Development of a new cold hardiness prediction model for grapevine using phased integration of acclimation and deacclimation responses

Abstract

Cold injury in plants limits distribution of perennial agricultural crops, though replacement of plants and other management practices allow for some damage to be tolerated. Grapes (Vitis spp.) represent a crop that is grown in many marginal areas due to its profitability. However, monitoring of cold hardiness or damage for deployment of cold damage mitigation and management practices is a laborious process. To that effect, a model was published in 2014 to predict cold hardiness using data from Washington state in the US. Although that model works well regionally, it has been found to not predict adequately in other regions. In part, it is known that its description of some dormancy aspects as they relate to cold hardiness are not accurate. Therefore, the objective of this work was to develop a new model that incorporate previous research on cold hardiness dynamics, such as to increase realism. Cold hardiness data from V. labruscana 'Concord', and V. vinifera 'Cabernet Sauvignon' and 'Riesling' from Geneva, NY, USA were used. Data were separated in calibration (~2/3) and validation (~1/3) datasets. The proposed model uses 3 functions to describe acclimation, and 2 functions to describe deacclimation, with a total of 9 optimized parameters. A shared response to chill between acclimation and deacclimation provides a phased integration where acclimation responses decrease over the course of winter and are overcome by deacclimation. The new model outperforms the currently available model, reducing RMSE by up to 37% depending on cultivar. The new model also tends to slightly underpredict cold hardiness, as opposed to the overprediction from the current model. The underprediction can be preferred as damage mitigation methods may be deployed prior to damage occurring with greater safety. Some optimized parameters were shared between cultivars, suggesting conserved physiological aspects of the process were captured