Use of water potential measurements for assessing water stress in Vitis vinifera L. cv. Tempranillo grown in Southern Oregon

Mestrado Vinifera EuroMaster - Instituto Superior de AgronomiaA study was conducted to investigate the practical use of four methods for measuring plant water potential (!) of Vitis vinifera L. cv. Tempranillo (syn. Tinta Roriz, Aragónez) under field conditions within the Southern Oregon American Viticultural Area. Predawn leaf water potential (!pd), midday leaf water potential (!l-md), midday stem water potential (!s-md) and early morning stem water potential measured between 07:00h and 08:00h solar time (!s-em) were each measured on vines before and after the initiation of irrigation. Measurements were conducted on the same vines and on the same dates both before and after veraison. Irrigation was applied based on an estimation of vineyard evapotranspiration (ETc) and consisted of four treatments: 70% of ETc continuously (70-70), 35% of ETc continuously (35-35), 70% ETc before veraison and 35% ETc after veraison (70-35) and 35% before veraison and 75% ETc after veraison (35- 70). Irrigation was initiated based on !l-md. When assessing vineyard variability before the initiation of irrigation, with !pd ranging between -0.05 and -0.53, all four measurement types were able to distinguish between high vigor (HV) and low vigor (LV) zones and able to categorize vines under previously established thresholds. However, !l-md showed a tendency to underestimate vine water status at levels greater than -0.9 Mpa. When comparing !s-em, !s-md and !l-md, early morning measurements showed significant differences between irrigation treatments on 3 of 4 measurement dates while midday measurements were able to distinguish significant differences on only 1 of 4 dates. !s-em measurements were generally able to distinguish differences between vine water status even when morning cloud cover was present. Linear regression analysis of !pd versus !s-em, !s-md and !l-md at one site resulted in significant r2 values of 0.62, 0.69 and 0.58, respectively. Linear regression analysis of !s-md versus !l-md using data from both sites resulted in r2 values of 0.88. Overall, !s-em seemed to provide a better option for differentiating plant water status of Tempranillo grapevines in Southern Oregon compared to !s-md and !l-md within the range of water deficit levels studied here. !s-em was able to show differences between the water status of vines before the initiation of irrigation as well as differentiate between irrigation treatments later in the year. However, questions remain about the environmental and physiological factors that might impact the results of this method before water deficit threshold levels can be clearly define