How big is the gap between lab predictions and real-world emissions?
The gap can be enormous. A 2021 study tested ten construction machines in Nanjing, China, using a portable emissions measurement system to capture real-world emissions of carbon monoxide (CO), hydrocarbons (HC), nitrogen oxides (NOx), and particulate matter (PM) during idling, moving, and working modes [1]. The official Chinese guidebook for emission inventories underestimated real-world emissions by 24% for CO, 120% for NOx, and 66% for PM on newer Stage II machines. For older Stage I machines, the underestimation soared to 126% for CO, 1,066% for NOx, and 559% for PM [1]. This means that even with modern engine technology, real-world conditions—like machine deterioration and varying operational modes—can cause emissions to be more than double what models predict, and for older equipment, more than ten times higher.
This gap matters for fusion pilot plants because they will also operate under real-world conditions—variable loads, startup/shutdown cycles, and aging components. The same study found that working mode (the most demanding) produced the highest emissions for all pollutants except CO, while idling mode had the least variation but 43% higher average CO emissions than moving mode [1]. So, a fusion plant's emission profile will depend heavily on how it's operated and maintained, not just its design specs.
Does fusion plant design matter more than real-world operation?
Design is critical, but it doesn't guarantee real-world performance. A 2019 study on compact spherical tokamak fusion pilot plants argued that smaller, high-performance reactors using high-temperature superconductor magnets could be feasible, potentially offering a faster route to fusion power [4]. This design-focused work suggests that fusion plants could be built to be inherently efficient and low-emission. However, the real-world evidence from the construction machine and CO2 capture studies [1][2] shows that even well-designed systems can underperform if not operated and maintained correctly. For example, the CO2 capture study found that nitrosamine and nitramine releases were below detection limits—a design success—but heat-stable salt accumulation was a persistent operational issue [2].
The construction machine study [1] also highlighted that stricter emission standards (Stage II vs. Stage I) reduced real-world emissions significantly, but the gap between predicted and actual emissions remained large. This suggests that while better design (like tighter standards) helps, it doesn't eliminate the need for real-world validation and adaptive management. For fusion pilot plants, this means that even the best design must be paired with robust monitoring and maintenance protocols to ensure emissions reductions are realized in practice.
About These Sources
This answer is built on 5 peer-reviewed studies — published from 2012 to 2023, 2 in Q1 journals, collectively cited 108 times — selected as the most relevant from 5 studies that passed quality screening, drawn from 68 papers retrieved from a database of over 500 million.
Sources used in this answer
Real-world emissions of construction mobile machines and comparison to a non-road emission model
Real-world tests of ten construction machines in Nanjing found that official emission models underestimated NOx by 120% for newer engines and up to 1,066% for older ones, showing a large gap between lab predictions and actual performance [1].
Solvent degradation and emissions from a CO2 capture pilot at a waste-to-energy plant
A mobile CO2 capture pilot at a Danish waste-to-energy plant showed that an acid wash removed up to 83% of gaseous degradation products from flue gas, but heat-stable salt accumulation was a persistent operational challenge [2].
Characterizing emission factors and oxidative potential of motorcycle emissions in a real-world tunnel environment
A tunnel study in Taipei measured real-world emission factors for motorcycles, light-duty, and heavy-duty vehicles, finding that fresh traffic emissions were less toxic than aged aerosols but still posed health risks [3].
Towards a compact spherical tokamak fusion pilot plant.
A design study argued that compact spherical tokamak fusion pilot plants using high-temperature superconductor magnets could be feasible and offer a faster route to fusion power [4].
Nitrous oxide emissions from the oxidation tank of a pilot activated sludge plant.
Continuous monitoring of a pilot wastewater treatment plant showed that nitrous oxide emissions were highly variable, with low dissolved oxygen and high nitrogen loads causing peak emissions [5].