The spatial variation of minimum near-surface temperature in complex terrain: Marlborough vineyard region, New Zealand
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
The economic impact of frost on agriculture remains a global problem. It is a particular concern for the New Zealand wine industry, where the consequences of an unexpected spring frost can be disastrous. Marlborough is located in the north-eastern corner of the South Island and is the largest grape-growing region in New Zealand. The region is surrounded by complex mountainous terrain that gives rise to extremes of climate, particularly large spatial variations of minimum temperature and the frequent occurrence of spring frost. The high spatial variation of near-surface minimum temperature can lead to under-preparedness among grape growers who rely on accurate frost forecasts as part of their frost mitigation systems. Field campaigns of the 1980’s and 90’s extended the understanding of the physical meteorological processes that affect cooling in complex terrain. More recent modelling efforts continue to refine this knowledge, although much less attention is given to the effects of different cooling processes on near-surface temperature. Agricultural developments in areas of complex terrain would benefit from an increased understanding of the meteorological processes that govern near-surface cooling, as this will help with the local prediction of frost. The spatial variation of near-surface minimum temperatures is first explored by identifying relationships with synoptic weather patterns using the Kidson (2000) synoptic classification scheme. Analysis revealed that Kidson types associated with the largest daily variations in near-surface minimum temperature (T, TNW and H) are not always associated with the occurrence of frost. Frost is more likely to occur during the cooler airflows of Kidson type HW, HNW and SW, or during the settled anticyclonic conditions that follow cooler airflows. The relationship between the spatial variation of near-surface minimum temperature and regional airflow patterns is explored using numerical weather prediction (NWP) modelling. Results indicated that a high σ Tmin around the region is a product of interaction between the region’s complex terrain and ambient meteorology, and it could occur in both settled weather and more dynamic synoptic conditions. A high regional σ Tmin during light ridge top winds could occur as a function of a location’s relative susceptibility to ventilation from thermally-induced drainage winds, and it may also occur as a result of the simultaneous ventilation and stagnation of near-surface air layers as synoptic wind interacts with local topography. The influence of the vertical structure of the nocturnal boundary layer (NBL) on nearsurface minimum temperature was investigated with the University of Canterbury Sonic Detection And Ranging (SODAR). Measurements confirmed the formation of low-level jets (LLJ’s) in the Awatere and Wairau Valleys during settled weather conditions, and that shear-induced turbulence beneath the jets was sufficient to mix warmer air to the surface and increase local temperatures. The process is sufficient to reduce frost risk to some of the region’s upper valleys during clear settled weather. In stronger ridge top winds development of the LLJ’s can be suppressed or eliminated and this was found to reduce shear-induced turbulence near the surface, allowing increased near-surface cooling. While results from this study are of greatest value to the prediction of near-surface minimum temperature and frost in Marlborough, the results could be applied to improved prediction of near-surface minimum temperature in complex terrain around the world. Further research could be directed toward the interaction of synoptic winds with thermally-induced airflows, as the transition zone between these wind systems is believed to govern the temporal and spatial evolution of near-surface stagnation, and this is related to episodes of strong near-surface cooling.