The relationship between sea surface temperature (SST) and wind stress has been widely used to characterize regional air-sea coupling, with distinct dependence emerging across spatiotemporal scales and geographic regions (Chelton et al., 2004; Xie, 2004; Small et al., 2008; O’Neill, 2012). At large scales (>1,000 km), atmospheric forcing dominates ocean variability, producing a negative SST-wind stress coupling primarily through wind-driven evaporation and latent heat release (Frankignoul and Hasselmann, 1977; Czaja and Frankignoul, 2002). This mechanism is particularly evident in the extratropic, where stochastic atmospheric forcing overwhelms oceanic feedbacks (Deser et al., 2010). At small scales (1–10 km), localized air-sea exchanges are influenced by surface waves (Sullivan and McWilliams, 2010), temperature gradients, and salinity variations. These interactions induce rapid (hourly to daily) modifications in boundary layer stability through turbulent mixing, predominantly affecting near-surface wind patterns (Gaube et al., 2015; Wenegrat and Arthur, 2018). However, their vertical influence is generally confined within the marine boundary layer, with limited horizontal extent. In contrast, mesoscale interactions (10–100 km) exhibit more persistent and systematic ocean-atmosphere coupling. Sustained SST anomalies associated with oceanic fronts and eddies reorganize boundary layer structure via thermodynamic adjustment processes over weekly to monthly timescales (Chelton et al., 2007; Frenger et al., 2015). These mesoscale features generate secondary circulations that significantly alter wind stress patterns (Cui et al., 2020) and can vertically propagate to influence free atmospheric processes, including cloud formation and precipitation. Horizontally, the impact of mesoscale SST anomalies can extend over hundreds of kilometers, creating coherent atmospheric responses. For instance, observational studies in the North Atlantic (Bryan et al., 2010), Gulf Stream (Park and Cornillon, 2002), and equatorial Pacific (Hayes et al., 1989) have demonstrated that winds blowing from colder to warmer SST regions enhance surface stress and heat fluxes (Wallace et al., 1989). Similar coupling patterns have been observed in the Antarctic Circumpolar Current (Frenger et al., 2013) and western boundary currents (Ma et al., 2016), highlighting the ubiquity of mesoscale air-sea feedbacks in dynamically active regions (Chelton et al., 2004; O’Neill et al., 2010; O’Neill, 2012).

Mesoscale air-sea interaction is a critical process governing the exchange of material and energy between the ocean and atmosphere (Small et al., 2008; Chelton and Xie, 2010), serving as a fundamental driver of variability in oceanic and atmospheric circulations (O’Reilly and Czaja, 2015; Ma et al., 2017). The ocean and atmosphere continuously exchange momentum, heat, and mass, with surface wind stress, thermal fluxes, and momentum fluxes playing pivotal roles in driving surface currents and redistributing heat across the global ocean (Bourassa et al., 2013; Cronin et al., 2019). Conversely, atmospheric responses to mesoscale SST anomalies feed back onto the ocean by modulating wind-driven mixing and upwelling, thereby influencing ocean circulation patterns (Chelton and Xie, 2010). The upper-ocean mixed layer acts as a critical mediator of these exchanges, with SST and wind stress emerging as key variables that reflect both thermal and dynamic coupling (Sun et al., 2020). Given their central role in regulating global climate and ocean circulation (Yu et al., 2020), accurately representing mesoscale air-sea interactions are essential for improving numerical predictions of weather and climate (Chen et al., 2022).

Mesoscale SST-wind stress coupling is most pronounced near oceanic fronts, where cross-frontal SST gradients induce atmospheric boundary layer (ABL) instability. This instability enhances vertical momentum mixing and increases surface wind stress over warmer waters, deepening the ABL and facilitating the downward transfer of momentum from the free atmosphere to the ocean surface, thereby accelerating surface winds (Small et al., 2008; Chelton and Xie, 2010; O’Neill et al., 2010). In contrast, colder waters stabilize the ABL and suppress near-surface winds (Chelton and Xie, 2010). As a result, a positive coupling coefficient between SST and wind stress is predominantly observed in frontal regions of the global ocean (Chelton et al., 2004). This coupling mechanism further generates wind stress curl (WSC) anomalies that are approximately proportional to crosswind SST gradients (CWSG) within frontal zones (Chelton et al., 2001; see Equations 1–3). Coupling coefficients, defined as the regression slope between WSC and CWSG, are commonly used to quantify the strength of the interaction (O’Neill et al., 2010). Observations consistently reveal substantial spatial variability in coupling coefficients across the global ocean (Wang and Castelao, 2016).

A critical factor modulating mesoscale SST-wind stress coupling is wind stress directional steadiness (WSDS), defined as the ratio of vector-averaged to scalar-averaged wind speed, which reflects the directional stability of surface winds (Chelton et al., 2007; see Equation 4). High WSDS, indicating stable wind directions, enhances SST-wind stress coupling by allowing the ABL to respond more consistently to underlying SST gradients (Chelton et al., 2001; Xie, 2004; Small et al., 2008). For instance, in the Brazil Current region, coupling coefficient is low during the onset of the monsoon but increases substantially as WSDS elevates (Castelao, 2012). Seasonal variations in WSDS have therefore been shown to drive pronounced intra-annual changes in mesoscale coupling intensity (Sun et al., 2020; Yu et al., 2020). In monsoon-dominated systems such as the Arabian Sea, the summer southwest monsoon induces strong coastal upwelling along the coasts of Oman and Somalia (Schott et al., 2009), generating prominent SST fronts that modulate surface winds (Vecchi et al., 2004; Xie, 2004). In contrast, the winter northeast monsoon is less favorable for frontogenesis and associated with weaker mesoscale air-sea interactions (Beal and Donohue, 2013).

Prior studies have elucidated key mechanisms governing mesoscale SST-wind stress coupling, including vertical momentum mixing, pressure adjustments, and WSDS (Chelton et al., 2004; Small et al., 2008; Castelao, 2012). Vertical momentum mixing refers to the turbulent exchange of momentum between the ocean surface and the ABL (Wallace et al., 1989; Hayes et al., 1989), which influences the intensity of the wind stress acting on the sea surface. This process can modify the feedback between SST and wind stress by altering the vertical profiles of velocity and turbulence (Small et al., 2003). Pressure adjustments involve SST anomalies induced changes in atmospheric pressure fields that can modulate wind patterns by adjusting the local pressure gradient force, thus affecting the strength and direction of wind stress near the ocean surface (Chelton and Xie, 2010). WSDS refers to the temporal variability of wind direction, which affects the development of air-sea coupling. When wind direction changes rapidly, the ocean has insufficient time to respond fully to atmospheric forcing, disrupting the coupling process (Chelton et al., 2004; Small et al., 2008). Conversely, more stable wind directions provide the necessary time for effective ocean-atmosphere interactions, thereby enhancing the strength of the coupling. Despite these advancements, substantial uncertainties persist in accurately quantifying mesoscale air-sea coupling and assessing its impacts on ocean dynamics (Chelton and Xie, 2010; Frenger et al., 2013). In particular, the impacts of vertical momentum mixing and pressure adjustments have been well investigated in former studies; however, the quantitative role of WSDS in modulating mesoscale air-sea interactions remains poorly understood, and the temporal evolution of these interactions requires further investigation. It is crucial to evaluate the influences of WSDS on the intensity, temporal evolution, and spatial variability of the coupling between wind stress and SST gradients, especially under the complex seasonal monsoon forcing and upwelling dynamics. The western Arabian Sea, characterized by strong seasonal monsoon variability (Schott et al., 2009), provides an ideal natural laboratory to investigate these processes.

In this study, we systematically evaluate mesoscale SST-wind stress coupling by synthesizing satellite observations and reanalysis products, with a particular focus on coastal frontal regions. We investigate the role of WSDS on modulating coupling strength, following observational frameworks established in recent studies (e.g., Oerder et al., 2016; Shao et al., 2019; Cui et al., 2020), which demonstrated that mesoscale SST gradients can substantially reshape surface wind stress patterns through ABL adjustments, with the degree of coupling strongly influenced by wind persistence and stability. Building on these insights, our findings aim to advance the understanding of mesoscale processes that are critical for marine ecosystem dynamics, regional climate resilience, and the predictability of the coupled Earth system.

The SST data are derived from the Moderate Resolution Imaging Spectroradiometer (MODIS), developed by NASA. MODIS provide data across 36 spectral bands ranging from 0.4 μm to 14.4 μm at different spatial resolutions (2 bands at 250 m, 5 bands at 500 m, and 29 bands at 1 km). It provides daily measurements with spatial resolution of 4km (Kilpatrick et al., 2015). The data is sensitive to provide measurements of mesoscale to large-scale global dynamics, aiding in the tracking of variability over time (Barnes and Salomonson, 1995). The time range of the data is 20 years from January 2003 to December 2022, covering a complete 20-year period that captures multiple annual cycles.

The wind stress data are sourced from the fifth-generation atmospheric reanalysis dataset ERA5, produced by the European Centre for Medium-Range Weather Forecasts (ECMWF) for global climate from January 1950 to the present. ERA5 represents nearly 80 years of ECMWF’s reanalysis of global climate and weather. This study utilizes daily averaged reanalysis data spanning from January 2003 to December 2022, consistent with SST dataset. The data are organized into a regular grid with a resolution of 0.25 degrees (Hersbach et al., 2020). To ensure spatial consistency, the original high-resolution SST data (4 km × 4 km) were interpolated to match the ERA5 wind field resolution (0.25° × 0.25°) using bilinear interpolation. This standardized grid system preserves the physical relationships between oceanic and atmospheric variables while enabling accurate calculation of CWSG relative to wind direction. The bilinear interpolation method effectively maintains both the magnitude and directional characteristics of SST gradients, which are crucial for robust air-sea interaction analysis.

WSC quantifies the rotational effect of wind stress on the ocean surface, indicating the tendency of the wind to induce cyclonic or anticyclonic circulation in the upper ocean. CWSG is the component of the SST gradient that runs perpendicular to the direction of wind. It reflects lateral variations in temperature across wind. The CWSG and WSC are calculated based on the method outlined by Wang and Castelao (2016):

τx and τy represent the zonal (eastward) and meridional (northward) components of wind stress, respectively, ∂∂x and ∂∂y denote the partial derivatives along longitude and latitude directions.

where ∇T is the SST gradient, ∂T∂x is the zonal (east-west) component of SST gradient, ∂T∂y is the meridional (north-south) component of SST gradient. τ^ is the unit vector of wind stress direction, × denotes the cross product.

WSDS is calculated as the ratio between the vector-averaged wind stress and magnitude-averaged wind stress (Bryan et al., 2010; O’Neill, 2012).

Where τ→=(τx,τy) is the wind stress vector, with τx and τy representing the zonal (eastward) and meridional (northward) components, respectively. |〈τ→〉|=〈τx〉2+〈τy〉2 is the magnitude of the vector-averaged wind stress. The angle brackets 〈·〉 denote a vector average (time or spatial mean) of the wind stress components τx and τy. 〈|τ→|〉=〈τx2+τy2〉 is the magnitude-averaged wind stress (time or spatial mean). The normalized value of WSDS constrains a value within the range of [0, 1], and thereby quantifying the directional variability relative to the overall wind stress magnitude. Specifically, when the wind stress is consistently oriented in a single direction over time, the wind vector is reinforced each other, i.e., with high WSDS approaching 1. In contrast, if the wind direction fluctuates significantly, the vector averaging causes partial cancellation among components, reducing |〈τ→〉| and indicating greater directional dispersion with WSDS approaching 0 (O’Neill, 2012).

All variables were averaged over 29-day periods at 7-day intervals, following the methodology commonly used in mesoscale air-sea interaction studies, to capture subseasonal variability while smoothing out high-frequency noise (e.g., Bryan et al., 2010; O’Neill, 2012). This temporal averaging scheme effectively isolates the ocean-atmosphere coupling signals associated with mesoscale eddies and fronts, as shorter periods may retain synoptic weather influences, whereas longer periods could obscure transient coupling dynamics (Chelton and Xie, 2010). All data were processed as anomalies by removing climatological means, thereby effectively isolating mesoscale signals from seasonal cycles and long-term trends, in consistent with standard procedures from former studies (Gaube et al., 2015).

Mesoscale air-sea coupling strength is quantified through a coupling coefficient combining binned averaging and linear regression techniques between CWSG and WSC (Gaube et al., 2015). This method captures the thermodynamic forcing of mesoscale ocean features on the ABL, where positive regression coefficients indicate that SST-induced pressure gradients elevate the spatial variation in wind stress that high SST is associated with high wind stress (Chelton and Xie, 2010; Renault et al., 2019). The coefficient with values being seasonally dependent due to variations in background wind fields (Small et al., 2008; Wikle, 2015).

The ocean-atmosphere interaction over the summer upwelling region in the western Arabian Sea examines with MODIS SST and ERA5 wind stress datasets. The topographic configuration of the region exerts a fundamental influence on coastal upwelling intensity during the southwest monsoon. Regions with narrow continental shelves and steep slopes, particularly along the Somalia coast where depths exceed 2,000 m within close proximity to the shoreline, promote highly efficient vertical transport of cold, nutrient-rich deep water. This topographically enhanced upwelling, driven by persistent alongshore monsoonal winds and associated Ekman transport, sustains elevated nutrient availability and primary productivity, in line with other upwelling-dominated systems (Beal et al., 2000; Lakshmi et al., 2020).

These physical processes are clearly expressed in the spatial distribution of SST anomalies. Notably, the Somalia upwelling front exhibits an exceptionally steep SST gradient (>7 °C per 100 km), indicative of an intense oceanic frontal system that enhances vertical nutrient flux and supports vigorous biological production (Schott et al., 2009; Lakshmi et al., 2020). In contrast, moderate SST gradients were observed off Oman (2–3 °C per 100 km), likely reflecting topographically mediated weakening of monsoon winds (Yi et al., 2018). In the central western Arabian Sea, SST gradients are minimal (<1 °C per 100 km), consistent with regions of wind stress convergence and weak surface forcing that limit vertical mixing and suppress thermal structure. Together, these regional patterns highlight the strong modulation of upwelling by both wind forcing and coastal bathymetry. Moreover, dynamic feedback exists as wind-driven SST gradients reshape ABL stability and in turn alter surface wind fields, reinforcing the complexity of air-sea coupling in the monsoon-dominated western Arabian Sea (Schott et al., 2009; Small et al., 2008).

The regional coupling patterns between WSC and CWSG further elucidate the spatial structure of monsoon-driven upwelling coupling in the western Arabian Sea, particularly across the Somalia and Oman upwelling regions (Figure 1d). In the Somalia upwelling region, this relationship is particularly pronounced (r ≈ 0.8), indicating that WSC responds sensitively to underlying SST fronts. This observation is in consistent with established feedback mechanisms, whereby sharp SST gradients modulate the overlying ABL, altering wind stress patterns and reinforcing ocean-atmosphere interactions (Chelton et al., 2004). However, this coupling weakens markedly in the northwestern Arabian Sea, particularly between 17°N and 20°N, where coupling coefficients decline significantly. This spatial disparity reflects the disruptive influence of coastal topography that prominent capes and coastal curvature can distort local wind stress patterns (Chelton et al., 2007), and numerical studies confirm the coastal terrain can attenuate wind stress magnitudes by 10–80% within 10–80 km offshore (Renault et al., 2016).

The exceptionally large coupling coefficients between WSC and CWSG observed in this study (R = 0.98) surpasses values previously reported for similar air-sea interaction systems (Chelton and Xie, 2010; O’Neill, 2012), offering compelling evidence for direct dynamical coupling between wind stress patterns and oceanic thermal fronts. This finding supports theoretical expectations from ABL models (Small et al., 2008) and further confirms the presence of mutual feedback mechanisms proposed in high-resolution coupled simulations (Renault et al., 2019). The bidirectional nature of this coupling suggests that WSC may not only drive but also be modulated by SST gradients, reinforcing the view of a tightly integrated air-sea feedback system. It also reflects a combination of unique geophysical and climatic factors. Specifically, the region’s steep continental slope enhances Ekman pumping efficiency by facilitating rapid vertical transport of subsurface waters (Schott and McCreary, 2001), while the strong and persistent southwest monsoon winds maintain background gradients conducive to intensified air-sea interaction (Currie et al., 2013). Moreover, minimal summer cloud cover (typically<20%) allows for high-precision satellite retrievals of both wind vectors and SST fields, which reduces observational uncertainty (Vialard et al., 2009). These regional characteristics collectively contribute to the pronounced WSC-CWSG coupling signal observed here, underscoring how topography and wind regimes jointly modulate air-sea dynamics, which can be dedicatedly depicted with high atmospheric clarity. Such a robust relationship implies close dependence between WSC and CWSG, with the potential to improve short-term oceanographic forecasts by 15–20%, consistent with predictive gains reported in other upwelling systems (Chelton and Xie, 2010; Renault et al., 2019).

The WSDS is confirmed to serve as a pivotal regulator of mesoscale air-sea interactions in the western Arabian Sea, exhibiting pronounced seasonal modulation under the influence of the South Asian monsoon system. During the boreal summer, persistently strong and highly unidirectional southwesterly winds foster exceptionally high WSDS values (>0.95), thereby creating an optimal dynamical regime for robust coupling between WSC and CWSG (Figure 3b). During periods of elevated WSDS (Figure 5), WSC and CWSG exhibit stronger coupling coefficients with an approximately linear relationship (Figure 4c), indicating a stable and direct dynamical linkage between them. The coupling coefficients between these two fields further exhibit a monotonic strengthening from May to September, reflecting the progressive intensification of monsoonal forcing and air-sea interaction (Figure 6). The persistence of stable monsoonal winds enhances Ekman pumping efficiency by prolonging atmospheric forcing, thereby reinforcing oceanic responses over time (Chelton et al., 2007). Observations indicate that coupling strength can increase by 30–40% during periods of high wind stability, such as the mature monsoon phase (Chatterjee et al., 2019).

By contrast, during the winter monsoon, the intrusion of weaker, variable northeasterly winds originating from the continental interior results in substantially lower WSDS (<0.6), leading to a marked collapse in the strength of WSC-CWSG coupling (Vialard et al., 2012). The suppression of upwelling, diminished SST gradients, and disorganized ABL structures collectively inhibit the maintenance of air-sea feedbacks, consistent with earlier observational syntheses (Shankar et al., 2002; Schott et al., 2009). Transitional seasons reveal an intermediate regime: during spring, the progressive buildup of WSDS initiates the strengthening of coupling; whereas in autumn, the decay of WSDS post-monsoon onset precipitates rapid de-coupling, even when residual SST anomalies persist (Figure 5). Spatially, stable wind directions promote the development of coherent thermal structures and facilitate more efficient momentum and energy transfer across the air-sea interface (Small et al., 2008; Belcher et al., 2012). In particular, the sharp decline and reversal of the WSC-CWSG correlation in October can be attributed to the seasonal transition of monsoonal wind forcing (Figure 6f). As the southwest monsoon weakens and transitions into the northeast monsoon, the prevailing wind direction shifts, reducing the alongshore wind stress that previously sustained upwelling (Schott et al., 2009). This transition disrupts the Ekman-driven offshore transport and weakens the coastal frontal gradients, thereby diminishing the coupling strength between WSC and CWSG (Xie, 2004). Additionally, the ocean adjusts to this forcing change through baroclinic processes such as coastally trapped waves and subsurface thermal structure adjustments, which further decouple the surface wind from the coastal upwelling signal (McCreary et al., 1993). These combined effects likely explain the breakdown in WSC-CWSG coherence observed in October.

In addition to the seasonal modulation, a striking spatial heterogeneity emerges, featuring a pronounced coastal intensification of WSC-CWSG coupling strength (Figure 1c). During the summer monsoon, large coupling coefficients exceeding 0.9 occurred within 100 km off the coast of Somalia, then decay rapidly offshore in an approximately exponential fashion (Figure 1d), aligning with previous observations of localized upwelling dynamics (Vialard et al., 2012). Several interacting coastal processes are simultaneously driving the dynamics. First, the steep coastal bathymetry significantly enhances Ekman pumping, under persistent cross-shore monsoonal winds, promoting strong upwelling (Figure 1a). Second, the shallow mixed layers (~20–30 m) observed during the early monsoon season increase SST sensitivity to atmospheric forcing, thereby accelerating the air-sea feedback loop (Small et al., 2008). Third, nearshore regions experience reduced lateral advection, allowing sharp thermal gradients to persist and reinforcing the coherence of mesoscale SST structures (Chelton et al., 2007). These gradients, in turn, help sustain atmospheric responses via ABL instabilities (Gaube et al., 2015). Together, these mechanisms explain the coastal intensification of coupling and underscore the critical role of regional bathymetry and hydrography in modulating air-sea interactions.

The western Arabian Sea presents fundamentally similar patterns to other major upwelling-dominated boundary currents, such as the California and Benguela systems (Renault et al., 2019), where strong nearshore coupling occurs in regions with sharp SST gradients and narrow coastal jets (Figure 1). However, it departs markedly in its temporal dynamics, displaying a binary seasonal transition between robustly coupled summer states and largely decoupled winter conditions. This behavior is fundamentally rooted in the unique monsoon-driven wind regime, which imposes sharp seasonality absent in trade-wind dominated systems (Figure 2).

A particularly novel aspect of the western Arabian Sea is the regional decoupling of downwind SST gradient (DWSG) and wind stress divergence (WSD) along the coast of Somalia, even in the presence of strong upwelling and sharp thermal fronts. Whereas classical ABL instability theory would predict strong DWSG-WSD coupling under such conditions (Chelton et al., 2010), we observe weak coupling coefficients (R<0.3), attributable to an approximate 90° misalignment between southwesterly monsoon winds and the orientation of SST fronts (Figure 1b). This geometric constraint effectively minimizes DWSG, undermining typical baroclinic adjustment mechanisms (Chelton et al., 2007; Tamsitt et al., 2016). Additionally, the strong remote forcing exerted by the Findlater Jet (Xie and Philander, 1994) likely overwhelms local SST-induced perturbations, further weakening DWSG-WSD coupling. This regional decoupling highlights the need for location-specific frameworks when diagnosing mesoscale air-sea interactions, challenging the universality of coupling paradigms developed in eastern boundary systems.

The demonstrated modulation of coupling strength by WSDS has significant implications for climate modeling and prediction. Most current coupled models lack an explicit treatment of WSDS variability, potentially limiting their ability to faithfully simulate seasonal SST structure and monsoon dynamics. Our findings highlight the need to incorporate WSDS based parameterizations into coupled models, particularly in monsoon influence. Specifically, quantifying the degree of air-sea coupling to explain wind direction persistence, coupled with high-resolution ocean-atmosphere grids, will be a crucial step forward (Small et al., 2008). From the perspective of physical mechanisms, variations in WSDS covary with other wind stress variables such as wind speed and WSC, making it difficult for existing signal separation and causal identification methods to effectively distinguish the independent contributions of each factor to vertical momentum mixing (Renault et al., 2016). Future work should develop advanced methods to separate the WSDS effect from other wind stress variables to better constrain the temporal evolution of mesoscale feedback mechanisms.

In summary, the study advances the understanding of mesoscale air-sea interactions by quantifying WSDS as a dynamic regulator of coupling strength, elucidating the geometric constraints underlying regional decoupling, and articulating the implications for next-generation climate modeling frameworks. These insights contribute to a growing recognition of the complexity and regional specificity of ocean-atmosphere feedbacks in the global climate system.

This study highlights the significant role of WSDS in modulating ocean-atmosphere interactions in the western Arabian Sea, particularly during the monsoon seasons. Understanding these dynamics not only enhances our knowledge of regional climate processes but also contributes to broader climate models that incorporate ocean-atmosphere coupling mechanisms. WSDS is a pivotal yet overlooked factor in western Arabian Sea air-sea interactions. High wind steadiness during summer facilitates strong mesoscale coupling via upwelling fronts, while transitional seasons exhibit phase-dependent coupling tied to monsoon dynamics. Future work should explore biogeochemical responses and extend this framework to other monsoon basins.