The dataset is constructed using atmospheric variables from the ERA5 global reanalysis (Hersbach et al., 2020) produced by the European Centre for Medium-Range Weather Forecasts. ERA5 provides a physically consistent representation of the atmosphere at a horizontal spatial resolution of 0.25° × 0.25° and an hourly temporal resolution. ERA5 data were used at the native 0.25° × 0.25° horizontal resolution without regridding. We extracted Z500 and PV between 150 and 500 hPa for the Southern Hemisphere at 00:00, 06:00, 12:00 and 18:00 UTC from 1979–2024, corresponding to a vertical spacing of 25 hPa between 150 and 250 hPa and 50 hPa between 250 and 500 hPa. Although ERA5 provides hourly output, we use 6-hourly fields because these correspond to fully assimilated analysis time steps, whereas the intervening hourly fields are short-range forecasts and are therefore less observationally constrained. This choice also maintains consistency with Pinheiro et al. (2019) and is well suited to the analysis of atmospheric blocking, which evolves on timescales of days to weeks. This period offers 46 years of homogeneous data suitable for constructing long-term climatologies and robust event statistics. The 30-year period from 1991–2020 was selected as the climatological base for all methods that require a reference state. The Z500 level is widely used for blocking diagnostics because it captures the dominant large-scale structure of anticyclonic ridges. Vertically integrated PV provides a complementary dynamical perspective of PV gradients and upper-level wave breaking, which strongly influence blocking formation in the Southern Hemisphere (Dong and Colucci, 2005). The 150–500 hPa layer has been established as the region where PV anomalies most clearly reflect blocking dynamics (Pinheiro et al., 2019).

Blocking detection is performed using four methods applied to both Z500 and PV (the gradient method is only applied to Z500). After each detection step, the same minimum area threshold and temporal persistence criterion are applied to each method via TempestExtremes DetectBlobs and StitchBlobs, so that differences in blocking statistics across methods reflect the detection logic rather than post-processing inconsistencies. For threshold-based methods, smoothly varying climatological reference fields derived from 1991–2020 are used so that blocking detection reflects both the seasonal cycle and regional variability.

The blocking directory contains the 6-hourly instantaneous blocking masks for the 1979–2024 period. These fields are the direct output of the detection methods described in Sect. 2. Each file consists of a binary grid where a value of one indicates blocked flow and a value of zero indicates unblocked flow. The fields preserve the native 0.25° × 0.25° ERA5 grid and allow users to examine the spatial and temporal evolution of blocking episodes. These data are intended for diagnostic analyses, comparison of blocking structures across methods, and integration with other atmospheric variables.

The blocking_id directory contains labeled blocking fields generated using the StitchBlobs tracking algorithm. Each blocked grid cell is assigned an event identifier (ID), whereas non-blocked grid cells are assigned a value of zero. Tracking is based on spatial overlap between blocking regions at consecutive time steps. In this dataset, regions are considered temporally connected when they overlap by at least one grid cell. StitchBlobs does not impose a one-to-one correspondence between structures. Thus, blocking regions that split from a parent region or merge into a single region can retain a common ID through the overlap-based stitching procedure. The dataset does not include a separate catalog of merging and splitting events, but these interactions can be reconstructed from the temporal evolution of the labeled fields. The fields support event-based diagnostics such as duration, spatial extent, displacement, intensity, and the evolution of blocking structures.

The time_series directory provides temporally aggregated time series of blocking metrics for the Southern Hemisphere. These time series describe the temporal evolution of blocking characteristics evaluated over fixed aggregation intervals, including monthly, seasonal, and annual periods, and are derived consistently for each blocking detection method. Three variables are provided: blocking frequency, blocking intensity, and blocking persistence.

Blocking frequency represents the proportion of time steps classified as blocked relative to the total number of time steps within each aggregation period. Blocking intensity reflects the characteristic magnitude of the underlying field during blocked conditions and is calculated as the mean value of the relevant variable over all blocked time steps within each aggregation period. For the Z500-Grad method, blocking intensity is defined as the mean absolute value of the geopotential height gradients GHGN and GHGS during blocked conditions. Blocking persistence represents the mean duration of blocking events affecting each grid cell and is calculated as the total blocked time divided by the number of blocking events passing through that grid cell within the aggregation period. Blocking intensity and persistence are undefined where no blocking events occur.

These aggregated time series enable analyses of temporal variability, long-term behavior, and potential trends in blocking characteristics. Because the time series are aggregated consistently for all methods, they allow direct comparison of interannual and decadal blocking variability across detection approaches and between the Z500- and PV-based formulations.

The clim directory contains long-term climatological fields derived from the aggregated time series over the 1979–2024 period. These climatologies summarize the mean characteristics of blocking and are provided at annual, seasonal, and monthly temporal resolutions for each detection method. Each timescale includes three variables: blocking frequency, blocking intensity, and blocking persistence.

The climatological blocking frequency describes the average fraction of time blocked at each grid cell. The climatological number of blocking events represents the mean occurrence of distinct blocking events affecting each location. Climatological blocking intensity and persistence characterize the typical magnitude and duration of blocking events at each grid cell, based on the definitions used in the time series products.

These climatological fields provide a method-consistent summary of blocking behavior and serve as a reference for assessing large-scale circulation variability and for evaluating how atmospheric and climate models reproduce key aspects of blocking, including its spatial distribution, seasonal cycle, event frequency, and characteristic duration.

All files adhere to standardized naming conventions that encode the variable, detection method, temporal scale and the Southern Hemisphere domain. Metadata include units, coordinate definitions, calendar information, and method identifiers to ensure long-term reproducibility and clarity. The consistent internal structure across detection methods ensures that users can substitute one method for another without requiring adjustments to data handling workflows. This design enables comparative analyses of blocking sensitivity across methods, facilitates the use of the dataset in modeling studies, and supports integration with complementary Antarctic climate datasets.

The dataset organization also ensures compatibility with common scientific workflows. The 6-hourly masks allow high-resolution diagnostics; the event ID fields provide a foundation for object-based analysis; the climatological fields permit long-term spatial comparison; and the time series summarize domain-integrated blocking behavior. Together, these components provide a comprehensive and flexible dataset supporting Antarctic circulation and extreme-event research.

The annual mean blocking frequency derived from the different detection methods is shown in Fig. 4. The methodological diversity among approaches leads to substantial differences in the large-scale spatial patterns of blocking across the Southern Hemisphere. Nevertheless, most methods identify some common broad features, with the notable exception of the Z500-Grad approach. Several methods indicate that blocking activity is preferentially located at mid- to high-southern latitudes, while enhanced frequencies form circumpolar bands in a subset of methods, particularly Z500-SD and PV-SD. Other approaches show blocking maxima centered over the Antarctic continent or a more spatially homogeneous distribution, as in the Z500-Perc and PV-Perc methods, respectively.

The Z500-Grad method depicts a narrow and zonally continuous band of blocking confined primarily to the mid-latitudes of the Southern Ocean (Fig. 4a). Blocking identified by this method is concentrated roughly between 40 and 60° S, reflecting its sensitivity to meridional reversals of the mid-latitude westerlies rather than broad anticyclonic anomalies extending into the polar cap. As a result, blocking frequencies over Antarctica itself are relatively low, with the climatological signal strongest in the mid-latitude regions near southern Africa and Australia. Within this circumpolar band, clear longitudinal asymmetries are evident, with relatively higher blocking frequencies over the South Pacific sector and weaker occurrence over the South Atlantic and Indian Ocean sectors. This pattern highlights the close association of the gradient method with Southern Ocean storm track and the mean flow, rather than with high-latitude continental blocking structures. These low Antarctic frequencies reflect the specific flow-geometry criterion targeted by Z500-Grad, namely reversal of the meridional Z500 gradient, rather than broad persistent anticyclonic anomalies. The contrast with the threshold-based methods is therefore an informative result of the dataset, indicating how diagnosed Antarctic blocking depends on the underlying detection concept. Z500-Grad is retained as a conservative flow-geometry baseline and to enable direct comparison with published studies using gradient-based diagnostics (e.g., Pinheiro et al., 2019).

In contrast with the gradient method, the threshold-based Z500 methods (Z500-Abs, Z500-SD, and Z500-Perc) suggest broader and more spatially extensive blocking patterns, with clear differences among the individual formulations. The Z500-Abs method identifies widespread blocking across much of the Antarctic continent and the surrounding Southern Ocean, with enhanced frequencies extending from the coastal margins into the interior (Fig. 4b). Local maxima are evident over East Antarctica and in the Amundsen Sea sectors, highlighting the sensitivity of this method to large amplitude geopotential height anomalies that affect both continental and oceanic regions. The Z500-SD method exhibits the strongest annular structure among the Z500-based methods (Fig. 4c). High blocking frequencies form a near continuous ring around Antarctica, with maxima concentrated over the Southern Ocean and a minimum over the Drake Passage. This pattern reflects the influence of climatological variability on the anomaly-based threshold, which preferentially highlights regions of large geopotential height fluctuations associated with the polar front and storm track. Blocking occurrence over the continental interior is less frequent in the threshold methods than in the Z500-Abs method. The Z500-Perc method produces a relatively uniform background blocking frequency over the Southern Ocean with a pronounced and spatially extensive enhancement over East Antarctica (Fig. 4d). Blocking over the continental interior is more widespread than in both the Z500-SD and Z500-Abs methods, consistent with the locally defined nature of the percentile threshold, which reduces the dominance of regions with inherently large variability and enhances sensitivity to persistent continental anomalies.

The PV-based methods exhibit lower blocking frequencies and weaker spatial contrasts than the Z500-based methods, reflecting both methodological and dynamical differences. Methodologically, the sensitivity analysis shows that lowering the fixed PV minimum threshold from 1.1 to 0.6 PVU substantially increases detection frequency (Fig. S6). Dynamically, vertically integrated PV anomalies associated with blocking are generally more spatially compact than Z500 anomalies because the vertical integration retains only coherent upper-tropospheric features (Schwierz et al., 2004; Hauser et al., 2024, 2026). A systematically lower PV blocking frequency relative to Z500 under comparable settings has also been reported for the Southern Hemisphere by Pinheiro et al. (2019). The PV-Abs and PV-SD methods show similar spatial patterns, with greater blocking frequencies aligned along the outer edge of the polar vortex, particularly in the mid- to high-latitude Southern Ocean (Fig. 4e, f). In contrast, the PV-Perc method produces the most spatially homogeneous distribution among the PV-based approaches, with relatively uniform blocking frequencies across the polar cap and weaker longitudinal structure (Fig. 4g). Blocking over the Antarctic interior is present but more attenuated than in the Z500-based methods.

The seasonal modulation of blocking frequency for the Z500- and PV-based methods is shown in Fig. 5 and reveals a method-dependent seasonal structure rather than a uniform summer-winter contrast (Fig. 5).

During December–January–February (DJF), the Z500-based methods display marked differences in the spatial organization of blocking rather than a uniform pattern. The Z500-Grad method identifies limited blocking over Antarctica itself, with activity largely confined to the mid-latitudes (Fig. 5a), whereas Z500-Abs and Z500-Perc indicate more frequent blocking over coastal and interior Antarctic regions (Fig. 5b, d). The Z500-SD method emphasizes a circumpolar ring of highest blocking activity over the Southern Ocean, with comparatively weaker signals over the continental interior (Fig. 5c).

In June–July–August (JJA), the spatial distribution of blocking over high latitudes changes markedly. The Z500-Grad method shows localized enhancements in blocking over the Amundsen and Bellingshausen Sea sectors, where the mid-latitude jet weakens and becomes less zonally continuous over the South Pacific (Trenberth and Mo, 1985; Renwick and Revell, 1999) (Fig. 5e). This pattern suggests that regional ridging can influence the polar circulation during winter, preferentially emerging in regions where the westerlies and associated storm track are less dominant. The Z500-SD method shows relatively enhanced blocking over parts of the Antarctic continent and adjacent seas, indicating a redistribution of blocking activity toward the polar cap (Fig. 5g). The Z500-Abs and Z500-Perc methods similarly show winter blocking concentrated in the high latitudes, particularly in regions where large-scale anticyclonic height anomalies project onto the polar circulation, although these occurrences are relatively fewer and more spatially confined in JJA compared to DJF (Fig. 5f, h).

The PV-based methods display a broadly consistent seasonal modulation but with systematically lower blocking frequencies and a more muted spatial structure. In DJF, the PV-Abs and PV-Perc methods show weak and diffuse blocking activity over the Antarctic region, with limited organization over the Southern Ocean and little penetration into the continental interior (Fig. 5i, k). In contrast, the PV-SD method highlights a clearer circumpolar structure, with enhanced blocking frequency over the Southern Ocean storm track (Fig. 5j). During JJA, blocking detected via PV-based methods remains relatively weak overall, with modest increases over the Southern Ocean and localized enhancements near the Antarctic margin, but without the pronounced continental signals seen in blocking detected by the Z500-based methods (Fig. 5l–n). This reduced spatial contrast reflects the greater dynamical constraint of vertically integrated PV and its sensitivity to large-scale, vertically coherent circulation anomalies.

To complement the spatial climatologies, we selected three broad Antarctic regions: West Antarctica (75–180° W, 50–72° S), East Antarctica (40° W–110° E, 50–65° S), and continental East Antarctica (10–150° E, 70–90° S), depicted by the colored polygons in Fig. 5. Figure 6 summarizes blocking frequency averaged over each domain, with temporal mean and standard deviation of area-weighted regional averages derived from annual and seasonal climatologies.

The regional averages confirm the methodological contrasts apparent in the spatial climatologies. The Z500-Abs, Z500-SD, and Z500-Perc methods yield the highest blocking frequencies across all three regions, with annual values generally between approximately 6 % and 10 %. In contrast, Z500-Grad indicates frequent blocking mainly in West Antarctica, particularly in JJA, while yielding near-zero frequencies over East Antarctica and continental East Antarctica. This low continental signal should be interpreted in light of the sensitivity tests: enhanced continental detection emerges only when both the GHGS threshold and area filter are relaxed (Text S1.1; Fig. S4), whereas the retained configuration provides a conservative flow-geometry baseline. The PV-based methods generally produce lower frequencies and weaker seasonal contrasts, consistent with their sensitivity to vertically coherent, large-scale circulation anomalies.

The largest temporal variability occurs for DJF frequencies over continental East Antarctica, particularly for the Z500-based methods, indicating substantial temporal variability in the regional mean blocking frequency. Seasonal differences are most evident in West Antarctica and East Antarctica, whereas continental East Antarctica shows weaker mean seasonal contrasts but considerable temporal variability.

The results presented in Sect. 4.1 and 4.2 show that the frequency and spatial variability of Antarctic blocking are fundamentally shaped by methodological choices in both variable and threshold selection. Blocking based on Z500 captures a broad spectrum of anticyclonic regimes, including high-amplitude anomalies projecting into the polar cap, consistent with their sensitivity to persistent, quasi-stationary highs (Renwick, 1998; Grotjahn et al., 2016). These methods are effective at identifying regions of weakened or reversed zonal flow, but may conflate true dynamical blocking with other persistent positive anomalies (Davini et al., 2012). In contrast, blocking based on PV identifies a more restricted subset of events, emphasizing large-scale, coherent wave breaking and vortex interactions over localized amplitude extremes, resulting in lower, more zonally constrained frequencies (Masato et al., 2013; Polster and Wirth, 2023). This reflects a stricter focus on anomalies with direct dynamical connection to the hemispheric potential vorticity gradient.

The seasonal cycle further illustrates these diagnostic sensitivities observed at the annual scale. The observed redistribution in blocking from summer to winter reflects differing responses to the strengthening and increased zonal symmetry of the polar vortex (Nakamura and Huang, 2018), as well as intraseasonal modulation of Southern Hemisphere wave activity that can favor clustered blocking during specific circulation states (Henderson et al., 2018; Liu and Wang, 2024). Potential mechanisms relevant to winter high-latitude blocking include regional ridge amplification driven by horizontal wave propagation and anticyclonic vorticity advection linked to Rossby wave activity, as documented in recent studies of extreme Antarctic circulation anomalies, as well as interactions with the stratospheric circulation that can reinforce or prolong blocking events (e.g., Gorodetskaya et al., 2023; Lim et al., 2026; Tang et al., 2026).

To quantify the relative importance of the blocking metric and diagnostic variable, we calculated two uncertainty measures from the 1979–2024 blocking-frequency time series at each grid point. Metric uncertainty, σmetric, was calculated as the time-mean standard deviation across the three methods shared by both variables (absolute threshold, standard deviation-based anomaly, and percentile-based anomaly), separately for Z500 and PV; the two resulting values were then averaged. Variable uncertainty, σvariable, was calculated as the time-mean standard deviation between Z500- and PV-based blocking frequencies, separately for each of the three shared methods; these values were then averaged across methods. The gradient method was excluded because it has no PV counterpart. The dominance ratio, R, was obtained by dividing σmetric by σvariable. Values of R above one indicate that uncertainty is dominated by the choice of blocking metric, whereas values below one indicate that it is dominated by the choice of diagnostic variable (Fig. 7).

The resulting dominance ratio shows that, over most of Antarctica and across all seasons, uncertainty is primarily controlled by whether blocking is diagnosed from Z500 or PV, rather than by the specific technique used to quantify blocking. This dominance of variable choice is particularly evident over the Antarctic interior and large sectors of the surrounding Southern Ocean, where Z500- and PV-based diagnostics yield systematically different blocking frequencies. In contrast, localized regions along the Antarctic coastal margins and parts of the storm-track belt exhibit dominance ratios closer to unity, indicating a more comparable influence of metric and variable choices.

These systematic contrasts confirm that methodological differences are not merely inconsistencies but instead reveal complementary physical phenomena. The Z500-based methods provide a comprehensive view of persistent anticyclonic conditions, while the PV-based methods isolate dynamically coherent upper-tropospheric disruptions. A multi-metric framework is therefore essential for a complete understanding of persistent high-latitude circulation regimes, their seasonal evolution, and their impacts on surface climate and sea ice.

Figures 8 and 9 summarize the annual climatology of blocking event number and persistence derived from the Z500- and PV-based detection methods, respectively. The results demonstrate that different blocking definitions lead to markedly different climatological patterns and event characteristics, reflecting the sensitivity of blocking diagnostics to the choice of variable and threshold logic. Rather than indicating the superiority of any single approach, these contrasts illustrate that blocking is a definition-dependent phenomenon and that each method emphasizes different aspects of persistent circulation anomalies.

For the Z500-based methods (Fig. 8), the mean number of detected blocking events per year varies markedly among methodologies. The Z500-Abs and Z500-SD methods identify the largest number of events (Fig. 8b, c), whereas the Z500-Grad method identifies the fewest (Fig. 8a), reflecting fundamental differences in detection methods. The Z500-SD method is particularly sensitive to spatially extensive but moderate departures from the climatological background, allowing multiple distinct events to be identified over a given region, while the Z500-Perc method produces intermediate event counts by focusing on locally extreme and persistent anomalies.

Blocking persistence shows a different ordering among methods, with the threshold-based Z500 methods indicating greater persistence over the Antarctic continent than over the surrounding Southern Ocean (Fig. 8f–h). The Z500-Perc method produces the longest mean event durations, particularly over Antarctica, suggesting that once extreme anomalies relative to the local distribution are established, they can remain quasi-stationary for extended periods (Fig. 8h). The Z500-SD method shows intermediate persistence, while the Z500-Abs method yields shorter mean durations despite higher event counts, consistent with a tendency to fragment blocking activity into multiple shorter-lived events.

The corresponding diagnostics for the PV-based methods are shown in Fig. 9 and reveal some similarities with their Z500 counterparts when comparable thresholds are imposed (i.e., Abs, SD, and Perc), alongside contrasts between different methodological families. Overall, fewer blocking events are detected by PV-based methods than Z500-based methods, reflecting the more conservative nature of PV diagnostics and their focus on dynamically coherent circulation features. Blocking events detected by the PV-Abs and PV-SD methods show similar large-scale spatial distributions, with enhanced event counts aligned along the outer edge of the polar vortex, while blocking detected by the PV-Perc method has the lowest event counts but the longest mean persistence. Persistence derived from vertically integrated PV shows less spatial variability than in the Z500-based methods, consistent with the reduced small-scale spatial variance of vertically integrated PV compared with single-level PV fields. Vertical integration over 150–500 hPa averages out PV anomalies confined to narrow layers, such as tropopause folds and PV streamers, while retaining anomalies that are coherent through a substantial depth of the upper troposphere and lower stratosphere.

Differences in event number and persistence point to a trade-off among detection methods. Methods yielding high event counts, such as Z500-Abs, tend to identify shorter-lived blocking features, suggesting sensitivity to a broader spectrum of positive anomalies rather than consolidated, quasi-stationary blocks. This behavior is consistent with earlier studies noting that some threshold-based methods may conflate true dynamical blocking with other persistent ridge structures (Tibaldi et al., 1997; Scherrer et al., 2006). In contrast, methods emphasizing flow reversal or local extremes tend to identify fewer but more persistent blocking events, more closely aligned with the canonical blocking paradigm reported in the early literature. Pronounced persistence over the Antarctic continent in percentile-based methods is consistent with studies showing that extreme geopotential height anomalies in high southern latitudes can remain anchored by topographic and thermal influences (Hoskins and Woollings, 2015; Clem et al., 2025).

Blocking intensity is not analyzed here because its definition is inherently method-dependent, particularly for absolute threshold approaches. For completeness, intensity diagnostics (excluding the absolute methods) are provided for both Z500 and PV, where they are used only to illustrate relative contrasts among anomaly-based metrics (Figs. S14 and S15).

PV-based diagnostics identify fewer and generally less persistent blocking events than Z500-based methods because they target a different dynamical signature. PV methods are specifically sensitive to large-scale wave breaking and potential vorticity streamer formation, processes that are typically more localized and transient than the broad, quasi-stationary anticyclonic anomalies captured by height field methods (Hoskins et al., 1985; Berrisford et al., 2007). Within this framework, the PV-Perc method isolates the most extreme anomalies and yields the longest-lived events, suggesting that the most robust PV structures exhibit enhanced coherence and duration.

These diagnostics provide complementary perspectives on Antarctic blocking. The Z500-Grad method highlights canonical blocked flow patterns, anomaly-based Z500 methods emphasize regions of recurrent high-amplitude anomalies, and PV-based methods preferentially identify upper-tropospheric wave-breaking signatures. Together, they span a spectrum of persistent circulation regimes, from near-surface high pressure to upper-tropospheric and stratosphere–troposphere coupling events. These preliminary results underscore the value of a multi-diagnostic framework for synthesizing differing interpretations in the literature and for evaluating the representation of high-latitude blocking in climate models (Woollings et al., 2018; Hanna et al., 2024).

Figure 10 presents three representative blocking events from March 2015, February 2020, and March 2022, illustrating how the different methods capture the evolution and spatial structure of individual events. These periods correspond to well-documented Antarctic extreme events, including episodes over the Antarctic Peninsula (Bozkurt et al., 2018; Xu et al., 2021) and over East Antarctica (Wille et al., 2024a). For the threshold-based Z500 and PV methods, there is strong agreement on the timing and general location of blocking across all cases.

In March 2015 (Fig. 10a–c), the methods consistently identify a blocking structure over the Antarctic Peninsula, although the Z500-based methods delineate a broader region than the PV-based methods. In February 2020 (Fig. 10d–f), the underlying Z500-Abs, Z500-SD, and Z500-Perc masks initially identify a separate, short-lived anticyclonic ridge over the South Atlantic, east of South America. This feature does not merge with the blocking event over the Antarctic Peninsula and weakens at later time steps, while the Peninsula feature becomes more consolidated. In March 2022 (Fig. 10g–i), the PV-based methods identify a compact, dynamically coherent block near 90–120° E, whereas the Z500 threshold methods capture a broader ridge signature.

Differences arise primarily in the delineation of event boundaries and spatial extent. The Z500-based threshold methods tend to produce larger and more expansive blocking regions, while the PV-based methods identify more compact blocked regions that closely follow the centers of the underlying dynamical anomalies. In all cases shown in Fig. 10, gradient-based features do not satisfy the minimum areal requirement applied uniformly across all methods and are therefore filtered out during event construction. When considered prior to areal filtering, the Z500-Grad method captures blocking-like structures associated with these events, but with spatial extents that are substantially smaller than those identified by threshold-based approaches (Fig. S16). Similar behavior has been reported in previous studies using alternative implementations of blocking diagnostics. For example, Wang et al. (2024) reported higher blocking frequencies when spatial constraints were relaxed. This comparison highlights the sensitivity of blocking diagnostics to methodological choices and the importance of applying spatial constraints to isolate large-scale features.

Despite these differences in spatial representation and sensitivity to areal filtering, the temporal evolution of the events detected by the threshold-based methods is highly consistent. The growth, mature, and decay phases of each block are captured coherently across each method, reinforcing confidence that the dataset presented here reliably represents the life cycle of major Antarctic blocking episodes. The case studies therefore demonstrate that methodological differences primarily influence the spatial characterization and quantitative metrics of blocking, while the identification of impactful large-scale events remains robust within the unified tracking framework.