Please enable JavaScript.

Coggle requires JavaScript to display documents.

Review of Nitrogen Attenuation in NZ seepage wetlands Rutherford (2017)

- - - - Hence, compaction and channelization reduces mixing
  - - - Individual Wetland studies
        
        RC wetland
        
        the annual export of total nitrogen were recalculated
        as 35.6-37.6 kg y-1
        
        Assuming the catchment comprised ‘improved’ pasture (23-28 kgN ha-1 y-1) and an effective area of 3 ha,
        attenuation was 46-58%
        
        McKergow et al. (2012) document very high concentrations of organic nitrogen leaving the wetland during
        ‘livestock events’ which occurred infrequently but accounted for c. 30% of the annual nitrogen export
        
        when those events are excluded, attenuation was 61-68% without changing rainfall events, 75-79% when replaced with baseflow tho this is likely over-estimate
        
        The ‘most likely’ scenario is considered to be that the catchment pasture was ‘improved’ and the ‘effective'
        catchment area draining to the wetland was 2.7 ha. The ‘most likely’ attenuation was 46-58% (average 52%)
        when cattle had access to the wetland and 61-79% (average 70%) if cattle had been excluded.
        
        Sukias and Collins - removal rate did not decrease at high flows
        
        The
        OVERSEER wetland module under-estimates the effectiveness of the RC wetland by 50-70%
        
        McKergow study: Catchment inflows estimated, outflows calculated
        Sukias and Collins: piezometers top and bottom
        
        Baseflow, rainfall and livestock events occurred 81%, 18% and 8% of the time, but made similar
        contributions to total nitrogen export
        
        JS wetland
        Sukias and Collins
        
        Piezometer study
        
        Nett removal rates for shallow wells were on average 3 times (range 2.6-3.8) higher than for the deep wells
        
        suggests shallow soils are more
        biologically active than deep soils
        
        DIN and total nitrogen concentrations
        decreased with distance along the wetland
        
        Removal rates were highest where drainage first entered the wetland but decreased with distance downstream.
        
        Tracer injection study
        
        Sukias & Collins reported the mean pore water velocity to
        be 100 cm d-1 based on the time of travel of the bromide centroid. This study re-calculated the velocity of the
        centroid to be 60 cm d-1. Both estimates are higher than the 30 cm d-1 reported by Burns & Nguyen (2002) in
        Barkers wetland, Whatawhata.
        
        areal nitrate removal rates of 190-425 mg m-2 d-1 estimated from this tracer experiment occurred at lower nitrate concentrations than typically flow into seepage wetlands draining pasture, and hence
        may under-estimate typical wetland removal rates
        
        Sukias & Collins Removal was estimated
        to be 96-98% of the added nitrate
        
        Nitrate removed from the sub-surface flow
        during the 6-8 days it took to
        travel past the downstream well.
      - TUT catchment studies
        
        wetlands
        
        wetlands occupied 5% of the catchment but
        that 11% of runoff passed through wetlands as baseflow
        and another 8% during rainfall events
        
        Based on the observation that the RC and JS wetlands
        removed 95% or more of incoming nitrate,
        then wetlands probably attenuate 10-18% of catchment nitrate runoff
        
        Assuming catchment losses were 17 ± 5 kg ha-1 y-1 and stream yields
        were 5 ± 1 kg ha-1 y-1 then catchment attenuation would be 70 ± 25 %.
        
        Riparian zones
        
        Piezometers were installed in pasture within 2m of the buffer fence line
        and in the riparian zone within 2m of the stream edge
        
        Sampled GW >25cm, <50cm
        
        average nitrate
        attenuation of 75%.
        
        At 11 sites the attenuation ranged from 36-99% but at 4 sites there was a gain of nitrate
        
        The average dissolved organic nitrogen concentration across all 13 sites increased significantly between pasture
        (548±40 mg m-3) and riparian (967±101 mg m-3) piezometers which indicates an average gain of 75%. At 11
        sites there were gains of 13-275% but at 4 sites there were loses of 6-52%. The 4 sites where nitrate increased
        were not the same 4 sites where dissolved organic nitrogen decreased.
        
        A key finding of this study was that the riparian buffer zone could be a source of dissolved organic nitrogen
        
        1 more item...
        
        The finding that nitrate concentrations in shallow groundwater
        decreased significantly as it flowed through the
        seepage wetlands in riparian zones was consistent with other
        New Zealand studies and overseas literature
    - - Steep, seepage wetland, Whatawhata
        
        Nitrate removal from surface flow<10cm)
        
        vertical mixing
        between surface flow and the
        wetland soils to a depth of c. 10 cm
        
        From 3 to 20 hours the N/Br ratio
        decreased linearly with time
        
        removal rate increased with concentration
        
        equivalent to 430-560 (average 495) mg m-2 d-1 at 20C.
        
        OVERSEER would under-estimate nitrate removal in Barkers wetland by c. 60%.
        
        lithium bromide tracer
        
        Soils were organically
        enriched, anaerobic and had
        high denitrification enzyme activity
        
        potential to remove nitrate by
        denitrification to N2 gas)
        
        Nitrate removal from sub-surface
        
        10-20cm
        
        99% of samples had a lower nitrate/bromide
        ratio than the injected tracer
        indicating nitrate removal
        
        90% of
        the added nitrate removed along the 100 cm flow path, with the majority removed within the first 30 cm
        
        areal removal rates of equivalent to 365, 61 and 21 mg m-2 d-1 at 20C) for the control volumes 0-30, 0-60 and
        0-100 cm respectively (vo
        
        The maximum value of 365 mg m-2 d-1 at
        a nitrate
        concentration of 2.05 g m-3 is
        approximately twice the OVERSEER value for Barkers
    - - Sewage treatment wetland
        Rotorua
        
        Pines could take up lots of P
        but not enough N
        to meet lake load limits
        
        postulated that wetland nitrate removal at
        Whakarewarewa is low because there is insufficient contact time between microbially active wetland soils and
        high nitrate water draining from the irrigated hillslopes
      - Surface injection experiment
        
        Tracer travelled quickly to outlet when injected at surface
        
        Observed nitrate
        concentrations (minus background) were lower than ‘expected’ concentrations, indicating nitrate removal
        
        Surface: 17% (Channel R) and 23% (Channel L) removal from surface flow
        
        denitrification enzyme activity suggests an order of
        magnitude lower than tracer
        3.3 ± 0.6 g m-2 d-1 vs
        30-40 g m-2 d-1
        
        In the Channel R the difference between the integrals over time of observed and ‘expected’ nitrate
        concentrations in the sub-surface collector at the outlet indicated 73% removal from sub-surface flow.
        
        observed removal rates from surface flow in the two channels (25-30 g m-2 d-1) were 2 orders of magnitude
        higher than the maximum rate in OVERSEER (250 mg m-2 d-1).
      - Sub-surface injection experiment
        
        Assuming the wetland removed 36% (229 g) of the added tracer (642 g) over 9 days then the average areal
        removal rate was 3900 mg m-2 d-1.
        
        The
        weighted average
        removal was 36%.
        
        Bromide injected 20-30cm below surface
        
        provided further evidence of vertical mixing within wetland soils
        
        The estimated areal uptake rate of 3900 mg m-2 d-1 is higher than the value in OVERSEER even though it was
        measured at nitrate concentrations below those typical of inflows to pasture wetlands
    - - 75% of the
        area occupied by dairy farms with a stocking rate of ~3 cows ha-1
      - Most of the flow at the lower weir must enter the wetland from sub-surface and
        groundwater flow.
      - median nitrate concentrations decrease
        significantly between the upper and
        lower weirs
        
        by an order of magnitude in baseflow and during summer/autumn, and by a factor of 3 during
        winter/spring
        
        total nitrogen concentrations also decrease by a factor of 2-3
        
        However, more data needed to determine if all inflows have similar concentrations
        
        upper wetland area
        contributes <10% of total wetland flow.
        
        decrease may indicate significant attenuation within the wetland
        (50-66% for total nitrogen and 66-90% for nitrate) but only if inflows between the weirs have similar
        concentrations to inflows above the upper weir
        
        indicates c. 95% removal of nitrate possibly through denitrification
    - - Scotsman Valley
      - 56-100% of nitrate removal
        occurred in organic riparian soils even though these occupied only 12% of the stream bank
      - active soils were
        located at the base of hollows and 37-81% of groundwater flow passed through them to the stream
        
        The soils were anoxic, and high in both denitrifying enzyme activity and available
        carbon
      - At an intermediate site denitrification rate averaged 250 mg m-2
        h-1 (6100 mg m-2 d-1) and nitrate concentration 218 mg m-3
        
        Cooper concluded that in many parts of the wetland
        denitrification rates were limited by the supply of nitrate.
      - Lower than OVERSEER close to the stream but much higher
        in the upper banks further away from the stream
    - - Near Toenepi
        
        Non riparian seepage zone
      - Irrigated with effluent (Oct-Mar)
      - Mass
        balance calculations of monthly load indicated
        that the wetland reduced nitrate by 70-95% except during
        occasional high flow events.
        
        both ammonium and TKN concentrations in the wetland
        outflow were frequently higher than in the inflows and mass
        balance calculations indicate that the wetland was,
        at times, a source of ammonium, dissolved organic nitrogen
        and particulate nitrogen
        
        ammonium may have originated from the decay of
        wetland plants and/or the dissimilatory reduction of nitrate to ammonium (DRNA)
        
        ammonium is readily
        absorbed by aquatic plants in receiving waters (viz., is readily bioavailable)
        
        while even low concentrations are
        toxic to fish at high pH.
- - - - labile organic nitrogen
        
        Decay of dead pasture and wetland
        vegetation
        # #
      - particulate nitrogen
      - Bioavailability of DON and PN leaving seepage
        wetlands has not been quantified, but is likely
        lower than for nitrate and ammonium
      - Ammonium
- - - - additional field investigations across a range of wetland sizes, types and locations needed
- - - - constant removal rate may lead to OVERSEER under-estimating
        nitrate removal by small wetlands.
    - - This simulation suggests that assuming a constant removal rate is unlikely to lead to
        significant errors when using OVERSEER to assess the potential of wetlands to remove nitrate
- - - - denitrification to gaseous nitrogen or nitrous oxide
      - dissimilatory reduction to ammonium (transformation)
      - uptake by
        plants (temporary storage