ICDM 2017: IEEE International Conference on Data Mining
New Orleans, Louisiana, USA
November 18–21, 2017

Seventh Workshop on
Data Mining in Earth System Science (DMESS 2017)

Co-conveners: Forrest M. Hoffman, Auroop R. Ganguly, Jitendra Kumar, and Richard Tran Mills

Pontalba Room on the Mezzanine Level of
The Roosevelt New Orleans, A Waldorf Astoria Hotel
New Orleans, Louisiana, USA | November 18, 2017

Final Workshop Program (November 18, 2017)

Printable Workshop Program (PDF)

Time Title Speaker Presentation Type Paper Authors
8:30 Introduction to Data Mining in Earth System Science (DMESS)
Forrest M. Hoffman Introductory Presentation and Panel Charge Forrest M. Hoffman, Auroop R. Ganguly, Jitendra Kumar, and Richard Tran Mills
9:00 Precipitation Estimate from Multi-Satellite Remote Sensing Measurements using Machine Learning Methods
Abstract | Slides
Kuolin Hsu Invited Keynote Presentation Kuolin Hsu and Soroosh Sorooshian
9:30 Convolutional Neural Network Approach for Mapping Arctic Vegetation using Multi-Sensor Remote Sensing
Abstract | Slides
Forrest M. Hoffman Contributed Paper Presentation SP19205 Zachary Langford, Jitendra Kumar, and Forrest M. Hoffman
10:00 Coffee Break
10:15 Resolution Reconstruction of Climate Data with Pixel Recursive Model
Abstract | Slides
Sookyung Kim Invited Paper Presentation SP19203 Sookyung Kim, Sasha Ames, Chengzhu Zhang, Jiwoo Lee, and Dean Williams
10:45 Quantifying Seasonal Patterns in Disparate Environmental Variables Using the PolarMetrics R Package
Abstract | Slides
Bjørn-Gustaf Brooks Contributed Paper Presentation SP19206 Bjorn Brooks, Danny Lee, Ankur Desai, Lars Pomara, and William W. Hargrove
11:15 Vital Role of Training and Education in Big Data Applications
Abstract | Slides
David A. Yuen and Gabriele Morra Invited Keynote Presentation David A. Yuen and Gabriele Morra
11:45 Lunch
13:00 A Machine Learning Approach to Non-uniform Spatial Downscaling of Climate Variables
Abstract | Slides
Soukayna Mouatadid Contributed Paper Presentation SP19207 Soukayna Mouatadid, Steve Easterbrook, and Andre Erler
13:30 Scalable Algorithms for Clustering Large Geospatiotemporal Data Sets on Manycore Architectures
Abstract | Slides
Vamsi Sripathi Invited Keynote Presentation Vamsi Sripathi
14:00 Deriving Data-driven Insights from Climate Extreme Indices for the Continental US
Abstract | Slides
David Sathiaraj Contributed Paper Presentation SP19201 Xinbo Huang, David Sathiaraj, Lei Wang, and Barry Keim
14:30 How Can Physics Inform Deep Learning Methods in Earth System Science?: Recent Progress and Future Prospects
Abstract | Slides
Anuj Karpatne Invited Keynote Presentation Anuj Karpatne
15:00 Coffee Break
15:15 DMESS Panel Discussion
16:30 Adjourn Workshop
19:15 DMESS Workshop Dinner at the Red Fish Grill (115 Bourbon Street). Meet in hotel lobby at 7:15 p.m. You are responsible for your own food and drinks.

Workshop Description:

Spanning many orders of magnitude in time and space scales, Earth science data, from point measurements to process-based Earth system model output, are increasingly large and complex, and often represent very long time series, making these data difficult to analyze, visualize, interpret, and understand. An “explosion” of heterogeneous, multi-disciplinary data–including observations and models of interacting natural, engineered, and human systems–have rendered traditional means of integration and analysis ineffective, necessitating the application of new analytical methods and the development of highly scalable software tools for synthesis, assimilation, comparison, and visualization. For complex, nonlinear feedbacks among chaotic processes, new methods and approaches for data mining and computational statistics are required for classification and change detection, model evaluation and benchmarking, uncertainty quantification, and incorporation of constraints from physics, chemistry, and biology into analysis. This workshop explores various data mining approaches and algorithms for understanding nonlinear dynamics of weather and climate systems and their interactions with biogeochemical cycles, impacts of natural system responses and climate extremes on engineered systems and interdependent infrastructure networks, and mitigation and adaptation strategies for natural hazards and infrastructure and ecosystem resilience. Encouraged are original research papers describing applications of statistical and data mining methods that support analysis and discovery in climate predictability, attributions, weather extremes, water resources management, risk analysis and hazards assessment, ecosystem sustainability, infrastructure resilience, and geo-engineering. Rigorous review papers that either have the potential to expose data mining researchers to commonly used data-driven methods in the Earth sciences or discuss the applicability and caveats of such methods from a machine learning or statistical perspective, are also desired. Methods may include, but are not limited to cluster analysis, empirical orthogonal functions (EOFs), extreme value and rare events analysis, genetic algorithms, neural networks and deep learning methods, physics-constrained data analytics, automated data assimilation, and other machine learning techniques. Novel approaches that bring new ideas from nonlinear dynamics and information theory, network science and graphical methods, and the state-of-the-art in computational statistics and econometrics, into data mining and machine learning, are particularly encouraged.

Previous workshops:

Program Committee Members:

Paper Submission:

Authors are invited to submit manuscripts of up to 10 pages reporting unpublished, mature, and original research and recent developments/theoretical considerations in applications of data mining to Earth sciences by August 7, 2017 August 28, 2017, in IEEE 2-column format. Accepted papers will be printed in the conference proceedings. Additional details and a link to the manuscript submission system will be provided in the near future. Submission implies the willingness of at least one of the authors to register and present the paper.

Please submit your paper via the website at https://wi-lab.com/cyberchair/2017/icdm17/scripts/submit.php?subarea=SP19&undisplay_detail=1&wh=/cyberchair/2017/icdm17/scripts/ws_submit.php.

Important Dates:

Full paper submission: August 7, 2017 August 28, 2017Deadline extended!
Author notification: September 4, 2017
Camera-ready paper submission: September 15, 2017 September 18, 2017Deadline extended!
DMESS 2017 Workshop: November 18, 2017
ICDM 2017 Conference: November 18–21, 2017


URL: http://www.climatemodeling.org/workshops/dmess2017/
E-mail: dmess2017 at climatemodeling dot org

Contribution to Computational Science:

This workshop will contribute to the field of Computational Science by creating a forum for original research papers and presentations from leading computational and Earth scientists who are applying data mining techniques on advanced computing platforms (HPC systems, clusters, grids and clouds) to distill knowledge from the massive—and growing—data sets created by the Earth science community.

About the Workshop Co-conveners:

Forrest M. Hoffman is a Senior Computational Climate Scientist at Oak Ridge National Laboratory (ORNL). As a resident researcher in ORNL’s Climate Change Science Institute (CCSI) and a member of ORNL’s Computational Sciences & Engineering Division (CSED), Forrest develops and applies Earth system models (ESMs) to investigate the global carbon cycle and feedbacks between biogeochemical cycles and the climate system. He applies data mining methods using high performance computing to problems in landscape ecology, remote sensing, and large-scale climate data analytics. He founded the workshop series on Data Mining in Earth System Science (DMESS) in 2009 and has served as lead convener for all six prior workshops. Forrest is also a Joint Faculty Professor in the University of Tennessee’s Department of Civil & Environmental Engineering in nearby Knoxville, Tennessee.

Auroop R. Ganguly is a civil and environmental engineer who works at the intersection of three broad areas: (1) Climate Extremes and Water Sustainability, (2) Infrastructural Resilience and Homeland Security, and (3) Applied Data and Computational Sciences. Prior to his current position as a faculty at Northeastern University in Boston, MA, he was at the US Department of Energy’s Oak Ridge National Laboratory for seven years, at Oracle Corporation for five years, and at a startup subsequently acquired by Oracle for a year. In addition, he has a dual interest in ancient history and science fiction, i.e., the forgotten past and the unknown future. While he has nothing particularly against the here and now, he rarely gets any time to spend there and then. Once upon a time he obtained a PhD from MIT, and currently, other than his day job as the Principal Investigator of the SDS Lab at Northeastern, he takes a bunch of undergraduate kids across India to study climate change, and happens to be the Chief Scientific Adviser for a startup, risQ Inc. (http://www.risq.io/), co-founded with one of his former PhD students.

Jitendra Kumar is a computational hydrologist at Oak Ridge National Laboratory and a Joint Assistant Professor at the University of Tennessee, Knoxville. He conducts research at the intersection of high performance computing, environmental and Earth sciences, and systems analysis and data mining. His research entails data mining, large-scale global optimization, computational hydrology and hydrogeology, landscape ecology, remote sensing, and development of parallel algorithms for large-scale supercomputers.

Richard Tran Mills is an HPC Earth System Models Architect at Intel Corporation, where he leads efforts related to weather, climate, and Earth System models and associated analysis tools on current and next-generation high-performance computing architectures. Prior to joining Intel in 2014, he spent a decade as a research scientist at Oak Ridge National Laboratory, and also held a joint faculty appointment at the University of Tennessee, Knoxville. His work has spanned high-performance scientific computing, geospatiotemporal data mining, computational hydrology, and climate change science. He is one of the original developers of PFLOTRAN, an open-source code for massively parallel simulation of hydrologic flow and reactive transport problems, and has also contributed to the development of PETSc, the Portable, Extensible Toolkit for Scientific Computation, a suite of solvers, data structures, and associated routines for the solution of a wide variety of scientific computing problems. He earned his Ph.D. in Computer Science in 2004 at the College of William and Mary, where he was a Department of Energy Computational Science Graduate Fellow. Prior to that, he studied geology and physics at the University of Tennessee, Knoxville as a Chancellor’s Scholar.

This Site vi powered For assistance or additional information, contact Forrest Hoffman (forrest@climatemodeling.org)
Last Modified: Wednesday, 09-May-2018 14:46:16 EDT
Warnings and Disclaimers