An official website of the United States government
Abstract. Recently, deep learning (DL) has emerged as a revolutionary andversatile tool transforming industry applications and generating new andimproved capabilities for scientific discovery and model building. Theadoption of DL in hydrology has so far been gradual, but the field is nowripe for breakthroughs. This paper suggests that DL-based methods can open up acomplementary avenue toward knowledge discovery in hydrologic sciences. Inthe new avenue, machine-learning algorithms present competing hypotheses thatare consistent with data. Interrogative methods are then invoked to interpretDL models for scientists to further evaluate. However, hydrology presentsmany challenges for DL methods, such as data limitations, heterogeneityand co-evolution, and the general inexperience of the hydrologic field withDL. The roadmap toward DL-powered scientific advances will require thecoordinated effort from a large community involving scientists and citizens.Integrating process-based models with DL models will help alleviate datalimitations. The sharing of data and baseline models will improve theefficiency of the community as a whole. Open competitions could serve as theorganizing events to greatly propel growth and nurture data science educationin hydrology, which demands a grassroots collaboration. The area ofhydrologic DL presents numerous research opportunities that could, in turn,stimulate advances in machine learning as well.
more »
« less
A trans-disciplinary review of deep learning research and its relevance for water resources scientists
Deep learning (DL), a new‐generation of artificial neural network research, has transformed industries, daily lives and various scientific disciplines in recent years. DL represents significant progress in the ability of neural networks to automatically engineer problem‐relevant features and capture highly complex data distributions. I argue that DL can help address several major new and old challenges facing research in water sciences such as inter‐disciplinarity, data discoverability, hydrologic scaling, equifinality, and needs for parameter regionalization. This review paper is intended to provide water resources scientists and hydrologists in particular with a simple technical overview, trans‐disciplinary progress update, and a source of inspiration about the relevance of DL to water. The review reveals that various physical and geoscientific disciplines have utilized DL to address data challenges, improve efficiency, and gain scientific insights. DL is especially suited for information extraction from image‐like data and sequential data. Techniques and experiences presented in other disciplines are of high relevance to water research. Meanwhile, less noticed is that DL may also serve as a scientific exploratory tool. A new area termed “AI neuroscience,” where scientists interpret the decision process of deep networks and derive insights, has been born. This budding sub‐discipline has demonstrated methods including correlation‐based analysis, inversion of network‐extracted features, reduced‐order approximations by interpretable models, and attribution of network decisions to inputs. Moreover, DL can also use data to condition neurons that mimic problem‐specific fundamental organizing units, thus revealing emergent behaviors of these units. Vast opportunities exist for DL to propel advances in water sciences.
more »
« less
- Award ID(s):
- 1832294
- PAR ID:
- 10077611
- Date Published:
- Journal Name:
- Water Resources Research
- ISSN:
- 0043-1397
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Long-term environmental research networks are one approach to advancing local, regional, and global environmental science and education. A remarkable number and wide variety of environmental research networks operate around the world today. These are diverse in funding, infrastructure, motivating questions, scientific strengths, and the sciences that birthed and maintain the networks. Some networks have individual sites that were selected because they had produced invaluable long-term data, while other networks have new sites selected to span ecological gradients. However, all long-term environmental networks share two challenges. Networks must keep pace with scientific advances and interact with both the scientific community and society at large. If networks fall short of successfully addressing these challenges, they risk becoming irrelevant. The objective of this paper is to assert that the biogeosciences offer environmental research networks a number of opportunities to expand scientific impact and public engagement. We explore some of these opportunities with four networks: the International Long-Term Ecological Research Network programs (ILTERs), critical zone observatories (CZOs), Earth and ecological observatory networks (EONs), and the FLUXNET program of eddy flux sites. While these networks were founded and expanded by interdisciplinary scientists, the preponderance of expertise and funding has gravitated activities of ILTERs and EONs toward ecology and biology, CZOs toward the Earth sciences and geology, and FLUXNET toward ecophysiology and micrometeorology. Our point is not to homogenize networks, nor to diminish disciplinary science. Rather, we argue that by more fully incorporating the integration of biology and geology in long-term environmental research networks, scientists can better leverage network assets, keep pace with the ever-changing science of the environment, and engage with larger scientific and public audiences.more » « less
-
Abstract Advancements in computing power have recently made it possible to utilize machine learning and deep learning to push scientific computing forward in a range of disciplines, such as fluid mechanics, solid mechanics, materials science, etc. The incorporation of neural networks is particularly crucial in this hybridization process. Due to their intrinsic architecture, conventional neural networks cannot be successfully trained and scoped when data are sparse, which is the case in many scientific and engineering domains. Nonetheless, neural networks provide a solid foundation to respect physics-driven or knowledge-based constraints during training. Generally speaking, there are three distinct neural network frameworks to enforce the underlying physics: (i) physics-guided neural networks (PgNNs), (ii) physics-informed neural networks (PiNNs), and (iii) physics-encoded neural networks (PeNNs). These methods provide distinct advantages for accelerating the numerical modeling of complex multiscale multiphysics phenomena. In addition, the recent developments in neural operators (NOs) add another dimension to these new simulation paradigms, especially when the real-time prediction of complex multiphysics systems is required. All these models also come with their own unique drawbacks and limitations that call for further fundamental research. This study aims to present a review of the four neural network frameworks (i.e., PgNNs, PiNNs, PeNNs, and NOs) used in scientific computing research. The state-of-the-art architectures and their applications are reviewed, limitations are discussed, and future research opportunities are presented in terms of improving algorithms, considering causalities, expanding applications, and coupling scientific and deep learning solvers.more » « less
-
Abstract. Long-term environmental research networks are one approach toadvancing local, regional, and global environmental science and education. Aremarkable number and wide variety of environmental research networks operatearound the world today. These are diverse in funding, infrastructure,motivating questions, scientific strengths, and the sciences that birthed andmaintain the networks. Some networks have individual sites that wereselected because they had produced invaluable long-term data, while othernetworks have new sites selected to span ecological gradients. However, alllong-term environmental networks share two challenges. Networks must keeppace with scientific advances and interact with both the scientific communityand society at large. If networks fall short of successfully addressing thesechallenges, they risk becoming irrelevant. The objective of this paper is toassert that the biogeosciences offer environmental research networks a numberof opportunities to expand scientific impact and public engagement. Weexplore some of these opportunities with four networks: the InternationalLong-Term Ecological Research Network programs (ILTERs), critical zoneobservatories (CZOs), Earth and ecological observatory networks (EONs),and the FLUXNET program of eddy flux sites. While these networks were foundedand expanded by interdisciplinary scientists, the preponderance of expertise andfunding has gravitated activities of ILTERs and EONs toward ecology andbiology, CZOs toward the Earth sciences and geology, and FLUXNET towardecophysiology and micrometeorology. Our point is not to homogenize networks,nor to diminish disciplinary science. Rather, we argue that by more fullyincorporating the integration of biology and geology in long-termenvironmental research networks, scientists can better leverage networkassets, keep pace with the ever-changing science of the environment, andengage with larger scientific and public audiences.more » « less
-
The advent of artificial intelligence and machine learning has led to significant technological and scientific progress, but also to new challenges. Partial differential equations, usually used to model systems in the sciences, have shown to be useful tools in a variety of tasks in the data sciences, be it just as physical models to describe physical data, as more general models to replace or construct artificial neural networks, or as analytical tools to analyse stochastic processes appearing in the training of machine-learning models. This article acts as an introduction of a theme issue covering synergies and intersections of partial differential equations and data science. We briefly review some aspects of these synergies and intersections in this article and end with an editorial foreword to the issue. This article is part of the theme issue ‘Partial differential equations in data science’.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Journal Article:
- https://doi.org/10.1029/2018WR022643
