Big data for sustainable water management

Agriculture uses nearly 70% of the water consumed each year. Three ways data helps manage water: fewer pesticides, fewer losses in municipal networks, and smart meters for citizens.

A water level through the seasons, with the stretch under the low-water mark boxed.

Called “blue gold”, water is a resource necessary for human survival and indispensable to agriculture and some industries. With nearly 70% of the water consumed each year, agriculture is the sector that uses the most water in the world, and the one that pollutes most by discharging water mainly polluted by excessive amounts of fertilisers and pesticides. Its scarcity makes water a resource to preserve, whose management has to be rethought entirely, in developing and developed countries alike.

Artificial intelligence and machine learning are emerging as answers to today’s water management problems. The progress these technologies can bring is of two kinds: on the one hand, helping to reduce water losses, and on the other, eliminating the water pollution caused by the discharge of pollutants.

The applications of artificial intelligence and/or machine learning to water management keep evolving and keep bearing fruit.

Let us now review the various innovative and promising practices in the fight against water waste and pollution.

1 – Reducing unnecessary pesticides and herbicides in agriculture with machine learning

The almost systematic use of pesticides, especially in intensive farming, is one of the main sources of water pollution. The effluents, loaded with chemical components, then run off into the surrounding waterways, polluting other sources as well. Moreover, treating this polluted water is long and costly for the local authorities who generally take charge of its treatment. That is why it seems relevant to tackle the problem at its source and to minimise the quantities of chemical inputs applied on the farms themselves.

Here, machine learning makes it possible to assess precisely the needs of each plant at a given time, in order to fight the pests that threaten its growth.

To do so, as much information as possible about the plants must first be collected. All kinds of precise, real-time assessment technologies can be applied, such as drones fitted with hyperspectral cameras able to fly over farmland. The aim is to collect enough data, transmitted directly to data science operators. Once processed by the data scientists, the data can then be synthesised and simplified to help farmers spread fertilisers, pesticides and herbicides optimally across the farm, according to the needs of the plants identified beforehand through machine learning. This kind of artificial intelligence application fundamentally reduces the quantities of inputs spread in agriculture, and therefore limits the discharge of pollutants into the water.

2 – Reducing municipal water losses with Big Data

Another cause of water waste lies in the age of some municipal pipe systems, whose installations leak water in ways that could largely be anticipated and avoided.

Big Data emerges as a solution to water losses, as new intelligent technologies can now keep local authorities better informed about the state of their cities’ pipes.

For example, some metropolitan areas already have an intelligent analysis platform able to give them a real-time, 360-degree overview of their municipal pipe systems. With immediate access to the state of municipal infrastructure, local authorities can locate the origin of water losses in real time and therefore act on these specific leaks. Here, the efficiency of Big Data enables public agents to improve the management of their infrastructure. This approach is part of an effort to preserve natural resources such as water.

3 – Giving citizens smart measuring tools to encourage the right everyday habits

Some regions are particularly exposed to drought and water shortages, which makes water a strategic, even precious, resource there for all local players: farmers, manufacturers and households.

Raising public awareness of water management issues, and preventing waste, have become common practice in developed and developing countries alike. Moreover, prevention works best when people are precisely informed of their own water consumption.

Here, Big Data serves as a way to collect households’ water consumption data in real time and to report it back to users through various possible means, such as a connected box installed in the home, able to receive the information and display it on a screen the users can consult directly. Thanks to this technology, people can easily see their water consumption and be alerted in case of a failure or a leak in their home.

Better still, when users see their daily water consumption compared with that of their neighbourhood, they are encouraged to reduce their water consumption and adopt simple, less water-hungry habits.