Skip to content

Mini-wetlands (MW) — Constructed Wetlands at Sub-catchment Level

Summary

Small constructed wetlands built to treat drainage water from agricultural sub-catchments. Unlike the field-level WL measure, MWs are indexed by their location (mw1/2/3) within an ID15 sub-catchment (ret). They intercept and denitrify N from the collective drainage of surrounding fields.

Three size classes

Class Catchment area N effect Fixed cost Land cost
MW1 20 ha mw1eff = 94.4 kg N/yr 27,270 DKK CostMW_ha_N(ret) × 0.20 ha
MW2 50 ha mw2eff = 236 kg N/yr 36,983 DKK CostMW_ha_N(ret) × 0.50 ha
MW3 100 ha mw3eff = 472 kg N/yr 53,171 DKK CostMW_ha_N(ret) × 1.00 ha

Note: N effects scale linearly with catchment area (94.4 × 1, 2.5, 5), suggesting a fixed loading rate assumption.

Decision variables

MW1x(mw1), MW2x(mw2), MW3x(mw3) ∈ {0,1}
Binary. Each location is either built or not.

N Reduction

MW_Red_ret(ret) = (Σ mw1eff×MW1x + Σ mw2eff×MW2x + Σ mw3eff×MW3x) × (1 − SurfRet_ID15(ret)/100)
Surface retention of the ID15 is applied (added Nov 2024). MW N reduction then adds to kr(k) via KystReductionN.

P Reduction

MW P effects are aggregated at ret level: 

MW_red_P(ret) = Σ_sizes [area × (P_Effects_MW_Matrix(ret) + P_Effects_MW_macropore(ret)) × MW_binary]
Both matrix and macropore pathways (each with a 0.45 factor). Only fields with macropore loss > 0 contribute to the ret-level averages.

Cost

MWCost_ret(ret) = Σ MW1x × (27,270 + CostMW_ha_N(ret) × 0.20)
                + Σ MW2x × (36,983 + CostMW_ha_N(ret) × 0.50)
                + Σ MW3x × (53,171 + CostMW_ha_N(ret) × 1.00)
CostMW_ha_N(ret) = area-weighted average prodcost(i) over all fields in the ret. (Gross margin added Dec 2024.)

Potential cap (VP2 linkage)

CatchmentMW(ret)..  MW_precalc1(ret) ≤ VP2_adapt(ret)
VP2_adapt(ret) = max(MW_precalc2(ret), VP2MW_per_ret(ret))
- MW_precalc2(ret) = natural potential from MiniPot files (dark green + 10% light/dark yellow) - VP2MW_per_ret(ret) = already implemented MW area from VP2

This prevents VP3 from selecting more MW than is physically or policy-appropriate, but ensures VP2 implementations are "grandfathered in."

Tripartite Agreement

MW_eq..  Σ [MW areas in catchments with N targets] ≥ 40,000 ha equivalent

Data sources

  • MW potentials: MiniPot1_darkgreen_new_2018.inc, MiniPot2_lightgreen_new_2018.inc, MiniPot3_darkyellow_new_2018.inc
  • MW indexing: setmw1.inc, setmw1ret.inc, setmw2.inc, setmw2ret.inc, setmw3.inc, setmw3ret.inc + VP2 variants (setmw*_VP2.inc, setmw*ret_VP2.inc)
  • VP2 MW area per ret: VP2MW_per_ret.inc
  • VP2 MW decisions: loaded from GDX file VP2_OCT2021_Raphael_collective_measures
  • N + P effects, costs: hard-coded in TargetEcon 2026.gms

Catalog source

DCA Rapport nr. 174 (Eriksen et al., 2020): Chapter "Minivådområder med åben vandflade" (p. 301–314). Authors: Carl Christian Hoffmann, Bo Vangsø Iversen et al. This chapter covers the open water surface (åben vandflade) type of mini-wetland.

(Note: A second chapter covers mini-wetlands with filter matrix (filtermatrice), covered in SR379 p.106–133. The model's MW measure appears to use open water type values.)

N effect — confirmed from catalog (Tabel 1): 13 Danish monitoring sites (2013–2018): - N loading: 512–5,096 kg TN/ha MW/yr - N removal: 112–1,007 kg TN/ha MW/yr - Efficiency: 13–37% - Mean: 472 kg TN/ha MW/yr at 22% efficiency

This translates directly to the model's three size classes: - mw1 (0.20 ha, 20 ha catchment): 0.20 × 472 = 94.4 kg N/yr ✅ - mw2 (0.50 ha, 50 ha catchment): 0.50 × 472 = 236 kg N/yr ✅ - mw3 (1.00 ha, 100 ha catchment): 1.00 × 472 = 472 kg N/yr ✅

All three model values match the catalog mean N removal exactly. ✅

Confidence: *** — "Estimaterne anses for rimeligt sikre og er baseret på et velafprøvet datagrundlag" (estimates are reasonably reliable, based on a well-tested data basis).

P effect — confirmed from catalog (Tabel 3): Same 13 sites: 45% P efficiency → 31 kg TP/ha MW/yr average. - The model uses 0.45 factor for both matrix and macropore P pathways from the MW — consistent with catalog.

Cost — comparison with catalog (Tabel 8):

MW size Catchment Catalog (no pump, kr/yr) Catalog (with pump, kr/yr) Model fixed cost
mw1 (0.20 ha) 20 ha 25,420 26,474 27,270
mw2 (0.50 ha) 50 ha 33,617 39,330 36,983
mw3 (1.00 ha) 100 ha 50,391 60,262 53,171

(Catalog "with pump" total includes anlæg + maintenance + land loss; "no pump" excludes pump cost.)

Model values sit between "without pump" and "with pump" cases — consistent with an average across lowland (pump) and upland (no pump) locations. Reasonable. ✅

Catalog eligibility criteria: - ≥80% of drainage catchment must be drained - N removal at outlet ≥300 kg N/ha MW/yr - TN concentration in drain water ≥4 mg TN/l - MW = 1–1.5% of drainage catchment (≥20 ha; ≥50 ha if water is pumped in)

Side effects from catalog: - Pesticides: slight retention of drain water pesticides (+) - Biodiversity: positive (+), especially with planted vegetation - P: 45% removal (+) - Climate: 1.6 t CO₂-eq/ha from cessation of cultivation (neutral overall due to denitrification N₂O)

Overlap note (catalog): Mini-wetlands overlap with IBZ (same function but IBZ used on smaller catchments <25 ha with sufficient slope). Surface measures (e.g. catch crops) on the contributing fields reduce the hydraulic and nutrient load to the MW, reducing MW efficiency.

Open questions

  1. SurfRet_ID15(ret) — is this a separate dataset or derived from field-level SurfRet(i) averaged over the ret?
  2. The MW potential color categories (dark green / light green / dark yellow) — what do these represent ecologically?
  3. Is there a cost for the actual land of the MW, or only for the 0.20/0.50/1.00 ha directly occupied? (Catalog Table 8 includes "Dækningsbidragstab" — yield loss — but this is a small component of total cost.)
  4. Does the model distinguish open-water MWs from filter-matrix MWs? The catalog shows filter-matrix MWs remove ~50% N (higher efficiency than open water's 22%). If the model uses 22% (open water), this is conservative for sites where filter matrix could be used.
  • WL — field-level wetland (N-targeted; different spatial unit — field vs ID15 catchment)
  • PWET — P-targeted wetland (field-level; sedimentation mechanism)
  • IBZ — in-stream buffer zone (smaller catchments <25 ha; same function at finer scale)
  • Retention types — MW uses SurfRet_ID15 (surface retention at sub-catchment level)
  • P loss pathways — MW reduces matrix + macropore P at catchment outlet
  • N Reduction — MW_Red_ret equation
  • P Reduction — MW_red_P equation
  • Constraints — VP2 linkage / potential cap
  • Tripartite Agreement — MW floor: ≥40,000 ha equivalent