A dynamic link between spring Arctic sea ice and the Tibetan Plateau snow increment indicator

Arctic sea-ice anomalies have been recognized as an indicator of local and remote climate extremes because they directly impact on the atmosphere through thermodynamic and radiative processes. In recent decades, the Arctic has witnessed significant sea-ice loss. The physical processes involved in Arctic sea-ice loss include oceanic warming, lapse rate feedback, Planck, albedo feedbacks, atmospheric moisture and energy transport, some of which may exacerbate Arctic warming and sea-ice melting via increased surface heating. Arctic sea-ice variability is influenced by tropical and mid- to high-latitude ocean-atmospheric processes and global warming. Tropical air-sea interactions, e.g., ENSO and the associated Pacific North American teleconnection, can impact summer and autumn Arctic sea-ice variability through atmospheric moisture and heat transports. Arctic atmospheric anomalies, such as Ural blocking and variations in the polar vortex, exert prominent impacts on winter Arctic sea ice through changes in downward longwave radiation caused by water vapor, temperature changes and sea-ice drift. These studies have focused mainly on summer, autumn and winter Arctic sea ice variability, but relatively few studies have examined the origin of spring Arctic sea-ice variability. Moreover, the Arctic sea-ice spring predictability barrier signifies the importance of exploring the precursors of spring Arctic sea ice variability. Previous studies suggest that preceding Tibetan Plateau (TP) snow cover anomalies could be one of the precursors to extratropical circulation systems because the TP snow cover has cross-seasonal climate impacts. For instance, late autumn TP snow cover anomalies may lead to a remote winter Pacific North American teleconnection response through a persistent snow forcing from autumn to winter. The spring TP snow cover could influence the simultaneous Aleutian Low. Given that variations in these circulation systems might play a role in Arctic sea-ice variability, Arctic sea-ice variability may be linked to TP snow conditions via atmospheric circulation systems. However, it is unclear whether there is a physical pathway through which the TP snow conditions can influence Arctic sea ice variability. This study aims to establish a linkage between the preceding TP snow cover and spring Arctic sea-ice variability. A recent study notes that the persistence of October–November–December (OND, late autumn) TP snow cover extent (SCE) anomalies is typically limited to within two months, suggesting that the OND averaged TP SCE is not a good indicator of the TP snow climate effect, especially for the impact during ensuing seasons. To address this issue, the TP SCE increment within the OND is examined, which can improve the standard deviation by 72.6% and diabatic cooling by 89.7% over the entire Tibetan Plateau domain. The snow increment indicator reflects both the degree of snow cover and the velocity of snow cover accumulation. In addition, the snow increment indicator associated snow anomalies can persist into the following early spring and thus establish a physical linkage with the following early spring Arctic sea-ice concentration (SIC) through sea-ice dynamic processes.

Time period: 1979–2021. Observational and reanalysis datasets: (i) Northern Hemisphere weekly snow cover extent (SCE) on 25 km grids from NSIDC (NSIDC-0046). (ii) Weekly SCE from NOAA Climate Data Record (CDR). (iii) Daily snow depth from the Big Earth Data Platform for Three Poles, converted to monthly 0.25° grids. (iv) Monthly snowfall on 0.25° grids from ERA5. (v) Polar Pathfinder daily sea-ice motion vectors (25 km) from NSIDC (NSIDC-0116). (vi) Monthly surface wind stress, Ekman transport, sea-ice thickness, and sea-ice concentration from ECMWF OCEAN5/ORAS5 (0.25°). (vii) Monthly mean SIC and SST from HadISST (1°). (viii) Atmospheric circulation (winds, geopotential height, specific humidity; 2.5°) and surface energy fluxes (shortwave radiation, sensible heat; T62 Gaussian) from NCEP-DOE Reanalysis-2. Indices and filtering: The OND SCE increment index (SCEII) is defined as the TP-area-weighted SCE difference between December and October and high-pass filtered with an 8-year Gaussian filter to isolate interannual variability. For comparison, the OND SCE index (SCEI) is computed from OND-mean TP SCE. Extreme years are selected by SCEII exceeding ±0.9 standard deviations; ENSO effects are controlled using Niño 3.4 index in composite/partial correlations when specified. Diagnostics: Wave activity flux (Takaya–Nakamura) is used to diagnose quasi-stationary Rossby wave propagation; Rossby wave source (Sardeshmukh–Hoskins) to identify wave generation; barotropic and baroclinic energy conversion terms quantify maintenance pathways; Eliassen–Palm (EP) flux and divergence diagnose vertical and meridional planetary wave propagation and stratosphere–troposphere coupling. Causal inference uses Liang–Kleeman information flow to quantify information transfer from SCEII to SIC. Model experiments: CESM v1.1.1 with CAM5.0 atmosphere (2.5° × 1.875°, 31 levels), CLM4.0 land, and CICE4.0 sea ice. Two 20-year experiments: (1) Control with 1980–2010 climatological monthly SSTs. (2) Perturbation identical to control but with increased TP land surface albedo (+0.4 capped at 0.85) over 76°E–103°E, 28°N–38°N from October to April to mimic persistent high-albedo snow forcing. Responses are computed as perturbation minus control using the last 19 years. Statistical methods: 8-year high-pass Gaussian filtering for interannual signals; extraction of zonal wavenumbers 1–3 via Fourier decomposition; partial correlations and ENSO removal using Niño 3.4; Student’s two-tailed t-test at 5% significance. Information flow significance assessed via Liang-Kleeman framework. Spatial regressions and composites used to relate SCEII to atmospheric, oceanic, and cryospheric fields.

- A late-autumn (OND) Tibetan Plateau snow cover extent increment index (SCEII), defined as December minus October SCE, better represents interannual TP snow variability than the traditional OND-mean SCE index (SCEI). Interannual standard deviation of SCE increment (NOAA dataset) is enhanced by 81.8% (eastern TP), 84.7% (northern), 82.4% (western), 57.4% (central), and 72.6% (entire TP). SCEII and SCEI correlate at 0.49 (p<0.01) but identify largely different extreme years (only 4 overlapping of 14 each), underscoring distinct physical meaning (accumulation speed vs. amount). - SCEII captures whole-TP cooling via diabatic processes: compared to SCEI, SCEII-associated increments show stronger surface energy budget responses and atmospheric cooling. Promotion percentages for SCEII increments over TP: air temperature cooling 89.7%, USWR +42.5%, ULWR −96.8%, SH −82.6%; LH changes are weak. Regressions show SCEII-linked cooling across the entire TP, whereas SCEI fails to represent cooling in the western TP. - Persistence: Positive TP SCE anomalies associated with OND SCEII persist into DJF and FMA, enabling cross-seasonal teleconnections. - Arctic sea-ice response: No significant simultaneous OND SIC correlation; negative Arctic SIC anomalies emerge in DJF and become significant in FMA across the northern Barents, northern Kara, Laptev Seas, extending toward the Arctic center. Information flow from SCEII to SIC is weak in OND/DJF but significant in FMA over the same regions, supporting causality. Results are robust to ENSO removal and largely independent of tropical Indian and North Atlantic SST influences. - Mechanism: SCEII-induced diabatic cooling excites a TP–Arctic hemispheric wave train guided by the subtropical westerly jet. Upper-level negative geopotential height over the TP extends eastward across the North Pacific/Atlantic, reaching the Arctic by FMA. CESM simulations forced by higher TP albedo reproduce the observed wave train and spring Arctic response. - Dynamics dominate SIC reduction: In FMA, a barotropic cyclonic anomaly over the northern Barents–Kara–Laptev region intensifies wind stress and Ekman transport, driving anomalous sea-ice drift out of the Arctic center toward the North Atlantic, thinning ice and reducing SIC. Moisture and thermal anomalies over the Arctic are weak, indicating a secondary thermodynamic role. - Wave train formation and maintenance: Negative RWS over the TP, aided by snowfall and moisture convergence and the SWJ waveguide, initiates the train. Maintenance involves barotropic energy conversion from the mean flow and transient eddy activity; baroclinic conversion supports snow–atmosphere interaction over and east of the TP. Stratosphere–troposphere coupling, seen in EP flux (upward Jan–Feb, downward in March), explains the northward migration and deepening of negative geopotential height from the North Atlantic into the Arctic. - Regional nuance: East of Greenland, opposing dynamical (ice export reduces SIC) and thermal (cold anomalies increase SIC) effects reduce statistical significance. - Trend: The OND TP SCE increment shows a weak long-term trend, implying limited contribution to the multi-decadal Arctic SIC decline dominated by greenhouse gas–driven warming. The original (unfiltered) SCEII correlates strongly (0.89) with filtered SCEII and yields similar FMA SIC patterns.

The study addresses the origin and predictability of spring Arctic sea-ice variability by identifying a dynamically meaningful Tibetan Plateau snow indicator that links late-autumn TP snow evolution to spring Arctic SIC anomalies. By focusing on the OND SCE increment (SCEII) rather than OND mean SCE, the authors demonstrate stronger interannual variability, enhanced diabatic cooling, and more coherent atmospheric circulation responses over the TP. These local effects persist into winter and spring, exciting a hemispheric-scale Rossby wave train along the subtropical westerly jet that projects into the Arctic. The resulting springtime (FMA) Arctic cyclonic anomalies strengthen wind stress and Ekman transport, drive anomalous sea-ice drift out of the central Arctic, and reduce ice thickness and concentration, particularly in the northern Barents–Kara–Laptev Seas. Causal analysis (information flow) supports that SCEII leads SIC changes in FMA. The mechanism is primarily dynamic; thermodynamic contributions over the Arctic are weak. Formation of the wave train is linked to negative Rossby wave source over the TP associated with snowfall and moisture convergence, while maintenance involves barotropic energy extraction and transient eddies, with baroclinic feedbacks over and east of the TP. Troposphere–stratosphere coupling facilitates the poleward extension and deepening of negative geopotential height from the North Atlantic into the Arctic. These findings reveal a robust TP–Arctic linkage that is independent of ENSO and largely insensitive to tropical Indian and North Atlantic SST influences. Consequently, the SCEII provides a physically based predictor for spring Arctic SIC variability, potentially helping to mitigate the spring predictability barrier by incorporating antecedent TP snow dynamics into prediction systems.

The OND Tibetan Plateau snow cover extent increment (SCEII) is a superior indicator of TP snow’s climatic impact compared to traditional OND-mean SCE. It exhibits larger interannual variability, stronger diabatic cooling and energy budget signatures, and a more coherent link to hemispheric circulation. SCEII-driven diabatic cooling over the TP excites a TP–Arctic wave train guided by the subtropical westerly jet, culminating in springtime Arctic cyclonic anomalies that dynamically export sea ice and reduce SIC and thickness in the northern Barents–Kara–Laptev Seas and toward the Arctic center. Causal analysis supports that antecedent SCEII anomalies lead spring SIC changes, and the relationship is robust to ENSO and major SST influences. This work highlights a dynamic bridge from the TP to the Arctic via atmospheric waveguides and sea-ice drift processes, offering a promising precursor for spring Arctic sea-ice forecasts. Future research should further examine regional compensation of thermal versus dynamic effects (e.g., east of Greenland), refine the role of tropical Indian Ocean SST in modulating wave pathways, and assess the integration of SCEII into operational prediction systems and coupled data assimilation for improved spring sea-ice predictability.

- The statistical linkage east of Greenland is weakened by competing dynamical (ice export) and thermal (cold-air) influences, reducing significance in that region. - While largely independent of ENSO and key SST patterns, removing tropical Indian Ocean SST signals alters some wave train pathways, indicating sensitivity that warrants further investigation. - The OND TP SCE increment exhibits a weak long-term trend and thus likely contributes little to the multi-decadal Arctic SIC decline dominated by greenhouse warming; implications for climate-change–time-scale predictability are limited. - Results rely on reanalysis and model diagnostic frameworks with inherent uncertainties (e.g., fluxes, sea-ice motion), and CESM experiments employ idealized albedo perturbations rather than fully interactive snow processes.