Time-series foundation modeling enables accurate lake ecosystem forecasting
- 1Kyoto University
Abstract
Accurate ecological forecasting is increasingly essential for understanding and managing ecosystem responses to climate variability and anthropogenic pressures; however, prediction remains difficult in lakes because key biological variables, such as phytoplankton biomass, exhibit nonlinear dynamics, observational noise, data sparsity, and non-stationarity. This study aimed to test whether large pre-trained time-series foundation models can improve forecasts of lake phytoplankton dynamics, focusing on chlorophyll-a (Chl.a) and water-quality variables in two ecologically contrasting Japanese lakes, the deep-stratified Lake Biwa and the shallow nutrient-rich Lake Kasumigaura. Using extensive monthly monitoring records spanning up to 30 years, we benchmarked two time-series foundation models - Transformer-based Chronos-T5 and probabilistic Lag-Llama - against 15 statistical (AR, ARIMA, SARIMA, Prophet), machine learning (Random Forest, XGBoost, KNN, SVR), and deep learning (LSTM, CNN, TCN, and SSA-hybrid) approaches. Chronos-T5 achieved superior reproduction accuracy and stability across diverse environmental variables and outperformed all other tested models. This performance stemmed from its ability to capture long-term dependencies and complex temporal patterns through large-scale pre-training. We further identified an optimal training window of approximately 14 years, balancing data sufficiency with regime stability, beyond which model accuracy diminished owing to ecological regime shifts. Our findings highlight the transformative potential of time-series foundation models for ecological forecasting, providing a scalable, data-driven framework that can underpin robust early warning systems and adaptive management strategies for aquatic ecosystems worldwide.
Related
- Accurate data as the foundation of equitable cancer control: Lessons from global cancer statistics 2024 and the complexity of ovarian cancer
- LoGoPPI enables fast and accurate protein protein interaction mapping at scale
- SpaMOAL is a deep learning method that enables accurate spatial domain identification from multi-omics data
- Accurate Hippocampal Electric Field Modeling for Electroconvulsive Therapy: Numerical Accuracy and Convergence
- Neural noise enables accurate internal simulation of rare events
BibTeX
@misc{matsushita2026timeseries,
title = {{Time-series foundation modeling enables accurate lake ecosystem forecasting}},
author = {A. Matsushita and H. Doi},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.13.751332},
url = {https://www.biorxiv.org/content/10.64898/2026.09.13.751332v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Time-series foundation modeling enables accurate lake ecosystem forecasting
AU - A. Matsushita
AU - H. Doi
PY - 2026
DA - 2026/09/13
PB - bioRxiv
DO - 10.64898/2026.09.13.751332
UR - https://www.biorxiv.org/content/10.64898/2026.09.13.751332v1
AB - Accurate ecological forecasting is increasingly essential for understanding and managing ecosystem responses to climate variability and anthropogenic pressures; however, prediction remains difficult in lakes because key biological variables, such as phytoplankton biomass, exhibit nonlinear dynamics, observational noise, data sparsity, and non-stationarity. This study aimed to test whether large pre-trained time-series foundation models can improve forecasts of lake phytoplankton dynamics, focusing on chlorophyll-a (Chl.a) and water-quality variables in two ecologically contrasting Japanese lakes, the deep-stratified Lake Biwa and the shallow nutrient-rich Lake Kasumigaura. Using extensive monthly monitoring records spanning up to 30 years, we benchmarked two time-series foundation models - Transformer-based Chronos-T5 and probabilistic Lag-Llama - against 15 statistical (AR, ARIMA, SARIMA, Prophet), machine learning (Random Forest, XGBoost, KNN, SVR), and deep learning (LSTM, CNN, TCN, and SSA-hybrid) approaches. Chronos-T5 achieved superior reproduction accuracy and stability across diverse environmental variables and outperformed all other tested models. This performance stemmed from its ability to capture long-term dependencies and complex temporal patterns through large-scale pre-training. We further identified an optimal training window of approximately 14 years, balancing data sufficiency with regime stability, beyond which model accuracy diminished owing to ecological regime shifts. Our findings highlight the transformative potential of time-series foundation models for ecological forecasting, providing a scalable, data-driven framework that can underpin robust early warning systems and adaptive management strategies for aquatic ecosystems worldwide.
ER -