Models for the beginning of sour cherry blossom

Area of Science:

• Phenology
• Climate Change Impact Assessment
• Agricultural Meteorology

Background:

• Accurate prediction of plant phenological events like sour cherry blossom is vital for agriculture.
• Traditional growing degree day (GDD) models have limitations in predicting phenological shifts due to climate change.
• Developing robust models is essential for understanding and adapting to changing climate conditions.

Purpose of the Study:

• To compare seven different model approaches for calculating sour cherry blossom onset in Rhineland-Palatinate, Germany.
• To evaluate the performance of pure forcing models versus combined chilling-forcing (CF) models.
• To develop and validate models capable of projecting phenological shifts under climate change scenarios.

Main Methods:

• Compared three pure forcing models (GDD-based) and four combined CF models.
• Optimized model parameters including temperature accumulation start dates, base temperatures, and forcing requirements.
• Incorporated daylength (DL) and different chilling metrics (chilling hours vs. chill portions) into CF models.
• Performed internal and external validation across different European and North American regions.

Main Results:

• Pure GDD models with early start dates showed limitations; M2DL (with daylength) improved performance over M2.
• Combined CF models without daylength (M3, M4) failed validation.
• CF models incorporating daylength (M3DL, M4DL) demonstrated meaningful parameter estimations and significantly reduced prediction errors.
• Models M3DL and M4DL showed successful external validation in diverse geographical locations.

Conclusions:

• Combined chilling-forcing models that include daylength are superior to pure GDD models for predicting cherry blossom onset.
• These advanced models provide reliable parameter estimations and accurate predictions, even across different regions.
• The validated models are suitable for projecting phenological shifts in sour cherry due to climate change.