Environmental Engineering: Clean Water In, Clean Water Out

You drink a glass of water from a tap and never think about it. That glass was a river or an aquifer a day or two ago, carrying mud, microbes and farm runoff. A plant you have never seen coagulated, filtered and disinfected it, and a pipe network kept it clean for a long walk to your kitchen. You flush the toilet, and the same thing happens in reverse.

Taken together, this is the quiet machine that made cities safe to live in. Before it, cities were where people went to die of cholera and typhoid. This course teaches the one tool that runs the whole machine, and then walks through its main parts: water, wastewater, air and waste.

Not medical or regulatory advice: this piece explains how engineering works. Local drinking-water rules differ by country.

Mass balance: the one tool that runs everything

Matter is not created or destroyed, so for any region you can draw a box around, what builds up inside equals what comes in, minus what goes out, plus what is made, minus what is destroyed. That is a mass balance. A lake, a treatment tank, a room, a river reach, a landfill and the atmosphere over a city are all boxes. The only questions are which flows and reactions you include.

This is the "well-mixed tank" version. Two numbers describe it. The residence time is V/Q, the time to flush one tank-volume through. And at steady state, when build-up stops, the concentration settles at C_in · Q / (Q + kV): it is lower if the water is flushed fast or the pollutant decays fast. The same equation shows why a small lake with a slow outflow is easily poisoned, and why it takes so long to recover: the clean-up follows the same exponential curve down, at the same rate.

Drinking water: stack the barriers

River water is a hazard soup: clay, algae, organic matter, bacteria, viruses, protozoa such as Giardia and Cryptosporidium, and chemicals. No single step removes all of it, so a plant is a chain of barriers, each catching what the others miss. The classic chain is five steps.

1. Coagulation. Tiny particles carry a charge that makes them repel each other and stay suspended for ever. Add an aluminium or iron salt and the charge is neutralised.
2. Flocculation. Gentle stirring lets them collide and glue into flakes (flocs).
3. Sedimentation. In a quiet tank the heavy flocs settle out.
4. Filtration. What is left passes through sand or other media, which catches the finest particles, including most protozoa.
5. Disinfection. Chlorine, chloramine, ozone or ultraviolet light inactivate what remains, and a small chlorine residual guards the pipes all the way to your tap.

Success is counted in logs. One "log" removal means 90% gone; two logs, 99%; four logs, 99.99%. Regulators set targets in logs for each organism class (the US surface-water rules, for example, require 3-log for Giardia and 4-log for viruses overall), and every barrier gets credit for its share. Disinfection follows a simple rule of thumb (Chick–Watson): kill, in logs, grows with the concentration of disinfectant times the contact time, called CT. The catch is that organisms differ by orders of magnitude. Viruses fall to a little chlorine; Cryptosporidium shrugs off any practical dose, so for it the plant relies on filtration and UV. And chlorine reacts with natural organic matter to form by-products, so the dose is a balance of two risks, with the harm from leaving microbes in the water much larger. That balance is why the system uses several barriers rather than one big dose.

John Snow, 1854, and the quiet machine

In the 1850s most doctors held that cholera came from miasma, foul air from rot and filth. In late August 1854 an outbreak began in Soho, London, around Broad Street. The physician John Snow, who suspected the water, went door to door, marked the deaths on a map, and found them clustered around one hand pump. The Reverend Henry Whitehead, a local curate who had doubted him, helped trace the source to a household where a baby had cholera and a leaking cesspool near the well. The council removed the pump handle on 8 September.

Here the popular story needs two corrections. The outbreak was already declining before the handle came off, and Snow himself said so. And the authorities did not accept his theory at the time: the Board of Health blamed miasma. His stronger evidence was a "natural experiment" from the same year. Two water companies supplied the same streets, house by house; one, Southwark and Vauxhall, drew its water from the sewage-laden lower Thames, while the other, Lambeth, had moved its intake upstream. Households supplied by the first suffered many times more cholera than those by the second, with the poor and the rich on both. Over the following decades London built its sewers and filtered its water, and cholera retreated. Hundreds died in the Broad Street outbreak itself (the number depends on the area and dates counted; one standard summary gives 420).

The same story is still being written. The WHO/UNICEF Joint Monitoring Programme reported in August 2025 that about 2.1 billion people, roughly one in four, lacked safely managed drinking water in 2024, including 106 million drinking directly from untreated surface water. It also noted that 2.8 billion people gained safely managed sanitation between 2000 and 2024. The engineering for what remains is known. The hard parts are money, maintenance, institutions and trust.

Wastewater and the river's oxygen sag

What goes down the drain is mostly water, with organic matter that bacteria will eat. Treating it is a copy of what a healthy river does, in a tank, faster. Primary treatment lets solids settle. Secondary treatment (usually the activated-sludge process) bubbles air through a tank of microbes that consume the dissolved organics. Tertiary steps remove nitrogen and phosphorus and disinfect. The measure of how much "food" the water carries is the BOD (biochemical oxygen demand): the oxygen that microbes will use up, in mg/L, usually measured over 5 days at 20 °C.

Why does that matter? Because the microbes breathe the river's oxygen. Put a polluted outfall into a river and two things happen at once. Bacteria consume oxygen as they eat, so dissolved oxygen (DO) falls. The air tops the river back up through its surface at a rate proportional to the deficit from saturation. At first the eating wins and DO sags; downstream the food runs out and the air wins, so DO recovers. That curve is the oxygen sag, described by H. W. Streeter and Earle Phelps in 1925 from Ohio River data collected in 1914–15.

Fish begin to struggle somewhere below 4 to 5 mg/L. Temperature hits this twice. Warm water holds less oxygen at saturation (about 9.1 mg/L at 20 °C and 7.6 at 30 °C, in fresh water at sea level), and warm bacteria eat faster. The same sewage that a river handled in cool weather can push a warm river to zero.

Flint, 2014: when the chemistry was changed and the pipes were not

On 25 April 2014 the city of Flint, Michigan, as a cost-saving step, switched its drinking water from Detroit's supply to the Flint River. The river water was more corrosive, and the city did not add the corrosion inhibitor (a phosphate) that would have helped the old pipes keep their protective coating. Lead pipes and solder in older homes began to leach lead. Treatment plants do not only clean water; they set its chemistry so that it does not eat the pipes it travels through.

What followed is a case study in how a system fails. Residents complained of brown, foul water. In 2015 a Virginia Tech team led by Marc Edwards found that at least a quarter of sampled households exceeded the federal action level for lead of 15 parts per billion, and some homes far more. The paediatrician Mona Hanna-Attisha reported that the share of children with blood lead above 5 µg/dL had risen from 2.4% to 4.9% (in the highest-risk areas, 4% to 10.6%); a later 2018 follow-up found a smaller rise that was not statistically significant after correction, so the size of the effect is itself debated, and this is a good example of a small-numbers problem. There was also an outbreak of Legionnaires' disease in the county, with 87 cases and 12 deaths between June 2014 and November 2015; the state never confirmed a definitive link to the water, though later studies pointed to low chlorine levels. Flint returned to Detroit's water on 16 October 2015. An EPA report in 2017 placed most of the blame for the oversight failures on the state's environmental agency. The last lead pipe in the city's replacement programme was replaced on 1 July 2025, as reported by the Wikipedia article on the crisis.

Take three lessons from it. A chemistry change is an engineering change. Regulation that only tests a few samples can miss a problem that families can smell. And data from people who live there should be taken seriously early.

Air pollution: a plume spreads sideways and up

Smoke from a chimney is a mass balance in motion. The wind carries it downwind and turbulence spreads it sideways and upwards, so the concentration at any point falls the further it goes, and a bell-shaped (Gaussian) profile fits both directions. Ground-level concentration behind a stack depends on the emission rate, the wind speed (more wind, more dilution), the stack's effective height, and above all on how turbulent the air is. In a sunny, unstable afternoon the plume spreads fast; on a calm clear night in a stable layer it stays thin and travels far.

Raise the stack and the worst ground-level concentration falls steeply, since the plume has to spread much more before it reaches the ground. In the toy below, doubling the stack from 60 m to 120 m in neutral air at 5 m/s cuts the peak by a factor of about 5. But the lesson cuts both ways: a tall stack sends the problem further away rather than removing it. That is why modern rules also limit emissions at the source and why acid rain crossed borders. "Dilution is the solution to pollution" has been abandoned for a reason.

Solid waste: what we throw away still exists

Mass balance applies to the bin too. Everything that is bought is eventually thrown away, burned, recycled or composted, and nothing disappears. The standard ladder, the waste hierarchy, ranks the options: prevent waste, reuse, recycle, recover energy, and only then dispose. A modern engineered landfill is a lined, drained, capped box whose rotting waste makes methane, so the engineer collects the gas (some plants burn it for power) and treats the liquid that seeps out, the leachate.

The World Bank estimated global municipal solid waste at 2.56 billion tonnes in 2022 and projected 3.86 billion by 2050 if nothing changes, a rise of about 50%. In low-income countries and across Sub-Saharan Africa and South Asia, the great majority is openly dumped, with collection rates in some regions as low as 31% to 67%. The world's waste problem is largely a collection and funding problem, then a design problem. Recycling rates differ widely, and what "recycling" counts depends on the country's rules, so check any number's definition before comparing.

What climate changes

Environmental engineers design for a "design storm", a "design drought", a "design river temperature". Climate change moves each of them. Warmer water holds less oxygen and speeds the bacteria, so the sag deepens. A warmer atmosphere holds more moisture, so heavy downpours are heavier, which can overwhelm the combined sewers that carry rain and sewage in the same pipe in many older cities, spilling untreated sewage into rivers. Drought leaves less water to dilute a discharge. Rising seas push salt into rivers and aquifers. Treatment itself has a footprint: aeration is energy-hungry, and wastewater and landfills release methane and nitrous oxide. The graduate recognises that the past record no longer sets the design for the future, the same point as in Hydraulics & Water Resources (34/67).

What a graduate just sees

• Draw a box; ask what comes in, what goes out, what is made, what is destroyed. Everything else follows.
• Residence time V/Q tells you how fast a lake, a tank or a room clears itself, and how slowly it recovers.
• No single barrier cleans water: coagulation, filtration and disinfection each catch what the last one missed.
• Treatment is measured in logs, and organisms differ enormously: what kills a virus may do nothing to Cryptosporidium.
• A river is a reactor: waste eats oxygen, air brings it back, and the lowest point is downstream of the outfall, not at it.
• Warm water holds less oxygen and bacteria work faster: summer is the dangerous season.
• Water treatment sets chemistry as well as cleanliness, and Flint shows what happens when you ignore the pipes.
• A tall stack reduces nearby pollution but only moves it; the fix is at the source.

Where it shows up next

• Hydraulics & Water Resources (34/67): the supply, the river's flow and flood risk that decide how much dilution a river offers.
• Corporate Finance (34/51, engineering economics) and Systems Engineering (34/22): treatment is a choice between costs, risks and who pays.
• Chemistry and biology sit under every step here, and a graduate often moves toward public health, ecology or policy.
• The real world: municipal water and sewer utilities, industrial wastewater, remediation of contaminated sites, air-permit consulting, and humanitarian water supply after disasters.

If you want to go deeper

• 📘 Textbook: Mackenzie L. Davis and David A. Cornwell, Introduction to Environmental Engineering (McGraw-Hill): the standard survey of water, air and waste.
• 📘 Textbook: James R. Mihelcic and Julie B. Zimmerman, Environmental Engineering: Fundamentals, Sustainability, Design (Wiley): mass balance first, with sustainability and global examples.
• 📘 Reference for the plant: Metcalf & Eddy, Wastewater Engineering: Treatment and Resource Recovery (McGraw-Hill).
• 📘 Gentler: Steven Johnson, The Ghost Map (2006): the Broad Street outbreak as a gripping story.
• ▶️ Free course: MIT OpenCourseWare 1.061 Transport Processes in the Environment (Fall 2008), an undergraduate course on how substances move through rivers and lakes (mass balance, diffusion and exchange). For treatment itself, 1.85 Water and Wastewater Treatment Engineering is a graduate course.
• 🌐 Easier first read in this library: Hydraulics & Water Resources (34/67) for the water side.