impact on local thermohaline structure are analyzed using altimeter data, the Simple Ocean Data Assimilation (SODA) data, and the Medium-Range Weather Forecasts (ECWMF) data. The results show that the eddy pair is a seasonal phenomenon usually found in the summer to fall. It can reach downward to about 400 m depth but is strongest at the upper 100 m depth. In addition, the eddy pair displays a remarkable interannual variability with periods of 5.6 and 3.6 years. The local wind stress curl plays an important role in the interannual variability of eddy pair. Moreover, the eddy pair has an obvious impact on the thermohaline structure of the local upper ocean. The eddy pair can lead to stronger temperature variability in the upper 380 m but stronger salinity variability only in the upper 150 m depth. When the eddy pair is strong, the anti-cyclonic eddy deepens the thermocline, while the cyclonic eddy shoals the thermocline. Taking the depth of 20 °C isotherm as the thermocline depth, we find that in September, the thermocline drops downward to 110–140 m depth in the anti-cyclonic eddy but rises up to 50–80 m depth in the cyclonic eddy.

Quantifying cross-shelf and vertical nutrient flux in the Coastal Gulf of Alaska with a spatially nested, coupled biophysical model

The Coastal Gulf of Alaska (CGOA) is productive, with large populations of fish, seabirds, and marine mammals; yet it is subject to downwelling-favorable coastal winds. Downwelling regions in other parts of the world are typically much less productive than their upwelling counterparts. Alternate sources of nutrients to feed primary production in the topographically complex CGOA are poorly known and difficult to quantify. Here we diagnose the output from a spatially nested, coupled hydrodynamic and lower trophic level model of the CGOA, to quantify both horizontal and vertical nutrient fluxes into the euphotic zone. Our nested model includes both nitrogen and iron limitation of phytoplankton production, and is driven by a fine-scale atmospheric model that resolves the effects of local orography on the coastal winds. Results indicate significant ‘‘rivers’’ of cross-shelf nitrogen flux due to horizontal advection, as well as ‘‘fountains’’ of vertical transport over shallow banks due to tidal mixing. Using these results, we constructed a provisional budget of nutrient transport among subregions of the CGOA. Contrary to expectations, this budget reveals substantial upwelling of nutrients over major portions of the shelf, driven by local wind-stress curl. These effects are large enough to overwhelm the smaller downwelling flux at the coast throughout the growing season. Vertical mixing by winds and tides, and horizontal flux from the deep basin, are other substantial contributors to nutrients above the 15-m horizon. These findings help to explain the productivity of this coastal ecosystem.

Bifurcation of Pacific North Equatorial Current at the surface

The grid altimetry data between 1993 and 2006 near the Philippines were analyzed by the method of Empirical Orthogonal Function (EOF) to study the variation of bifurcation of the North Equatorial Current at the surface of the Pacific. The relatively short-term signals with periods of about 6 months, 4 months, 3 months and 2 months are found besides seasonal and interannual variations mentioned in previous studies. Local wind stress curl plays an important role in controlling variation of bifurcation latitude except in the interannual timescale. The bifurcation latitude is about 13.3°N in annual mean state and it lies at the northernmost position (14.0°N) in January, at the southernmost position (12.5°N) in July. The amplitude of variation of bifurcation latitude in a year is 1.5°, which can mainly be explained as the contributions of the signals with periods of about 1 year (1.2°) and 0.5 year (0.3°).

Numerical study on the summer upwelling system in the northern continental shelf of the South China Sea

A three-dimensional baroclinic nonlinear numerical model is employed to investigate the summer upwelling in the northern continental shelf of the South China Sea (NCSCS) and the mechanisms of the local winds inducing the coastal upwelling, associated with the QuikSCAT wind data. First, the persistent signals of the summer upwelling are illustrated by the climatological the Advanced Very High-Resolution Radiometer (AVHRR) Sea Surface Temperature (SST) image over 1985–2006 and field observations in 2006 summer. Then, after the successful simulation of the summer upwelling in the NCSCS, four numerical experiments are conducted to explore the different effects of local winds, including the wind stress and wind stress curl, on the coastal upwelling in two typical strong summer upwelling regions of the NCSCS. The modeled results indicate that the summer upwelling is a seasonal common phenomenon during June–September in the NCSCS with the spatial extent of a basin-scale. Typical continental shelf upwelling characteristics are clearly shown in the coastal surface and subsurface water, such as low temperature, high salinity and high potential density in the east of the Hainan Island, the east of the Leizhou Peninsula and the southeast of the Zhanjiang Bay (noted as the Qiongdong-QD), and the inshore areas from the Shantou Coast to the Nanri Islands of the Fujian Coast (noted as the Yuedong-YD). The analysis of the QuikSCAT wind data and modeled upwelling index suggests that the local winds play significant roles in causing the coastal upwelling, but the alongshore wind stress and wind stress curl have different contributions to the upwelling in the Qiongdong (QDU) and the coastal upwelling in the Yuedong (YDU), respectively. Furthermore, model results from the numerical experiments show that in the YD the stable alongshore wind stress is a very important dynamic factor to induce the coastal upwelling but the wind stress curl has little contribution and even unfavorable to the YDU. However, in the QD the coastal upwelling is strongly linked to the local wind stress curl. It is also found that not only the offshore Ekman transport driven by the alongshore wind stress, the wind stress curl-induced Ekman pumping also plays a crucial effect on the QDU. Generally, the wind stress curl even has more contributions to the QDU than the alongshore wind stress.

The Role of Bottom Vortex Stretching on the Path of the North Atlantic Western Boundary Current and on the Northern Recirculation Gyre

Abstract The mechanisms affecting the path of the depth-integrated North Atlantic western boundary current and the formation of the northern recirculation gyre are investigated using a hierarchy of models, namely, a robust diagnostic model, a prognostic model using a global 1° ocean general circulation model coupled to a two-dimensional atmospheric energy balance model with a hydrological cycle, a simple numerical barotropic model, and an analytic model. The results herein suggest that the path of this boundary current and the formation of the northern recirculation gyre are sensitive to both the magnitude of lateral viscosity and the strength of the deep western boundary current (DWBC). In particular, it is shown that bottom vortex stretching induced by a downslope DWBC near the south of the Grand Banks leads to the formation of a cyclonic northern recirculation gyre and keeps the path of the depth-integrated western boundary current downstream of Cape Hatteras separated from the North American coast. Both south of the Grand Banks and at the crossover region of the DWBC and Gulf Stream, the downslope DWBC induces strong bottom downwelling over the steep continental slope, and the magnitude of the bottom downwelling is locally stronger than surface Ekman pumping velocity, providing strong positive vorticity through bottom vortex-stretching effects. The bottom vortex-stretching effect is also present in an extensive area in the North Atlantic, and the contribution to the North Atlantic subpolar and subtropical gyres is on the same order as the local surface wind stress curl. Analytic solutions show that the bottom vortex stretching is important near the western boundary only when the continental slope is wider than the Munk frictional layer scale.

Upper ocean heat content and atmospheric anomaly fields in the off-equatorial North Pacific related to ENSO

We have investigated interannual-scale variations of oceanic and atmospheric anomaly fields, such as upper ocean heat content (OHC), sea surface temperature (SST), latent heat flux (LHF) through the sea surface, sea level pressure (SLP) and wind stress curl (WSC) in the tropical Pacific and their relationships to El Nino/Southern Oscillation (ENSO) events. The results reported here show that the OHC and SST anomalies are almost in phase and lead LHF anomalies in the western tropical Pacific (WTP) region, which are preferable to the generation of subsequent atmospheric anomalies in the WTP. We also describe linear relationships between the amplitudes of these variables in the WTP. In addition, the results show that the both WSC and LHF anomalies are in phase with the temporal trend of OHC anomalies in the WTP, and suggest a combined effect of the local WSC and LHF anomaly in the WTP and ENSO-related, off-equatorial, westward propagating OHC anomaly to generate a large OHC anomaly in the WTP. In contrast to the WTP, OHC and SST anomalies are not in phase to the east of the WTP. The results also indicate that OHC anomalies in the WTP have a potential effect on the generation of an equatorial OHC anomaly via both a reflection of waves at the western boundary and atmospheric variations, which force the enhancement of western equatorial OHC anomaly. Therefore, the WTP is a key region where ENSO events are significantly modulated, and OHC anomalies in the WTP play an important role in the subsequent ENSO event.

A model study of the summer Southeast Vietnam Offshore Current in the southern South China Sea

The western boundary current in the southern South China Sea (SCS) in summer does not always flow northward along the Indo-China Peninsula, it leaves the southeast coast of peninsula around 10–14°N, forming a strong eastward jet called “Vietnam Coastal Current” or “Southeast Vietnam Offshore Current” (SVOC). It is known that the wind stress curl is the major driving factor responsible for this current. In this paper, we carry on the study of the separation position, strength and forming time of this current. A connected single-layer/two-layer model is employed here to study these problems. According to the numerical experiments and analyses of the vorticity dynamics, it is found that, the local wind stress curl (including the northern cyclonic and the southern anticyclonic wind forcing curl), the nonlinear term, the topographic effect, the planetary vorticity advection and the water exchange between the SCS and Java Sea via the Sunda Shelf have an important effect on both the position where this current leaves the coast and its strength; when there is an inflow via the Sunda Shelf, the current is stronger and the separation position is more northward; whereas the water stratification, the coastline and the inflow of Kuroshio have little effect on its separation. In fact, two opposite flowing currents, the northward SVOC and the southward western branch of the cyclonic eddy to its north near the Indo-China Peninsula, collide with each other, and the strength of these two currents determine the separation position of the SVOC. Origin of the SVOC may be driven by the local negative wind stress curl in the middle SCS in mid-spring, this current flows along the coast of the Indo-China Peninsula and leaves the coast at high latitude, flowing northeastward; once the local positive wind stress curl near the northern Indo-China Peninsula or the negative one near the southern Indo-China Peninsula is large enough, this current will begin to leave the coast at low latitude.

Evolution of model systematic errors in the Tropical Atlantic Basin from coupled climate hindcasts

Significant systematic errors in the tropical Atlantic Ocean are common in state-of-the-art coupled ocean–atmosphere general circulation models. In this study, a set of ensemble hindcasts from the NCEP coupled forecast system (CFS) is used to examine the initial growth of the coupled model bias. These CFS hindcasts are 9-month integrations starting from perturbed real-time oceanic and atmospheric analyses for 1981–2003. The large number of integrations from a variety of initial states covering all months provides a good opportunity to examine how the model systematic errors grow. The monthly climatologies of ensemble hindcasts from various initial months are compared with both observed and analyzed oceanic and atmospheric datasets. Our analyses show that two error patterns are dominant in the hindcasts. One is the warming of the sea surface temperature (SST) in the southeastern tropical Atlantic Ocean. This error grows faster in boreal summer and fall and peaks in November–December at round 2°C in the open ocean. It is caused by an excessive model surface shortwave radiative flux in this region, especially from boreal summer to fall. The excessive radiative forcing is in turn caused by the CFS inability to reproduce the observed amount of low cloud cover in the southeastern ocean and its seasonal increase. According to a comparison between the seasonal climatologies from the CFS hindcasts and a long-term simulation of the atmospheric model forced with observed SST, the CFS low cloud and radiation errors are inherent to its atmospheric component. On the other hand, the SST error in CFS is a major cause of the model’s southward bias of the intertropical convergence zone (ITCZ) in boreal winter and spring. An analysis of the SST errors of the 6-month ensemble hindcasts by seven coupled models in the Development of a European Multimodel Ensemble System for Seasonal-to-Interannual Prediction project shows that this SST error pattern is common in coupled climate hindcasts. The second error pattern is an excessive deepening of the model thermocline depth to the north of the equator from the western coast toward the central ocean. This error grows fastest in boreal summer. It is forced by an overly strong local anticyclonic surface wind stress curl and is in turn related to the weakened northeast trade winds in summer and fall. The thermocline error in the northwest delays the annual shoaling of the equatorial thermocline in the Gulf of Guinea remotely through the equatorial waveguide.