Freeze events Across the Midwestern and Northeastern United States: Trend and Teleconnection Analysis

Keywords: 
Abstract Type: Poster Abstract

Abstract

Freeze events, especially those occurring in the late spring, are among the most impactful weather hazards for fruit production. The extent of fruit crop damage resulting from cold temperature extremes is mainly determined by the minimum temperature and the duration of freezing conditions, both of which are sensitive to changes in daily minimum air temperature and its associated variability. The aim of this study is to investigate historical freeze variability across the Midwestern and Northeastern regions of the United States using a high-resolution (800 m), homogenized, gridded historical temperature dataset (TopoWx). Our goal is to improve our understanding of freeze risk by quantifying the temporal variability and trends in several freeze-related metrics. These include traditional measures such as the date of the last spring freeze and first fall freeze and the time interval separating them (i.e., the frost-free season), as well as novel metrics, such as the date of the penultimate spring freeze, the length of the time interval separating the last spring freeze and penultimate spring freeze, as well as the accumulation of heat during this interval. Trends are estimated using non-parametric methods (Theil-Sen) that account for temporal autocorrelation and field significance. Self-Organizing Maps (SOM) are constructed to analyze teleconnections between freeze events and synoptic circulation patterns enhances our ability to understand and predict the timing and locations of these freeze events with greater accuracy.