Recent waning snowpack in the Alps is unprecedented in the last six centuries

Snow cover is crucial for high-elevation and high-latitude regions, influencing hydrological cycles and climate. It impacts surface energy balance, affecting glaciers and permafrost. Snowpack acts as a water storage reservoir, vital for downstream environmental and human needs. The Alps, Europe's most important water-supplying mountain range, are particularly vulnerable to climate change, with cascading impacts on downstream areas. The reduction in snow cover duration negatively affects snow-dependent species, winter tourism, and related socio-economic activities. In the Alps, seasonal mean snow depth declined by 8.4% per decade (1971-2019), with a parallel reduction in snow cover duration. To understand the unusual nature of recent snow cover dynamics, long-term data are crucial, but such comprehensive data are scarce for the Alps. Existing instrumental records are relatively short, and tree rings, while useful for other climate variables, have limitations for reconstructing snow conditions in this region due to the long resting period during winter and negligible moisture-limiting conditions. This study aims to address this gap by creating a long-term reconstruction of snowpack duration in the Alps.

Several studies highlight the decline in snow cover in the Alps. Matiu et al. (2021) documented an 8.4% per decade decline in seasonal mean snow depth and a 5.6% per decade reduction in snow cover duration between 1971 and 2019. Other studies focus on shorter periods or specific aspects, such as earlier snowmelt or later snow onset. There is a long tradition of weather record collection in the Alps, with instrumental pressure, temperature, and precipitation series extending back to the mid-eighteenth century. Tree-ring studies have reconstructed summer temperatures, but few have focused on snowpack duration due to the challenges mentioned earlier. The absence of continuous long-term records of snowpack duration hampers the accurate assessment of present-day shrinking trends and their context within historical variability.

This study reconstructs snowpack duration using 572 ring-width series from high-elevation (*>2,000 m*) prostrate *Juniperus communis* shrubs in the Val Ventina, Southern Alps. Unlike upright trees, these shrubs' growth is directly inhibited by snow cover, making ring width a potential proxy for snow cover duration. Preliminary tests assessed the robustness of long-term trends and addressed growth peculiarities (missing and wedging rings, eccentric development). A snow degree-day model, using daily instrumental precipitation and temperature data (1834–2018), estimated snowpack duration. The model considers both solid precipitation and snowmelt, based on temperature thresholds. This approach provides a reliable long-term record (1834–2018) for calibration and verification against the ring-width proxy. The final reconstruction (SALP) covers 1400–2018 CE. The calibration/verification of ring width against modeled snowpack duration is strong and stable (median bootstrap calibration/verification statistics of -0.688 and -0.687, respectively). The high- and low-frequency components of temperature, precipitation, and juniper ring-width indices were analyzed using fast Fourier transform to determine their contributions to snowpack variability. Comparisons were made with documentary records, other snow cover studies, and glacier dynamics reconstructions.

The SALP reconstruction shows high year-to-year variability in snow cover duration, ranging from nearly year-round snow cover to durations of around 2 months less than the long-term mean (251 days). A clear decline in snow cover duration started around the end of the nineteenth century. The first two decades of the 21st century recorded the lowest mean snow cover duration (215 days), 36 days less than the long-term mean. This recent decrease is unprecedented in the last 600 years. The shrinkage trend in the last five decades (-5.4 ± 2.5 d decade⁻¹) aligns well with observational data (-6.67 days decade⁻¹). The reconstruction's variability is driven by a combination of high-frequency (cool-season precipitation) and low-frequency (yearly temperature) signals. Comparisons with other data sources, including documentary records and glacier dynamics reconstructions, show qualitative similarities, but also inconsistencies which suggests a complex interaction between temperature and precipitation influencing snowpack duration. The study indicates that some years with extreme conditions in precipitation and temperatures show a remarkable snow cover duration anomaly of >35 days, which closely corresponds in SALP to a decile departure from the mean estimated over the same 500-year period. The reconstruction demonstrates a complex interplay of temperature and precipitation in affecting snow cover. While some extreme cold spells or exceptional snowfalls do not necessarily lead to persistent snow cover.

The study's findings highlight the unprecedented decline in Alpine snowpack duration. The significant decrease in recent decades surpasses the natural variability observed over the past six centuries, underscoring the impact of climate change. The use of *Juniperus communis* ring-width as a proxy for snow cover duration provides valuable insights into long-term snowpack dynamics, complementing existing shorter-term instrumental records. While snow amount (water equivalent) is a widely used parameter in snow hydrology, this study focuses on duration, providing a different perspective. The prostrate growth habit of the juniper, its sensitivity to crown emergence from the snowpack, makes it an effective proxy for duration, not necessarily snow depth. The complex interplay of temperature and precipitation, as observed in the reconstruction, highlights the need for comprehensive modeling approaches that incorporate both factors when assessing future snowpack conditions. The study's findings highlight the importance of understanding the combined effects of climatic variables and highlight the urgent need for adaptation strategies in the Alps.

This study provides the first long-term (600-year) reconstruction of snow cover duration in the Alps, revealing an unprecedented decline in recent decades. The use of *Juniperus communis* ring widths as a proxy successfully captures this long-term trend. This work demonstrates the significant impact of climate change on Alpine snowpack and emphasizes the urgency of developing adaptation strategies for sensitive environmental and socio-economic sectors. Future research could expand the geographical scope of this approach, using similar proxies in other snow-prone regions to provide a broader understanding of global snowpack changes and improve climate models.

The study's findings are based on a single site in the Southern Alps, so the generalizability to other parts of the Alps may be limited. The snow degree-day model, while effective, is relatively simple and may not fully capture the complexities of snowpack dynamics in all situations. Also, the study relies on proxy data that might not perfectly capture the subtleties of snowpack conditions, leading to a slightly smoothed reconstruction of the snowpack duration. The use of the regional curve standardization (RCS) detrending technique might imply a loss of high-frequency variability in the data.