Design a constructed wetland that uses plants, soil and microbes to polish wastewater. Adjust the flow, inlet water quality, wetland type and temperature to see the required area, HRT and removal efficiency update in real time using the Kadlec-Knight k-C* model — and size a low-cost piece of green infrastructure.
Parameters
Flow rate Q
m³/day
Influent BOD Ci,BOD
mg/L
Influent TN Ci,TN
mg/L
Target effluent BOD Ce,BOD
mg/L
Discharge limit. EU secondary treatment is BOD ≤ 25 mg/L
Influent (left) flows through the rhizosphere and gravel medium (HSSF). Microbial activity removes BOD particles along the way. Colour shows removal progress (red → orange → green).
HRT: hydraulic retention time (day). d: effective depth (≈0.5 m for subsurface flow), n: porosity (≈0.4 for gravel). η: removal efficiency. Compute A for BOD and TN separately and use the larger value.
人工湿地 (Constructed Wetland) による排水処理設計
🙋
A constructed wetland sounds like "plant some reeds in a pond and pour wastewater on top". Does that really treat sewage?
🎓
Visually that's it, but the inside is a tidy biology-chemistry-physics machine. Emergent plants like Phragmites or Typha carry dense biofilms on their roots, and as the water passes through, (1) heterotrophic microbes oxidise BOD, (2) nitrifiers convert ammonia NH4 to NO3, and (3) denitrifiers turn that NO3 into N2 gas that escapes to the atmosphere. So it's really "natural self-cleaning, packaged into a designed vessel". Because gravity drives the flow there are no aerators or sludge pumps, so operating costs can be one-tenth of conventional activated sludge.
🙋
If it's that cheap, why don't all wastewater plants use constructed wetlands?
🎓
The biggest drawback is the land footprint. Try the defaults on the left — 100 m³/day (about 500 people equivalent) already needs about 3,700 m² of wetland, roughly half a soccer field. That's hopeless in central Tokyo or New York. Constructed wetlands shine for small villages, agricultural runoff, stormwater treatment, or as a polishing stage after a city plant — situations where "land is available but money and energy are not".
🙋
Switching between HSSF, VSSF and FWSF changes the area a lot. How do I pick the right one?
🎓
It depends on the goal and the site. FWSF is a shallow pond — beautiful and great for biodiversity, but it needs the largest footprint. HSSF flows horizontally through gravel, is hidden underground, and is strong for BOD removal. VSSF doses water from the top so the bed re-aerates between pulses — that's why it's by far the best for nitrification (NH4 → NO3). In practice you often see a hybrid system: VSSF first (nitrify), then HSSF (denitrify). Look at the "Required area by wetland type" chart to compare them.
🙋
When I dropped the temperature from 15 °C to 5 °C the required area jumped. How do you operate one in winter?
🎓
That's the classic objection to constructed wetlands in cold climates. Microbial kinetics follow the Reed equation k_T = k_20 × 1.06^(T-20), so going from 20 °C to 5 °C drops k to about 0.42×, meaning roughly 2.4× more area. In Hokkaido or northern Europe you must design for the coldest monthly water temperature (worst-case design). The snow cover actually insulates the bed surprisingly well, but if you don't oversize for winter the spring melt overloads the system. Swedish and Finnish design guidelines treat this winter k-correction as mandatory.
🙋
And what does "7.4 m²/PE per person" actually mean — is that a lot or a little?
🎓
International rules of thumb: HSSF for BOD removal only needs about 3-5 m²/PE, secondary-grade treatment 5-10 m²/PE, and including nitrogen removal 10-20 m²/PE. So 7.4 m²/PE is reasonable when nitrogen is included. Denmark's Riemann guideline recommends 5 m²/PE for HSSF, while Vymazal (Czech Republic) suggests 1-3 m²/PE for VSSF. For small communities below 500 people, this "m²/PE" is the single most important indicator — it sets both the construction cost and the land you have to secure.
Frequently Asked Questions
Use the Kadlec-Knight k-C* first-order plug-flow model: A = (Q/k_T)·ln(C_i/C_e), where Q is the flow rate (m³/day), C_i is the inlet concentration, C_e is the target outlet concentration and k_T is the temperature-corrected areal rate constant (m/day). The Reed equation k_T = k_20·1.06^(T-20) accounts for slower microbial activity in winter. Compute A for both BOD and total nitrogen (TN) and take the larger value.
FWSF (Free Water Surface Flow) keeps water above ground in a shallow pond — high biodiversity and low maintenance but it freezes in cold climates and needs the largest area. HSSF (Horizontal Subsurface Flow) sends water horizontally through gravel — a mixed aerobic/anaerobic environment that is strong for BOD removal but weaker for nitrification (NH4 to NO3). VSSF (Vertical Subsurface Flow) doses water from the top and lets gravity infiltrate it — oxygen-rich, the most area-efficient for nitrification, but it needs dosing pumps and intermittent loading cycles.
Pilot-scale measurements are best, but typical literature values for BOD removal are HSSF 0.06-0.2 m/day, VSSF 0.5-1.0 m/day and FWSF 0.03-0.1 m/day. The default 0.3 m/day in this tool represents a mid-range VSSF wetland. Apply the Reed correction 1.06^(T-20): a 10 °C drop multiplies the required area by about 1.79. The TN rate constant is one order of magnitude smaller, typically about 0.05 m/day at 20 °C.
For subsurface flow wetlands HRT = (A·d·n)/Q, where d is the depth and n the porosity (about 0.4 in gravel). 3-5 days is typical for BOD removal and 5-10 days for nitrogen (denitrification). If HRT drops below 3 days microbes don't have time to mineralise the load and the effluent exceeds target. This tool warns when HRT < 3 days and rates the design OK above that. FWSF wetlands typically achieve 1-2 times longer HRT at the same depth, which helps in cold climates or with recalcitrant pollutants.
Real-world applications
Small communities and decentralised sewerage: villages and islands of 50-2,000 people often can't justify the cost of extending sewers to a central plant, so constructed wetlands are the first choice. Denmark, southern Germany, the Czech Republic and France have run national programs since the 1990s, with thousands of installations operating today. In Japan, agricultural-village wastewater systems often use constructed wetlands as secondary treatment and discharge to receiving farmland.
Tertiary polishing of municipal plants: a one-hectare wetland appended to a conventional plant can shave another 30-60% off the residual BOD, TN and SS — a powerful tool against eutrophication in closed water bodies such as lakes and inner bays. Examples include Lake Kasumigaura and Lake Biwa in Japan, and Florida's Everglades Stormwater Treatment Area (over 40,000 ha) — one of the largest constructed wetlands in the world.
Agricultural and livestock wastewater: piggery and dairy effluents have BOD of 1,000-5,000 mg/L and TN of 200-500 mg/L, so activated sludge becomes expensive. Hybrid VSSF + HSSF systems are common, and the treated water can be recycled as irrigation, giving a triple benefit of water, nutrients and plant biomass (such as harvested reeds).
Stormwater and road-runoff treatment: urban road runoff carries heavy metals, oil and tyre dust that pollutes rivers if discharged untreated. Placing an FWSF wetland in a low-lying area provides both peak-flow attenuation and pollutant removal at once. The US EPA's Green Infrastructure Plan recommends this approach, and it has been adopted widely in Seattle, Portland and the Tokyo waterfront.
Common misconceptions and pitfalls
The biggest trap is "taking the maximum literature value for k". Papers often highlight conditions favourable to good performance (warm summer, young wetland) and quote k = 0.5-1.0 m/day. In winter, or in a wetland clogged after 10 years of operation, k can drop below half. Always design with the coldest monthly water temperature and an end-of-life k. This tool uses a conservative default of k = 0.3 m/day; in cold climates or with high organic loads drop it to 0.1 m/day. Kadlec and Wallace's "Treatment Wetlands, 2nd ed." (2009) is the worldwide reference.
Next, the assumption that "nitrogen removal comes for free with BOD removal". In reality denitrification is an order of magnitude slower than BOD removal, with k_TN ≈ 0.05 m/day. In this tool the TN-driven area is usually more than four times the BOD-driven area. Denitrification also needs an electron donor (organic carbon), so if BOD is over-removed denitrification stalls too. Where nitrogen limits are tight, build a multi-stage system: VSSF (nitrify) → HSSF (denitrify) → FWSF (polish). It's flowsheet design, not just adding more area, that gets you there.
Finally, the myth that "the plant species makes a huge difference". Across 20+ years of comparative studies the difference between Phragmites, Typha and Cyperus is only about 10-20%, nowhere near doubling the area or k value. The plant's main jobs are (1) oxygen transport through the roots, (2) providing surface area for biofilm, and (3) visual landscape value — direct uptake of pollutants accounts for only a few percent. Pick fast-growing native species, and assess invasion risk carefully before planting alien species (Phragmites is itself invasive in North America).
How to Use
Enter influent flow rate (m³/day) and initial BOD concentration (mg/L) using the sliders or numeric inputs for qNum, qRange, bodNum, and bodRange.
Set total nitrogen input (mg/L) via tnNum and tnRange, then specify your target BOD effluent level (mg/L) using targetBodNum and targetBodRange.
The simulator calculates required wetland area (m²/ha), hydraulic retention time (HRT in days), BOD removal percentage, and construction cost (USD) based on soil type, plant species, and microbial kinetics.
Worked Example
Design a wetland for a municipality treating 500 m³/day with influent BOD of 280 mg/L and total nitrogen of 45 mg/L, targeting 30 mg/L BOD effluent. The simulator returns: required area 2.8 ha (0.56 m²/PE for 5,000 population equivalent), HRT of 4.2 days using Phragmites australis beds, BOD removal of 89%, and land cost of USD 168,000 (USD 60/m²). Adjust surface loading rate or depth to optimize performance.
Practical Notes
Free water surface (FWS) wetlands require 5–8 m²/PE but perform better for TN removal; subsurface flow systems need 1–2 m²/PE with lower mosquito risk.
Cold climates (below 10°C) reduce microbial activity by 40–50%; increase HRT from 3 days to 5–6 days or add supplemental heating zones.
Media porosity (40% sand/gravel) and plant density (6–10 shoots/m²) directly control oxygen transfer; insufficient planting reduces BOD removal below 70%.
Combine wetland output with UV disinfection if pathogen removal below 2 log units; fecal coliform reduction in wetlands alone is typically 1–2 logs.