Incorporation of winter tree physiology in forecasting models of orchards’ bloom and yield under climate shifts
Research team:Dr. Tarin Paz-Kagan (PI) Prof. Maciej A. Zwieniecki (CI), and Dr. Or Sperling (CI)
MSc: Oren Lanterman
This study addresses the challenge of detecting almond spatiotemporal flowering phenology using multispectral satellite imagery across California's Central Valley. In response to the impact of climate change on deciduous crop yields, our research focused on exploring the potential of remote sensing to model accurately and predict flowering periods in almond orchards in California. We utilized Sentinel-2 imagery, enhanced vegetation indices, and in situ time-lapse camera data from 2019-2022 to develop a methodology for identifying peak bloom times. The study area encompassed approximately 30,000 orchards, precisely located using the Almond Industry Map. We employed the Enhanced Bloom Index (EBI) to quantify bloom intensity, while greenness measurement relied on the Normalized Difference Vegetation Index (NDVI). These indices were standardized and interpolated to daily resolution in the time series analysis. Our approach achieved a mean absolute error (MAE) of 1.9 days in detecting peak bloom, demonstrating the accuracy of our satellite-based phenological monitoring. Additionally, our study revealed significant spatial and temporal patterns in flowering phenology, underlining the influence of regional climatic conditions on crop development. In conclusion, our results highlight the potential of remote sensing and satellite imagery to accurately detect peak bloom in almond orchards and monitor phenological patterns on regional and field scales. This research has important implications for improving agricultural practices and assisting the almond industry in decision-making processes. By advancing phenological monitoring techniques, our study offers a scalable approach to managing perennial plant species in changing climates.
