What question did this study set out to answer?

The central aim is to improve mapping accuracy of tea plantations using AI and satellite data in challenging environments.

February 12, 2026Open Access

AI-Driven Precision Mapping of Tea Plantations Using AlphaEarth Foundations: A Scalable Solution for Smart Agricultural Monitoring

Key Points

The central aim is to improve mapping accuracy of tea plantations using AI and satellite data in challenging environments.
Utilized Google AlphaEarth Foundations satellite embeddings for feature extraction.
Compared various classification algorithms to identify the most effective method for mapping.
Conducted 12 classification scenarios with robust sampling strategies.
AEF-augmented scenario achieved overall accuracy of 92.69% and Kappa of 0.90 in Rizhao.
Producer's accuracy reached 97.47%, minimizing omission errors effectively.
Model exhibited high generalizability to the unseen Qingdao region without requiring retraining.

Abstract

Accurate mapping of tea plantations in fragmented, mountainous landscapes faces challenges from spectral confusion, cloud-induced data gaps, and limited model transferability. To address these issues, this study proposes a data-driven approach leveraging 64-dimensional Google AlphaEarth Foundations (AEF) satellite embeddings as core predictive features, integrated with Sentinel-2 spectral, textural, and topographic variables. Prior to feature optimization, comparative experiments confirmed that Random Forest outperformed Gradient Boosting Trees, Classification and Regression Trees, and Support Vector Machines in stability and accuracy, serving as the core classifier. Leveraging a robust sampling strategy, this study evaluated 12 classification scenarios. Results showed that the AEF-augmented scenario achieved the best performance in Rizhao (Overall Accuracy 92.69%, Kappa 0.90), with a high Producer’s Accuracy of 97.47% that effectively minimized omission errors. SHapley Additive exPlanations (SHAP) analysis revealed the model’s physically interpretable logic: utilizing embeddings as “exclusion filters” to separate tea from non-target classes by encoding latent phenological patterns, while relying on original spectral bands to capture canopy biological signals. Crucially, the model demonstrated exceptional generalizability when transferred to the unseen Qingdao region without retraining. This study validates AEF embeddings as a robust, scalable feature representation for regional crop monitoring in label-scarce and heterogeneous environments, offering a transferable data foundation for precise agricultural inventory and sustainable development planning.

KI fragen

Bookmark

View Full Paper