Two real, independent laws combine inside every UV disinfection reactor:
Beer-Lambert (light absorption by water):
I(r) = I₀ · e^(−α·r), α = −ln(UVT) / L
Chick-Watson (microbial inactivation vs. dose):
D = ∫ I(r) dt [UV dose, mJ/cm²]
log₁₀(N/N₀) = −D / D₉₀ [D₉₀ = dose for one log of kill]
Each glowing particle is a microorganism drifting past the central UV-C lamp. Its distance r from the lamp sets the intensity it receives via Beer-Lambert absorption — particles that pass close to the lamp, or through clearer water (higher UVT), pick up dose fast. That dose accumulates over the particle's residence time and is converted into a log-inactivation via the Chick-Watson dose-response line; particles fade from live (cyan) to inactivated (dim grey) as they cross the regulatory 4-log threshold.
- Lamp UV-C intensity — output at the lamp surface, mW/cm² at 254 nm.
- Water transmittance (UVT) — the fraction of UV-C light that survives a 1 cm path through the water; turbid or organic-rich water has a lower UVT and shields organisms from the lamp.
- Flow rate — faster flow means less residence time under the lamp and a lower dose per pass.
- Target organism — sets D₉₀. Illustrative values used here (mJ/cm² per log₁₀ reduction, ordered from EPA UV Disinfection Guidance Manual dose tables): E. coli ≈ 1.3, Giardia ≈ 1.9, Cryptosporidium ≈ 2.9, Adenovirus ≈ 47. Adenovirus is why real reactors are sized for far higher dose than bacteria alone would need — it is markedly more UV-resistant, while Cryptosporidium and Giardia (both notoriously chlorine-resistant) are actually easy to inactivate with UV.
This is a simplified linear (single-slope) Chick-Watson model for teaching clarity; real validated dose-response curves often show a flattening "tailing" effect at very high log-reductions that this simulation does not reproduce.