Collaborative Modeling Philosophy
This website chronicles ongoing development of a collaborative modeling framework conceived as a joint effort between the University of Hawaii Water Resources Research Center (UHWRRC) and the American Samoa Power Authority (ASPA). Recent advancements in cloud-computing technologies provide new opportunities for collaboration and communication between existing institutions that can more efficiently share resources thereby allowing for direct application of models designed to solve water management challenges. The intended outcomes for the models developed within this framework are commensurate with the motivations behind the participatory and collaborative modeling movement that has, in recent decades, become a highly utilized approach in environmental management. These outcomes are centered around addressing the need for enhanced researcher – stakeholder engagement and producing, practical, defenseable models that sufficiently address stakeholder needs and promote model use in guiding important water management decisions.
"The collaborative modeling process includes developers, decision-makers, stakeholders and others working together to develop a shared understanding of [the region’s] management objectives and the model’s role in supporting those objectives." Moran (2016)
|
CyberinfrastructureThe cloud based cyber-infrastructure framework used in the UHWRRC-ASPA modeling process was imperative for facilitating communication, providing the ability to collaboratively code, and for taking care of the basic cloud-computing needs that would have been overwhelmingly resource intensive for either participant to develop independently. Almost every step in the Tutuila modeling framework project is compiled in Jupyter Notebook format (.ipynb). These notebooks can then directly integrated into the GitHub and Binder or Azure platforms to allow remote access for sharing, developing, and collaboration.
|
This case-study demonstrates a long-term, process-based groundwater modeling approach, that as of this writing is evolving and under active development. While goals, methods, and preliminary results have been established, there remain numerous planned, and likely many unplanned additions and modifications to be made as stakeholders weigh in, and as our experience with these tools grows. Planned future objectives include continued development with the groundwater modeling component, as well as some additions to the monitoring network.
Workflow
The modeling framework developed for this project integrates weather station data, streamflow data, water budget modeling, and groundwater modeling into a seamless data-to-model workflow. The workflow is made entirely open-source, reproducible, and dynamic by using innovative cloud-computing tools such as Jupyter Notebooks, GitHub, and Binder/Azure. These tools manage the data-science infrastructure, so the project team can focus on communicating with each other and developing models that are scientifically relevant and useful for water resources management. While this framework was deployed in American Samoa it could be easily scaled to other islands or localities.