Why OpenAgua?

As water management challenges mount globally, innovative tools are needed to help society plan and manage its water resources in ways that are sustainable, resilient and transparent. OpenAgua is an innovative, web-based software platform for modeling water resources systems to help meet this need, a result of collaborations initially between researchers at the University of California Davis, Tecnológico de Monterrey, the University of Manchester, and now the University of Massachusetts Amherst. And this is its blog.

In general, we anticipate this blog will be a place for general writings about OpenAgua, covering the application (see http://test.openagua.org for now), related technologies and methods, announcements, and musings. Of course, we’ll update via postings on Twitter, so follow us.

Detailed documentation about the OpenAgua project, including key innovations, about the OpenAgua team and funding partners, a user guide, technical documentation, etc., can be found at http://docs.openagua.org.

With that in mind, I thought I would write about some of the technical motivations behind OpenAgua, in terms of the state of water resources modeling today, and the general software development philosophy. Of course, the grander motivation – more socially and economically efficient water management – is more important, and drives long term development efforts. But in the near term OpenAgua is a decidedly technical endeavor.

Water system modelers, as perhaps with modelers in all disciplines, are at a very interesting moment in time. For the past few decades, water system modeling has been conducted primarily on desktop computers, using either custom-built decision support systems or off-the-shelf applications (e.g., WEAP, RiverWare, Aqua Tool, and MIKE PLANNING). However, using desktop-based software for all but basic studies in a reasonable amount of time isn’t pretty, as demonstrated by this image showing a setup for running WEAP in parallel (this can also be done on a single machine using virtual machines, as I have done – still not pretty):


Although desktop computers have become increasingly speedy, they have not been able to keep up with several key changes in society spanning multiple spheres of human endeavors, as follows:

  1. The physical, social, and economic basis for water resources management is rapidly changing, calling for the need to explore an increasingly complex array of management conditions to better inform infrastructure planning. This includes increasing populations and resultant increasing demand for water, diminishing water quality (e.g., from pollution or changing salinity) and quantity (e.g., from groundwater depletion or upstream diversions), changing societal preferences, and so on. Furthermore, we are increasingly able to project future baseline conditions (of climate, populations, etc.), yet with a great degree of uncertainty, further necessitating advanced scenario assessments.
  2. The sheer volume of data being generated in various domains relevant to water management is increasing rapidly, encouraging us to explore water management possibilities using this data in ways that traditional desktop applications are poorly suited for. In the parlance of today, this means “big data”, available from a wide array of sources (e.g., satellites, ubiquitous environmental sensors, simulated climatic data, etc.). While this does not mean we have enough data–indeed, much of the developing world remains data poor, and even some advanced economies have major data deficiencies–it does mean that where data is becoming “big” we lack easy-to-use tools and methods to leverage this data.
  3. Even in cases where the academic water resources planning community has the methodological and computational tools to be able to perform complex exploratory analyses, these tools are generally cumbersome to use or are otherwise not designed to be used directly by the broader water planning community without case-by-case customization by experts. This inhibits the broader adoption of even some of the more popular modeling and analytic methods (e.g., genetic algorithm and other evolutionary algorithms). The few desktop systems that do exist for water planning use a fairly narrow set of traditional modeling approaches, limiting the possibilities for broader adoption of newer, better approaches.
  4. There has been a recent explosion in the development and availability of easy-to-use, web-based information technologies that can help address some, if not all, of the issues described above. Some are free, while others require paid licenses; most are open source. In the field of data analytics, two examples include Plotly (for plotting on a variety of platforms, including the web) and Tableau (for business intelligence and analytics). Many other examples exist, many of which will never be as visible as these from a branding perspective but that nonetheless enable rapid development of powerful websites.
  5. Cloud computing (Amazon Web Services, Google Cloud Platform, Microsoft Azure, etc.)–also known as Infrastructure as a Service (IaaS)–is growing up rapidly (see this article on the state of cloud computing as of early 2016), enabling on-demand computing for problems too large for traditional desktop systems. Of course, these technologies also enable other exciting features not available with a traditional desktop-based application, such as realtime or near-realtime collaboration between different water management participants, among others. Cloud computing is already used by the water planning community, but not for easy-to-use water system modeling; one interesting example of the promise of the cloud for water system modeling includes Insight Maker, a “systems thinking” app similar to Stella Professional.

The availability (and easy customizability) of rich website technologies combined with easy-to-use and relatively affordable cloud computing resources represents a fundamental turning point in terms of possibilities available to water systems modelers to help understand complex water management challenges. However, to-date, these remain possibilities, and have not been leveraged in a way that fully exposes methodological advances, data availability, and computing  possibilities in a way that is easily accessible to the broader water planning community.

It is our hope that OpenAgua will provide this exposure. Beyond this immediate need, however, we also envision OpenAgua as a platform that can absorb changes in water planning methods, web technologies and the physical, social and economic contexts of water planning. In other words, as technologies and water management needs change, OpenAgua as a modeling platform should be able to adapt. The open source aspect of OpenAgua is critical toward that end, and we expect a long term engagement with the water management community, as well as the broader information technology development community.

We would be thrilled if you could join us in this endeavor!

One thought on “Why OpenAgua?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s