Downscaling is a critical step to bridge the gap between large-scale climate information and local-scale impact assessment. This study presents a novel deep learning approach: Super Resolution Deep Residual Network (SRDRN) for downscaling daily precipitation and temperature. This approach was constructed based on an advanced deep convolutional neural network with residual blocks and batch normalizations. The data augmentation technique was utilized to address overfitting that is due to highly imbalanced precipitation and nonprecipitation days and sparse precipitation extremes. Synthetic experiments were designed to downscale daily maximum/minimum temperature and precipitation data from coarse resolutions (25, 50, and 100 km) to a high resolution (4 km). The results showed that, during the validation period, the SRDRN approach not only captured the spatial and temporal patterns remarkably well, but also reproduced both precipitation and temperature extremes in different locations and time at the local scale. Through transfer learning, the trained SRDRN model in one region was directly applied to downscale precipitation in another region with a different environment, and the results showed notable improvement compared to classic statistical downscaling methods. The outstanding performance of the SRDRN approach stemmed from its ability to fully extract spatial features without suffering from degradation and overfitting issues due to the incorporations of residual blocks, batch normalizations, and data augmentations. The SRDRN approach is thus a powerful tool for downscaling daily precipitation and temperature and can potentially be leveraged to downscale any hydrologic, climate, and earth system data.

Most environmental data come from a minority of well-monitored sites. An ongoing challenge in the environmental sciences is transferring knowledge from monitored sites to unmonitored sites. Here, we demonstrate a novel transfer-learning framework that accurately predicts depth-specific temperature in unmonitored lakes (targets) by borrowing models from well-monitored lakes (sources). This method, meta-transfer learning (MTL), builds a meta-learning model to predict transfer performance from candidate source models to targets using lake attributes and candidates’ past performance. We constructed source models at 145 well-monitored lakes using calibrated process-based (PB) modeling and a recently developed approach called process-guided deep learning (PGDL). We applied MTL to either PB or PGDL source models (PB-MTL or PGDL-MTL, respectively) to predict temperatures in 305 target lakes treated as unmonitored in the Upper Midwestern United States. We show significantly improved performance relative to the uncalibrated PB General Lake Model, where the median root mean squared error (RMSE) for the target lakes is 2.52°C. PB-MTL yielded a median RMSE of 2.43°C; PGDL-MTL yielded 2.16°C; and a PGDL-MTL ensemble of nine sources per target yielded 1.88°C. For sparsely monitored target lakes, PGDL-MTL often outperformed PGDL models trained on the target lakes themselves. Differences in maximum depth between the source and target were consistently the most important predictors. Our approach readily scales to thousands of lakes in the Midwestern United States, demonstrating that MTL with meaningful predictor variables and high-quality source models is a promising approach for many kinds of unmonitored systems and environmental variables.

Explaining deep learning models operating on time series data is crucial in various applications of interest which require interpretable and transparent insights from time series signals. In this work, we investigate this problem from an information theoretic perspective and show that most existing measures of explainability may suffer from trivial solutions and distributional shift issues. To address these issues, we introduce a simple yet practical objective function for time series explainable learning. The design of the objective function builds upon the principle of information bottleneck (IB), and modifies the IB objective function to avoid trivial solutions and distributional shift issues. We further present TimeX++, a novel explanation framework that leverages a parametric network to produce explanation-embedded instances that are both in-distributed and label-preserving. We evaluate TimeX++ on both synthetic and real-world datasets comparing its performance against leading baselines, and validate its practical efficacy through case studies in a real-world environmental application. Quantitative and qualitative evaluations show that TimeX++ outperforms baselines across all datasets, demonstrating a substantial improvement in explanation quality for time series data. The source code is available at https://github.com/zichuan-liu/TimeXplusplus.