Potential of Time-Series Sentinel 2 Data for Monitoring Avocado Crop Phenology

Abstract

The ability to accurately and systematically monitor avocado crop phenology offers significant benefits for the optimization of farm management activities, improvement of crop productivity, yield estimation, and evaluation crops’ resilience to extreme weather conditions and future climate change. In this study, Sentinel-2-derived enhanced vegetation indices (EVIs) from 2017 to 2021 were used to retrieve canopy reflectance information that coincided with crop phenological stages, such as flowering (F), vegetative growth (V), fruit maturity (M), and harvest (H), in commercial avocado orchards in Bundaberg, Queensland and Renmark, South Australia. Tukey’s honestly significant difference (Tukey-HSD) test after one-way analysis of variance (ANOVA) with EVI metrics (EVImean and EVIslope) showed statistically significant differences between the four phenological stages. From a Pearson correlation analysis, a distinctive seasonal trend of EVIs was observed (R = 0.68 to 0.95 for Bundaberg and R = 0.8 to 0.96 for Renmark) in all 5 years, with the peak EVIs being observed at the M stage and the trough being observed at the F stage. However, a Tukey-HSD test showed significant variability in mean EVI values between seasons for both the Bundaberg and Renmark farms. The variability of the mean EVIs between the two farms was also evident with a p-value

Plant phenotyping relevance

Sentinel-2時系列データからアボカド樹冠反射を取得し、開花・生育・成熟・収穫という植物フェノロジー状態を識別・評価する手法が研究の中心である。

Code and data availability

The supplied blocks describe Sentinel-2 EVI time-series analysis of avocado phenology, but no public phenotype dataset, imagery, code, or supplement with authors' deposit URL is mentioned. Phenological observations came from growers, Sentinel-2 data were processed via GEE, and no code or data availability statement is.

No evidence-backed public reproduction asset is currently recorded.