Yiqun Xie

Streamflow prediction is essential for water resources management, flood forecasting, and climate resilience. Long short-term memory (LSTM) networks have advanced large-sample hydrology through cross-basin learning, but their recurrent architectures have limited ability to capture long-range temporal dependencies, particularly for medium-range forecasting. Meanwhile, existing applications of attention-based models in hydrology have largely relied on historical streamflow observations as model inputs and have been developed for individual basins, leaving their potential for regional rainfall–runoff prediction largely unexplored. We present HydroTFT, an attention-based rainfall–runoff model based on the Temporal Fusion Transformer that performs multi-horizon streamflow prediction using only antecedent meteorological forcings, static catchment attributes, and hydrologically informed engineered features. Trained jointly across 531 Catchment Attributes and Meteorology for Large-sample Studies basins, HydroTFT consistently outperforms entity-aware LSTM, LSTM, PatchTST, and iTransformer for 1 d, 7 d, and 14 d forecasting, while also exceeding both regionally and basin-wise calibrated process-based hydrologic models for 1 d prediction. Ablation experiments identify the attention mechanism as the primary contributor to the performance gains, and interpretability analyses show that the model learns physically meaningful representations of antecedent wetness and hydrologic memory. These results demonstrate that attention-based architectures provide an accurate, interpretable, and scalable framework for regional rainfall–runoff prediction across diverse hydro-climatic conditions.

1. Observational data indicate that the Arctic climate system has been undergoing a fundamental transition in recent years as air temperatures have increased twice as fast as the global average (In...

Agroecosystem, which heavily influenced by human actions and accounts for a quarter of global greenhouse gas emissions (GHGs), plays a crucial role in mitigating global climate change and securing environmental sustainability. However, we can't manage what we can't measure. Accurately quantifying the pools and fluxes in the carbon, nutrient, and water nexus of the agroecosystem is therefore essential for understanding the underlying drivers of GHG and developing effective mitigation strategies. Conventional approaches like soil sampling, process-based models, and black-box machine learning models are facing challenges such as data sparsity, high spatiotemporal heterogeneity, and complex subsurface biogeochemical and physical processes. Developing new trustworthy approaches such as AI-empowered models, will require the AI-ready benchmark dataset and outlined protocols, which unfortunately do not exist. In this work, we introduce a first-of-its-kind spatial-temporal agroecosystem GHG benchmark dataset that integrates physics-based model simulations from Ecosys and DayCent with real-world observations from eddy covariance flux towers and controlled-environment facilities. We evaluate the performance of various sequential deep learning models on carbon and nitrogen flux prediction, including LSTM-based models, temporal CNN-based model, and Transformer-based models. Furthermore, we explored transfer learning to leverage simulated data to improve the generalization of deep learning models on real-world observations. Our benchmark dataset and evaluation framework contribute to the development of more accurate and scalable AI-driven agroecosystem models, advancing our understanding of ecosystem-climate interactions.

Carbon monitoring is needed to track the exchange of carbon between the land and atmosphere. However, comprehensive monitoring of terrestrial ecosystems is challenging, partially because of the fine-scale spatial heterogeneity of land cover and vegetation structure. While forest ecosystems are relatively well studied, knowledge of trees outside forests (TOFs) is limited, even though TOF contribute substantially to many ecosystem services. In this study, we leveraged advances in remote sensing and modeling to quantify TOF and their contribution to the 2011–2023 Maryland state carbon budget. Specifically, the Ecosystem Demography model was initialized with 1 m airborne optical and LiDAR data. These datasets were combined with 30 m National Land Cover Database maps to identify and monitor TOF statewide. In the base year, TOF contributed to 22.86% (2795.59 km ² ) of the total tree cover, 14.70% (15.98 Tg C) of the total aboveground live tree carbon stocks, and 22.56% (0.39 Tg C yr ⁻¹ ) of the corresponding carbon fluxes in the state, with variation by land cover class. TOF on developed and planted/cultivated lands contributed approximately 90% of the total TOF tree cover and carbon stocks, as well as over 96% of the total TOF carbon fluxes. Most TOF occurred as polygons smaller than 1 km ² and were within 600 m distance from forests. From 2011–2023, persistent TOF carbon stocks were estimated to increase by 5.15 Tg C, representing 23.71% of the state’s total increase in tree carbon stocks over the interval, with substantial interannual variability in carbon fluxes. These results suggest TOF are an important contribution to statewide tree cover and produce carbon stocks and fluxes that should be highlighted in carbon budgets for effective monitoring and assessment. Moving forward, efforts are needed to expand the availability and application of high-resolution remote sensing data and models needed for TOF quantification in carbon budgets.

Environmental modeling faces critical challenges in predicting ecosystem dynamics across unmonitored regions due to limited and geographically imbalanced observation data. This challenge is compounded by spatial heterogeneity, causing models to learn spurious patterns that fit only local data. Unlike conventional domain generalization, environmental modeling must preserve invariant physical relationships and temporal coherence during augmentation. In this paper, we introduce Generalizable Representation Enhancement via Auxiliary Transformations (GREAT), a framework that effectively augments available datasets to improve predictions in completely unseen regions. GREAT guides the augmentation process to ensure that the original governing processes can be recovered from the augmented data, and the inclusion of the augmented data leads to improved model generalization. Specifically, GREAT learns transformation functions at multiple layers of neural networks to augment both raw environmental features and temporal influence. They are refined through a novel bi-level training process that constrains augmented data to preserve key patterns of the original source data. We demonstrate GREAT's effectiveness on stream temperature prediction across six ecologically diverse watersheds in the eastern U.S., each containing multiple stream segments. Experimental results show that GREAT significantly outperforms existing methods in zero-shot scenarios. This work provides a practical solution for environmental applications where comprehensive monitoring is infeasible.