Abstract. Recently, deep learning (DL) has emerged as a revolutionary and versatile tool transforming industry applications and generating new and improved capabilities for scientific discovery and model building. The adoption of DL in hydrology has so far been gradual, but the field is now ripe for breakthroughs. This paper suggests that DL-based methods can open up a complementary avenue toward knowledge discovery in hydrologic sciences. In the new avenue, machine-learning algorithms present competing hypotheses that are consistent with data. Interrogative methods are then invoked to interpret DL models for scientists to further evaluate. However, hydrology presents many challenges for DL methods, such as data limitations, heterogeneity and co-evolution, and the general inexperience of the hydrologic field with DL. The roadmap toward DL-powered scientific advances will require the coordinated effort from a large community involving scientists and citizens. Integrating process-based models with DL models will help alleviate data limitations. The sharing of data and baseline models will improve the efficiency of the community as a whole. Open competitions could serve as the organizing events to greatly propel growth and nurture data science education in hydrology, which demands a grassroots collaboration. The area of hydrologic DL presents numerous research opportunities that could, in turn, stimulate advances in machine learning as well.

The behaviors and skills of models in many geosciences (e.g., hydrology and ecosystem sciences) strongly depend on spatially-varying parameters that need calibration. A well-calibrated model can reasonably propagate information from observations to unobserved variables via model physics, but traditional calibration is highly inefficient and results in non-unique solutions. Here we propose a novel differentiable parameter learning (dPL) framework that efficiently learns a global mapping between inputs (and optionally responses) and parameters. Crucially, dPL exhibits beneficial scaling curves not previously demonstrated to geoscientists: as training data increases, dPL achieves better performance, more physical coherence, and better generalizability (across space and uncalibrated variables), all with orders-of-magnitude lower computational cost. We demonstrate examples that learned from soil moisture and streamflow, where dPL drastically outperformed existing evolutionary and regionalization methods, or required only ~12.5% of the training data to achieve similar performance. The generic scheme promotes the integration of deep learning and process-based models, without mandating reimplementation.

Expression of the acute-phase response genes, such as that for alpha-i acid glycoprotein (AGP), involves both positive and negative transcription factors. A positive transcription factor, AGP/EBP, and a negative transcription factor, factor B, have been identified as the two most important factors responsible for the induction of the AGP gene. In this paper we report the purification, characterization, and identification of a B-motif-binding factor from the mouse hepatoma cell line 129p. The purified factor has been identified as nucleolin by amino acid sequence analysis. Biochemical and functional studies further established that nucleolin is a transcription repressor for regulation of AGP and possibly other acute-phase response genes. Thus, in addition to the many known functions of nucleolin, such as rRNA transcription, processing, ribosome biogenesis, and the shuttling of proteins between the cytoplasmic and nuclear compartments, it may also function as a transcriptional repressor.The initiation of transcription in eukaryotes is an intricately controlled process. Short sequence motifs in the promoter regions of genes interact in a specific manner with DNAbinding transcription factors. These bound factors interact with general transcription factors and thereby result in gene transcription. Not only transcriptional activators but also repressors are important in the controlled regulation of gene expression. For a given gene, the combinations of cis elements and the trans-acting factors are major determinants of transcriptional activity. Protein-protein interactions and posttranslational modifications are important for regulating the activities of these factors. An array of transcriptional activators and repressors have been identified and characterized (9,12,14,15,18,24,35,37).Tissue injury and infection produce significant alterations of the host metabolic and immune homeostasis (19). It has become increasingly clear that many of these changes result from a complex cascade of mononuclear phagocyte-derived endogenous mediators, in particular, cytokines. Injection of purified lipopolysaccharide (LPS) into laboratory animals leads to the development of many biological activities with similarities to those that follow tissue injury and infection. These can range from an acute-phase response to shock with lethal outcome. The well-studied biological activities of LPSinduced liver gene expression are mediated by multiple cytokines, including interleukins 1, 6, and 11, leukemia inhibitory factor, and tumor necrosis factor alpha (13,19,30,31). We used LPS-induced transcription of the alpha-1 acid glycoprotein (AGP) gene as a model for studying the regulation of gene expression during the acute-phase response (9, 24). Transcription of the AGP gene in response to LPS treatment is regulated by both a positive factor, AGP/EBP (C/EBP-P), and a negative factor, factor B (24). During the acute-phase * Corresponding author. Mailing address: Institute of Molecular Medicine, College of Medicine, National Taiwan University, Taip...

Dissolved oxygen (DO) reflects river metabolic pulses and is an essential water quality measure. Our capabilities of forecasting DO however remain elusive. Water quality data, specifically DO data here, often have large gaps and sparse areal and temporal coverage. Earth surface and hydrometeorology data, on the other hand, have become largely available. Here we ask: can a Long Short-Term Memory (LSTM) model learn about river DO dynamics from sparse DO and intensive (daily) hydrometeorology data? We used CAMELS-chem, a new data set with DO concentrations from 236 minimally disturbed watersheds across the U.S. The model generally learns the theory of DO solubility and captures its decreasing trend with increasing water temperature. It exhibits the potential of predicting DO in “chemically ungauged basins”, defined as basins without any measurements of DO and broadly water quality in general. The model however misses some DO peaks and troughs when in-stream biogeochemical processes become important. Surprisingly, the model does not perform better where more data are available. Instead, it performs better in basins with low variations of streamflow and DO, high runoff-ratio (>0.45), and winter precipitation peaks. Results here suggest that more data collections at DO peaks and troughs and in sparsely monitored areas are essential to overcome the issue of data scarcity, an outstanding challenge in the water quality community.

No abstract