We are sure that in the future most of water resources software will be open source, but somehow there are doubts about the operating system, the platform and some requirements of these future software. It might be that concepts of this article are against the mainstream on the development of water resources evaluations, but we are talking about the future, a interesting stage in our concern for the sustainable management of water resources, and a stage with lots of uncertainties, challenges and trends.

The problem of predicting the future

The concern about future is inherit to the humankind. From the stone age, humans were trying to understand the physical environment in order to survive and preserve the available resources. On ancient times, people suffer from climate change scenarios that sometimes ruled the end of civilizations, migrations, and political rearrangements. We don’t know if people in the past were so aware with the concept that future can much different in time frames of less than 30 years.

However, by tradition, humans have never been able to predict the future on a right way. Issac Asimov failed to conceptualize computers on the fiction book Steel Caves (1953). In the film Back to the Future 2, on year 2015 there will be flying cars and hover boards. The concept of today Smartphone is almost absent on the science fiction books.

Which are the challenges of water resources software?

There is software for surface water, groundwater, and river hydraulics, however, future software have to integrate all components from the water cycle together with an assessment of main parameters as climate change, land use change, flow/quality interactions. These kind of simulation requires huge baselines, and multiple stakeholder participation for the construction, calibration and simulation. Results of these simulations will be required by regulatory agents.

In our opinion, numerical models will be (or are already) the key for coming political discussions about water resources management and environmental protection. More requirements on the simulations with more involvement from communities would lead to complex models where the uncertainties have to be reduced.

What is Jupyter?

Project Jupyter (http://jupyter.org) is a non-profit organization that develops open-source software, open standards, and services for interactive computing across multiple programming languages. This project developed Jupyter Notebook, a open source web-application that allows the creation of documents with code, equations, visualization and text.

This article deals with Jupyter Notebook, however, any future software that has the same or upgraded capabilities would be useful.

A single ring that rules all rings

On the Lord of the Rings (J.R.R Tolkien), the Ring was able to rule other Rings. We conceptualize the future of water resources software as rings that will be ruled by a master ring as Jupyter. We need a common platform where all software can run and interact in between each other.

GUI vs Batch Commands

On a discussion from groundwater modeler teachers, there was perspective that it doesn’t matter which Graphical User Interface (GUI) is used to learn MODFLOW, the most important thing is to understand the groundwater flow regime and the numerical tools and options available to simulate it. However, when it comes to have a wider understanding of the model output, water balances, and impact assessment, it might be that any GUI is a good tool but not a optimized tool for this task.
GUI requires the manual selection and processing of tabular and spatial data, and many steps for the output representation and even the interaction with GIS software for results interpretation.

Run numerical models as batch commands is not for beginners, it require a higher knowledge of the model tools and sometimes special APIs to create the model. However, when it comes to speed in model construction and analysis, this type of modeling can be really handy. Jupyter can run any model as batch commands or with the use of APIs, and its interactive tools allow the model run on different scenarios and a immersive experience on the output analysis.