Journal cover Journal topic
Hydrology and Earth System Sciences An interactive open-access journal of the European Geosciences Union
https://doi.org/10.5194/hess-2018-168
© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.
Opinion article
09 Apr 2018
Review status
This discussion paper is a preprint. It is a manuscript under review for the journal Hydrology and Earth System Sciences (HESS).
HESS Opinions: Deep learning as a promising avenue toward knowledge discovery in water sciences
Chaopeng Shen1, Eric Laloy2, Adrian Albert3, Fi-John Chang4, Amin Elshorbagy5, Sangram Ganguly6, Kuo-lin Hsu7, Daniel Kifer8, Zheng Fang9, Kuai Fang1, Dongfeng Li9, Xiaodong Li10, and Wen-Ping Tsai1 1Civil and Environmental Engineering, Pennsylvania State University, University Park, PA 16802
2Institute for Environment, Health and Safety, Belgian Nuclear Research Centre, Mol, Belgium
3Energy Technologies Area, Lawrence Berkeley National Laboratory, Berkeley, CA 94720
4Department of Bioenvironmental Systems Engineering, National Taiwan University, Taipei, 10617, Taiwan
5Dept. of Civil, Geological, and Environmental Engineering, University of Saskatchewan, Saskatoon, Canada
6NASA Ames Research Center/BAER Institute, Moffett Field, CA 94035
7Civil and Environmental Engineering, University of California, Irvine, Irvine, CA 92697
8Computer Science and Engineering, Pennsylvania State University, University Park, PA 16802
9Civil Engineering, University of Texas at Arlington, Arlington, TX 76013
10State Key Laboratory of Hydraulics and Mountain River Engineering, Sichuan University, Sichuan, China
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 water science has so far been gradual, but the related fields are now ripe for breakthroughs. This paper proposes that DL-based methods can open up a viable, complementary avenue toward knowledge discovery in hydrologic sciences. In the new avenue, machine-learning algorithms present competing hypotheses that are consistent with data for scientists to further evaluate. Interrogative studies are invoked to interpret DL models. In addition, we lay out several opinions shared by authors: (1) deep learning may bring forth transformative progress to the field of hydrology due to its ability to assimilate big data and identify commonalities and differences; (2) The community may benefit greatly from a variety of shared datasets and open competitions; (3) Big hydrologic data can be obtained via various ways including data compilation and working with citizen scientists, which offers the co-benefits of education and stakeholder engagement; (4) Water sciences, and hydrology in particular, offer a unique set of challenges that can, in turn, stimulate advances in machine learning; and (5) An urgent need for research is hydrology-customized methods for interpreting knowledge extracted by deep learning.
Citation: Shen, C., Laloy, E., Albert, A., Chang, F.-J., Elshorbagy, A., Ganguly, S., Hsu, K.-L., Kifer, D., Fang, Z., Fang, K., Li, D., Li, X., and Tsai, W.-P.: HESS Opinions: Deep learning as a promising avenue toward knowledge discovery in water sciences, Hydrol. Earth Syst. Sci. Discuss., https://doi.org/10.5194/hess-2018-168, in review, 2018.
Chaopeng Shen et al.
Chaopeng Shen et al.
Chaopeng Shen et al.

Viewed

Total article views: 283 (including HTML, PDF, and XML)

HTML PDF XML Total BibTeX EndNote
214 66 3 283 3 9

Views and downloads (calculated since 09 Apr 2018)

Cumulative views and downloads (calculated since 09 Apr 2018)

Viewed (geographical distribution)

Total article views: 283 (including HTML, PDF, and XML)

Thereof 283 with geography defined and 0 with unknown origin.

Country # Views %
  • 1

Saved

Discussed

Latest update: 20 Apr 2018
Publications Copernicus
Special issue
Download
Short summary
Recently, deep learning (DL) has emerged as a revolutionary and versatile tool transforming industry and scientific disciplines. We argue that DL can offer a complementary avenue toward advancing water sciences because it can efficiently learn from all data and generate new data. New methods are being developed to interpret the knowledge learned by deep networks. We argue open competitions, more data sharing, data collection from citizen scientists will be helpful in incubating advances.
Recently, deep learning (DL) has emerged as a revolutionary and versatile tool transforming...
Share