2021 Volume 90 Issue 2 Pages 181-189
The number of leaves in ‘Delaware’ grape is an excellent indicator to determine the optimal timing for gibberellic acid (GA3) application to fruit clusters to induce seedless berries. To predict the optimal timing for GA3 application to ‘Delaware’ grape, we aimed to develop a method to estimate the accumulated leaf number (LN) on the second shoot (which grows from the second dormant bud from the end of the fruiting mother shoot) from the air temperature. This was undertaken using leafing data from a vineyard in Osaka, Japan from 1987 to 2019. Base temperature (Tb, °C), ranging from −2.5 to 12.5°C with step increases of 2.5°C, was estimated using the coefficient of regression determination (R2) between LN and effective accumulative temperature (EAT, °C·day) and two types of root mean square error (RMSE) (RMSE of LN and RMSE of the day when LN reached 10.0 leaves). The LN showed a good fit in linear regression with EAT at every Tb (−2.5 to 12.5°C); in particular, a Tb of 7.5°C yielded the highest R2 and the lowest RMSE. From these results, we determined that the Tb for leaf emergence of ‘Delaware’ grape was 7.5°C. We then compared the estimation accuracy in predicting leaf emergence using EAT (Tb of 7.5°C) (proposed method) and constant leaf emergence rate per day of 0.36 leaves·day−1 (conventional method). As the RMSE when using the proposed method was lower than the RMSE when using the conventional method, the proposed method was more suitable to predict of optimal timing for GA3 application than the conventional method. The findings of this study could be used to predict the optimal timing for GA3 application with high accuracy using LN on the measurement date and forecast values for temperature. Furthermore, the coefficient of the developed equation suggests that the ‘Delaware’ grape develops 4.0 leaves per EAT of 100°C·day with a Tb of 7.5°C.
