KM and the environment: Water management uses analytics, big data and collaboration to handle complexity.
Water is a vital resource with pervasive effects on the environment, health and the economy. More than a billion people are facing water shortages, and many contend with problems such as flooding and pollution. The complexity of the scientific, social and political issues that surround water management makes the enabling technologies of knowledge management, such as analytics and collaboration, a good match for addressing those issues.
The Internet of Things (IoT) is providing a growing number of inputs from sensors, adding to the available data. Greater computing power and more sophisticated scientific models, combined with better integration of information from diverse sources, now allow water management officials to make more precise, data-driven decisions that help ensure the most positive outcome.
A typical set of issues is illustrated in the management of a river basin in Greece. The Spercheios River winds its way through central Greece, passing through villages and agricultural land before it empties into the Maliakos Gulf, just off the Aegean Sea. The river basin presents several environmental challenges. Frequent flooding was affecting residents living near the river, and water quality problems had arisen because of agricultural chemicals and wastewater. Toxic blooms of Chatanella algae had caused a massive fish kill after flowing from the river into the Gulf, seriously impacting the fishing industry.
The Hellenic Center for Marine Research (HCMR) wanted to provide a holistic solution to those issues. With funding from the General Secretariat for Research and Technology in Greece and support from DHI, a Danish firm specializing in water issues and water modeling and management software, the Center wanted to develop a holistic, comprehensive solution to address all those issues in a coordinated way.
The goal was to create a decision support system (DSS) that would integrate data from multiple sources, including weather predictions and sensors that measure water quality. The monitoring, forecasting and early warning systems could help officials make valid decisions about how to manage the river and mitigate adverse environmental effects.
Timely information essential
Elias Moussoulis, hydrologist and environmental scientist at DHI, says, “The HCMR needed to aggregate all the data into one platform so they could model it against different scenarios and provide web forecasts to the authorities and the public. The system would then support making the right decisions about actions that could be taken to mitigate the effect of pollution incidents and extreme rain events.” Such an integrated system would allow better decision making to respond to potential flooding and improve management by farmers in the surrounding area to minimize fertilizer runoff that causes riverine pollution and fish kills in the Gulf.
In addition, the system needed to operate in near real time so it could predict what would happen within the next few hours. “Timely information would allow water management officials to take actions, such as using various hydraulic structures to divert or impede some of the upstream water flow, to mitigate flooding in the valley,” Moussoulis says, “or to advise farmers about the best time to use fertilizer to minimize nutrients runoff.”
DHI used its MIKE Operations platform and its water modeling software products MIKE She and MIKE Flood to implement the system. Its capabilities encompass information/data management, planning and forecasting. The planning function addresses risks, cost/benefits, scenario analysis and key performance indicators (KPIs). The operational component handles forecasting/early warning, real-time operations and optimization. Data management provides analysis, reporting, GIS, spreadsheet and scripting functionality and sharing/publishing results.
The HCMR installed stations to monitor water level and quality in the river basin and the Gulf. Data is obtained from those sensors and from meteorological radar that is installed in the river basin. MIKE Powered by DHI software integrates and presents the data into a single platform." “Users do not have to open up individual applications such as a GIS system or databases containing measurements, because all the information is integrated into one interface,” Moussoulis says.
He continues, “Data is imported in an automated way. The models are then updated with real-time data.” The system automatically sends alerts and publishes information to a website that documents the status of the river water, river basin and Gulf water. “We can now optimize the management of flood protection measures in the river, and advise farmers about when it is safe to put down fertilizer,” Moussoulis adds. The overall impact will be a river basin that is safe for residents, productive for farmers and protective of the fish population.
Big data streaming from environmental sensors
The number of sensors collecting environmental data is growing rapidly, and the collected data offers the potential for a wealth of insights. The Environmental Law Institute provides a sampling of big data projects in its report “Big Data and Environmental Protection: An Initial Survey of Public and Private Initiatives.” The U.S. Geological Survey, for example, maintains the National Water Information System (NWIS), which provides information from more than 1.5 million sites involving water. Each site inventory includes about 300 components. Information is provided on the NWIS website in the form of a searchable database that is available to the public.
The National Weather Service, which is part of the National Oceanographic and Atmospheric Administration (NOAA), collects billions of observations per day to generate its weather forecasts, including predictions for rainfall, flooding and hurricanes. The National Weather Service has been collecting data for about 60 years. Its data collection capabilities have increased over time, along with the sophistication of its models, which have allowed for steadily improving preparation and recovery from weather-related incidents. The data feeds models that generate forecasts and warnings for government agencies and the public.
One change in the use of environmental data is in its accessibility. “Historically a lot of data has been collected by government agencies, but it was not widely accessible,” says Linda Breggin, a senior attorney with the Environmental Law Institute and one of the authors of its report on big data. “We are now in a different era, because the user interfaces have improved and people are more comfortable interacting with data on their own.” Greater usability benefits public officials and citizens seeking information.
The report also points out that crowdsourcing is generating large amounts of data through various apps through which users can contribute. The Creek Watchapp was developed by IBM and the California State Water Resources Control Board’s Clean Water Team. It allows citizens to photograph waterways with their mobile phones and rate the waterway on the amount of water, rate of flow and trash levels.
The information is provided to water control boards, watershed groups, government agencies and scientists. The app has caught on and is being used by more than 4,000 people in 25 countries. “There are many situations in which ‘citizen scientists’ can gather data,” Breggin says, “and this can be used to augment data already being collected by the government at the federal, state and local levels.”