Managing storm water with the Internet of Things

By | 2 minute read | August 14, 2017

Storm water management

Storms are destructive beasts, especially to cities and built-up areas. Even your average bout of heavy rain can lead to overflowing rivers and flooding; damaging infrastructure, rendering roads impassible and halting transport until the damage is repaired. So can the Internet of Things help manage excess storm water, and even put it to good use? IBM engineer Matt Paschal thinks so.

In the spring of 2016, Matt assembled a team to address this problem.  Matt, Eric Jenney, Perry Dykes, Dave Draeger, and Tracy Harris put together a demonstration at the Rochester, MN Place Makers Prototyping Festival to show how the IoT can manage excess storm water. Using the IBM Bluemix Platform and a network of sensors, the team reused storm water to irrigate local plant life, reducing the amount sent straight to the river.

The problem of drainage

Why is heavy rain such a problem? The answer lies in increasingly built-up infrastructure. Without large fields and bodies of soil, excess water – which would usually filter down through the soil – has nowhere to go. Instead, it’s collected by pipes, transported to drains and dumped in the river. There are two problems with this: first, storm water isn’t clean. Dirty water means higher temperature water, which affects aquatic life. Second, the sheer volume of water can put a strain on river banks and potentially lead to flooding.

Collecting water data to make smart decisions

As part of a small scale demo, the team connected moisture sensors, turbidity (water quality) sensors, flow sensors and electronic valves to Photon microcontroller boards connected to the IBM Bluemix Platform. Bluemix analyses the data from these various components and displays it on simple dashboards so that it’s easy to see patterns emerging, make decisions based on the data, and set automatic responses to different scenarios.

Redirecting storm water

To decrease the pressure on rivers, Matt diverted excess water to local plant life. Moisture sensors in the soil sent data to Bluemix and told the system when to deliver water to the plants. The system itself is flexible: not only can you be selective about which plants are watered, you can automatically set thresholds to determined when the water supply is allowed to run, and when it is shut off. When the moisture sensors detect too much water in a particular area, that data will be sent to Bluemix and the water supply to that area will be automatically stopped.

Cleaning it up: smart filtering

Even if local vegetation is particularly thirsty, there’s still going to be water left over, and inevitably it’s going to end up in the river. So can it be cleaned beforehand? And what role does the IoT play in the process?

Turbidity sensors measure the quality of the water – and by that, I mean how dirty it is. Storm water is dumped into an inexpensive fabric filter to remove dirt particles, and the filtered clean water is checked for quality by the sensors, before being diverted back into the river.

Take a look at the video to see the process in action.

Scaling up

Within the confined of a small-scale demonstration, we’ve seen how Watson and the Internet of Things can help manage water flow. On a larger scale, simple measures like these could protect river banks and city infrastructure during storms, potentially saving money, time, and hassle.