Water Quality Forecasting
Danish SME WaterWebTools, Baylor University, and the water supply Tarrant Regional Water District are working to provide short-term forecasts for water quality in the Richland Chambers reservoir in Texas. The forecasts, which are to support the water supply's daily decision-making processes and operations, are generated via a new forecasting platform developed by WaterWebTools. The goal is to spread the solution and contribute to solving ever-increasing challenges with water quality throughout the United States.
As the term “The Lone Star State” suggests, Texas is almost an independent country. In the southern state, a significant part of the drinking water supply for the almost 30 million inhabitants is based on surface water from hundreds of reservoirs. However, as in the rest of the world, urbanization, the intensification of agriculture and the contours of a changing climate are leading to a deterioration in water quality. An ever-increasing problem in these reservoirs is algae blooms and oxygen depletion – it challenges confidence in the water and requires extraordinary planning or treatment before distribution to consumers. WaterWebTools is a spin-out company from Aarhus University and since the fall of 2019 has worked with Baylor University in Texas on the issue of the water quality in one of the reservoirs supplying the million-strong city of Fort Worth with drinking water.
The Richland Chambers Reservoir itself is located in the northeastern part of Texas and is the third largest in the state when it comes to surface area. The reservoir is owned by the local water supply Tarrant Regional Water District (TRWD) and though its primary purpose is water distribution, the reservoir is also widely used for recreational purposes, irrigation and as a flood buffer. TRWD has a total of seven reservoirs in their supply network with a total water surface area of over 400 km2 and has an annual budget of 190 million dollars, which in addition to drinking water distribution, also covers flood control and recreational initiatives.
TRWD supplies drinking water for about 2.3 million people, and this includes the water from Richland Chambers as an important element that is pumped via a 2 m wide pipeline to water treatment plants or to other reservoirs, as part of the overall supply network.
The Danish-Texan partnership with Baylor University is responsible for setting up a water quality model for the reservoir. There are some observational data from the reservoir, which is part of the model setup, and profiling probes have been installed for a year, which further complements the insight into the reservoir's water quality and dynamics. The water quality model is implemented in the forecasting platform that WaterWebTools has built. The platform's digital tools automate all steps in an operational model forecast with connection to different weather services, whereby forecasts can be delivered for different key aspects relating to water quality.
In principle, the length of the weather forecast sets the framework for how far ahead in time forecasts can be made in the platform. The US based Global Forecast System (GFS) provides up to 16 days of weather forecast. But it is important to keep in mind that the increasing uncertainty in the weather forecast, the further ahead one looks, naturally also has an effect on the water quality forecasts. We all know for ourselves that when you check the weather forecast for the coming weekend on a Monday morning, it can vary considerably in relation to what the weather actually is like when the weekend occurs. The focus is therefore on detailed knowledge of the water quality for a few days ahead and more general signals when looking at 14 day forecasts, for example.
Why water quality forecasts? According to Thad Scott, associate professor at Baylor University, one significant challenge in Texas, which also applies to most of the United States, is eutrophication of surface water, where algae concentrations are rising due to an increased supply of nutrients. Eutrophication directly challenges the drinking water supply with the so-called HABs, harmful algal blooms, as well as large fluctuations in oxygen concentration, where definite oxygen depletion can be experienced in all or parts of the bottom water in the supply reservoirs. HABs are characterized by the fact that they may contain toxin-producing blue-green algae (cyanobacteria), which can make the water harmful to health, cause fish death and contribute to odor and taste issues, which can be difficult and expensive to remove via purification.
When asked what he sees in terms of benefits and opportunities in water quality forecasts, Scott said, “Most water supplies in the U.S. remain focused on long-term forecasts (+50 years). But they are beginning to discover and recognize the importance of short-term forecasts, especially when they have the capacity for adaptive management.”
Specifically for the water supply that is part of the collaborative project, he said: “TRWD can redirect water from all their reservoirs if the water exceeds specific quality criteria for e.g. dissolved oxygen or chlorophyll-a., but it requires monitoring and this could be optimized based on short-term forecasts.”
Scott predicts that there will be a sharp increase in demand for this type forecasts over the next 10 years.
The forecast platform from WaterWebTools provides predictions of e.g. algae concentrations down through the water column, and the first version of the operational forecast for the Richland Chambers reservoir can already be followed via. www.asap-forecast.com. The platform also provides insight into the reservoir's oxygen levels, whether the reservoir is stratified and whether there is a prospect of destratification during the forecast window. By activating the platform's level alarms, a warning can be given if a level in the forecast window is exceeded, and this can be used in adaptive management (diversion of water, etc.), as well as to plan and optimize times for monitoring.
In particular, the collaborative project provides a concrete understanding of the challenges that the water supply in Texas faces. This understanding and insight are continuously translated into further developments of the platform, so that the water quality forecasts are targeted at the needs in the market, and in the future, can actively assist in the water supply's daily operation and supply of drinking water. With the many reservoirs and increasing water quality challenges – not just in Texas, but in large parts of the USA – WaterWebTools sees good opportunities for the concept of forecast information to be spread more widely in the U.S. market for the benefit of water supply companies and consumers.