Zero-Shot Transfer Capabilities of the Sundial Foundation Model for Leaf Area Index Forecasting
This demonstrates that pretrained time series foundation models can serve as effective plug-and-play forecasters for agricultural and environmental applications, offering a potential solution for remote-sensing time series prediction without task-specific tuning.
This work tackled the problem of forecasting Leaf Area Index (LAI) in agricultural monitoring by evaluating the zero-shot capabilities of the Sundial foundation model, finding that it outperforms a fully trained LSTM when using a sufficiently long input context window covering more than one or two seasonal cycles.
This work investigates the zero-shot forecasting capability of time series foundation models for Leaf Area Index (LAI) forecasting in agricultural monitoring. Using the HiQ dataset (U.S., 2000-2022), we systematically compare statistical baselines, a fully supervised LSTM, and the Sundial foundation model under multiple evaluation protocols. We find that Sundial, in the zero-shot setting, can outperform a fully trained LSTM provided that the input context window is sufficiently long-specifically, when covering more than one or two full seasonal cycles. We show that a general-purpose foundation model can surpass specialized supervised models on remote-sensing time series prediction without any task-specific tuning. These results highlight the strong potential of pretrained time series foundation models to serve as effective plug-and-play forecasters in agricultural and environmental applications.