Connecting Large-Scale Atmospheric Circulation to Spring Frost Events in Marlborough Vineyards

Spring frost events pose a significant risk to viticulture (Poni S. et al., 2022). In Marlborough, New Zealand's leading wine region, the economy is heavily dependent on vineyard yields, making spring frost events a very important threat to consider. 

This risk is exacerbated in the context of climate change, as the increase in warmer temperatures has led to the advancement of phenological stages in grapevines, notably bud break (Van Leeuwen C., et al., 2016). This advancement makes grapevines more vulnerable to frost, as new growth is then exposed to potential late frost events. Despite these critical implications, frost risk in Marlborough has not been thoroughly assessed since 2018, emphasizing the need for an updated analysis.

Initial research on frost occurrence patterns was conducted by Clark and Sturman in 2009 and revealed significant variations in observed climate across New Zealand's primary vineyard regions. This suggests that temperature changes are not uniform throughout the country, with Marlborough experiencing an increase in frost occurrence in contrast to other areas. These regional disparities have been attributed to changes in large-scale atmospheric circulation and its interaction with New Zealand's complex topography (Sturman, A. et al., 2013).

Given the high sensitivity of the wine industry to weather fluctuations and the considerable regional variability in trends of key parameters, such as temperature, a deeper understanding of the spatial and temporal variability of New Zealand's weather regimes is proving to be highly valuable for viticulture. This study investigates the variability of spring frost events in Marlborough by linking large-scale atmospheric circulation patterns to local climate impacts. 

We use a combination of K-means clustering and Self-Organizing Maps (SOMs) on high-resolution WRF model data spanning three nested domains (27 km, 3 km, and 1 km resolution). While K-means is widely used to identify regional weather regimes (Polo, I. et al., 2011), its tendency to oversimplify the continuum of atmospheric variability is mitigated here by the Self-Organizing Maps, which preserve the topological relationships of the data and capture gradual transitions between climate states. 

The proposed methodology aims to elucidate the large-scale and synoptic drivers of frost events by exploring their spatial and temporal distribution with fine precision. The results identify the role and recurrence of specific circulation patterns, responsible for triggering spring frost events. By improving the identification of weather patterns responsible for frost events, this research sets the groundwork for developing targeted frost risk forecasts and management strategies, ensuring resilience in the face of climate change for the New Zealand wine industry. This methodology is transferable to other regions around the world and can be applied to a wide range of crops and agricultural systems. 

 

Stefano, P. et al. ; Am J Enol Vitic. 2022. doi:10.5344/ajev.2022.22011.

Van Leeuwen, C. et al. ; Journal of Wine Economics. 2016. doi:10.1017/jwe.2015.21.

Sturman, A. et al. ; Academic Journal. 2013. doi:10.1002/joc.3608.

Polo, I. et al. ; J. Climate. 2011. doi:10.1175/2011JCLI3622.1.