The quiet risk inside drinking water systems

Drinking water treatment plants are becoming more complex at the exact moment our industry is losing the people who understand how they work end-to-end. When I began my career as a process engineer, it was common to gain hands-on experience across the entire treatment train, understanding how decisions in one part of a plant influence everything downstream.

Today, due to the way available research money has influenced early career education, many engineers specialise at the start of their careers, focusing on a single treatment technology that they may have studied during their academic training. That shift may improve efficiency in project delivery, but it also carries a quieter risk: the gradual loss of the holistic understanding required to keep drinking water systems operating reliably and sustainably into the future.

When I entered the profession in the late 1980s, the industry was undergoing a period of significant regulatory change. The US Environmental Protection Agency was developing a new framework of drinking water rules, including the Lead and Copper Rule and the Surface Water Treatment Rule (and its successor Microbial/Disinfection By-Product (M/DBP) cluster of regulations), which reshaped how treatment plants were designed and operated.

Engineering firms responded by building dedicated process teams, where engineers gained hands-on experience across multiple treatment technologies while working on new plants and major upgrades. Over time, the pace of regulatory change slowed and the focus shifted towards optimisation and cost efficiency. That shift has driven greater specialisation, with engineers often working within narrower roles to deliver projects more quickly and with less labour. From conversations with peers across the industry, this is not unique to one region. The same pattern is playing out in other mature water markets, including the UK.

Early in my career, I had the opportunity to lead pilot testing work for New York City, which fundamentally shaped how I understand water treatment. Pilot plants allowed us to test ideas in real time, experiment with different process configurations and see how small changes in raw water quality or chemical dosing could affect overall performance.

It was one thing to understand these interactions in theory but seeing how they actually behaved in practice was very different. Being able to push processes beyond conventional assumptions, and sometimes to the point of failure, provided insights that are difficult to gain in any other way. That experience reinforced a simple point: treatment plants do not operate as isolated steps, but as systems that must be understood as a whole.

Water treatment plants are often described as a sequence of individual processes, such as coagulation, clarification, filtration and disinfection. In practice, they behave more like interconnected systems. Decisions made upstream influence everything that follows, from how effectively particles are removed to how disinfectants perform and how water behaves in the distribution system.

Without understanding how these physical, chemical and biological interactions play out, it is easy to create treatment trains that do not quite fit together. Plants may perform well under stable conditions but become much harder to manage when those conditions change.

That level of system understanding is becoming more important, not less. Climate variability is pushing many treatment plants beyond the assumptions used during their original design. Flood events can introduce sediment loads that plants were never intended to handle, while warmer temperatures and increased nutrient loading are driving more frequent and severe harmful algal blooms.

At the same time, contaminants of emerging concern such as PFAS are forcing utilities to add new treatment steps, often at significant cost. These technologies work, but how well they perform depends heavily on what is happening upstream. Getting the best out of them requires a clear understanding of how the whole system behaves, not just how individual components perform.

The industry will continue to have engineers who understand individual processes. What may become harder to find is the practical experience needed to understand how those processes work together. Over time, experienced engineers develop ways of thinking that go beyond individual technologies.

They learn how to diagnose complex interactions, how to select approaches that meet multiple objectives and how to design plants that can adapt to changing conditions. Many of the engineers who developed those instincts are now approaching retirement. At the same time, a generation of academics who taught this broader approach is also leaving the profession. What we risk losing is not just the technical skill set. It is the way of thinking that allows engineers to understand how an entire treatment system works together.

There is, however, an opportunity to preserve that expertise in new ways. Over the course of my career, I have developed mental frameworks for interpreting water quality data and working out what types of treatment approaches are likely to perform best.

More recently, I have begun mapping those decision-making processes, identifying the rules, constraints and questions that guide those choices. By translating that thinking into structured decision trees and models, it becomes possible to retain not just the outcome of those decisions, but the reasoning behind them.

Machine learning tools can also help analyse operational data to improve plant performance, but they need to be used carefully. In drinking water systems responsible for public health, human judgement still matters.

Preserving system-level expertise will require deliberate action:

• Create opportunities for broad, hands-on experience
  Young engineers need exposure to multiple treatment processes, not just narrow specialisations, to build real system understanding.
• Reinvest in pilot testing and controlled experimentation
  Engineers need safe environments where they can test assumptions and learn from failure.
• Embed mentorship that explains both how and why
  Passing on experience means sharing decision-making logic, not just process steps.
• Capture expert knowledge using structured frameworks and digital tools
  Decision trees, design logic and operational insights should be documented and embedded before they are lost.
• Support industry and academic collaboration on training infrastructure
  Innovation hubs and permanent pilot facilities can help rebuild the practical learning environments that once developed generalists.

Over the next decade or more, drinking water systems will face increasing pressure from new contaminants, climate-driven changes in source water, ongoing regulatory requirements and funding challenges.

At the same time, much of the current generation of experienced engineers and academics will transition out of the workforce. New technologies will continue to play an important role, but they will not be enough on their own.

Capturing and transferring the knowledge of experienced professionals will be just as important as the technologies we deploy. The tools to do this already exist. What remains is the willingness to invest the time and effort required to capture that expertise before it disappears.

Drinking water systems depend on more than infrastructure and regulation. They depend on understanding how treatment processes actually behave together under real-world conditions. Preserving that knowledge is critical to keeping those systems working reliably.