Engineering & Technology · Environmental Engineering & Resource Protection

Monitoring, lifecycle assessment, uncertainty and environmental justice

A regional average improves after a project, yet one neighbourhood’s local conditions worsen. A technically correct average can still answer the wrong question for the people bearing the change. Environmental assessment needs both sound measurements and a clear account of whose service, exposure and choices are represented.

Use Process Balances & Design and Fluid Engineering for flow, load, storage and transport. Chemistry’s solutions and concentration and Biology’s ecosystems and biogeochemistry supply further preparation. No example establishes drinking-water suitability, discharge permission, site cleanup completion or a health threshold.

Back to the chapter

After reading, you can

Design decision-relevant monitoring and compare lifecycle and distributional consequences without hiding uncertainty or unequal burdens in an average.

A data plan begins with the decision

State the quantity, location, time window, population or ecosystem and comparison that matter. A sample’s analytical precision does not establish that its place and time represent the intended boundary. Plan methods, preservation, instrument checks, blanks, duplicates and data review as appropriate to the question. Preserve identifiers, conditions, exclusions and reported limits.

A nondetect is a result of a defined detection or reporting process, not proof of a zero concentration. For example, a supplied report of “less than 0.01 mg/L” must retain that qualifier. Averaging it as an exact zero would add unsupported information. Interpret apparent changes alongside method consistency, sampling coverage, natural variation and uncertainty.

Follow impacts across the lifecycle

Lifecycle assessment defines a comparable function, system boundary and inventory of relevant flows, then uses appropriate models to consider potential impacts. Treating one cubic metre of wastewater to a different quality is not automatically equal service. A lower energy total can coexist with greater chemical use, sludge transport or another impact category.

Data quality and omitted stages can change a comparison. State allocations, replacement assumptions, operating life, geography and the relevant uncertainty or scenarios. An inventory flow is not itself a complete ecological or health effect. The same mass emitted at different places and times can involve different pathways and receptors; a total should not erase those distinctions.

Make a bounded lifecycle comparison

An original classroom comparison gives both fictional treatment routes the same influent and the same service: deliver 1000 m³ of water at the same stipulated outlet quality, location and time. This is a supplied equivalence, not evidence that either real treatment achieves it. Allocate construction and consumable-production electricity to that service; include operating electricity and electricity for final residue handling. Count final dry residue once at its final management destination, rather than counting its internal transfers again.

The table gives the electricity inventory. Route A totals 80+500+20=600 kWh per service; B totals 160+300+40=500 kWh. B is lower by 100 kWh, or 100/600≈16.7% relative to A. Separately supplied final dry-residue totals are 40 kg for A and 70 kg for B. Thus B uses less electricity but leaves 30 kg more dry residue in this account. Kilograms cannot be added to kilowatt-hours to manufacture an overall score.

This boundary follows selected flows across lifecycle stages; it is not a complete impact assessment. Fuel use, the electricity supply mix, residue composition, releases to air and water, land effects and local exposure are not supplied. They are missing, not assumed zero. A kilogram of one residue need not have the effect of a kilogram of another. Report the two inventory results, the omissions and the service assumption before considering potential impacts or affected people.

Original electricity inventory for one equal 1000 m³ service; all values are invented kWh, not impact scores.
Lifecycle contribution	Route A (kWh)	Route B (kWh)
Allocated construction and consumable production	80	160
Operation	500	300
Final residue handling	20	40
Total per equal service	600	500

Who experiences the change matters

Our original two-area example supplies group-average air concentrations, with fixed populations of 900 in A and 100 in B. Before, A is 10 and B 30 µg/m³; after, A is 5 and B 35 µg/m³. Their population-weighted concentration summary falls from 12 to 8 µg/m³. Yet B’s supplied mean rises by 5, or about 16.7%. These invented values describe neither a real population nor personal dose.

Environmental justice asks how benefits, burdens and opportunities for meaningful involvement are distributed. An overall improvement does not decide whether an increased local burden is acceptable or whether people had a fair opportunity to shape the alternatives. Include affected communities’ knowledge early, communicate what is open to influence and explain how concerns changed the decision. A public meeting count alone is not evidence of that influence.

Area A population 900: mean falls 10 to 5 µg/m³. Area B population 100: mean rises 30 to 35. Population-weighted summary falls 12 to 8, but B rises 16.7%. Same common concentration axis used for all rows. Invented area means and weights imply no real personal exposure or health conclusion.

Report the average and the unresolved burden

The before weighted summary is (900×10+100×30)/1000=12 µg/m³. After, it is (900×5+100×35)/1000=8 µg/m³, a one-third decrease. The four area means are all retained in the figure, so the rise in B remains visible rather than disappearing inside the summary.

This calculation assumes the area means and population weights are appropriate for that specified summary. It does not establish individual exposure, causation, health risk or fairness. A useful project review would investigate B’s changed conditions, alternatives and missing evidence, alongside the full lifecycle effects and the views of people whose circumstances the average conceals.

Try the reasoning

Can more precise laboratory readings repair an unrepresentative sampling plan?
Show the answer and reasoning
No. Analytical precision and coverage of the relevant population, place and time are different properties.
Does the lower weighted average establish improvement for both areas?
Show the answer and reasoning
No. Area B rises from 30 to 35 µg/m³ in the supplied example.
What makes participation meaningful rather than a count of events?
Show the answer and reasoning
People can understand the decision, contribute while alternatives remain open and see how their input was considered.
Keep the same 1000 m³ service, but change B’s operating electricity from 300 to 450 kWh; its other entries and both dry-residue totals stay fixed. What comparison follows, and which conclusion remains unsupported?
Show the answer and reasoning
B now totals 160+450+40=650 kWh, exceeding A’s 600 by 50 kWh (about 8.3% relative to A). The electricity ordering reverses under this changed assumption. The stipulated dry residues remain 70 versus 40 kg. These two inventories still cannot establish an overall health, ecological or fairness ranking: the missing flow, composition, supply and exposure evidence must be retained, and equivalent delivered service must actually be verified in a real comparison.

What to carry forward

Build monitoring around the decision, compare equivalent lifecycle service and preserve uncertainty, unequal outcomes and meaningful participation.

References

Teaching resources and research supporting this lesson’s explanations and examples.

U.S. EPA4 sources