Edge of Field Workshop

Edge of field workshop

Tue 17th March 2015

Workshop Panel

Grameme Doole (Uni of Waikato)

John Quinn (NIWA)

Louis Schipper (Uni of Waikato)

Chris Tanner (NIWA)

Lucy McKergow (NIWA)

Malcolm McLeod (Landcare R)

Annette Semademi-Davies (NIWA)

Grant Blackie (WRC)

Rebecca Eivers (WRC)

Jonathan Cowie (WRC)

Session #1 – into and goals

GD - Healthy Rivers economic and catchment modelling, current does not well represent of how to identify opportunities to capture and remove sediments at both farm & catchment levels. Workshop aims is to support this possible improvement to the model.

Discussion questions include; how we can represent them (in a model), which technologies are relevant, what attenuations expected for for each treatment option, how to locate them, and associated costs.

GD - 3 key goals of workshop:

Protocol’s to help guide the modellers to identify potential locations where key mitigation options (eg wetlands, riparian fences, etc) may be best located.
Costs of the key mitigations (eg cost/ha for wet land development)
Efficacy of each key treatment.

CT – Need to classify landscapes to better define hydrological connections, to define edge of field opportunities (slopes, shallow groundwater, topography, soil, infiltration characteristics, human engineering, drainage). Would help define 4 or so categories of land, which would drive better prediction of edge of field opportunities.

All - agreed there may be limitations to the protocols approach (such as ignoring small accord streams, which may be important from nitrogen perspective) – but it’s a good start.

GB/LS - Need to excluded mitigations that are already in place, or will throw out the economic modelling outcomes.

Treatment Options / Drivers / Identification criteria
Headwater & riparian seeps / Change in slope.
Drainage class. (very poor is peat – ignore). / Agricultural land only.

>15deg, <35 – alone.
Slope 7-15deg + drainage class (poor & imperfect).

Constructed wetland larger [JC1](intercept surface flow) / Drainage
Slope /

3-5deg slope, well drained, ag soils.

Cons’ wetland (small scale – surface & sub-surface flow) / Drainage
Slope /

0-15deg – very poor/poor/ imperfect), dairy or intensive sheep & beef.

Denitrification beds/bio reactors / Same as for tile drains (also possible in surface drains – clog risk) / Poor & imperfect drainage (hence limited applicability in Waikato).
Sediment traps / Flow rates (mean/flood).
Tile drains.
Lower slope. / 0-15 deg
+ flow & flow ratio[JC2].
+ or ag land, + order 1 streams, + very poor/poor/imperfect drains.
DELETE
Detention pond. / DELETE –
Slope Ephemeral streams only. / DELETE

>15 < 25deg, + <10Ha service area, + ag land.

Detention Bund - wetland.[JC3] / Slope /

7-25deg, ag land, 10Ha service area

Session #2 – Wetland & Sediment traps options

General modelling considerations.

Potential options to best approach modelling/identifying areas of interest:

MM/CT - Key approach to participationthe catchment by slope, by soil type and drainage class. This will indicate risk of sediment loss, and drive identification of key mitigation options and areas.
LM – Maximise what you get out of the data, by picking the key areas, then matching the technology to best suite it – ie hydrax soils will indicate likelihood of tile drains – indicating effective use of infield wetlands.
RE - WRC already mapped up priority areas in the Waikato region, particular focus on sediments.
Mapped wetlands down to 0.5Ha & also seepage areas. Data exists, however at a higher resolution (good, but not to farm levels).
ACTION – RE to package together whatever GIS layers currently available, to give to ASD.
Determine the max potential impact of the mitigation options and work backwards....

All–Discussion relating to which unit of measure to use for efficacy of mitigation options:

LS - use grams of contaminant (N, P, SS), and not %. For sediment may be best to base it on sediment deposited/m2/day – and annualised. ACTION – CT, to review efficacy figures for retention bunds and wetlands.

CT/GD - Rate measure would not work for wetlands where a concave function for removal needed – as more efficient with increasing Nitrogen, until reaches max point).

ASD – can use above rules (above table), then develop pivot table indicating % of sub-catchment, which is indicative of where shallow seep will occur.
GD/others – Discussion regarding possibility of determining source of seepage flows, then how to apply an appropriate efficacy to it, for each treatment type. Unresolved, with concerns raised you may have high efficacy for the flow that interacts, however 70% of the flow may bypass the structure.

Difficulties discussed:

Difficulty in modelling nutrient losses from mitigations such as wetlands, as you may remove 90% of the nutrient flowing through the treatment, but 70% of the total flow bypassing the wetland.
Difficult to determine the source areas of seepage (only areas where there is a greater likelihood of occurrence).
Difficulty in known how many areas are intact, which areas already have engineering controls, which areas have low stocking rates.

Existing large wetlands

CT - Also of large wetlands in Waikato region, but these are often peat based, and therefore could be ignored from a surface runoff perspective.

Shallow springs – seepage wetlands - Edge of field/in-field wetlands

Brown glasses - Edge of field/in-field wetlands – shallow springs flowing some time through the year have a high potential for allowing nutrients to transport.

JQ - Often in head of valley, which has been cleared for cropping. Impacted by stream power – at low flows they can be highly effective, however at high flows they ‘blow up’ (often losing 10 years of stored up sediments).
ASD – Knowledge of depth of ground would help to identify location of these shallow springs.
RE[JC4] – already working on how to best identify seepage wetlands through, soil drainage classes, use of LIDAR looking for depressions & sat imagery. This could be coupled with accord/non-accord streams to help define critical areas where first-flush of nutrient may occur.

Constructed wetlands – intercept flow of surface water(and tile drains) into a constructed wetland to remove sediments & nutrients.

CT – classified as a “surface flow wetland” – vegetation planted in a shallow depressions. Can be in stream channel, or adjacent bank, or on-farm overland flow. General agreement this project should focus on off-stream flow areas.
CT – performance of wetlands more optimal when there is a slow flow of water through them, however over they are bypassed by drainage lines, etc. The best and cheapest bang for buck is to restore & improve upon existing wetlands.
CT – often important to have a high flow by-pass. This avoids blow out during extreme weather, therefore protect against wall failure, or flushing any stored sediments, etc out.
CT – considerations for locations for where best to install - bottom of catchments (sometimes head of catchment). Less effective in pumice country – as can’t intercept water. From a landscape perspective - landscapes to guide install. Undulating, rolling landscapes will increase cost effectiveness to build.
CT – considerations for improving effectiveness – not too deep (0.2 – 0.5m, - a couple of deep water areas OK to distribute water), contain a sediment trap at beginning, slow through flow (if too fast, then insufficient time to causeattenuate, 1-5% of the catchment to attenuate nitrogen (upto 40 nitrogen reduction, less if lower catchment area)
CT/JQ/LS – a novel approach is to have a controlled sub-surface drain (backfilled with Al or organic matter). General agreement that this is unproven technology, which may not be effective. Idea abandoned.
CT –considerations in regards to efficacy for wet lands:

1% catchment / 2.5% catchment
N / P / SS / E.Coli / N / P / SS / E.Coli
Surface / 30 / 55 / 60 / 80* / 60 / 70 / 80 / 90*
Sub-surface / 30 / 0 / 40 / 80* / 60 / 0 / 55 / 90*

* During peak flows (considering some ponds increase e coli during low flow due to ducks etc).

Note – assumes P coming in with the sediment load.

CT/LM – in regards to costs to construct wetlands:

GRAEME NOTE – Chris & Lucy have referred you to NIWA Waituna Reports for capital abatement curves for – capital, operating costs & lost opportunity costs. This relates to surface flow wetlands only.

Sediment traps and dams – low land areas.

RE – 20- 50Ha catchments are ideal for each sediment trap, but depends on slopes, soil type, etc. Only work under certain flow & slope conditions (esp medium to low flows). If flows too great, the sediment traps can fail.

ASD – asked if can use estimated flows to predict location of a sediment trap, and RE agreed this is sensible, but depends on the hydrograph (i.e. large flows leading to failure). JQ – ratio of median /flood ratio – may overcome this.

RE – expertise lies more in low land, low flow areas.

ACTIONS – RE to provide advice on effectiveness of well preforming sediment traps.

Sediment traps and - Higher areas (15 - 25deg) – Detention ponds.

JQ/Grant - Closure up towards head waters. Often have duel function such as being a duck ponds, which also function as sediment capture. Note – best designed as a leaking wall, and not as impermeable dam (such as used to stock water impoundment).

WW - Modelling to assume includes maintenance (or sediments would eventually be lost, NB – may need a high flow by-pass, or excessive inflows may cause it to blow out and flow downstream

Session #3 – De-nitrification filters & walls

De-nitrification Walls

LS – only useful for where confining impermeable layers giving defined subsurface flow paths. Especially useful where plumes of contaminated groundwater flows are known. It was stated by LS (and agreed in general) that due to the Waikato geology, its de-nitrification walls unlikely to be a viable option (close enough to the surface, and where the water would not just flow around it). Only likely usage would be if:
There is a specific nitrate plume – almost like a contaminated site.
As a novel approach – put alongside a tile drain, but need replacing within 10yrs, would degree aerobically during dry spells further reducing lifespan.

De-nitrification Bed/Bioreactor

When you have control of hydrology by pumping water into a large bed which is full of wood chips
CT - NIWA building on in Southland - $18K for 9Ha of area, aiming at being sized for 80% of the median flow.
LS – stating from a journal (Morman, 2015) 20-30% nitrogen removal using 0.3% of water shed.
LS - 7x nitrate reduction technologies in mid-west USA. Bioreactors - $2US/Kg nitrogen removed. Wetlands $2.5-3/Kg nitrogen removed.
CT – research on wastewater treatment suggests >90% removal of E Coli. Phosphorous <10% removal. Not designed for sediment removal (as it clogs up the system). Note – makes improvements to Al and other metals, however insufficient current info to project an efficacy.

Session #4 – Other attenuation media.

Phosphorous-sorption Filters – unlikely from a cost & field verification perspective.

LS - Restricted to pre-filtered, or sub-surface drainage. Also accumulates metals (unlike bio-reactors) so eventually clogs and needs removal.
Glenbrook smelter slag (note – may require pre washing to remove the boron).
CT – some cast studies in NZ. Worked for 5-10yrs before becoming exhausted.
CT/LM – NIWA currently conducting a trial on Aqual-P at Waitura. Design basis available in 4-6weeks, but results for 2yrs.
CT/JQ – Alum or ferric chloride – reuse of spent water treatment plan media (otherwise goes to landfill), and if place in areas such as bottom of wetlands, it would increase P-sorption capacity.
LS/LM – shall provide some references, including Ballantyne Review Paper (NZ Journal of Agricultural Research) 2010.

Floating wetlands – installation of floating wetlands in public waterways and lakes to form a floating sub straight to grow aquatic plants. Has been shown to have impact in reducing fine sediments that are in suspension. Creates a zone of high oxygenation, which promotes de-nitrification. Water depth must be >0.8m to avoid roots anchoring to stream-bed.

$200-300/m3, more expensive then a constructive, however likely to be similar to better in efficacy. An Opus report reviewing cost and benefits of surface flow treatment options, suggested floating wetlands would be 10x cost of alternatives.
Advantage – not taking up agricultural land.
Longevity of 10-20yrs.
General acceptance of limited application in agricultural setting, relative to costs. CT – “expensive & high risk”. Has been considered, but not worth progressing to modelling or costing.

Saturated Batters – 1 study by Dr Dan Jayms, 2014 (Iowa) suggests tile drainage diverted to run parallel to stream and seep through riparian soil to enhance de-nitrification. Operated under low flow.

Diverted 35% of flow from a 10Ha plot (300m long parallel drain).
Relatively inexpensive, $2US/kg nitrogen with a 20yr life expectancy.
Group’s considered opinion – concept has merit, and should progress to NZ trials. Some modelling could be done to assess the potential impact of this treatment option (using Jayms, 2014 at 50% efficiency), to justify further investment and research (potentially inclusive of backfilling with organic matter). Not to be proposed to CSG as a viable option, as its still not even at a proof of concept stage.

Double Ditching – not considered suitable due to flooding damage which is common in the Waikato region.

[JC1]JQ/CT – larger wetlands are more complex, engineering input, total cost, etc - however may be cheaper per unit area & impact.

[JC2]ASD – need advice on max flow rates before sed trap fails.

[JC3]Similar in concept to a dry storm-water pond, with allows water to pool up behind it – thusly additional benefit of acting at some level as a constructed ‘headwater wetland’ (hence N & sed’ attenuation). Simple to construct (<1m high bund), if can get earthmoving equip in area.

GD/JQ/ADS – for modelling determine - max area in sub-catchment (ag land, 15-25slope), then divide by 10, to determine max number of detention drains. This would divided into 2 soil types (well drained, and poor/imperfect) as this will effect development of wetland, efficacy & cost (inc fencing).

Note - long-term effectiveness is based on wiliness of farmers to do annual maintenance.

[JC4]Opportunity – existing work by WRC to identify shallow seepage wetlands, and key areas of high nutrient first flush.