Enabling Collaborative Numerical Modeling in Earth Sciences using Knowledge Infrastructure: Landlab Notebooks
|Authors:||Christina Bandaragoda · Anthony Michael Castronova · Jimmy Phuong · Erkan Istanbulluoglu · Sai Siddhartha Nudurupati · Ronda Strauch · Nicole Gasparini · Eric Hutton · Greg Tucker · Daniel Hobley · Katherine Barnhart · Jordan Adams|
|Owners:||Christina Bandaragoda · Jimmy Phuong|
|Created:||Aug 07, 2017 at 3:29 p.m.|
|Last updated:||Oct 29, 2018 at 9:41 p.m. by Christina Bandaragoda|
The ability to test hypotheses about hydrology, geomorphology, and atmospheric processes is invaluable to research in the Earth and planetary sciences. To swiftly develop experiments using community resources is an extraordinary emerging opportunity to accelerate the rate of scientific advancement. Knowledge infrastructure is an intellectual framework to understand how people are creating, sharing, and distributing knowledge -- which has dramatically changed and is continually transformed by Internet technologies. We are actively designing a knowledge infrastructure system for earth surface investigations. In this paper, we illustrate how this infrastructure can be utilized to lower common barriers to reproducing modeling experiments. These barriers include: developing education and training materials for classroom use, publishing research that can be replicated by reviewers and readers, and advancing collaborative research by re-using earth surface models in new locations or in new applications. We outline six critical elements to this infrastructure, 1) design of workflows for ease of use by new users; 2) a community-supported collaborative web platform that supports publishing and privacy; 3) data storage that may be distributed to different locations; 4) a software environment; 5) a personalized cloud-based high performance computing (HPC) platform; and 6) a standardized modeling framework that is growing with open source contributions. Our methodology uses the following tools to meet the above functional requirements. Landlab is an open-source modeling toolkit for building, coupling, and exploring two-dimensional numerical models. The Consortium of Universities Allied for Hydrologic Science (CUAHSI) supports the development and maintenance of a JupyterHub server that provides the software environment for the system. Data storage and web access are provided by HydroShare, an online collaborative environment for sharing data and models. The knowledge infrastructure system accelerates knowledge development by providing a suite of modular and interoperable process components that can be combined to create an integrated model. Online collaboration functions provide multiple levels of sharing and privacy settings, open source license options, and DOI publishing, and cloud access to high-speed processing. This allows students, domain experts, collaborators, researcher, and sponsors to interactively execute and explore shared data and modeling resources. Our system is designed to support the user experiences on the continuum from fully developed modeling applications to prototyping new science tools. We have provided three computational narratives for readers to interact with hands-on, problem-based research demonstrations - these are publicly available Jupyter Notebooks available on HydroShare.
When you open this resource with the CUAHSI JupyterHub server (upper right, click on Open With, Select JupyterHub), you will launch a Welcome Notebook that will connect you to the CyberGIS virtual machine on the ROGER super computer at the University of Illinois, Urbana-Champagne. When you execute (Run Step 1 and Step 2 only) in the Jupyter Notebook cells on the Welcome Notebook, you will download related data and three Notebooks designed to explore hydrologic research problem solving using data and model integration in HydroShare. Skip Step 3 "Welcome" tutorial steps unless you want to explore how to do work and Save back to HydroShare.
The problem: Researchers need a modeling workflow that is flexible for developing their own code, with easy access to distributed datasets, shared on a common platform for coupling multiple models, usable by science colleagues, and with easy publication of data, code, and scientific studies.
The emerging solution: Collaborate with the CUAHSI HydroShare community to use and contribute to water data software and hardware tools, so that you can focus on your science, be efficient with your time and resources, and build on existing research in multiple domains of water science.
Notebook 1: Educate by exploring rainfall driven hydrographs with Landlab
Notebook 2: Watershed subset within regional Landlab landslide model to explore fire impacts
The resource was originally derived from a reproducible demonstration of the landslide modeling results from: Strauch, R., Istanbulluoglu, E., Nudurupati, S. S., Bandaragoda, C., Gasparini, N. M., and Tucker, G. E.: (2018) A hydro-climatological approach to predicting regional landslide probability using Landlab, Earth Surf. Dynam. Discuss., https://doi.org/10.5194/esurf-6-49-2018.
Notebook 3: Reuse ecohydrology model for exploring climate scenarios
The gridded meteorology forcings are pre-processed in the Notebook in NewMexico_observatory_gridmet.ipynb.
How to cite
This resource is shared under the Creative Commons Attribution CC BY.http://creativecommons.org/licenses/by/4.0/
|Coordinate System/Geographic Projection:||WGS 84 EPSG:4326|
|Coordinate Units:||Decimal degrees|
|Christina Bandaragoda||University of Washington|
|Anthony Michael Castronova||CUAHSI||MA, US||4357970852|
|Jimmy Phuong||University of Washington||Washington, US|
|Erkan Istanbulluoglu||University of Washington|
|Sai Siddhartha Nudurupati||University of Washington - Seattle||WA, US||4026170098|
|Ronda Strauch||University of Washington|
|Nicole Gasparini||Tulane University|
|Katherine Barnhart||University of Colorado at Boulder||CO, US|
|Jordan Adams||Tulane University||Louisiana, US||6107397582|
|Erkan Istanbulluoglu||University of Washington|
|David Tarboton||Utah State University||Utah, US||4357973172|
|Dan Ames||Brigham Young University|
|Zhiyu (Drew) Li||Brigham Young University|
This resource was created using funding from the following sources:
|Agency Name||Award Title||Award Number|
|National Science Foundation|
Please wait for the process to complete.