Original source · versioned page text
RFI - Attachment F - Engineering Report
Very large file: it may take a few minutes
The original of this document is very large (15.7 MB). Opening the original or downloading it may take a few minutes. Please be patient and don't click again. The page text is quick to read. If you are on mobile data, it is best to wait until you are on Wi-Fi.
MPN CONSULTING PTY LTD
STRUCTURAL AND CIVIL ENGINEERS
Level 5, 39 Sherwood Road, Toowong, Qld, 4066
P O Box 328, Toowong, Qld, 4066
Phone 61 7 3335 4555
Email [email protected]
Web www.mpnc.net.au
ABN 39 062 191 799
File: 7700
15 May 2019
Noosa Shire Council
9 Pelican Street
TEWANTIN QLD 4565
Dear Sir / Madam
RE: 36 & 40 HOFFMAN DRIVE – NOOSAVILLE
RESPONSE TO REQUEST FOR FURTHER INFORMATION
MCU19/0017
Please find enclosed a response to the civil engineering items within the request for further
information from Noosa Council dated 28 February 2019.
Issue - Earthworks
The proposed earthworks have the potential to impact on the adjoining properties, especially the
building located adjacent to the southern boundary. The submitted engineering report does not
include sufficient comment to demonstrate how the potential impacts can be addressed.
Information Required
25. Provide an amended engineering report that includes comment on the construction methods
that could be employed to address the potential to impact on the adjoining properties.
Response
The civil engineering report has been amended to provide comment on construction methodology of the
boundary retaining for the basement. The retaining/shoring will be undertaken as a Design and Construct
package as part of detailed design and will ensure that the neighbouring buildings are supported during
both the construction phase and operational phase.
Excellence With Economy In Engineering Document Control
Approved
See the original
The page text is free to read. Viewing or downloading an original file needs an account.
Log in to see the originalLog in to download the original file
Very large file: the download may take a few minutes. If you are on mobile data, it is best to wait until you are on Wi-Fi.
Searchable page text hides email addresses. Original files are unchanged and may show email addresses.
Full text of pages 2 to 102 (of 118)
Page 2
15 May 2019
Issue – Acid Sulfate Soils
O1 of the Water Sensitive Design Code requires downstream ecosystems, properties and
infrastructure are not adversely impacted by either changes to flow, water quality or ecosystem
health caused by development; and
O25 of the Natural Hazards and Natural Resources Overlay Code requires Works do not disturb
acid sulfate soils or are managed to avoid the release of acid or associated metal contaminants.
The Stormwater Management Plan states:
The runoff from the basement subsurface drainage pump out sump will be treated to neutralise any pH
increases due to the acid sulfate soils prior to discharge from site. This system will be separate to the
treatment of the stormwater runoff from the exposed surfaces of the site which will be treated by the
filtration cartridges.
This comment implies that the development will require continual dewatering of groundwater. It is
unclear if the cone of depression that will be caused as a result of the dewatering will extend into
and convert Potential Acid Sulfate Soils into Actual Acid Sulfate soils and consequently release
metals to the receiving environment. Simply neutralising waters pumped from the basement does
not act to prevent release of metal contaminants from conversion of PASS to ASS, additionally,
neutralising of the water does not necessarily mean the pH and other chemistry requirements of
the receiving environment will remain unaffected.
Information Required
27. Demonstrate how construction and operation of the development will be managed to
prevent, or manage all the impacts associated with conversion of Potential Acid Sulfate Soils
to Actual Sulfate soils as a result if dewatering required for the basement. The response must
have regard to the ecological needs of the environment receiving any discharge waters or
likely to be impacted by chemistry changes within the soil. The response must also have
regard to the impacts of any changes in soil chemistry on the surrounding built environment
as a result impacts to acid sulfate soil from dewatering.
28. The response must be supported by an Acid Sulfate Soils Investigation prepared in
accordance with the Queensland Acid Sulfate Soils technical manual: soil management
guidelines and the Guidelines for sampling and analysis of lowland acid sulfate soils.
Response
An acid sulfate soils investigation has been completed on site with a report prepared. The report has been
appended to this response.
The Stormwater Management Plan has been amended to suit the outcomes of the Acid Sulfate Soils
investigation.
Excellence with Economy in Engineering 2
l:\7000-7999\7700-7799\7700\letter\15-05-19 noosa council.docx
Page 3
15 May 2019
Issue – Water Quality
This issue is in relation to the following three Specific Outcomes of the Water Sensitive Design
Code.
O1 Downstream ecosystems, properties and infrastructure are not adversely impacted by either
changes to flow, water quality or ecosystem health caused by development
O10 Development design and operation facilitates nil increase in gross pollutants entering
stormwater systems.
O7 Natural treatment strategies, including grass swales and bio-retention systems, vegetated
filtration strips and gross pollutant traps, are used wherever possible
The application proposes the use of a Stormwater 360 proprietary product to achieve the required
stormwater quality pollution reduction targets to achieve the Specific Outcomes.
No information has been provided as to whether the nodes for the proposed Stormwater 360
proprietary devices within the MUSIC model have been tested and verified against the Stormwater
Australia (2018) Stormwater Quality Improvement Device Evaluation Protocol. Version 1.2 December
2018. Propitiatory devices are currently only considered in lieu of vegetated stormwater
treatment strategies retain or allow enhancement of existing environmental values on the site. In
these situations, their performance must be verified prior to being considered for a trial within
Noosa Shire. Trials are only considered where:
Peer Review from a stormwater professional, endorsed by Stormwater Australia, has been
provided verifying that the device has been tested in accordance with the Stormwater Australia
(2018) Stormwater Quality Improvement Device Evaluation Protocol. Version 1.2 December to
generate accurate and reliable data used within the MUSIC node for the proprietary device,
and retesting against the protocol under Noosa rainfall conditions and development site is
agreed to; and
a significant environmental benefit is demonstrated by allowing a trial of the proprietary
device at the development site (i.e. an area with existing environmental value will be retained
in lieu of installation of traditional stormwater quality improvement device such as bio
retention basins); and
Opportunity is retained within the developed property to retrofit of traditional stormwater
quality improvement devices (i.e. bioretention basin) to achieve the required pollution
reduction targets. This is necessary for the instance where testing of the nominated
proprietary product under local conditions against the Stormwater Australia (2018) Stormwater
Quality Improvement Device Evaluation Protocol. Version 1.2 December 2018 demonstrates that
the proprietary device within the developments treatment train fails to achieve the required
pollution reduction targets.
Information Required
29. Information in regard water quality improvement must demonstrate how the runoff from the
development will deliver water quality and quantities consistent with the requirements of
downstream ecology, as detailed in relevant flora and fauna management plans developed
for the Shire Business Centre.
30. Provide MUSIC modelling consistent with the Water by Design Music Modelling Guidelines
to also demonstrate that the pollution reduction targets of 80% reduction in Total
Suspended Solids, 60% reduction in Total Phosphorous and 45% reduction in Total Nitrogen
are achieved.
Excellence with Economy in Engineering 3
l:\7000-7999\7700-7799\7700\letter\15-05-19 noosa council.docx
Page 4
15 May 2019
31. In order for the Stormwater 360 device to be considered any further within this proposed
development, provide:
a) Independent verification from a suitably qualified and experienced professional to
confirm that the parameters within the node for the device in MUSIC modelling have
been tested and verified as per Stormwater Australia (2018) Stormwater Quality
Improvement Device Evaluation Protocol. Version 1.2 December 2018 to achieve the
predicted outcomes in the MUSIC model for the development. For the purpose of this
information request, the suitably qualified and experienced professional must be able to
demonstrate a minimum 10 years urban stormwater testing and modelling experience
and be verified in writing by Stormwater Australia as having an appropriate reputation
and track record within the stormwater industry to provide such independent advice.
b) A MUSIC model along with plans that demonstrates how a biobasin or other vegetated
stormwater quality treatment asset could be retrofit into the proposed development to
achieve the required pollutant reduction targets in the event the proprietary device fails
to perform under the local conditions.
c) Agreement that any proprietary device would be considered as a trial only with the view
to retrofit with a bio retention basin in the event the proprietary device fails to perform as
required under the local conditions.
Response
30. Stormwater runoff from the site will be treated to the levels required by the State Planning Policy
and Noosa Council.
Runoff quantities from site will be detained so that the peak runoff volume from site in the
developed scenario is no greater than the pre-development scenario which currently discharges into
the downstream system.
The above will ensure that the post-development quality and quantity of runoff discharging into the
downstream system is consistent with the runoff that currently discharges from site pre-
development.
31. The MUSIC models for the assessment have been provided with the Stormwater Management Plan
which achieves the required targets of pollutant reduction from the State Planning Policy.
32(a) Independent testing and peer review assessment have been completed for the stormfilter devices
and the MUSIC nodes have been provided by Ocean Protect/Stormwater 360. The data and nodes
have been reviewed and approved within many South East Queensland Council areas along with
other councils across Australia.
(b) MUSIC models of the stormfilter and enviropod devices have been submitted with the revised
Stormwater Management Plan. It is not proposed to retrofit bioretention basins within the site as
the development layout and footprint will not allow for large bioretention areas. The mechanical
treatment devices will be monitored and maintained to ensure the quality treatment of the
stormwater runoff is compliant with the requirements of the State Planning Policy and Noosa
Council.
(c) proposed treatment devices have been assessed under many conditions with independent analysis
and peer reviews completed. As such it is considered that the treatment devices will function as
designed within the local condition. The tank has sufficient size to add more filters, if required,
however this is unlikely. The mechanical devices will be maintained and monitored to ensure the
runoff is compliant with the requirements of the State Planning Policy and Noosa Council.
Excellence with Economy in Engineering 4
l:\7000-7999\7700-7799\7700\letter\15-05-19 noosa council.docx
Page 5
15 May 2019 (7 1O M consulting Issue - Impacts on the receiving environment Specific Outcome O6 of the Water Sensitive Design Code requires that within the Shire Business Centre site, as shown on Map SBC of the Noosa Plan, on-site detention measures are contained within the developable areas of the site, clear of drainage lines and Open Space Precincts. Information Required 32. Demonstrate that on-site detention measures are wholly contained within the developable areas of the site, clear of drainage lines and Open Space Precincts. Response 33 the onsite detention measures are wholly contained within the developable areas of the site as per the concept plans and Stormwater Management Plan. Should you require any further information do not hesitate to contact the undersigned. Yours faithfully Lachlan Stephenson For and on behalf of MPN CONSULTING PTY LTD IIs:dh Excellence with Economy in Engineering 5
Page 6
QUT 2 George Street GPO Box 2434 Brisbane Qld 4001 Australia Queensland University of Technology Faculty of Built Environment Phone 07 3864 1433 Fax 07 3864 1529 www.qut.edu.au and Engineerin 9 9 CRICOS No. 00213J ABN 83 791 724 622 15" March 2011 KURANDA STORMWATER TREATMENT SYSTEM FIELD EVALUATION | have peer reviewed the report dated 15" October 2010, on the Kuranda Stormwater Treatment System field evaluation prepared by Stormwater 360. | find the report to be a factual evaluation of the data obtained from the field study. The data evaluation and the conclusions derived have been presented using clear and precise terminology which is easy to read and understand. It is also important to note that any identified shortcomings in the system has been highlighted and discussed in an open manner. It is quite evident that the data collection has been based on a very rigourous and technically demanding monitoring program. This adds further credibility to the field evaluation undertaken. | have personally inspected the field set up at Kuranda and also assessed the sample collection and testing protocols and can confirm the validity of the information presented in the document. Prof, Ashantha Goonetilleke School of Urban Development Email: [email protected] Tel: +61 7 3138 1539 mum ~Campuses Gardens Point Kelvin Grove Carseldine
Page 7
SEPTEMBER 2011 RRP $16.95 inc. GST AWA australian association PUTTING A PRICE ON WATER... sel ae eleg * on page’50 = =. ~_ oe B WASTEWATER TREATMENT ¢ SMALL & ONSITE SYSTEMS ¢ STORMWATER REUSE 2 S Ss Ss g x ee} 6 aN Nn a a a o > Ss ec a a <= BE Dn 3 Os LE 2 a a
Page 8
W at e Journal of the Australian Water Association ISSN 0310-0367 Volume 38 No6 September 2011 contents jas See EL tk RGR Ty Residents at Buru in Queensland map their water system. See page 73. An aerial view of the Suncoast Sewage Treatment Plant. See page 58. TECHNICAL FEATURES sm as ita Baga WASTEWATER TREATMENT al Defining Treatment Plant Capacity Equivalent Population is more sensible than Average Dry Weather Flow MP Thomas 58 ‘| Biogas Production Potential from Meat Processing Plant (@E=uE= Results showed good potential for biogas production MOthman&S Woon 64 INTERNATIONAL AID PROJECTS | Analysis of Breakdown of Technology Transfer in Cambodia (@RSHIET! Positioning ceramic water filters as a fundamental service rather than a luxury C Browne 67 SMALL SYSTEMS The Community Water Planner Field Guide A package to assist Indigenous communities R Grey-Gardner, R Elvin, P Taylor & MAkeroyd = 73 [The Potential for Membrane Distillation of Industrial Wastewaters (SHE! Using waste heat to recover high-quality water from a waste stream NDow&MDuke 78 STORMWATER TREATMENT EI Nutrients and Solids Removal by an Engineered Treatment Train Field evaluation of a gully pit insert and cartridge media filter water M Wicks, N Vigar& M Hannah 83, WATER TREATMENT [ Application of Chlorination for Cyanobacterial Toxin Control Chlorination can be an effective final treatment barrier for a range of cyanotoxins L Ho etal 89 WATER RECYCLING |&) Rapid Cross-Connection Detection by Portable Fluorescence Spectroscopy (™@featal” Recycled water is typically able to be distinguished from potable water by a factor of 4 to 5 AC Hambly et al, 96 ASSET MANAGEMENT Water Infrastructure Benefits of asset management-centred organisations ZSlavnic 102 WATER BUSINESS New Products and Business Information 104 Advertisers’ Index 112 2 SEPTEMBER 2011 water
Page 9
refereed paper stormwater treatment
NUTRIENTS AND SOLIDS
REMOVAL BY AN ENGINEERED
TREATMENT TRAIN
Field evaluation of a gully pit insert
and cartridge media filter
M Wicks, N Vigar, M Hannah
Abstract one-third of concentrations reported Environment and Climate Change (DECC)
in the literature as typical for urban 2007 recommends Suspended Solids (SS)
The performance claims for individual
catchments in Australia. 85%, Total Phosphorus (TP) 65%, and Total
stormwater treatment devices is often
Nitrogen (TN) 45%) and/or Event Mean
open to debate, as much of the data One storm was also analysed
Concentrations (EMCs) in any stormwater
available has not been subjected to for an expanded suite of nitrogen
discharged into natural ecosystems (e.g.
robust scrutiny and/or the claims are analytes, which determined that more
ANZECC 2000 recommends turbidity
unable to be replicated. The following than half the load was in soluble form.
article summarises the results from a field 2-15 Nephelometric Turbidity Units (NTU),
Furthermore, results from the field trial
trial of two such devices: an EnviroPod® TP 0.01 mg/L and TN 0.15 mg/L for river
and research project indicated that this
and a StormFilter®, arranged in series systems in tropical Australia).
treatment train system has the potential
(or a ‘treatment train’) treating runoff to achieve meaningful load reductions In general, each pollutant is removed
from a small road catchment on Streets of Suspended Solids (up to 99%), from the water column using a specific
Creek, Kuranda, west of Cairns Phosphorus (up to 70%) and Nitrogen physical, chemical or biological process.
in Far North Queensland. (up to 45%) through the use of Arranging these processes in sequence
conventional screening, filtration and provides a treatment train approach that
This field trial complements an earlier
research project undertaken on the ion-exchange removal technologies. addresses and treats the whole pollutant
same system by James Cook University. load. There is, however, a paucity of
Introduction published peer-reviewed scientific
Data was collected from six storm
events, predominantly during the dry Livingston and McCarron (1992) identified information validating the removal
seasons of 2008 and 2009, and includes that pollution loads (gross pollutants, efficiency of each element or device used
simultaneous sampling of both the flow sediment and nutrients) in stormwater within a treatment train – let alone the
rate and water quality on the inflows increase proportionally with the degree performance of the treatment train itself.
to, and outflows from, the treatment of urbanisation in the catchment. Most The research referred to herein provides
train for a suite of particulate and consent authorities in Australia have information to validate the performance
soluble stormwater pollutants. Influent established pollution removal efficiencies claims of an EnviroPod® gully trap and a
concentrations for both Phosphorus to be achieved prior to discharge from the StormFilter ® cartridge arranged in series
and Nitrogen were found to be half to urban catchment (eg, NSW Department of as a treatment train.
Photo: ©Google, ©GeoEye ©Whereis® Sensis Pty Ltd ©Digital Globe
Figure 1. Location of the Kuranda Test Site.
water SEPTEMBER 2011 1
Page 10
stormwater treatment refereed paper
Background
This field trial follows a previous research
project undertaken by the School of
Earth and Environmental Sciences,
James Cook University (JCU), as part of
a wider investigation into the impacts of
road runoff on the Kuranda Range Road
watershed, near Cairns (Munksgaard and
Lottermoser, 2008), which discharges
into the sensitive environment of Streets
Creek. JCU reported on the quality of
the watershed’s receiving waters, the
chemical characterisation of the road
runoff and the performance of the system
over four runoff events.
JCU found that the system “had a
high retention capacity for suspended
Figure 2. Schematic of the SYSTEM treatment train.
sediment and by implication particulate
metals”. Conversely, they reported that achieved substantial removal of protection on the steel components.
the “treatment train” had only a “modest Total Nitrogen (45%), Total Phosphorus Given the substantial removal of
retention capability for dissolved (filtered) (70%), Total Aluminium (71%), Total Nickel suspended solids, nutrients and total
metals”. In addition, JCU identified that (73%), Total Lead (60%) and Total Copper metals, it appears unlikely that the
the treatment train system was, in fact, (58%). On the other hand, it identified dissolved copper and zinc, observed
responsible for a significant net export of potential releases of Suspended Solids in the outflows, was associated with
zinc. On the basis of their data, nutrient under 500 microns, as well as dissolved a release of the under-500 micron
levels in the road runoff were low, and zinc and copper. sediment fraction.
do not constitute a water quality concern
at Streets Creek. However, they also One explanation for the above- It was largely to address these issues
reported significant retention of both TN mentioned releases is that they could and better understand the sources of
and TP. The JCU study, which, in their be related to the anaerobic conditions these copper and zinc releases that
own words “do[es] not constitute a full present in either the standing water Stormwater360 undertook a further field
evaluation of the EnviroPod/StormFilter within the wet-sump or, in the case of evaluation of the treatment train system,
treatment system”, found the system zinc, corrosion of the exposed galvanised which is the subject of this evaluation.
Table 1. Water quality analytical parameters.
Parameter Abbreviation Analytical Method* Units Limit of Reporting Analysed by
Electrical Conductivity EC APHA 2510B µS/cm 1 Cairns Water
pH pH APHA 4500-H+ - 0.1 Cairns Water
Suspended Solids above 500 micron sieve &
SS > 500 micron mg/L 1 Cairns Water
500 microns APHA 2540B
Volatile Suspended Solids 500 micron sieve &
SS Vol. > 500 micron mg/L 0.1% Dry Solids Cairns Water
above 500 microns APHA 2540E
Suspended Solids APHA 2540B; equiv.
SS < 500 micron mg/L 1 Cairns Water
below 500 microns ASTM D-3977-97
Volatile Suspended Solids
SS Vol. < 500 micron APHA 2540E mg/L 0.1% Dry Solids Cairns Water
below 500 microns
Suspended Solids SS Calculated mg/L - -
Volatile Suspended Solids SS Vol. Calculated mg/L - -
Total Phosphorus TP APHA 4500-P mg/L P 0.02 Cairns Water
Total Nitrogen TN APHA 4500-N mg/L N 0.05 Cairns Water
Total Kjeldahl Nitrogen TKN Calculated mg/L N - -
Ammonia Nitrogen
NH3-N APHA 4500-NH3 mg/L N 0.05 Cairns Water
(Ammonium Nitrogen)
Nitrate/Nitrite
NO3-/NO2--N APHA 4500-NO3 mg/L N 0.01 Cairns Water
(Total Oxidised Nitrogen)
Total Organic Carbon TOC APHA 5310-B mg/L 1 ALS
Dissolved Organic Carbon DOC APHA 5310-B mg/L 1 ALS
Particle Size Distribution
PSD Malvern Mastersizer S micron 0.05 QUT
(Laser Diffraction)
2 SEPTEMBER 2011 water technical features
Page 11
refereed paper stormwater treatment
Antecedent dry period was not identified as a constraint, due
to the impervious nature of the catchment and the absence of a
base flow; however, at least a three-day antecedent dry period
was preferred. If the storm was deemed to qualify, Stormwater360
would inform Cairns Water and Waste Laboratory Services
(Cairns Water, NATA accreditation # 14204) that samples required
VEE NOTCH
WEIR
collection and analysis. Analysis was performed by Cairns
PRESSURE
TRANSDUCER
Water and Waste Laboratory Services, ALS Laboratory Group
– Brisbane (ALS, NATA accreditation # 825). All water quality
OUTLET PIPE
parameters for qualifying storms were sent to an independent
ROAD
peer reviewer at Queensland University of Technology (QUT),
ensuring transparency of data. Test methods for water quality
analysis used for this study are provided in Table 1.
DRAIN
SAMPLING TUBE
Gross pollutants were not monitored as part of this
study, although significant quantities were captured. Previous
AUTOSAMPLER
monitoring by White et al. (2001) demonstrated that the
AUTOSAMPLER
FLOW METER
Enviropod® filter retained all (100%) litter up to an approach
flow of 100L/sec.
NEL
CHAN
PLAN LAYOUT
Results and Discussion
The system was installed at the Streets Creek site in March 2006
Figure 3. Schematic of the sampling location. and remained an active treatment and sampling site for four
years until being decommissioned in March 2010. Stormwater360
Sampling Procedure and Equipment monitored the system from April 2008 to December 2009. During
A graphical representation of the system is shown in Figure 2. this time, the unit was maintained annually, prior to the onset
The direction of flow through the gully pit insert (EnviroPod®) of each dry season. Complete maintenance involved removing
and into the cartridge media filter (StormFilter®) is shown in all sediments and debris from the system, gully pit insert and
sequence from 1 to 4. The gully pit insert is intended to treat replacing the cartridge media. The gully pit insert required
most flows and filter solids above 100 µm while containing additional manual maintenance approximately once per year.
contaminants in a dry state. Maintenance frequencies for the study were conducted
in line with the systems standard operational lifecycle. Due to
After treatment by the gully pit insert, water is filtered radially
the nature of the catchment and size, there was an absence of
through the media cartridge (outside to inside). The media
a base flow or dry weather flows. Potential pollutant leaching
cartridge had a nominal flow rate of 0.95 L/s (at 46 cm head,
of soluble contaminants was, however, still accounted for;
when the cartridge is primed) and a peak flow rate of ca. 1.3 L/s
organic debris left within the system was allowed to break
(at maximum 0.88 m head prior to bypass). The ZPGTM media
down between maintenance periods and permitted to be
used was a proprietary blend containing perlite (50%), granular
sampled by the effluent sampler during storm events.
activated carbon (GAC, 10%) and zeolite (40%).
A summary of the principal analytes sampled is contained
The system samples were collected using automated
in Table 2.
influent and effluent samplers (Figure 3), collecting continuous
flow and precipitation data and water quality simultaneously. Suspended Solids
The influent sampler was programmed to send an SMS alert
to Stormwater360, via the GSM cellular network, when the ANZECC (2000), DECC (2007) and Fletcher et al. (2004)
have identified suspended solids as a stressor of aquatic
sampling program was triggered. A dial-up connection was
ecosystems. In addition, many of the other pollutants, such
then made to each sampler to download data for analysis.
as metals, hydrocarbons etc, are transported attached to the
To qualify as a representative sample, the following criteria suspended solids and sediment. The system achieved an SSC
were specified.
Table 2. Summary of results.
I. Collection of at least three
simultaneous influent and Range of Median Range of Median Mean Removal
No. of
effluent samples per storm; Analyte Influent EMCs Influent EMC Effluent EMCs Effluent EMC Efficiency
events
(mg/L) (mg/L) (mg/L) (mg/L) (Sum of Loads)
II. Samples must have been
SSC 6 75 to 4384 1181 8 to 63 20 99%
collected while the treatment
system operated within SSC < 500
6 48 to 180 105 8 to 62 20 78%
design flow rates (not in micron
bypass); TP 6 0.08 to 0.19 0.123 0.02 to 0.15 0.055 47%
III. The sampled portion of the TN 6 0.6 to 1.5 1.045 0.2 to 0.9 0.615 44%
storm event must represent 0.175 to
TKN 6 0.6 to 1.2 1.007 0.515 49%
at least 60% of the storm 0.800
total flow volume; NH3-N 6 0.05 to 0.15 0.050 0.05 to 0.07 0.050 31%
IV. A minimum of six data sets TOC 6 3 to 16 7 3 to 10 5 32%
must be collected for a full
DOC 6 3 to 12 7 3 to 11 6 21%
performance evaluation.
water SEPTEMBER 2011 3
Page 12
stormwater treatment refereed paper
between 71 mg/L (forested catchments)
and 232 mg/L (urban roads). Fletcher et
al. (2004) recommend using a value of
ca. 120 mg/L for roads and ca. 100 mg/L
for most other land uses. Both sources
propose a median value of ca. 40 mg/L
for forested catchments. The influent
concentration of Suspended Solids at
Streets Creek is within the typical range of
average annual EMCs proposed within the
literature; however, no data was collected
during large wet-season storm events.
Consequently, the median influent EMC
reported herein should not be regarded
as indicative of an annual median value.
Effluent EMCs recorded for SS <500
micron were in the range of 8 to 62 mg/L.
The median effluent EMC was 20 mg/L.
Mean removal efficiency for SS<500
Figure 4. SS <500micron data (JCU + SW360). micron, calculated by aggregate load Figure 6. Total Nitrogen (SW360 and
aggregate load reduction of 99%. SSC
reduction, was 78%. It is evident from JCU combined).
(ie, SSC is defined as the sum of SS Figure 4 that the Stormwater360 and JCU Particle size distribution (PSD) by laser
<500 micron and SS >500 micron) is data sets are in relatively good agreement diffraction was performed for the SS <500
‘suspended’ in the sense that all these with each other, with the exception of micron fraction for three storms during
particles were sufficiently suspended the JCU outlier, which represents the the Stormwater360 evaluation. Inspection
to reach the system. However, SS first storm from JCU’s research project. of the three cases analysed consists of
<500 micron represents what is more This storm was deemed an outlier for all particles between ca. 10 microns and 200
commonly understood by the term water quality parameters due to possible microns in diameter. There is substantial
‘suspended solids’. It excludes coarse sampling errors and has been removed variation between the three events.
settleable sediment, which, while being from the analyses. The box plot in Figure • Storm 2 influent PSD centred at ca.
a management issue, does not represent 4 shows that the combined dataset is 20 microns for a removal efficiency
such an acute threat to water quality. also clustered around an influent EMC of approximately 65%;
Figure 4 shows influent and effluent of ca.100 mg/L and an effluent EMC of
• Storm 3 influent PSD centred at ca.
data (Stormwater360) for SS <500 ca.20 mg/L. In practical terms, 10 mg/L
100 microns for a removal efficiency
micron, together with the results approximates the system’s irreducible
of approximately 85%;
published by JCU. In the scatter plot, EMC for under-500 micron suspended
the filled-in circles represent data from solids. The box plot in Figure 4 indicates • Storm 6 influent PSD centred at ca.
the trial reported herein, and open circles that, over the course of two trials, the 35 microns for a removal efficiency
represent data from the previous JCU’s effluent EMCs from the system, were of approximately 75%.
research project. The exception is the typically within the range of 10 to 40 mg/L. Generally, the higher removal efficiency
JCU outlier represented as an open would be expected for the coarser
square, which has not been included in samples, and this was the case for all
this evaluation. The line of best fit shown three storms sampled.
as a solid straight line was calculated by a
least squares linear regression for all data Total Nutrients
points except the JCU outlier (intended
to be informational only). Its relative slope The system achieved an aggregate load
provides an appreciation of the trend of reduction for total phosphorus (TP) of
the removal efficiency for the treatment 47% (note, JCU recorded a load reduction
train. The dotted curves represent of 70%), the median influent and effluent
the 95% confidence limits for these EMCs for TP were 0.123 mg/L and 0.055
mg/L respectively (refer to Table 2).
same data points. The true statistical
Duncan (1999) and Fletcher et al. (2004)
significance of the regression lines is
recorded EMCs within a similar range and
open to interpretation and requires further
Fletcher (2004) recommends mean TP
investigation, due to the limited number
concentrations of between 0.25 and 0.50
of data points available for this analysis.
mg/L for most land uses. Similarly, BMP
Over the six storms analysed by Database (2010) suggests that a typical
Stormwater360, the influent EMC for SS range for TP concentrations in stormwater
<500 micron was in the range of 48 to is from 0.11 to 0.38 mg/L, across a range
180 mg/L with a median influent EMC of of land uses. In this context it is apparent
105 mg/L. Duncan (1999) literature review that the influent TP concentration at the
determined that the median concentration Figure 5. Total Phosphorus (SW360 and JCU Kuranda site is towards the very low end
for most land uses (roofs excepted) lies combined). of published data. Consequently, the 47%
4 SEPTEMBER 2011 water technical features
Page 13
refereed paper stormwater treatment
Table 3. Nitrogen results from Storm 6.
Influent EMC Effluent EMC Mean Removal Efficiency
Phase Analyte
(mg/L) (mg/L) (Sum of Loads)
TN 0.8 0.4 50%
Total TKN 0.8 0.34 58%
(dissolved and NH3-N 0.15 0.07 53%
particulate) Org-N 0.65 0.27 58%
NO3-/NO2--N 0.01 0.06 -500%
TN 0.4 0.3 25%
TKN 0.39 0.23 41%
Dissolved NH3-N 0.16 0.073 54%
Org-N 0.23 0.157 32%
NO3-/NO2--N 0.01 0.07 -600%
TN 0.4 0.1 75%
TKN 0.41 0.11 73%
Particulate
NH3-N 0 0 N/A
(by calculation)
Org-N 0.41 0.11 73%
NO3-/NO2--N 0 0 N/A
reduction recorded in the Stormwater360 biological uptake or denitrification Given the removal efficiency for
trial could be related to the difficulty in (in an anaerobic environment). suspended solids, the high removal
removing TP at very low influent EMCs, Consequently, the consistent removal of particulate organic nitrogen is
and a much higher removal rate (similar of TN exhibited by the system deserves understandable. Removal mechanisms
to the 70% recorded by JCU) could be further consideration. The majority (ca. for dissolved organic nitrogen are less
expected as the influent EMC increased. 95%) of the total nitrogen load at Kuranda obvious. It is possible that there is
is TKN and a breakdown of TN species some adsorption to the ‘schmutzdecke’
The system achieved an aggregate load is contained in Table 3.
reduction for total nitrogen (TN) of 44%, (bio-film) that develops on the cartridge;
while the median influent and effluent A small proportion of this TKN load another possibility is removal under
EMCs for TN were 1.045 mg/L and 0.615 (ca. 5%) is ammonia nitrogen, which the anaerobic conditions within the
mg/L respectively (Table 2). Again, this implies that ca. 90% of the total nitrogen standing water within the wet-zones,
influent EMC is low with respect to most load is present as organic nitrogen, in being the wet-sump and around the
either soluble or particulate forms. An base of the cartridge.
of the published data and, according to
expanded nitrogen suite analysis was
Duncan (1999), it correlates well with the When runoff first enters the StormFilter®,
conducted for Storm 6, and filtered (0.45
median for data from forested catchments it initially displaces the standing water
micron) and unfiltered samples were
(0.95 mg/L), but is significantly lower than in the wet-zones. Any pollutants in the
processed in order to establish whether
the median for roads (2.2 mg/L) or urban standing water are sampled by the effluent
the removal processes, for this event,
catchments (2.5 mg/L). Fletcher et al. sampler (once they have passed through
involved particulate removal or removal of
(2004) recommends using a typical total dissolved species. Essentially, the entire the StormFilter® cartridge), but they are
nitrogen value of at least 2 mg/L for most TN load was present as TKN and ca. 20% not sampled by the influent sampler.
land uses, with the exception of forested of this was ammonia-N (Table 3). Furthermore, the last of the runoff to
catchments. enter the cartridge during a storm event
The entire ammonia-N load was
The total nitrogen results from JCU soluble, and the treatment train system does not necessarily pass through the
and SW360 are presented in Figure 6. achieved 54% removal of this species. filter cartridge during that event and
The spread of influent EMCs is broad, The remainder (ca. 80%) of the TN/TKN may be retained within the wet-sump
but removal efficiency appears relatively load was present as organic nitrogen, of until the next storm event, whereupon
consistent and substantial. This is in spite which ca. 35% was dissolved. Overall, it is displaced. When the (particulate or
of the low influent concentrations. TN is 73% removal of particulate organic dissolved) organic nitrogen converts to
generally considered to be predominantly nitrogen and 32% removal of dissolved ammonia in the anaerobic wet sump, it can
soluble, which is best removed by organic nitrogen was achieved. be removed as ammonia-N by the zeolite.
Table 4. Grab samples from wet sump.
Antecedent Dry Diss. Cu Diss. Zn DOC Diss. N Diss. NH3-N Diss. NOx--N
Date Report #
Period (days) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L)
07/07/2008 8 40627 0.011 0.053 17 - - -
20/02/2009 6 42998 0.001 0.016 - 2.4 2.39 <0.01
06/05/2009 19 43826 0.005 0.082 16 7.2 5.85 0.72
21/07/2009 79 44703 0.004 0.083 20 3.4 2.24 0.025
water SEPTEMBER 2011 5
Page 14
stormwater treatment refereed paper
Periodic grab samples from the found that oxidised nitrogen comprised The Authors
wet-sump indicate that most of the only ca. 10% of the TN load, across
TN load in the standing water is present three different land uses, and most of
as ammonia-N at concentrations that the TN load was present as TKN and
are two orders of magnitude higher a significant proportion of this was
than typical influent ammonia-N particulate in nature. Consequently, the
concentrations. As such, ammonia-N measured TN load from the Gold Coast
Michael Wicks (email: michaelw@
is, possibly, generated in the wet-zones catchments is similar to that measured at
stormwater360.com.au) is Technical
by anaerobic decomposition of organic the Streets Creek, Kuranda site, providing
Director of Stormwater360 Australia.
nitrogen in the inter-storm event periods. applicability of Nitrogen removals to Nick Vigar is Research Manager and
This has two important implications: 1): various urban land uses. Mike Hannah is Technical Director, both
the load of ammonia-N passed to the
of Stormwater360 New Zealand.
StormFilter® cartridge is significantly Conclusions
higher than is suggested by the influent
The results from this field trial generally References
EMC, which implies that the removal
correlate well with an earlier study ANZECC, 2000: Australian and New Zealand
rates for ammonia-N removal may be
at this site by JCU (Munksgaard and Guidelines for Fresh and Marine Water Quality,
an under-estimate; and 2): by converting Volume 1. The Guidelines. Tables 3.3.4-3.3.5
Lottermoser, 2008). The data collection
organic nitrogen to ammonia-N in the Tropical Australia p.3.3-12 and Table 3.3.5
wet-zones and then removing this from this study has been based on a
p.3.3-13. National Water Quality Management
ammonia, the system has the potential rigorous and technically demanding Strategy, October 2000.
to remove soluble organic-N. monitoring program, which adds further
BMP Database, 2010: International Stormwater
credibility of the results (Goonetilleke,
Best Management Practices (BMP) Database
Discussion 2010). From an operational perspective, Pollutant Category Summary: Nutrients.
the system captured an appreciably large Prepared by Geosyntec Consultants Inc. and
The results for Storm 6 represent a
sediment load requiring annual cleaning Wright Water Engineers Inc. (available from
snapshot of one storm, and should not http://www.bmpdatabase.org).
to maintain its operational effectiveness.
be considered as comprehensive; they
Duncan HP, 1999: Urban Stormwater Quality: A
do suggest, however, that the main TN The EnviroPod®/StormFilter® Statistical Overview, Report 99/3, Cooperative
removal pathways for the treatment train treatment train achieved 78% removal Research Centre for Catchment Hydrology,
is the efficient removal of particulate for suspended solids under 500 microns, Melbourne, Australia. ETV (2004).
organic nitrogen, complemented by which approximates the long-term Fletcher T, Duncan H, Poelsma P & Lloyd S,
the sorptive removal of soluble environmental target recommended by 2004: Stormwater Flow and Quality, and the
ammonia-N and organic-N. NSW DECC (2007), QLD DERM (2010) Effectiveness of Non-Proprietary Stormwater
for South East Queensland (SEQ) and Treatment Measures – A Review and Gap
Very often TN removal is treated Analysis. Cooperative Research Centre for
consistent with the 80% reduction target
as a key performance benchmark for Catchment Hydrology, Technical Report 04/8.
stormwater treatment practices. This of many consent authorities in the US.
Goonetilleke A, 2010: Letter to Author, 15 March,
is potentially problematic, given the The runoff at Streets Creek contained 2010.
apparent variation in the nature of the very low levels of phosphorus and
Livingston EH & McCarron ME, 1992: Stormwater
TN load. In a comprehensive study of nitrogen. Total Phosphorus removal Management: A guide for Floridians.
nitrogen composition in Melbourne (Taylor was between 45% and 70% respectively Florida DER (71 pages).
et al., 2005), ca. 25% of the load was in both the Stormwater360 field trial
present as particulate organic nitrogen. Miguntanna NP, Goonetilleke A & Egodowatta
and the JCU research project, which P, 2010: Understanding nutrient build-up on
The remainder was soluble and, of these approximates the NSW DECC (2007) urban road surfaces. Journal of Environmental
species, oxidised nitrogen predominated and QLD DERM (2010) SEQ long-term Sciences, Vol 22(6), pp 806–812.
over dissolved organic nitrogen and environmental targets of 65% and Munksgaard NC & Lottermoser B, 2008:
ammonia-N. 60% respectively, and is better than Treatment of Road Runoff Waters, Kuranda
Taylor et al. (2005) inferred that either expected given the low influent EMCs. Range Project. Report for Queensland
Total Nitrogen removal was consistent, Department of Main Roads, School of Earth
‘removing’ the water by infiltration or
and Environmental Sciences, James Cook
denitrification (ie, in the anaerobic zone substantial and in agreement with the
University, Cairns, Queensland, Australia.
of bio-retention practices) would be NSW DECC (2007) and QLD DERM (2010)
SEQ 45% long-term environmental target, NSW Department of Environment and Climate
necessary to achieve significant TN
Change (DECC, 2007): Managing Urban
reduction. Fletcher et al. (2004) reported despite the proximity of the influent EMC Stormwater: Environmental Targets.
that the TN composition measured in wet to the irreducible concentration of the Consultation Draft – October 2007, Department
weather samples for various land uses treatment train. The removal of nitrogen of Environment and Climate Change NSW, p 4.
in the Sydney and Illawarra regions was was particularly noteworthy, given that QLD DERM, 2010: Urban Stormwater Quality
extremely variable. For urban catchments, the debris captured and stored within the Planning Guidelines 2010 – December 2010,
median oxidised nitrogen concentrations treatment train was not included in the Department of Environment and Resource
were in the range 0.09 to 0.42 mg/L, while influent load into the system, but may Management, Table 2.2 SEQ (2010).
the median TN concentration range was have been sampled as a soluble leachate Taylor GD, Fletcher TD, Wong THF, Breen PF &
0.65 to 2.32 mg/L. by the effluent sampler. Duncan HP, 2005: Nitrogen Composition in
Urban Runoff – Implications for Stormwater
The oxidised nitrogen represents
Acknowledgements Management. Water Research, Vol 39, pp
a much smaller proportion of the TN 1982–1989.
load than was observed by Taylor et al. The authors would like to acknowledge
White M & Pezzaniti D, 2001: Evaluation of Gully
(2005) for Melbourne data. In a study of the support of and contributions by Pit Inlet Control Systems Project Number:
nutrient build-up on urban roads in the Professor Ashantha Goonetilleke 2368261, Urban Water Resources Centre,
Gold Coast, Miguntanna et al. (2010) and Geoffrey Hunter. University of South Australia (20 pages).
6 SEPTEMBER 2011 water technical features
Page 15
Contaminant Investigation at Blackbutt Creek
for Ku-Ring-Gai Council
Prepared by:
Michael Hannah and Ian Hiles
Ingal Environmental Services
A Division of Industrial Galvanizers Corporation Pty. Ltd.
Sydney Office: 9/111 Old Pittwater Road, Brookvale N.S.W. 2100 Australia
Telephone: (02) 9907 1401 Facsimile: (02) 9907 1420
Auckland Office: 9/11 Arrenway Drive, Albany Auckland New Zealand
Telephone: (09) 476 5586 Facsimile: (09) 476 5582
1
Page 16
Table of Contents
Executive summary............................................................................................................. 1
Introduction......................................................................................................................... 5
Background ......................................................................................................................... 5
The Enviropod Filter....................................................................................................... 6
Project Methodology........................................................................................................... 7
Catchments...................................................................................................................... 8
Contaminant Characteristics ............................................................................................... 8
Enviropod Removal Loads ............................................................................................. 8
Removal Loads and Rainfall......................................................................................... 10
Removal loads and Intensity......................................................................................... 12
Removal Volumes......................................................................................................... 13
Grainsize Analysis ........................................................................................................ 14
Composition.............................................................................................................. 14
Seasonal Variation .................................................................................................... 17
Removal Load Curves............................................................................................... 19
Nutrient Analysis .......................................................................................................... 20
Organic Matter .......................................................................................................... 20
Phosphorous.............................................................................................................. 21
Nitrogen .................................................................................................................... 23
Calculated Loading Rates ................................................................................................. 24
Suspended Solids/Bed Loads/Gross Pollutants ............................................................ 24
Organic Content ............................................................................................................ 26
Nitrogen ........................................................................................................................ 27
Phosphorus.................................................................................................................... 29
3rd Party Modelling Comparison....................................................................................... 31
Conclusion ........................................................................................................................ 32
REFERENCES ................................................................................................................. 34
2
Page 17
Table of Tables
Table 1: Monthly Removal loads........................................................................................ 9
Table 2: Percentage below 1180 microns (Suspended Solids) ......................................... 14
Table 3: Percentage of above 1180 micron and below 4.75mm (bed Load) .................... 15
Table 4 Percentage above 4.75mm (Gross Pollutants) ..................................................... 15
Table 5 Organic content of samples.................................................................................. 20
Table 6 Concentrations of total nitrogen and total phosphorus (mg/kg) .......................... 21
Table 7: Loading Rates ..................................................................................................... 26
Table 8: Organic loading Rates......................................................................................... 26
Table 9: Nitrogen loading Rates ....................................................................................... 28
Table 10: Phosphorous loading Rates............................................................................... 30
Table 11 Average unit export rates (Middle Harbour Catchments) ................................. 31
Table of Figures
Figure 1: Sediment removal by the Gullypit Containing an Enviropod Filter with a 200
micron mesh (Tonkin & Taylor)................................................................................. 7
Figure 2 : Removal loads by land Use .............................................................................. 10
Figure 3: Monthly Loads with Rainfall ............................................................................ 11
Figure 4: Monthly Rainfall against Removal Load .......................................................... 11
Figure 5: Loads vs. Peak Intensity.................................................................................... 12
Figure 6: Gross Pollutants and Peak Intensity. ................................................................. 13
Figure 7: Volume Removal Curve................................................................................... 14
Figure 8: Particle Size Distribution vs. Land Use............................................................. 15
Figure 9: Average Composition all Enviropods .............................................................. 16
Figure 10: Season Variation of Gross Pollutants.............................................................. 17
Figure 11: Season Variation of Suspended Solids............................................................ 18
Figure 12: No Seasonal variation of Bed Load................................................................. 18
Figure 13: Gross Pollutant Loading Curve (CRC 1998) .................................................. 19
Figure 14: Gross Pollutant Loading Curve for Blackbutt Creek Catchment ................... 19
Figure 15: Total Phosphorous Total Organic Load .......................................................... 22
Figure 16: Monthly Phosphorous Loads.......................................................................... 23
Figure 17: Nitrogen Loads vs. Total Organic loads......................................................... 24
Figure 18: Total Loading Monthly Removal Rates .......................................................... 25
Figure 19: land Use Loading Rates................................................................................... 25
Figure 20: Organic Loading Rates by Particle size .......................................................... 27
Figure 21: Organic Loading Rates by land use................................................................. 27
Figure 22: Nitrogen Loading Rates for particle size groups............................................. 28
Figure 23: Nitrogen loading rates by land use .................................................................. 29
Figure 24: Phosphorous Loading Rates by land use......................................................... 29
Figure 25: Phosphorous Loading Rates for particle size groups ...................................... 30
3
Page 18
Executive summary
Introduction and background
Ingal Environmental Services (I.E.S) was contracted by Ku-Ring-Gai Municipal Council
(KMC) to undertake an investigation into the performance of the Enviropod units in
removing nutrients.
The aim was to investigate the contaminant characteristics associated with topography
and land use in a typical Ku-Ring Gai catchment. It was also intended to investigate how
nutrients are transported into the reticulation system and in turn into the receiving
environment.
Contaminates are transported in stormwater by four main mechanisms:
1. Dissolved.
2. In suspension (suspended solids).
3. Saltation or bed load (bouncing along the bottom).
4. Floating.
It is hypothesised that a considerable amount of the material in stormwater, is being
transported into the reticulation system as bed load or gross pollutants. This in turn
contributes a significant proportion of overall contaminants in stormwater. To date there
has been limited research carried out on quantifying the loads and effects of bed load or
gross pollutants
For the purpose of this study the following definitions for suspended solids, bed load and
gross pollutants have been adopted.
• Suspended Solids (TSS) <1mm
• Gross Pollutants (GP) >5mm
• Bed Load (BL) 1mm<x<5mm
Enviropod filters were retrofitted into 80 gully pits at the head of the Blackbutt Creek
Catchment. By installing the Enviropods in the gully pits, an extremely large screening
area and storage volume is obtained within the catchment. For the Blackbutt Creek
catchment the 80 Enviropods provided approximately 8.5 m3 of storage and 90m2
screening area.
1
Page 19
Removal Loads and Volumes
The highest loads were removed from the urban sub-catchment during the study period.
However this sub-catchment had the largest catchment area (12.75 ha) and the largest
number of Enviropods (52).
Removal Loads exhibited seasonal variations, with the highest loads encountered in
autumn and during months of high rainfall
Removal loads were related to rainfall. An analysis of monthly removal loads showed
that it had a log normal relationship with monthly rainfall. The analysis also indicated
that there may also be a relationship with gross pollutant loads and peak intensity of
rainfall during the month.
The in-situ density of removal material by the Enviropods in the study was determined to
be 176kg/m3 (damp). This is lower than the values determined by the CRC for gross
pollutants. However the CRC values were determined from wet gross pollutants from a
CDS unit.
Approximately 60m3 of material was removed from the Blackbutt creek catchment
during the year. This equates too 3.5 m3/ha/yr. The volume of material entrained in
urban runoff is an important factor in determining the appropriate sizing of a stormwater
treatment device
Considering the in-situ density and the available storage in the Enviropods filters it is
suggested that all Enviropods be cleaned after approximately 100mm of rainfall. A rain
gauge could be mounted in the catchment to help determine cleaning frequency.
Particle Size Analysis
It was observed that there were a greater percentage of suspended solid size particles as
you travel down the catchment. Inversely there were a greater percentage of gross
pollutant size particles as you travel up the catchment. It is suggested that finer particles
are being transported overland by wind and runoff.
On average and across the catchment the retained material trapped by the enviropod had
the following composition. 56% suspended solids, 32% bed load and 12% gross
pollutants. There is a considerable amount of material being transported as bed load
which is not usually considered when evaluating loads transported in stormwater.
As previously stated loads were related to the rainfall in the catchment. However particle
size composition of the loads was not related to rainfall. It was thought that higher
rainfall would cause a greater percentage of larger particles as these would be mobilized
during heavier rainfall. It was observed that there was a seasonal variation in suspended
solids and gross pollutants which was more influenced by wind than rainfall. Spring and
2
Page 20
summer months had lower suspended solid loading and high gross pollutant loading than
autumn and winter months. Winds are typically stronger in spring and autumn months in
Sydney. This observation has management implications for example. If mechanical
street sweeping is performed to lower suspended solids levels, this should be carried out
more in less windy months, alternatively if manual street sweeping is being performed to
reduce gross pollutants this should be done more frequently in windy months.
A gross pollutant vs monthly rainfall loading curve was developed. This was compared
with work carried out by the CRC for gross pollutant loads in Melbourne. A very similar
relationship was found for gross pollutants in Blackbutt creek catchment as was found in
Melbourne. However gross pollutant loads were only approximately 60% of those
encountered in Melbourne. This suggests that Blackbutt Creek has a low gross pollutant
loading.
Nutrient Analysis
The organic content of the retained material was surprisingly low in the retained material.
Only 10% of the retained material was organic. Overseas research has shown that
between 20% and 50% of suspended solids is organic. Work by the CRC in Melbourne
observed that 66% gross pollutants were organic and 80% of the material caught in a
CDS gross pollutant trap was organic. This may account for the low gross pollutant
loading discussed previously.
October was the month with the highest organic concentration. Generally organic
concentration was highest in particles over 1200 micron in size.
Surprisingly higher concentrations of phosphorous were found in larger particles. This is
contrary to the literature that suggests phosphorous is generally related to fine
particulates. It is suggested that since one of the principle source of phosphorous is from
rock and soil erosion that these particles are eroding into fine particulates as they travel in
the stormwater reticulation system.
Analysing the organic content of the retained material with the concentration of
phosphorous fund there was little relationship between them. Hawkesbury sandstones are
known to have low concentrations of phosphorous therefore it is hypothesised that runoff
and erosion from neighbouring properties are contributing to the phosphorous load in the
runoff from black but creek catchment. Educational activities could be carried out on
neighbouring property owners to encourage them to minimise erosion on there properties
in addition council should also ensure erosion of soils in not occurring within the road
berms.
High concentrations of nitrogen were encountered in October. This is associated with the
higher organic loadings during this month which has been attributed to the stronger
winds. Nitrogen concentrations in retained material were found to be related to the
organic content of the retained material.
3
Page 21
Loading Rates.
The highest loading rates for all sized particles were encountered in the commercial area.
High suspended solids loading rates were also observed in the base sub catchment. This
suggests that at source treatment is best applied in commercial areas and at the base of a
catchment as finer size particles often contain high concentrations of nutrients and heavy
metals.
High loading rates for nitrogen and phosphorous were encounter in the commercial area.
These high loading rates were present in large and small size particles. The high loading
rates for nitrogen in fine particles from the commercial area is attributed to mechanical
breakdown by traffic movement. It is suggested that street sweeping be performed in the
commercial area to prevent the break down of larger particles.
High loading rates were also encountered a base of the hill for nitrogen a phosphorous.
However the highest rates for phosphorous were in the larger particles.
3rd Party Modelling Comparison
Comparing the loading rates form 3rd part modelling by Hughes Truman engineering
consultants, it is estimated that the Enviropod is removing 65% TSS, 40% total
phosphorous, 21% total nitrogen and nearly 100% gross pollutants.
Not considered in the Hughes Truman’s export rates is the bed load. It was observed that
176 kg/ha of the exported load was bed load. This is a considerable mass not to be
considered. This has great implications when determining maintenance frequency of
treatment devices
Conclusion
Land use and topography greatly affect the loads and composition of the contaminates in
stormwater. Climatic conditions also have an effect on the loads and composition of
contaminates. Wind and rainfall have the ability to move and break down contaminants
making them more difficult to capture.
4
Page 22
Introduction
Ingal Environmental Services (I.E.S) was contracted by Ku-Ring-Gai Municipal Council
(KMC) to undertake an investigation into the performance of the Enviropod units in
removing nutrients.
The assessment involved collecting sediment and debris samples from the Enviropod
units during the first 12 months operation.
The aim was to investigate the contaminant characteristics associated with topography
and land use in a typical Ku-Ring Gai catchment. It was also intended to investigate how
nutrients are transported into the reticulation system and in turn into the receiving
environment.
The catchment was divided into 3 different areas based around land use and topography.
These were named as below;
1. Commercial: a commercial area with an arterial road at the top of the catchment,
2. Urban: a steep graded urban sub-catchment.
3. Base: an urban sub-catchment at the base of a hill, close to the receiving
environment, (Blackbutt Creek)
Background
Many factors influence the mobilisation and transportation of stormwater contaminates
from surfaces. Size, mass and density of the contaminants will greatly affect their ability
to be transported to the receiving water body. Contaminates are transported by four main
mechanisms:
5. Dissolved.
6. In suspension (suspended solids).
7. Saltation or bed load (bouncing along the bottom).
8. Floating.
Research by the CRC (CRC 1997) has identified a strong relationship between suspended
solids and more illicit contaminants such as nutrients, heavy metals and PAHs in
stormwater runoff. However sampling techniques which obtain data on suspended solids
loads tend not to measure bed or gross pollutants loads.
It is hypothesised that a considerable amount of the material in stormwater, is being
transported into the reticulation system as bed load or gross pollutants. This inturn
contributes a significant proportion of overall contaminants in stormwater.
To date there has been limited research carried out on quantifying the loads and effects of
bed load or gross pollutants. Work by Allison at the CRC (CRC 1998) examined gross
pollutant loads and identified a relationship between gross pollutants, organic matter and
5
Page 23
nutrients. While the concentrations of nutrients are generally lower in gross pollutants
they contribute a significant proportion to the overall transport sediment load.
There is much debate on the definition of suspended solids. Some agencies consider the
definition of suspended Solids to be < 63 micron (North Shore City Council, New
Zealand), while others have suggested 500 micron (City of Portland, USA). Some
agencies have adopted a finer definition of suspended solids as the finer fraction of
particles often have a higher concentration of contaminants which are also more bio-
available in the receiving environment. However, in this study sediment has been
obtained before it has entered the reticulation system and research (Pratt & Adams,
Hannah) has shown that particle size distribution of particles leaving the road surface are
often coarser than those sampled further down the reticulation system.
For the purpose of this study we will adopt the following definitions for suspended solids,
bed load and gross pollutants. These were obtained through correspondence with the
CRC for catchment hydrology.
• Suspended Solids (TSS) <1mm (T. Wong, CRC Catchment Hydrology, pers.com)
• Gross Pollutants (GP) >5mm (Allison, CRC 1998)
• Bed Load (BL) 1mm<x<5mm
The Enviropod Filter.
The Enviropod Filter is a gully pit insert comprising of a structural frame work, an
overflow mechanism and removal filter bag. The Enviropod filter was retrofitted into 80
gully pits at the head of the Blackbutt Creek Catchment. By installing the Enviropods in
the gully pits, an extremely large screening area and storage volume is obtained within
the catchment. For the Blackbutt Creek catchment the 80 Enviropods provided
approximately 8.5 m3 of storage and 90m2 of screening area.
The performance of the Enviropod Filter (200 micron bag) has been independently
evaluated buy Auckland City Council (ACC) in the laboratory (Tonkin & Taylor). This
was conducted using a range of synthetic sediment samples and sediment collected from
a road in Auckland. The Enviropod filter essentially removed all the sediments greater
than 100 micron in size. Removal of sediments less than 100 micron was approximately
20%. On road sediment the Enviropod Filter was found to remove approximately 97%.
Figure 1 shows the relative efficiencies for an Enviropod Filter fitted with a 200 micron
filter bag, at varying particle size ranges.
6
Page 24
Figure 1: Sediment removal by the Gullypit Containing an Enviropod Filter with a 200 micron mesh
(Tonkin & Taylor)
Project Methodology
The Blackbutt Creek catchment was separated into three different reticulated sub-
catchments. Each sub-catchment was characterised by the different types of land use and
geographic characteristics within the catchment. Different land uses and geographic
characteristics (in this case catchment grade) produce contaminants at different rates
It was proposed to clean all Enviropods in the catchment together, on approximately 2-
month cycles, however at the time of monitoring this was increased due to the high
loading of many of the Enviropods. Not all Enviropods were cleaned during each event
due to access problems associated with cars parked on pits and other obstructions.
As part of the cleaning, 5 samples were taken from the retained material from each sub-
catchment on every alternative clean i.e. 15 samples were collected from the Blackbutt
catchment. The 5 samples from each sub-catchment were combined in a composite
sample and sent to a NATA approved laboratory for grain size distribution. The grain
size distribution was conducted to examine the type of material collected in the
Enviropod filter and define the suspended solid load such that it is consistent with the
CRC catchment hydrology definitions of Suspended solids and the definitions for the
CRC “Music” modelling programme. The “Music” modelling package has assumed that
7
Page 25
particles less than 1mm are classified as suspended solids and particles greater than 5 mm
are considered gross pollutants ( T. Wong, CRC Catchment Hydrology, pers.com)
Total Nitrogen and Total Phosphorous was measured in particles under 1000 µm and
material over 1000 µm (1180µm screen actually used to meet the AS1289 3.6.1 Soils
analysis). Gross pollutants will be classified as particles over 4750 µm because of the
available sieve sizes.
Catchments
Three sub catchments were defined, commercial (Comm.), urban (Urban) and urban base
(Base), based on the differences in each catchment characteristics.
Urban and urban base sub catchments have essentially the same type of residential land
use, however the urban catchments is located on grade, with a moderate to steep
catchment grade. It was thought prior to the study that there would be a noticeable
difference in the type of material collected in each unit due to the steepness of the grade
mobilising different size particles. The other main difference with the urban sub
catchment is the pedestrian use. It was observed the urban sub catchment had a higher
pedestrian use the urban base sub catchment as pedestrians would use the area to park
cars.
The commercial sub catchment is situated at the top of the catchment and collects
stormwater runoff from retail outlets along the Pacific Hwy and from light
industrial/commercial properties in the north-west corner of the study area. This sub
catchment has high levels of pedestrians associated with the use of the retail areas and the
nearby train station.
Traffic flow in the area was observed to be moderate (based on a visual assessment of the
frequency of vehicles in comparison to other Enviropod projects) on all streets, with the
exception of several “No Though Road” streets, which had fewer vehicles and the Pacific
Hwy, which has high to very high vehicle movements.
Contaminant Characteristics
Enviropod Removal Loads
The following table indicates the total mass removed from all units as part of each
monitoring event. The total load removed during the study period was approximately
10.5 tonnes.
8
Page 26
Monitoring Base Commercial Urban Total
Event Removed
(kg)
March 249 387 493 1129
April 217 442 450 1109
May 327 379 887 1593
June 144.5* 272* 433.5* 850
July 148.92* 280.32* 446.76* 876
August 36 197 233 466
September 6 15 60 81
October 142 270 630 1042
November 104 154 359 617
December 96 216.5 337 649.5
January 134 401 534 1069
February 161 460 462 1083
Total 1765.42 3473.82 5325.26 10564.5
Subtotal for 1604.42 3013.82 4863.26 9481.5
Music
Modelling
Table 1: Monthly Removal loads
• Based on average composition
9
Page 27
Removal loads By Land Use
6000
5000
Rem oval Load (kg)
4000
3000
2000
1000
0
Base Commercial Urban
Land Use
Figure 2 : Removal loads by land Use
Figure 2 shows the highest loads were removed from the urban sub-catchment during the
study period. However this sub-catchment had the largest catchment area (12.75 ha) and
number of Enviropods (52).
Removal Loads and Rainfall
Removal Loads exhibited seasonal variations, with the highest loads encountered in
autumn and during months of high rainfall. The lowest loads were encountered in
August and September when there was low rainfall. Rainfall depth is not the only
mechanism involved in the transportation of contaminants in stormwater. Rainfall
intensity greatly affects the size and amount of contaminants transported in overland
flow.
10
Page 28
Monthly Removal Loads
1800 400
1600
Monthly Rainfall (mm)
1400
Rainfall
300
1200
Load (Kg)
Base
1000
800 200
Commercial
600
400 100 Urban
200
0 0 Total
Removed
Ju y
ne
Se Au ly
em er
ch
br r y
em t
ril
r
N cto r
D em r
ry
pt gu s
Ja be
O be
a
ov be
Ju
Ap
Fe nua
ec b
ua
ar
M
M
Month
Figure 3: Monthly Loads with Rainfall
Figure 4 below give a generalised removal curve for the Enviropods in the Blackbutt
catchment.
Removal load vs Rainfall
1800
1600
1400
Removal Load (kg)
1200
1000
2
800 R = 0.7358
600
400
200
0
0 100 200 300 400
M onthly Rainfall
Figure 4: Monthly Rainfall against Removal Load
11
Page 29
Removal loads and Intensity
Loads and Peak Intensity
Peak Rainfall Intensity
1800 10
1600 Total
Total loads (kg)
1400 8 Removoed
(mm/6mm)
1200 (kg)
1000 6
800 Peak
4 Intensity
600
400 2 (mm/6mins)
200
0J a be r 0
Ju e
S e u ly
y
Ap h
M il
nu r
e st
ay
De vem er
No c to r
ce be
r
ar
O mbe
n
c
pt gu
b
Ju
ar
m
M
A
Month
Figure 5: Loads vs. Peak Intensity.
It is apparent that there may be a relationship between rainfall and intensity. A
comparison of Figure 3 and Figure 5 shows that loads stayed the same in December,
despite the drop in rainfall; however there was a storm with high peak intensity in
December. This storm would have created high overland flow velocities which would
have mobilised larger particles into the Enviropods, especially for the steep grades
encountered in the Blackbutt catchment.
Figure 6 below show there may be some relation between peak intensity and gross
pollutant capture (R2=0.67). This correlation from a small data set suggests further
investigation could be carried out on the relationship between rainfall depth, intensity and
load. Possibly there is a threshold of intensity that will mobilise gross pollutants
12
Page 30
Gross Pollutants Vs Peak Intensity
200
Gross Pollutant Load
150
(Kg)
100 y = 100.4Ln(x) - 45.189
R2 = 0.6655
50
0
0 2 4 6 8 10
Peak Rainfall Intensity (mm/6 min)
Figure 6: Gross Pollutants and Peak Intensity.
Removal Volumes
Allison at the CRC suggests gross pollutants have a mass density (wet) of 250 kg/m3.
However the Enviropod retains captured material in a dry (damp) state with voids
between retained particles. During the study the level of fullness and weight of each
Enviropod was recorded at each service. A type A1 Enviropod contains 0.142 m3 of
storage. During the study it was recorded that the weights of a full type A1 Enviropod
were between 9 and 32 kg. Assuming a conservative value of 25 kg this gives an in-situ
density of 176 kg/m3.
Equating this to the total load removed over the year (10.5 tonnes) this equates to 60m3
of material was removed from the Blackbutt creek catchment or 3.5 m3/ha/yr. The
volume of material entrained in urban runoff is a important factor in determining the
appropriate sizing of a stormwater treatment device
Using this in-situ density, the following volume curve is derived Figure 7. In Blackbutt
Creek there are 80 Enviropods with a total storage volume of 8.6 m3, installed over a
total catchment area of 17.2 hectare. This equates to approximately 0.5m3 of storage for
each hectare. Assuming an Enviropod should be cleaned when it is 2/3 full, the
Enviropods should be cleaned after approximately 100mm of rainfall. This analysis
could be carried out on each sub-catchment to determine the most accurate frequency or
even on individual Enviropods. Ideally a rain gauge could be mounted within the
catchment to aid determination of cleaning frequency.
13
Page 31
Monthl Removal Voume vs Monthly Rainfall
0.6
Retained Volume (m3/ha)
0.5
0.4
0.3
0.2
0.1
0
0 100 200 300 400
Monthly rainfall (mm)
Figure 7: Volume Removal Curve
Grainsize Analysis
Composition
For comparison grain size analysis was limited to three main groups. Suspended solids
(particles below 1180 microns), bed load (particles between 1180 and 4750 microns) and
gross pollutants, particles over 4750 microns. Table 2, Table 3 and Table 4 present this
data below and Figure 8 compares the weighted averages for each particle size group in
relation to the land use.
Sub-Catchment
Month Urban Comm. Base
April 54% 55% 68%
June 64% 66% 66%
August 66% 46% 61%
October 49% 49% 66%
December 45% 33% 47%
Weighted Average 54% 51% 63%
Table 2: Percentage below 1180 microns (Suspended Solids)
14
Page 32
Month Sub-Catchment
Urban Comm. Base
April 41% 28% 25%
June 31% 25% 27%
August 25% 42% 31%
October 42% 24% 26%
December 27% 41% 46%
Weighted Average 35% 31% 29%
Table 3: Percentage of above 1180 micron and below 4.75mm (bed Load)
Month Sub-Catchment
Urban Comm. Base
April 5% 17% 7%
June 5% 9% 7%
August 9% 13% 8%
October 10% 27% 8%
December 28% 26% 7%
Weighted averages 11% 18% 7%
Table 4 Percentage above 4.75mm (Gross Pollutants)
Particle Size Distribution vs. Land Use
70%
60% Suspended Solids ( <1.08 mm)
50%
% of Load
40%
Bed Load (1.08mm<X>4.75mm)
30%
20%
10% Gross Pollutants ( >4.75mm)
0%
Comm. Urban Base
Land Use
Figure 8: Particle Size Distribution vs. Land Use
The data presented above indicate that material collected in the Enviropod units in each
of the areas tends to have a finer composition further down the catchment. Finer
particles are more easily mobilised by water and wind. Larger particles also settle out
easier as velocities lower. As flow reaches the base of the hill, the grade levels and flows
tend to lower. This causes large particles to settle. The settling would also occur as a
storm’s intensity subsides or if there was a change in surface.
15
Page 33
Inversely there were a greater percentage of coarse or gross particles higher in the
catchment. This suggests the large particles are not as easily transported overland
because of the greater mass. The percentage of bed load in the bed load did not seem
vary with topography.
This observation has considerable stormwater management implications as finer particles
are generally thought to have higher concentrations of contaminants which are more bio-
available and further investigation into the overland transportation of particles could be
carried out.
Figure 9 shows the average composition of the material retained across the catchment by
the Enviropod. As can be seen the majority of the material caught in the Enviropod can
be classified as suspended solids.
Only 12 % of the material caught in the Enviropod is gross pollutants. This has
significant management implications for stormwater treatment devices within Ku-ring-
gai, as removal of gross pollutants alone will result in the majority of the contaminants
entering the receiving environment.
32% of the material caught in the Enviropods can be classified as bed load. Bed load is
not usually considered when determining pollution loads in stormwater; however it is
obviously a large percentage of the material being transported into the stormwater
reticulation.
Average Composition of Material Removed by the
Enviropods From Blackbutt Creek
12%
Suspended Solids ( <1.08 mm)
Bed Load (1.08mm<X>4.75mm)
32% Gross Pollutants ( >4.75mm)
56%
Figure 9: Average Composition all Enviropods
16
Page 34
Seasonal Variation
The figures below examine the variation in composition of retained material throughout
the year. While loads are influenced by depth of rainfall, the composition of the load
does not seem to be effected by rainfall. It was though that higher rainfall would cause a
greater percentage of larger particles as these would be mobilized during heavier rainfall.
Figure 10 show the retained material had a higher percentage of gross pollutants in
October and December in the urban and commercial areas possibly this is because of
increase use of the shopping area in warmer months and Christmas. Conversely Figure
11 shows lower percentages of TSS during these months in these areas. It is assumed
that this phenomenon is caused by wind. Typically spring and summer months in Sydney
have stronger winds The stronger and more frequent winds in these months tend to move
suspended solids size particles from impermeable surfaces while generating more gross
pollutants for example ripping of leaves from trees. Figure 12 shows the bed load % of
the retained material is relatively consistent over all the months monitored.
Seasonal Variation of Gross Pollutants
% Gross Pollutants
50% 250
Rainfall (mm)
40% 200 Rainfall
30% 150 Urban
20% 100 Comm
10% 50 Base
0% 0
ne
ril
r
st
er
be
Ap
gu
ob
Ju
m
Au
ct
ce
O
De
Month
Figure 10: Season Variation of Gross Pollutants.
17
Page 35
Seasonal variation of Suspended Solids
70% 250
% Suspended
Rainfall
Rainfall (m m)
60% 200
Urban
Solids
150
50%
100 Comm
40% 50 Base
30% 0
ne
ril
r
st
er
be
Ap
gu
ob
Ju
m
Au
ct
ce
O
De
Months
Figure 11: Season Variation of Suspended Solids
No Seasonal Varation Of Bed Load
50% 250
40% 200 Rainfall
% Bed Load
Rainfall (mm)
30% 150 Urban
20% 100 Comm
10% 50 Base
0% 0
ne
ril
r
st
er
be
Ap
gu
ob
Ju
m
Au
ct
ce
O
De
Month
Figure 12: No Seasonal variation of Bed Load
This seasonal variation in suspended solids size particles and gross pollutants has
management implications. For example if mechanical street sweeping is performed to
lower suspended solids levels, this should be carried out more in less windy months,
alternatively if manual street sweeping is being performed to reduce gross pollutants this
should be done more frequently in windy months
18
Page 36
Removal Load Curves
Allison (CRC 1998) examined gross pollutant loads on a catchment wide base for
numerous storm events. From this work he derived the following (Figure 13) loading
curve for gross pollutants. This grading curve has become the basis for estimating gross
pollutants in the Music Model
Using the average results for gross pollutant loads on a monthly basis, a very similar
curve is derived for gross pollutants for the Ku-Ring-Gai Catchment; however loads are
approximately 40% less than those determined by the CRC. This suggests that the
Blackbutt Creek catchment has a low gross pollutant loading.
Figure 13: Gross Pollutant Loading Curve (CRC 1998)
Gross Pollutant Load Vs Monthly Rainfall
14
y = 2.3344Ln(x) - 3.3671
Goss Pollutant Load
12
R2 = 0.779
10
(Kg/hec)
8
6
4
2
0
0 50 100 150 200 250 300 350 400
Monthly Rainfall (mm)
Figure 14: Gross Pollutant Loading Curve for Blackbutt Creek Catchment
19
Page 37
Nutrient Analysis
Organic Matter
Organic concentrations encountered in the Blackbutt catchment were surprisingly low.
On average only 10% of the retained material in the Blackbutt creek was organic. There
has been a significant amount of research on organic content of stormwater contaminants.
Some of the organic concentrations observed in other studies are as follows:
• Ellis found that 30 to 55% of road sediments are organic,
• Kinglet Mitchell (ARC 1992) found that approximately 20% of suspended solids
in stormwater were organic.
• Lenhart (SMI) found that 50% particles retained in stormwater is organic.
• Allison found that 66% of the mass of gross pollutants is organic.
• Allison also found the 80% of the material caught in a CDS in Melbourne was
organic matter.
The low organic content may account for the low gross pollutant loading discussed
previously.
Commercial Area Urban Areas Bottom of Catchment
Sub1200 Over 1200 Sub1200 Over 1200 Sub1200 Over 1200
April NA NA NA NA NA NA
June 3.5% 2% 6.7% 5.9% 3.7% 3.7%
August 7.7% 5.5% 6.6% 10.5% 5.5% 11.8%
October 13.3% 19% 15.9% 24.4% 5.1% 22%
December 5.5% 3.8% 8.2% 8.4% 3.7% 8.5%
Average 7.50% 7.58% 9.35% 12.30% 4.50% 11.50%
Table 5 Organic content of samples
Kinglet Mitchell (ARC 1992) found that there is a general trend of increasing
concentrations of organic carbon in finer sediment in gutter dust. However higher
concentrations are found in coarser particulates in stormwater. This is possibly a
reflection of the floatable nature of plant debris.
Further, Kinglet Mitchell found catchments of low traffic movements have a higher
percentage of organics in coarser particles. This was also encountered in Blackbutt creek
with high organic carbon concentrations in particles over 1200 microns in the urban or
base catchments. This is because large organic particles such as leaves are easily
mechanically broken down by vehicle movement.
Table 5 details the observed percentage of organic material in the retained material
highest percentages were encountered in the month of October. This is assumed to be
because of the stronger winds in his month as discussed previously
20
Page 38
Phosphorous
Table 6 details the concentrations of phosphorous and nitrogen in the retained material.
Commercial Area Urban Areas Bottom of Catchment
Sub1200 Over 1200 Sub1200 Over 1200 Sub1200 Over 1200
Clean N P N P N P N P N P N P
Event
April 1600 553 2800 492 1700 421 2100 434 1400 859 2400 734
June 958 524 1762 623 710 662 1680 630 815 355 467 782
August 2840 692 1120 354 1760 445 2260 1060 1700 428 748 458
October 4520 735 6370 879 3930 694 7180 861 1410 452 10500 1390
December 746 47 1390 683 2800 738 4100 766 1330 550 1610 526
Average 2132 410 2688 517 2380 516 3464 672 1331 374 3145 645
Table 6 Concentrations of total nitrogen and total phosphorus (mg/kg)
Total phosphorus was, on average, greater in samples above 1200 microns, however the
values were inconsistent with research by the CRC that’s assumes 90% of Total
Phosphorous is associated with particles under 1000 microns. Kinglet Mitchell and
Allison found that dissolved and particulate concentration of nutrients was elevated from
wet sump GPT’s and sumped gully pits. This could be because of biological and
mechanical breakdown in the body of water during prolonged dry periods and turbulence/
re-suspension from a storm event. The Enviropod holds the material dry preventing
biological breakdown. This may be preventing the breakdown into suspended solid size
particles through the treatment device and or reticulation system. Further investigation is
warranted as this suggests that the nature of nutrients may change as it travels through the
stormwater system.
Analysis of nutrient concentration, organic matter concentration and particle size groups
revealed there is little relationship between concentration of phosphorous in the retained
sediment and organic concentration of the retained sediment Figure 15.
21
Page 39
TP Concentration vs Organic Content
1600
TP Concentration (mg/kg)
1400
1200
1000
800
600
y = 2588.3x + 423.54
400
R2 = 0.4279
200
0
0% 5% 10% 15% 20% 25% 30%
% Organic Content
Figure 15: Total Phosphorous Total Organic Load
Examination of phosphorous in stormwater shows that TP concentrations are related to
suspended solids concentrations (Kinglet Mitchell, CRC). However this is not necessary
organic material. While organic material contains phosphorous, often there majority of
phosphorous is generated by surface erosion. Weaver Et al
Phosphorous is a natural constituent in rock. Often dissolved phosphorus is created by
erosion. In Australia evidence has indicated that phosphorous arises from soil erosion
and is usually transported in episodic events (weaver et al), this was observed in large
storms during the study period. However Hawkesbury sandstones, lower in the
catchment are known to contain very low concentrations of phosphorus. Therefore it is
hypothesised that soil erosion from neighbouring pervious and landscaped garden areas
which drain to the Enviropods are contributing to the phosphorous load in the runoff.
Educational activities could be carried out on neighbouring property owners to encourage
them to minimise erosion on there properties in addition council should also ensure
erosion of soils in not occurring within the road berms.
Overland flow has the capacity to transport significant quantities of phosphorous in
various forms, while some of this is dissolved moving directly with the water, in some
catchments a much larger flux’s of phosphorous can occur via eroded particulate
material. In the Blackbutt Creek the steep slope would affect the overland flow velocity
and therefore the amount transported. This is also observed in Figure 16 where high
loads are experienced in months where there is high peak intensity (June and October). A
high intensity rainfall event will cause higher velocities of overland flow and therefore
greater erosion.
22
Page 40
Phospohorus Loads Sub 1200 Micron
140000
Comm
120000 Total P
(mg)
TP Load (mg)
100000
80000 Urban
Total P
60000 (mg)
40000 Base
20000 Total P
(mg)
0
June Augest October December Total
Months
Figure 16: Monthly Phosphorous Loads
Nitrogen
Table 6 shows high nitrogen concentrations in October in nitrogen in large particles.
This is related to the high organic content during these months
The sources of nitrogen are generally plant material and animal faeces. Kinglet Mitchell
found higher concentrations of nitrogen in coarse stormwater particles, the same was
found in the Blackbutt catchment, which was mostly likely related to leaf material
collected in the filter bags.
Figure 17, shows nitrogen concentrations in the Blackbutt creek catchment tend to be
related to the organic content of the sediment.
23
Page 41
Nitrogen Concentration vs. Organic Content
12000
Nitrogen Concentration
10000
8000
(mg/Kg)
6000
4000
2000 R2 = 0.7708
0
0% 5% 10% 15% 20% 25% 30%
Organic Content (%)
Figure 17: Nitrogen Loads vs. Total Organic loads
These observations have management implications as the removal of organic material or
all sizes can have a significant effect on nitrogen loads
Calculated Loading Rates
Loading rates were calculated for suspended solids, bed load, total nitrogen, total
phosphorus and gross pollutants. The total mass removed was used to estimate the loads
based on either the lab analytical data or from the grain size distribution.
Suspended Solids/Bed Loads/Gross Pollutants
Figure 18 and Figure 19 show that there was a variation in loading rate with the different
land use. The highest loading rates were encountered in the commercial area. This is
constant with the literature. High loadings were also encountered at the base of the hill.
This has a management implication for the use of at-source treatment. At-source
treatment should be focused in commercial areas and at the base of steep topography.
The figures also show that monthly loading rates across all land uses were proportional
and generally related to rainfall. This suggests that rainfall is important in determining
maintenance frequency.
24
Page 42
Monthly Loading Rates
Removal Load Kg/ha/Yr
180.0
160.0 Base
140.0
120.0 Commercial
100.0
80.0
60.0 Urban
40.0
20.0
0.0 Removal Rate
r
y
ch
ly
r
be
ay
be
ar
Ju
ar
M
m
nu
em
M
ve
Ja
pt
No
Se
Month
Figure 18: Total Loading Monthly Removal Rates
Land Use loading rates
700
Loading rate kg/hec/yr
600
500
TSS (kg/ha/yr)
400
BL (kg/ha/yr)
300
GP kg/ha/yr
200
100
0
Base Commercial Urban
Land Use
Figure 19: land Use Loading Rates
25
Page 43
Loading Rates Total
Urban Comm. Base Load
Particles < 1.18mm Loading Rates (kg/ha/yr) 227 660 516 328
Partials 1.18mm< X<4.75mm Loading Rates (kg/ha/yr) 147 394 238 196
Particles >4.75mm Loading Rates (kg/ha/yr) 44 234 60 76
Total Particles Load Rates For the Kg/ha/Yr 418 1287 814 600
Table 7: Loading Rates
Organic Content
Table 9, Table 10 and Table 8 details the compares the loading rates for Nitrogen,
Phosphorus and organic material with particle size and land use. High organic content
was encountered in the urban area, however low loads from this sub catchment resulted
in low loading rates
ORGANIC MATTER kg/ha/yr kg/ha/yr kg/ha/yr
Sub 1200 Microns Over 1200 Microns All Particles
Commercial 30 34 64
Urban 8 12 19
Base 60 67 128
Average 33 38 70
Table 8: Organic loading Rates
Higher organic material loading rate was encountered at the base of the hill. Organic
matter has lower specific density and is more easily transported by wind and water to the
base of the catchment. Also organic material can chemically absorb to the surface of fine
particles.
It was expected because of the high organic matter loading rates there would the base
would have had the highest nitrogen loading.
Nitrogen loading rates were generally higher in the bed load and Gross pollutant loads
26
Page 44
Organic Matter
80
70
60
50 Commercial
kg/ha/yr
40 Urban
30 Base
20
10
0
Sub 1200 Microns Over 1200 Microns
Category
Figure 20: Organic Loading Rates by Particle size
Orgainc Matter
140
120
100
Sub 1200 Microns
kg/ha/yr
80
Over 1200 Microns
60
All Particals
40
20
0
Commercial Urban Base
Catergory
Figure 21: Organic Loading Rates by land use
Nitrogen
NITROGEN Sub 1200 Microns Over 1200 Microns All Particles
kg/ha/yr kg/ha/yr kg/ha/yr
27
Page 45
Commercial 1.28 1.63 2.91
Urban 0.26 0.45 0.72
Base 0.68 1.66 2.34
Average 0.74 1.25 1.99
Table 9: Nitrogen loading Rates
The highest loading rate for nitrogen were obtained in the commercial area and at the
base catchment
The high nitrogen loading rate encountered in the commercial areas in all size particles
and at the base of the hill in the larger particles is attributed to the mechanical breakdown
of gross pollutants from the high traffic movements.
The high loading rate with fine particles is not consistent with the literature, however
these results show that in areas of high nitrogen loading and high traffic movements can
result in high concentrations in fine particulates. Street sweeping is encouraged in these
areas as it prevents the large organic gross pollutants from breaking down and becoming
more difficult to catch.
There was a low Nitrogen loading encountered at the base of the hill in suspended solids
size particles. There was also a low percentage of organic matter in this subgroup. This
is consistent with the literature (CRC 2003); however this suggests the Fine suspended
solids in the bottom of the catchment have not been generated by mechanical breakdown
of leaf material. The more probable source of the suspended solids size particles is
erosion and runoff from neighbouring properties and is further proven by the high
loading rates of phosphorous discuss latter.
Nitrogen Loading Rates
2.00
1.50
Commercial
kg/ha/yr
1.00 Urban
Base
0.50
0.00
Sub 1200 Microns Over 1200 Microns
Categories
Figure 22: Nitrogen Loading Rates for particle size groups
28
Page 46
Nitrogen Loading Rates
3.50
3.00
2.50
Sub 1200 Microns
Kg/ha/yr
2.00
Over 1200 Microns
1.50
All Particals
1.00
0.50
0.00
Commercial Urban Base
Categories
Figure 23: Nitrogen loading rates by land use
Phosphorus
The highest loading rates were equally encounter in the commercial and base sub
catchments as can be seen in Figure 24. There is a significant relationship between traffic
density and traffic loading. Mechanical break down of plant debris and earth will lead to
rapid release of Phosphorous (Cowen & Lee)
Phosporous Loading Rates
0.80
0.70
0.60
0.50 Sub 1200 Microns
kg/ha/yr
0.40 Over 1200 Microns
0.30 All Particals
0.20
0.10
0.00
Commercial Urban Base
Catorgory
Figure 24: Phosphorous Loading Rates by land use
29
Page 47
Surprisingly lower loading rates for phosphorous were encountered in the suspended
solid size particles than in the bed load and gross pollutants. This is inconsistent with the
literature. As suggested one of the principle sources of phosphorous is rock erosion. It is
possible that the eroded particles are continuing to erode and break down as they travel
with in the runoff. Capture of these particles before they enter the reticulation system
prevents further break down. All studies in the literature have tended to sample
stormwater at a downstream point for an entire catchment allowing entrained particles to
brake down in the reticulation system before they are sampled. This would explain the
higher loading rates in larger particles encounter in the Enviropod.
Phosphorous Loading Rates
0.50
0.40
Commercial
kg/ha/yr
0.30
Urban
0.20
Base
0.10
0.00
Sub 1200 Microns Over 1200 Microns
Category
Figure 25: Phosphorous Loading Rates for particle size groups
PHOSPOROUS Sub 1200 Microns Over 1200 Microns All Particles
kg/ha/yr kg/ha/yr kg/ha/yr
Commercial 0.36 0.38 0.74
Urban 0.07 0.08 0.15
Base 0.30 0.43 0.73
Average 0.24 0.30 0.54
Table 10: Phosphorous loading Rates
30
Page 48
3rd Party Modelling Comparison
The following information was supplied from Hughes Trueman. Hughes Trueman
conducted contaminant modelling using the Music model loading on various catchments
for KMC.
The following table indicates the average loading from the seven catchments used in the
modelling exercise. These values were simulated in a wet year, dry year and average year
to determine the variation based on rainfall.
Year Rainfall TSS TP TN GP kg/ha
mm kg/ha kg/ha kg/ha
Dry (10%ile) 868 203 0.4 3.4 45
Average 1383 539 1.2 9.5 73
Wet (90%ile) 1972 786 2 16.2 115
Blackbutt Creak Enviropod
Removal Rates 1127 294 0.36 1.4 68
Table 11 Average unit export rates (Middle Harbour Catchments)
The comparison of loading rates with the removal loading rates obtained in the Blackbutt
creek catchment show that significant removal was being obtained.
The CRC assume (CRC 2003) GPTs have a removal efficiency of 30% for Phosphorous
and 13% for Nitrogen. Comparing the loading rates form Hughes Truman, it is estimated
that the Enviropod is removing 65% TSS, 40% total phosphorous, 21% total nitrogen and
nearly 100% gross pollutants.
Applying the results of the ACC laboratory testing (Tokin & Taylor) and the CRC
adopted PSD for the Music model; the Enviropods should have obtained a removal of
48% for TSS. Observations on the site suggest that the particle size distribution in the
runoff from system is much coarser than those adopted for Melbourne and Brisbane.
The higher removal rate obtained by the Enviropod filter than a standard GPT device is
attributed to the following:
1. The Enviropod filter fitted with a 200 micron filter bag has the ability to remove
fine particulates. Auckland city council laboratory testing has shown 20%
removal efficiency for particles less than 100 micron in size and 98% efficiency
for particles 100 micron to 1000 micron in size. (Tokin & Taylor).
2. The hypothesis that soil and rock particles entrained in the runoff are eroding and
breaking down as they are travelling overland and in the reticulation system.
Capturing of the particulates at the source before they enter the reticulation
system limits this mechanic break down of the particles
31
Page 49
Not considered in the Hughes Truman’s export rates is the bed load. It was observed that
176 kg/ha of the exported load was bed load. This is a considerable mass not to be
considered. This has great implications when determining maintenance frequency of
treatment devices
Conclusion
It is apparent from this study that land use and topography have a large effect on
contaminant loading and composition. It was also found that climatic conditions also
influence loading and composition. Wind and rainfall have the ability to move and break
down contaminants making them more difficult to capture.
It was generally found that the highest rates for contaminants were found in the
commercial area. High loading rates were also encountered at the base of the hill on
which the sub-catchments were located.
Loads and volumes were related to rainfall and it is suggested that all Enviropods in the
catchment should be cleaned after 100mm of rainfall
Composition of the retained material was found to get finer as you travel down the
catchment. High overland flow velocities and wind is thought to be transporting particles
from the top of the catchment to the base of the catchment, down the steep slope. These
climatic forces could also be eroding the particles
The composition of the retained material showed a seasonal variation. The retained
material was coarser in spring and in early summer months. It is suggested this is from
wind as these months historically have stronger winds.
It was found that the concentration of nitrogen in retained material was related to the
organic content. It was also found that nitrogen and organic loadings were highest in
windy months when leaves could be ripped from trees
Phosphorous concentration was not related to organic content of the retained material. It
was therefore suggested that a possible source of phosphorous was erosion of soil and
rock from neighbouring properties. Higher phosphorus loads were encountered in large
particles than finer particles; this is contrary to previous research which has sampled
stormwater at the end of a reticulation system. It is thought that eroded soil and rock
particles are breaking down as they travel entrained in reticulation system.
The composition of material entering the enviropod is mainly made up of suspended solid
size particles. The Blackbutt creek catchment had a low gross pollutant load. This is
thought to be associated with the low organic load from the site. Approximately 32% of
the retained load (196 kg/ha/yr) was bed load. Bed load should be considered in export
32
Page 50
rates from catchments and considered when sizing treatment devices as it is a
considerable percentage of material entering the reticulation system
The use of gully pit inserts should be targeted in commercial areas and at the base of
steep slopes to obtain the highest removal. Use of other at-source treatment such as street
sweeping should consider climatic effects when determining there frequency of use
Comparison of third party modelling results show the enviropod Filter located at the
source of the contaminates is an effective method of capturing gross pollutants, bed load
suspended solids, nitrogen and phosphorus
33
Page 51
REFERENCES
Ellis, J.B. Revitt, D.M. 1989 Traffic related Pollution on highway Surfaces Sediments:
report 18 urban Pollution. Middlesex Polytechnic, research and consultancy. July 1989
Hannah, M M, 2001, Enviropod Wairau Road Study, Ingal Environmental Services, Feb
2001
Kinglet Mitchell and Associates, ARC 1992, An assessment of Stormwater Quality and
implications for treatment of Stormwater in the Auckland Region, January 1992
Allison (1998), From roads to rivers, Gross pollutant removal from Urban waterways,
CRC for Catchment Hydrology may 1998
ANZECC (2001), Water quality Guidelines for fresh and marine Waters, Australian and
New Zealand Environment and conservation council
CRC (2003), Music User Guide Version 2, CRC for catchment Hydrology, December
2003
CRC 1997, Mudgway, Duncan, McMahon and Chiew, Best Practice Environmental
Management Guide lines for Urban Stormwater CRC for catchment Hydrology October
1997
Duncan CRC (1999), Urban Stormwater Quality: a statistical Overview, February 1999
Pratt, C J: Adams, J R 1981, Sediment wash off into road side gullies, Second
international conference on urban drainage, Ubana, June 1981
Tonkin And Taylor 2002, Auckland Cities Field and Laboratory testing of Stormwater
Catchpit Testing., NZWWA Stormwater Conference, Rotorua, May 2003
Weaver, D. Phosphorous in the landscape, Diffuse Source to surface waters. Murray
Darling Basin Commission, May 1997.
Hunter, G. Procedure for determining the appropriate Size and maintenance of
requirements of stormwater quality Improvement Devices.
OZ Estuaries, Coastal Indicator Development and Information System, OZ Estuaries
website. http: //www.ozestires.orf.html
Cowen, W.F. & Lee GF 1973, leaves as source of Phosphorous, Environ Sci & Techno
vol7 no9.
34
Page 52
AUCKLAND CITY’S FIELD AND LABORATORY TESTING OF STORMWATER CATCHPIT FILTERS Kevin Butler, Principal, Tonkin & Taylor Lid Grant Ockleston, Manager Stormwater Projects, Auckland City Council Mark Foster, Tonkin & Taylor Ltd ABSTRACT The paper documents the resulis from a field and laboratory testing programme undertaken by Auckland City Council to compare the performance of commercial Catchpit Filter Systems (CFS). A catchpit filter takes the form of a fine-mesh filter bag that is inserted inside a standard catchpit. The need for the testing arose from the City’s plans to install several hundred CFS units if they met satisfactory field and laboratory performance targets for capturing sediment washed off roads and paved surfaces. Four suppliers volunteered CFS units for the field testing programme. Each was installed in a City street and observations made over a five-month period covering: ease of fitting, sediment retention, maintenance needs. rigidity/strength, ability to catch flows and the effects of litter/organics. Laboratory testing was carried out on CFS units from two suppliers. The testing, carried ont at the University of Auckland, sought to quantify sediment capture performance. It also sought to determine the head loss charactcristics of the filter fabric to establish its potential to limit the hydraulic capacity and cause flow to bypass the CFS unit. In addition, a catchpit without the CFS unit was tested. Testing was carried out for a range of flow rates and sediment cencentrations. In summary, for a composite “street sweep sediment” sample, the CFS units were found to capture between 789 to 97% of the sediments entering the catchpit. KEYWORDS catchpit filter systems, filter bag/fabric, performance, field observations, operation and maintenance, laboratory testing, sediment capture, head loss. 1. INTRODUCTION Auckland City lies on the isthmus between the Waitemata and Manukau harbours. The harbours and surrounding waters are used extensively for trade and recreation, and contain a number of important ecosystems. The treatment of stormwater runoff is essential if the harbour and its surrounding waters are to be preserved. Studies have shown that by removing the sediment contained in stormwater, a large amount of the pollutants will be removed (Dept. of Land and Water Conservation NSW). It is found that pollutants attach to the sediments contained in stormwater. Traditional methods of stormwater treatment, such as sediment retention ponds, swales, and infiltration trenches, work by removing the sediment from the stormwater runoff. They often require large arcas of land to function effectively. Because of the limited availability and high cost of land in the more developed parts of the City, retrofitting these large stormwater treatment devices is extremely difficult. Catchpit Filter Systems (CFS) are essentially fine mesh bags that are inserted into street catchpits to capture sediments washed off the road in wet weather. The use of CFS as a means of stormwater treatment does not require the purchase of additional land, or the use of recreational reserves that the council may already control.
Page 53
The City is considering the use of CFS devices to treat road runoff primarily in the central business district which combine high traffic volumes and limited space to employ other treatment devices. CFS units are a relatively new method of stormwater treatment aimed at capturing stormwater pollutants close to their source. Being a relatively new treatment method, limited information was available on the operation and sediment removal performance of CFS. Previous performance investigations on some products has shown it is very difficult to carry out conclusive field testing. The City sought to identify CFS products that met their performance targets before progressing to large scale installation of CFS. A study was carried out to compare the performance and the effectiveness of four different CFS. The four suppliers of the four different CFS all took part in the study on a voluntary basis. The study consisted of © A Field Trial e Laboratory Testing 2 FIELD TRIAL The aim of the field trial was to compare the operational performance of the four different CFS by observing their installation, operational performance and maintenance requirements. The field trial was a comparative investigation of the four CFS based on qualitative observations of their operational performance under similar weather conditions. The individual catchpits or contributing catchments were not identical, requiring observations to take account of the differing characteristics, The field trial looked at the qualitative performance of the CFS under conditions that the laboratory testing could not simulate. « The field trial was tindertaken by Tonkin & Taylor (T&T) in the Newmarket/Grafton area of Auckland city from 21 March to 20 August 2002. 2.1 FIELD TRIAL METHODOLOGY CFS product suppliers were invited to install two CFS units in two trial catchpits identified by Auckland Ci Prior to the installation of CFS, all catchpits were cleaned and downstream pipes were checked for blockages. During the trial period, suppliers were responsible for deciding when any maintenance (including clea the CFS was required. Suppliers were also requested to meet with Auckland City's maintenance contractor, discuss suppliers’ maintenance/cleaning requirements. Suppliers were encouraged to be present during maintenance operations so they could direct the maintenance contractors, : Operational performance observations for each CFS product were undertaken every fortnight and during and. after significant rain events. Inspections were carried out by the same two T&T staff members throughout the trial to ensure consistency of observations and records. . The performance of each of the CFS was compared on a number of attributes, as described below. (a) Ability of the CFS to capture and retain sediment (b) CFS maintenance characteristics (c) CFS physical fit within the catchpits (d) Rigidity and strength of CFS (e) CFS ability to catch road runoff (f) Effect of litter and organics on the CFS performance The information from the field observations was then used to summarise the operational perfc each of the above performance criteria. These results were then used to focus selection for: testing of the CFS products.
Page 54
2.2 FIELD TRIAL RESULTS Of the four CFS observed in the field trials, Auckland City determined that two CFS products met operational performance requirements for their project, and these were selected for further laboratory testing. The two selected CFS performed favourably on all six performance attributes listed in 2.1 above. Some interesting issues encountered with some of the CFS in the field trials were: e Unnecessary use of the CFS overflow mechanism, resulting in untreated flows. © Mechanical overflow mechanisms such as flaps were prone to failure by blockage. « Flows bypassing the filter bags due to gaps in the seal between the CFS and the catchpit wall. These gaps were a result of poor installation of the CFS. ¢ A small sediment storage capacity, which would resuit in a high cleaning frequency or loss of captured sediments during high flows. e One of the CFS did not have an overflow mechanism to bypass the filter bag under high flows. This resulted in localised ponding/flooding around the catchpit when flows exceeded the filter bags capacity. e One filter bag was seen to invert causing the sediment that had been captured to wash down the overflow mechanism and out of the outlet pipe. * The filter bag from one of the CFS was sucked down the catchpit outlet pipe, creating a blockage and decreasing the capacity of the outlet. 2.3 FIELD TRIAL OUTCOMES The two main outcomes of the field trial were 1. A CFS ability to function well is very dependant on how well it is installed. Careful specification of contract requirements and installation monitoring is required to ensure good quality control of CFS installation. vn It is essential that CFS have an efficient and reliable overflow mechanism. Mechanical flaps observed in the field trial generally were unsuccessful. 3 LABORATORY TESTING The laboratory study was conducted by Auckland Uniservices, in the School of Engineering at the University of Auckland. The laboratory tests were used to quantify the effectiveness of the CFS under different flow rates and with different sediment particle sizes and concentrations. This provided a simplified and controlled representation of field conditions to test the relative performance of the CFS. The laboratory data also enables the performance of these products to be predicted from a measured sediment particle distribution. Laboratory tests were also carried out to determine the head loss across the two synthetic filter fabrics to establish the potential for limiting catchpit hydraulic capacity and relative overflow potential between different CFS.
Page 55
3.1 LABORATORY TESTING METHODOLOGY The laboratory setup consisting of a rectangular channel, wooden apron, hopper, and catchpit containing t CFS is shown schematically in Figure 2. The laboratory catchpit was constructed of plexiglas to dimensio similar to units in the field (catchpit dimensions are presented in Figure 3). The catchpit received water from skewed apron (sloped in the longitudinal and transverse directions at 10° and 6°, respectively, to the horizon! plane) so that flow occurred primarily along the lower edge of the wooden apron. The test procedure involved feeding synthetic road sediment, derived by adding specified amounts of siz sediments to running water, to the catchpit at various flow rates. Tests were performed for five flow rates (C Vs, 1 Vs, 4 Is, 12 Vs, 20 Is), four sediment concentrations (50 mg/l, 150 mg/l, 250 mg/A, 400 mg/l) and fc particle sizes (<100 ft, 100-500 1, 500-1000 pL, 1000-10000 2), yielding a total of 80 tests for each of t following three cases: e Catchpit without a CFS (base case) *® Catchpit with an Enviropod CFS e@ = Catchpit with a Flogard CFS. Constant inflows of synthetic runoff to the CFS were maintained for a minimum of four hydraulic retenti times, corresponding to approximately six minutes for 20 l/s and 30 minutes for 0.5 i/s. The overall C performance was based on the efficiency of sediment capture under specified flow rates, particle sizes, a sediment concentrations. In addition, each of the three cases were tested using street sediments to simulate particle entrapment in- field. The street sediments were obtained by vacuuming a number of streets in Mt Roskill within the Oak Creek catchment. The CFS tests with street sediments were conducted for flows of 4 I/s, 12 l/s and 20 l/s, w an influent sediment concentration of 250 mg/l. Head loss tests were performed on the filter fabric by passing various flowrates across the filter fabric, measuring the head on either side of the filter fabric using a peizometer. Figure 1 below shows a schematic the laboratory set up. Figure 1: Laboratory Setup for Headloss Measurements Head Loss Filter Fabric
Page 56
Figure 2. Experimental setup of Catchpit
Inflow
Outflow Catchpit
Figure 3: Catchpit Dimensions
x
£60 15
pon 0
ate
0 15
« |
2
m CC
1795,
15,
t
A-A section
B-B section
Page 57
3.2
% Removal
LABORATORY TEST RESULTS
Test results for sediment removal for the three cases is presented in Figures 4, 5 and 6.
Figure 4 - Sediment Removal for the Base Case Catchpit With No CFS
100 Oana a
‘a i
‘ _ Particle Size (u)
80 ~———___-—.--__4_ = “00070000 7
4 | jae 500-1000
‘ ! ‘ , eet - 100-500
60 -| a. | fp @ 0-100 .
= a Street sweep
ee fee pee tr
a ay
40 + - i
*,
20 12. a : La _~— es
: ge ceecse ee ee ROT
° 777-9
) ‘ \
0 5 10 15 20
Flow Rate (I/s)
Figure 5 - Sediment Removal by the Catchpit Containing an Enviropod CFS with a 200 sun mesh
% Removal
100 +-@8-- =a = =
80) - -
—— 1000-10000
60 he _ eee tae ce - econ ee mea - —B— 500-1000 '
: i —a~ 100-500
- - 0-100
40 4 wor =
20 + wget Orne tees yt
*
4
0 t t T
0 5 10 15 20
Flow Rate (I/s)
Page 58
Figure 6 - Sediment Removal by the Catchpit Containing a Flogard CFS with a 400 jun mesh 100 +- #- a ; ea jet sme —-— epee ee _ | ee FP or remeron ty , Particle Size(uy 3 60 4 Ho nn tn onner menace onli 000 = 10,0001 2 i ' |+@—500 -1,000 £ . ‘|-na- 100 -500 X 40 4-- event be fe ie on te eft 2 < 100 i O_ Streetsweep 204 . fect obo Orr : es ~~ TTT ete ee ee eee eee ecb eT 0 + ¢ t + 0 5 10 15 20 Flow Rate (l/s) * Figure 4 above shows the results in the standard catchpit without a CFS installed. Sediments larger than 500u. were completely removed for flows up to 20 I/s. The removal of 100-500 particles decreases from 75% to 47% by increasing the flow from 0.5 I/s to 20 W/s, and only 18% of particles smaller than 100, are wapped by the catchpit. Approximately 58% of the sediment collected from street sweepings were removed at 4 I/s. Figures 5 and 6 show the laboratory performances of the Enviropod CFS and a Flogard CFS. The performance of both CFS was quite similar. The Enviropod CFS sediment removal efficiency was slightly greater than the Flogard for the finer sediment. These CFS remove essentially all the sediments >100y, while the removal of sediments <100p is similar to the standard catchpit with no CFS. The sediment removal efficiencies of the street sweep sediments at 4, 12 and 20 I/s are 97%, 89% and 88% respectively for the Enviropod CFS, and 86%, 79% and 78% respectively for the Flogard CFS. For the street sweep sediments the Enviropod CFS gave approximately 10% greater removal than the Flogard CFS due to more abundant fine sediments present in the street sweep sediments. 4 DISCUSSION 41 EFFECTIVENESS OF CATCHPITS The laboratory results indicate that catchpits without CFS installed are effective at removing coarser sediments. Results show they have the potential to remove nearly 100 percent of sediments greater than 500 microns in size. Tests undertaken with road sediments collected from the Oakley Creek catchment demonstrated this related to a sediment capture efficiency of 58%. It would be expected that these efficiencies would reduce as greater volumes of sediment accumulate within the catchpit and washout of the sediment occurs due to disturbance under high flows. Similarly, efficiencies could be maintained with regular cleaning of the catchpits. At present catchpits in Auckland City are cleaned three times per year. A more regular cleaning frequency or targeted cleaning of catchpits receiving high sediment loads may increase annual sediment capture volumes of
Page 59
existing catchpits. Alternatively, for new catchpits, changing the design of catchpits to increase sediment storage volume could increase sediment capture volumes. Clearly, the addition of CFS within catchpits provide a level of protection to accumulated sediments not provided in a standard catchpit. 4.2 ADDITIONAL BENEFIT OF CFS IN CATCHPITS The test results show that CFS can increase sediment removal above that expected from a standard catchpit. The additional benefit of these devices is primarily in the 500-100 micron particle range. The increase in the percent removal efficiency by removing this particle range will vary depending on the constituents of the stormwater runoff. In the tests using street sediments from the Oakley Creck catchment, the addition of a CFS to the catchpit increased the overall sediment removal efficiency from 58% to almost 90%. The removal of additional sediment may not be entirely due to sediment becoming trapped by the CFS filter material. Neither of the CFS tested had pore sizes as small as 100 microns yet both products still managed to remove particles below this size, Field trial observations found that the use of CFS resulted in less turbulence within the catchpit. The reduced turbulence may provide more favourable conditions for sedimentation and hence the removal of particles smaller than the filter bag perforations. Alternatively the particles may be adhering to the filter bag. As discussed above a valuable outcome of the installation of CFS would be the protection of accumulated sediment from washout during higher flows. 4.3 LABORATORY VS FIELD EFFICIENCY It is worth noting that the above CFS sediment removal efficiencies are based on laboratory conditions. These are the sediment removal efficiencies from stormwater runoff that the CFS unit receives. In a field situation not all stormwater runoff is received by the CFS, This is due to many factors including: * Poor installation resulting in stormwater bypassing the CFS and entering the catchpit directly. « High flows bypassing the filter bag via the engineered overflow. « Reduced filter capacity due to filling with sediment, increasing the incidence of engineered overflow. 4A FINDINGS * Modified catchpits or catchpits with CFS could be a practical and economical form of sediment capture in retrofit situations such as heavily built up urban areas. ¢ Existing standard catchpits are capable of high annual rates of sediment capture. Logically, the more frequently they are cleaned, the greater the annual volume of sediment capture. « At the time of this study, some CFS products compromised the hydraulic capacity of the catchpit. Clearly, any CFS system must not in any way compromise the hydraulic design capacity of an existing catchpit. ¢ Most current off the shelf CFS units are designed to be gross pollutant traps and/or fuel oil traps which trap only the coarser sediment particles. It appeared that most CFS products would be able to be modified to focus on sediment capture. © The field installation of CFS products needs to carefully specified and be well supervised, especially for new entrants (o the CFS market. 5 ENVIRONMENTAL BENEFITS There are many stormwater treatment systems that provide various single or combined levels of environmental benefit. These systems include coarse liner taps, fuel oil removal systems, chemical traps, nutrient removal systems, sediment traps and the like. This study has focused on the sediment capture efficiency of CFS products. Test results have shown that an appropriately maintained standard catchpit can remove the majority of sediment down to 500 microns in size.
Page 60
Test results show that with the addition of CFS, the majority of sediment down to 100 microns in size can be removed. If the removal of particles greater than 500 micron provides an opportunity to halt degradation or improve the quality of the Auckland environment then a focus on standard catchpits would be a step in the right direction. If the requirements are to remove down to 100 microns then the addition of a suitable CFS would equally be a step in the right direction. Identification of the relationship between sediment particle size removed and the volume of various pollutants removed has not been part of this study. DISCLAIMER The intent of this paper has not been to proclaim any particular CFS brand name or supplier as being “better” than another, The authors acknowledge that suppliers may have changed the character or form of CFS products since this study was carried out, and this paper should not be used as a basis for users to select one product aver another. REFERENCES DLWC (1998) ‘The Constructed Wetlands Manual’ Department of Land and Water Conservation, New south Wales, Volume 2. ABBREVIATIONS CFS — Catchpit Filter Systems T&T Tonkin & Taylor Ltd Environmental & Engineering Consultants Ltd
Page 61
Evaluation of Gully Pit Inlet Litter Control
Systems
Final Report
PROJECT NUMBER: 23 68261
Prepared for Brisbane City Council
GPO BOX 1432
BRISBANE QLD 4001
Att.: Mr. M Steer
Prepared by Mr. Michael White
Mr. David Pezzaniti
Urban Water Resources Centre
University of South Australia
MAWSON LAKES SA 5095
Telephone +618 8302.3652
Facsimile +618 8302.3386
Date of issue 10 April 2002
Distribution Client (2)
Urban Water Resources Centre
Authors (1)
Important Notice
This report applies only to the subject of the project. University does not accept
ISO 9001 responsibility for the conformity or non-conformity of any other subject to the findings
QEC6382 of this report.
This report is confidential and was prepared exclusively for the client named above. It is
not intended for, nor do we accept any responsibility for its use by any third party.
This report consists of one cover page, twenty pages of text and seven pages of
appendices. It may be reproduced only with permission of the copyright owner, and
only in full.
Page 62
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
TABLE OF CONTENTS
SUMMARY ....................................................................................................................... 3
1. INTRODUCTION AND PREAMBLE .................................................................. 4
2. TEST SPECIFICATION AND PROCEDURE......................................................... 5
3. TEST RESULTS AND OBSERVATIONS................................................................ 7
3.1 36MM HORIZONTAL BAR...................................................................................... 7
3.2 “MESHED” PIT INSERT........................................................................................ 10
3.3 TRAY BASKET ................................................................................................... 12
3.4 ENVIROPOD TESTING ........................................................................................ 14
3.5 CHAMBER WATER LEVEL PRE-TESTING ........................................................... 19
4. CONCLUSION ........................................................................................................... 20
APPENDIX A ................................................................................................................. A1
APPENDIX B ................................................................................................................. B1
Urban Water Resources Centre 2 University of South Australia
Page 63
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
SUMMARY
Stormwater pollution control is usually managed near source or at the
end of a drainage system. The Brisbane City Council and Gold Coast
City Council are currently investigating several options of managing
stormwater litter at the source utilising their Lip-in-Line Gully Pit (small
lintel).
A range of devices are currently available to collect litter using the gully
pit, however there is limited information regarding;
• The effect on the inlet hydraulic capture performance, and
• the effectiveness of litter capture.
Four litter control systems were evaluated for both on-grade and sag
conditions with varying approach flows up to 320 L/sec. Test results and
observations included:
• The horizontal bar placed across the inlet had little effect on
hydraulic capture for on-grade slopes. However for sag conditions
and at flows above 200 L/sec, the hydraulic capture capacity was
reduced. Depths at invert were up to 50mm greater for the case
without litter and up to 120mm greater with litter.
• Both the mesh and tray basket inserts were ineffective in collecting
and retaining litter. Both systems had no significant effect on the
hydraulic capture capacity of the gully pit.
• The Enviropod basket is capable of collecting and retaining
considerable amounts of pollution at flow rates up to 320 L/sec.
Resuspension of pollution occurred at flows above 100 L/sec,
however the unit still retained up to 70% of pollution at 1%
longitudinal slope at an approach flow of 320 L/sec. The Enviropod
basket affected the hydraulic capture capacity of the gully pit at 12%
grade due to blockage caused by litter trapped in the overflow outlet.
Urban Water Resources Centre 3 University of South Australia
Page 64
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
1. INTRODUCTION AND PREAMBLE
Brisbane City Council engaged the Urban Water Resources Centre to
undertake a hydraulic evaluation of various litter control systems for a
typical Lip-in-Line Gully Pit.
Hydraulic testing was undertaken using a short (2400mm) lintel inlet
with the roll top kerb and channel and flows ranging from 0L/sec to
340L/sec at longitudinal grades of 1%, 4% and 12%. Testing was
undertaken with the following gully pit litter control systems:
• A 36mm diameter horizontal bar placed across the mid point of
inlet opening (as shown in Appendix A),
• meshed pit insert,
• a tray basket (900x500x150), and
• standard Enviropod trash basket.
Testing was carried out using the full-size road test rig at the University
of South Australia. A standard litter sample was developed and used for
the evaluation. Litter comprised of 80% organics (leaves) and 20% paper,
plastic and sediment. A modified litter sample (Appendix B) was used
for the evaluation of the horizontal bar system. It contained a greater
proportion of larger litter, including plastic food and drink containers,
plastic bags and aluminium cans.
Briefly, the aim of the evaluation was to:
• Determine the litter capture performance characteristics of each
system, and
• determine the effect that each litter control system has on the inlet
hydraulic capture performance.
Urban Water Resources Centre 4 University of South Australia
Page 65
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
2. TEST SPECIFICATION AND PROCEDURE
Test specification:
• Approach flows ranging from 0 – 320 L/sec.
• Channel longitudinal slopes:
- 1%, 4% and 12% (3 off)
- “Sag” 0% slope for all configurations.
• Road pavement modelled to simulate asphaltic construction (hot-mix)
seal (Manning’s “n” = 0.014).
• Crossfall slope of 3.3%.
• Kerb type: Roll-top type kerb-and-channel currently set-up on Road
Rig.
• Lintel: Small (S) as per Gold Coast City Council (Drawing 59301).
• BCC perpendicular support bar grate
Procedure:
The test procedure involved:
(i) Add the pollution (10L) to an approach flow of 25 L/sec.
(ii) Observe and record the amount of pollution retained.
(iii) Increase approach flow and record capture flow.
(iv) Observe and record amount of pollution retained.
These procedures were repeated several times during each test to assess
the effect of increasing flows on litter and hydraulic capture performance.
Urban Water Resources Centre 5 University of South Australia
Page 66
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
Testing and evaluation for each of the litter control systems were slightly
different.
36mm diameter horizontal bar.
Testing of the horizontal bar was conducted with the modified litter
sample only. It was accepted that the inlet would collect most organic
litter and have no effect on the capture performance of the bar inlet
configuration.
Meshed Pit insert.
Testing was conducted to determine capture efficiency of the meshed
insert under fully blocked conditions. The insert was fabricated from
sheet metal hence simulating blocked conditions. Testing was conducted
using the standard litter sample.
Litter Crate Basket (900x500x150).
Assessment of collection performance, type of litter collected and
comments on re-suspension and effect on hydraulics (with basket fully
laden) for the standard litter basket was carried out.
Enviropod Basket.
The Enviropod insert required modifications to withstand the force of the
high approach flow. The plastic sides on the top of the basket were
replaced with sheet metal and strengthened.
NB: A new plastic was being tested during this trial. Subsequently a
plastic with increased strength and durability designed to withstand flows
greater than 320l/s is now used on all Enviropod units.
Urban Water Resources Centre 6 University of South Australia
Page 67
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
3. TEST RESULTS AND OBSERVATIONS
3.1 36mm horizontal bar.
Testing was undertaken to determine the effect on hydraulics and capture
performance of the lip in line gully inlet with a bar placed horizontally
across the inlet. The modified litter sample was used for this test and was
added to the gutter at an approach flow of 25L/s. The flow rate was
increased up to 320L/s and amounts of litter retained on the inlet were
observed at each increment. “Sag” results, calculated as depth at inlet, are
shown in Figure 3.
Test results and observations:
Litter Capture Performance.
The table below shows percentages of litter added that is retained on the
grate or by the bar for varying approach flow.
SLOPE APPROACH % LITTER
FLOW L/s RETAINED
25 100
1% 85 88
243 82
320 76
25 88
4% 60 82
120 76
180 18
320 18
25 94
12% 60 30
240 30
320 6
Urban Water Resources Centre 7 University of South Australia
Page 68
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
Discussion:
For the 1% longitudinal slope, 76% of the litter was retained at an
approach flow of 320L/sec (see Figure 1). The remaining 24% passed
through the inlet beneath the bar or bypassed the inlet. The high capture
efficiency is due to the lower approach flow velocities at flatter grades.
The majority of litter retained in front of the bar was plastic bottles and
milk cartons.
At the steep (12%) slope the inlet retained 30% of litter at an approach
flow of 60L/sec. The high velocity forced the majority of litter past the
inlet. At an approach flow of 320L/s only a plastic bag remained stuck
on the grate.
Figure 1: Pollution remaining after 320L/sec at 1% slope.
The horizontal bar caused a build up of litter at the downstream end of
the inlet. Water was observed ‘jetting’ over the downstream transition,
due to the litter caught, at higher approach flows and more significantly
at steeper slopes (see Figure 2).
Urban Water Resources Centre 8 University of South Australia
Page 69
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
Figure 2: Water ‘jetting’ due to build up of pollution.
Hydraulic performance.
On-grade tests:
The reinforcing bar did not effect the hydraulic capture performance of
the inlet for on-grade tests.
Sag tests:
Under sag conditions, the bar alone significantly reduced capture
capacity (see Figure 3). The flow pond depth at invert was up to 50mm
greater for the case with the horizontal bar. Testing was also conducted
to assess the effect on pond depths at invert with litter retained by the
horizontal bar. As expected, the litter reduced the capacity of the inlet to
capture flow, with pond depth increase up to 120mm greater for the same
approach flow. Litter retained by the horizontal bar further reduced the
hydraulic capacity of the gully pit.
Urban Water Resources Centre 9 University of South Australia
Page 70
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
250
bar with debris
bar no debris
no bar
200
Depth at Invert (mm)
150
100
50
0
0 50 100 150 200 250 300 350 400
Approach Flow (L/s)
Figure 3: “Sag” Reinforcing Bar “0” Grate.
3.2 “Meshed” pit insert.
The meshed pit inlet configuration (see Appendix A) was modified to
simulate fully blocked conditions. For this reason, a steel plate was used
instead of mesh.
Test results and observations:
Observations from the testing were:
Litter capture performance.
• At 1% and 4% slope, all litter was captured by the inlet (at 25L/sec),
however only 5% of the 10L of litter was retained by the insert for
both longitudinal slopes. This litter was resuspended and passed
through the inlet when flow was increased to 60L/sec.
• At 12% slope and approach flow of 25L/sec, 10% of the 10L of litter
was retained in the inlet box. This material was mostly semi-buoyant
litter (wet leaves) which either sank to the bottom or remained in
suspension. The grate captured a small proportion of litter. However
when the test was repeated at 25L/sec remobilisation of the captured
material in the inlet occurred and only 5% was retained. At 60L/sec
Urban Water Resources Centre 10 University of South Australia
Page 71
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
all litter had passed through the chamber except for an aluminium
can.
Hydraulic performance.
On-grade:
The mesh insert did not significantly effect the hydraulic capture
performance of the gully pit for all longitudinal slopes.
Sag:
At 0% (sag) longitudinal slope, the mesh insert caused the inlet to reach
‘orifice’ condition sooner (ie. at a smaller approach flow).
200
180
160
140
120
Depth at Invert (mm)
100
80
60
meshed pit
40
standard pit
20
0
0 50 100 150 200 250 300 350 400 450 500
Approach Flow (L/s)
Figure 4: Sag “0” Grate results – meshed pit insert
Discussion:
From the observations above, the ‘blocked’ meshed insert had no
significant effect on the hydraulic capture performance (on-grade) of the
inlet. All floatable litter passed through the chamber at 25L/sec and
semi-buoyant litter was remobilised at flows greater than 25L/sec.
For the sag condition, the insert affected the hydraulic capture between
150-340L/sec. Above 340L/sec it would appear that the capture (orifice)
was controlled by the inlet opening.
Urban Water Resources Centre 11 University of South Australia
Page 72
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
3.3 Tray Basket
The tray basket insert (Appendix A) was tested for capture and hydraulic
performance under similar conditions.
Test results and observations:
Litter capture performance.
On-grade
• 40% of the litter added to the approach flow (25L/sec) was retained
in the basket for all on-grade longitudinal slopes.
• Once the basket became blocked nearly all (90%) of pollution added
thereafter bypassed the basket.
• At the steeper (12%) grade the basket was more effective in capturing
the litter.
• As the approach flow increased, the litter capture performance of the
basket was marginally better at the 12% grade than for 1% and 4%.
However, when subjected to an approach flow of 330L/sec, only 2%
of the pollution was retained in the basket for all grades.
Sag
• The pit was bisected for the sag test to simulate flow from opposing
directions (see Figure 5). The basket retained 80% of the litter with
the approach flow set at 50L/sec. However it did not retain any of the
larger plastic litter.
Urban Water Resources Centre 12 University of South Australia
Page 73
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
X
Figure 5: Modified tray basket setup for Sag test.
Hydraulic capture performance.
On-grade:
The effect of the basket on the hydraulic performance of the pit was
insignificant.
Sag:
The basket caused the pond depth to increase, resulting in “orifice” flow
conditions at lower flows (see Figure 6).
Urban Water Resources Centre 13 University of South Australia
Page 74
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
200
180
160
140
120
Depth at Invert (mm)
100
80
60
40
Standard pit
20
Tray basket insert
0
0 50 100 150 200 250 300 350 400
Approach Flow (L/s)
Figure 6: Sag results - tray basket.
3.4 Enviropod Testing
Hydraulic testing, including high chamber water level (CWL) pre-testing,
and pollution capture performance testing was carried out for the
Enviropod basket. Two basket (collection bag) types were assessed; a
coarse and a fine fabric.
Test results and observations:
During initial tests, plastic sheeting on both the front and back sides of
the basket failed when subjected to high flows (>100L/sec). The sides
were subsequently replaced with sheet metal.
Litter Capture Performance
On-grade
During the initial tests, all of the litter (10L) was captured with approach
flow set at 25L/sec. It was decided to increase the amount of litter until
bypass occurred. 30L of litter was collected (at 25L/sec) before bypass.
Urban Water Resources Centre 14 University of South Australia
Page 75
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
1% grade:
• All litter was collected and retained up to an approach flow of
100L/sec for both collection baskets.
• Approximately 15% of the litter had bypassed the unit at 200L/sec
due to resuspension.
• The finer mesh basket retained 70% of litter after an approach flow of
320 L/sec and the coarse basket retained 65%.
• The mesh in both baskets became blocked after a flow of 320 L/sec
and held water for a short period after flow was stopped.
• As the flow dropped from 320L/sec to below 50L/sec some of the
pollutants attached to the fabric dislodged from the sides. The low
flow effectively washed the litter off the basket sides (see Figure 7).
Figure 7: Self cleansing action of basket at low flows.
Urban Water Resources Centre 15 University of South Australia
Page 76
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
4% longitudinal slope:
• At an approach flow of 200L/sec, the water level in the basket was
higher (100mm below invert) for the finer meshed basket than the
coarse mesh (200mm below invert). The higher water level in the pit
can be attributed to the reduced size in mesh of the basket.
• Both baskets retained 55% of debris after an approach flow of 320
L/sec.
12% longitudinal slope:
• The water level in the pit rose above the overflow level at 100L/sec
therefore resulting in small losses.
• After 320L/sec the coarse mesh basket retained 65% of debris while
the fine meshed basket retained 60%.
Sag
For the sag test, the basket was bisected in half to simulate flow
approaching from opposite directions. 20L of litter was used for the sag
litter tests because of the high capture capacity of the basket before
bypass. The following observations were made:
• With the coarse mesh the basket retained 50% of the 20L of litter
after being subjected to an approach flow of 340L/sec. The bag
became considerably coated with larger sized litter (mostly organics).
• With the fine mesh, only 10% of the 20L was retained after an
approach flow of 360L/sec (equivalent). The finer mesh experienced
less coating on the sides of the basket. It was observed that the flow
into the basket had the effect of ‘washing off’ some of the litter
attached to the sides and therefore allowing litter to bypass the basket
at higher flows.
Urban Water Resources Centre 16 University of South Australia
Page 77
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
Hydraulic Capture Performance
On-grade
When compared to the standard lip in line gully pit, the capture capacity
of the Enviropod basket was:
• The same for 1% and 4% longitudinal tests
• At 12% grade:
• 12L/sec less between 200 and 320 L/sec approach flow for the
fine mesh basket, and
• the same at 200L/sec for the coarse mesh basket, and 6L/sec
less at an approach flow of 320L/sec.
It was observed at the 12% grade that a plastic drinking bottle and plastic
bag were trapped in the overflow outlet, reducing the capture capacity of
the pit insert (see Figure 8).
Plastic bottle caught in outlet
Figure 8: Litter trapped in overflow outlet.
Urban Water Resources Centre 17 University of South Australia
Page 78
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
• At 1% and 4% grades, the fine mesh basket (with litter) captured up
to 5L/sec more than the coarse mesh basket at an approach flow of
320L/sec.
• At 12% longitudinal slope, the coarse mesh basket captured up to
11L/sec more than the fine mesh basket at an approach flow of
200L/sec and 8L/sec more with an approach flow of 320L/sec.
Sag
Four sag tests to determine hydraulic performance of the inlet were
conducted; two with fine mesh (with and without litter) and two with the
coarse mesh (with and without litter). This enabled a comparison of
performance between the two baskets under empty and blocked
conditions. The results were also compared to sag capture results with no
basket. Pond depth at invert versus approach flow characteristics are
provided in Figure 9.
Fine mesh sag test
At low to medium approach flows (up to 200L/sec) the addition of litter
reduced the capture capacity of the inlet, increasing the pond depth at
invert by up to 15mm. However at flows above 200L/sec there was no
difference in flow pond depth at invert for the case of with and without
litter.
Coarse mesh sag test
The effect of the coarse mesh was to increase the flow pond depth at
invert by up to 15mm at flows below 200L/sec, as for the fine mesh.
However the coarse mesh (with litter) showed similar flow pond depths
to the fine mesh (without litter) up to 200L/sec.
The results from the sag test are presented in figure 9 below.
Urban Water Resources Centre 18 University of South Australia
Page 79
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
200
160
120
Depth at Invert (mm)
80
fine mesh no pollution
fine mesh with pollution
40 coarse mesh with pollution
coarse mesh no pollution
standard inlet
0
0 50 100 150 200 250 300 350 400
Approach Flow (L/s)
Figure 9: Sag test results (depth at invert)
Discussion of results
Both baskets were effective in capturing and retaining a large proportion
of litter at flows up to 100L/sec. At 1% longitudinal slope the fine mesh
basket was slightly better in capture performance, whilst at the 12% slope
the coarse mesh basket was slightly better.
The fine mesh basket resulted in lower depths at invert at the same
approach flow than for the coarse mesh basket. This is due to the
resuspension of litter in the basket at high approach flows. Only 10% of
the 20L of litter added to the fine mesh basket was retained, where as
50% was retained by the coarse mesh basket.
3.5 Chamber Water Level Pre-testing
Testing was carried out to determine the effects on inlet capture as a
result of high chamber water level. Tests were conducted for the 1, 4 and
12% longitudinal grades. It was observed that with the short lintel
(2.4m) a high CWL of 150mm below the gutter invert did not affect the
capture performance.
Urban Water Resources Centre 19 University of South Australia
Page 80
Evaluation of Gully Pit Inlet Litter Control Systems– Brisbane City Council and Gold Coast City Council
4. CONCLUSION
The main findings from the testing programme are:
• The reinforcing bar placed across the inlet had little effect on
hydraulic capture for on-grade tests. However for sag conditions and
at flows above 200 L/sec, the hydraulic capture capacity was reduced.
Depths at invert were up to 50mm greater for the case with no
pollution and up to 120mm greater with pollution. A build up of
pollution on the grate, restricted by the bar, caused ‘jetting’ of water
above the kerb.
• The crate basket was ineffective in retaining pollution above a flow
rate of 25 L/sec for on-grade testing. Nearly all (90%) of pollution
added to the gutter after the basket had become blocked bypassed the
basket. After an approach flow of 320 L/sec only 2% of pollution
was retained for all longitudinal slopes. The basket was more
effective in retaining pollution under sag conditions.
• The mesh pit insert was inefficient in retaining pollution above a flow
rate of 25 L/sec. Only 5% of the pollution was retained at an
approach flow of 25 L/sec for the 1% and 4% longitudinal slope.
10% was retained for the 12% grade however all pollution had
bypassed the pit at an approach flow greater than 60 L/sec for all
slopes.
• The Enviropod basket was capable collecting and retaining large
amounts of pollution and had no significant effect on the hydraulic
performance of the gully pit. The basket retained up to 70% of
pollution added after an approach flow of 320 L/sec. At 12% grade,
litter was trapped in the overflow outlet, reducing the capture
capacity of the gully pit. This could also occur for flatter grades and
high approach flows. Initial testing of the basket led to some
modifications to increase support under high approach flows used in
the testing programme.
Urban Water Resources Centre 20 University of South Australia
Page 81
HUNTER ENVIRONMENTAL SERVICES
Stormwater and Catchment Management
Geoffrey J. Hunter (MIEANZ); Churchill Fellow 1996 — Soil & Water Management
Jillian M. Hunter; Post Graduate Diploma Adult & Community Education
52 Centre Crescent Blaxland NSW Australia 2774
Phone (mobile): 0437 502 952 E-mail: [email protected]
A.B.N. : 90 264 168 968
Our Ref.: 1410 15 September 2014
Technical Director
Stormwater 360 Australia
PO Box 792
BOTANY NSW 1455
Attention: Mr Michael Wicks
RE: Peer Review of Product Evaluation Report
Stormwater Management StormFilter® Stormwater Treatment System
Mitchell Community College Mooresville, NC
Thank you for commissioning Hunter Environmental Services to peer review the field study of
the StormFilter® technology undertaken by Contech Engineered Solutions within the parking
lot of the Mitchell Community College in Mooresville North Carolina USA.
The primary goal of the field study and subsequent report “was to collect performance data in
order to support the approval of the StormFilter system and its potential inclusion into the
North Carolina Stormwater BMP Manual. A major condition of the approval for continued
participation in the PEP [Preliminary Evaluation Period] program was the execution of an
additional field evaluation in accordance with the conditions outlined in the PEP program.”
Consequently monitoring and evaluation of the performance of the StormFilter® system,
installed in the Mitchell Community College carpark, was carried out to demonstrate its
effectiveness at removing Total Suspended Solids (TSS) and nutrients (Phosphorus and
Nitrogen) from runoff, as required by the PEP program.
Our review has concentrated on answering the key issues raised in your commissioning letter
of 5 August 2014, namely:
“Review of treatment system, sampling setup and equipment and sampling methodology.
Confirmation of independent sampling chain of custody external to Contech Engineered
Solutions.
Review of Report including calculations, discussions and conclusions. (refer to Discussion
and Conclusion sections within this Peer Review).
Comment on the comparisons of EMC’s from the study compared with local data ie
Duncan & Fletcher + others (if applicable) to help provide transferability of results.
Comment on the comparison of nutrient species or the composition of soluble species
compared with local data.
HES – StormFilter® System Peer Review Page 1
Page 82
REVIEW
1. Review of StormFilter®Treatment System, Sampling and Methodology
The StormFilter® system described in the report (Contech 2012) is made up of an eight (8)
cartridge StormFilter® preceded by a detention tank designed to capture 75% of the runoff
volume from the 1.0 inch (25.4 mm) event. The catchment to the system consists of a 1.08
acre (0.437 ha) carpark with an equivalent impervious area of 0.73 acres (0.297 ha)
representative of a catchment that is 68% impervious.
The layout provides for detention of the runoff with a controlled discharge to the StormFilter®
treatment system at a flow rate determined by the hydraulic capacity of eight (8) StormFilter®
cartridges. Each cartridge was filled with an activated alumina coated perlite media
(PhosphoSorb).
The layout of this treatment system allows the catchment hydrology to be modified based on
the hydraulics of the tank cartridges and media rather than the runoff characteristics of the
catchment. This system provides a generic layout suitable for applications where there are
major constraints on the available site area for treatment options. It could be expected that
the alumina coating on the perlite media would enhance the adsorption of pollutants onto the
exterior of the perlite media, while the perlite itself has a demonstrated ability to absorb
pollutants (Siber A et al, 2011)
Influent and effluent samples were collected using twelve (12), 1 litre sample bottles.
Collection of flow proportional samples commenced once the flow rate exceeded 5 g(US)pm
(0.32 L/s). The sample strainers intakes and flow measurement equipment were secured to
the invert of the influent and effluent pipes, which biases the samples towards the larger
fraction of the suspended solids. Collection and testing of the samples was carried out by
Pace and TestAmerica independent of Contech and oversight of the project was provided by
Ryan Winston, MS, extension Associate in the Department of Biological and Agricultural
Engineering at North Carolina State University.
Rinsate blank and replicate samples were collected, which is consistent with the sampling
guidelines detailed in NWQMS (2000) and demonstrates scientific rigour within the sampling
methodology as well as a good understanding of the vagaries of sampling rainfall/runoff
events.
2. Sampling “chain of custody”
Personnel from Pace and TestAmerica collected the samples and delivered them to the
laboratory for testing. No record of the use of “chain of custody” procedures for the samples
is referred to in the Porject Evaluation, submitted for this peer review. However,
correspondence with Contech Engineered Solutions has identified that “chain of custody”
procedures, similar to those outlined in Sections 4 and 5 (NWQMS, 2000) and Section 6
(EPASA, 2007), were employed as part of the sampling Project Plan. Further, Pace and
TestAmerica are accredited laboratories in accordance with the National Environmental
Laboratory Accreditation Conference (NELAC) and all samples were tested in accordance
with their nationally approved testing protocols (John Pedrick pers com).
HES – StormFilter® System Peer Review Page 2
Page 83
3. EMC Comparison with Local Data
Very little published data exists on pollutants in runoff from carparks worldwide, and in
particular in Australia. The very nature of the carpark and its use can vary both the amount
and type of pollutants present in stormwater and the high impervious area and small
catchment area generally favours a “first flush” effect, especially in regions with high rainfall
intensities and defined wet and dry seasons. Further, maintenance/cleaning procedures e.g.
mechanical sweeping, can artificially alter the pollutant species and their concentrations, in
runoff from carparks.
Morison P (2001) reported on a study that categorised pollutant loads from a commercial
carpark in western Sydney using a rainfall simulator calibrated for a 1 in six month storm of 15
minutes duration. The results of the study demonstrate a marked “first flush” effect for 10
minutes with an approximate Event Mean Concentrations (EMCs) for a duration of 15
minutes of:
Suspended Solids 95 mg/L;
Total Phosphorus 0.15 mg/L;
Total Nitrogen 1.85 mg/L.
The EMCs, determined by this study, approximate Typical Wet Weather Concentrations for
Roads provided in Tables 2.43, 2.44 and 2.45, by Fletcher T et al (2004), which are generally
accepted as typical EMCs for various land uses in Australia:
Suspended Solids range 90 to 800 with a typical value of 270 mg/L;
Total Phosphorus range 0.15 to 1.5 with a typical value of 0.5 mg/L;
Total Nitrogen range 1 to 5 with a typical value of 2.2 mg/L.
The EMCs reported in the StormFilter® Product Evaluation carried out over twenty (20)
months within the Mitchell Community College carpark are:
Suspended Solids range 10.3 to 98.2 with a mean value of 34.6 mg/L;
Total Phosphorus range 0.07 to 0.90 with a mean value of 0.22 mg/L;
Total Nitrogen range 0.35 to 2.95 with a mean value of 1.00 mg/L.
Comparison of Event Mean Concentration (mg/L) for Various Locations
Variable St Martins Carpark CRCCH TR04/8 Mitchell Community
NSW (Australia) College NC
Suspended Solids 95 (approx) 90 to 800 10.3 to 98.2
typical 270 mean 34.6
Total; Phosphorus 0.15 (approx) 0.15 to 1.5 0.07 to 0.90
typical 0.5 mean 0.22
Total Nitrogen 1.85 (approx) 1 to 5 0.35 to 2.95
typical 2.2 mean 1.00
The above comparison identifies significantly lower EMCs measured in runoff from the
Mitchell Community College, Mooresville NC than are generally accepted as typical for car
parks in NSW and Australia. With respect to the determination of an Efficiency Ratio (ER) for
a stormwater treatment system, the higher the EMC in the influent the easier it is to achieve a
particular removal efficiency. Consequently the ER determined for the Mitchell Community
College system could be expected to be much higher if the influent EMCs approached those
reported in the Australian literature.
HES – StormFilter® System Peer Review Page 3
Page 84
4. Nutrient Species Comparison with Local Data
There is limited information available relating to the categorisation of nutrients, into their
constituent species, within runoff from carparks and roads in Australia. However, a literature
search did reveal a range of EMCs for Ammonia which could be used for comparative
purpose. These are referenced below.
Morison P (2001) reported on the Ammonia EMC in runoff from the St Martins carpark which
was in the order of 1 mg/L.
The recommended typical EMC values for nutrients in runoff from various land uses, provided
in Fletcher T et al (2004), do not include a species breakdown for either Total Phosphorus or
Total Nitrogen. Other data, referencing Ammonia EMCs in runoff from urban catchments in
Sydney and the Illawarra, suggest a mean range of between 0.02 and 0.54 mg/L while similar
EMCs measured in Hornsby returned mean Ammonia EMC ranges of between 0.01 and 0.11
mg/L. Similarly, Ammonia EMCs in runoff from urban catchments in Grafton were found to be
between 0.02 and 0.41 mg/L (Fletcher T et al, 2004).
The Contech Engineered Solutions (2012) product evaluation identified an EMC range of
between 0.05 mg/L and 0.72 mg/L with a mean of 0.27 mg/L for the Ammonia in runoff from
the Mitchell Community College carpark.
Consequently it would appear that the Ammonia EMCs measured in runoff from the Mitchell
Community College carpark are similar to those reported in runoff from the St Martins carpark
and other urban catchments in the Sydney, Illawarra and Grafton regions of NSW as reported
in Fletcher T et al (2004).
DISCUSSION
Based on the information provided in Contech Engineered Solutions (2012) product
evaluation of the StormFilter® installed, within the Mitchell Community College carpark, it is
our opinion that the sampling and monitoring of the system has been undertaken with
sufficient scientific rigour to meet the objectives of the project. Further, the sampling
methodology and testing procedures used by Contech Engineered Solutions are consistent
with the Australian and New Zealand Guidelines, NQWMS (2000).
With regards to compatibility with Australian conditions:
Pollutant concentrations in the influent appear lower than those that may be expected
under Australian conditions. Consequently the Efficiency Ratios determined may be
expected to be higher if the same StormFilter® treatment system were installed on a
similar carpark in Australia;
The use of a storage tank, upstream of the treatment system, allows the required
treatment volume of runoff to be released to the treatment system at a flow rate
comparable to the Treatable Flow Rate of the cartridge filter system. Consequently
comparisons between rainfall intensities is not thought necessary as the storage
volume within the detention tank and the size of the outlet orifice control the flow rate
HES – StormFilter® System Peer Review Page 4
Page 85
into the StormFilter® rather than the relationship between catchment rainfall and
runoff;
The small catchment area and high impervious nature of the carpark could be
expected to result in a “first flush” effect with a high initial EMC in the influent
concentrations as demonstrated by Morison P (2001). Consequently treating the first
1 inch (25.4 mm) of runoff from the catchment should result in the majority of pollutants
being washed from the surface of the carpark into the system. Morison P (2001) was
only treating the first 10 mm of runoff and was observing a marked reduction in
pollutant loads after 10 minutes;
It would appear from the information provided in the Contech Product Evaluation
Report that by-pass of the system did not occur during the course of the study (John
Pedrick pers com) and as a result was not able to be included in the ER determination.
In general Australian requirements, for issuing approvals for specific stormwater
treatment systems, are based on an ER determined through the use of Average
Annual Rainfall and the total volume of runoff from the catchment. Consequently any
assessment of the pollutant removal by a system should include all of the runoff from
the catchment i.e. the influent volume and EMC as well as any by-pass volume and its
corresponding EMC;
For modelling purposes, in Australia, Suspended Solid (SS) loads have generally been
defined as that fraction of the Particle Size Distribution (PSD) that is ≤ 125 µm (Brodie
& Rosewell, 2008, Hunter G, 2008). The Contech Product Evaluation Report does not
include a PSD breakdown of the SS load and only refers to the < 2000 µm fraction and
the < 500 µm fraction. Pace and TestAmerica referred to the texture of the suspended
solids sampled as Loamy Sand (John Pedrick pers com), refer to Figures 1 & 2.
However, additional sieve analyses of the sediment, captured within the system,
undertaken by GeoTesting Express on 5 September 2014, has determined that in
excess of 80% of the sediment was < 75 µm in diameter (refer to Figure 3);
Both TSS and SSC testing (Kayhanian M, et al 2005) was undertaken to determine
ERs for the < 2000 µm and < 500 µm fractions of Suspended Solids. The results for
both testing procedures returned ERs in excess of 90%. The Suspended Solids EMC
recorded in the influent was significantly less than that recorded in the Australian
literature and consequently achieving an ER in excess of 90% demonstrates that the
system can achieve very high pollutant reductions from runoff containing lower than
expected levels of Suspended Solids contamination;
The detention tank and StormFilter® system is designed to drain completely within a
short period of time after completion of a rainfall/runoff event. Consequently the
captured solids are retained in a dry/semi-dry condition intended to maintain them in a
solid state, thereby minimising the opportunity for anaerobic decomposition and their
re-suspension in the effluent in a dissolved form (refer to Figure 4);
The intake to the automatic samplers, being located in the invert of the influent and
effluent pipes could bias the samples towards the larger fraction of suspended solids.
However, the sieve analysis, undertaken by GeoTesting Express on 5 September
2014 (refer Figure 3) suggests that the majority of the sediment retained within the
system was < 75 µm and was removed prior to the location of the effluent sampling
intake. Consequently the relationship between the influent and effluent EMCs for SS
indicate a significant reduction in the finer fraction of the Suspended Solids.
HES – StormFilter® System Peer Review Page 5
Page 86
CONCLUSION
Generally, ER criteria for stormwater treatment in the Sydney Region is based on removing a
minimum 85% of the Suspended Solids, 65% of the Total Phosphorus and 45% of the Total
Nitrogen average annual loads. The Efficiency Ratios determined by Contech Engineered
Solutions for the StormFilter® installed within the Mitchell Community College carpark are
related to thirteen (13) storm events over a twenty month (20) period, which resulted in the
removal of 90.4% of Total Suspended Solids, 86.1% of Total Phosphorus and 55.9% of Total
Nitrogen.
The procedures employed in collecting and testing the samples appear to be consistent with
those outlined in NWQMS (2000) and as such should provide a high degree of confidence
that the EMCs determined are correct for each of the samples collected. The determination
of the ERs is dependent on the difference between the influent and effluent EMCs.
Consequently, it is our opinion, that the ERs determined in the report are those that were
actually achieved by the system under field conditions, based on the sampling and testing
protocols employed, and similar ERs could be expected if the same system were installed
within a similar catchment in Australia.
Although a Particle Size Distribution was not carried out as part of the initial study, the texture
of the Suspended Solids was described as Loamy Sand. A Loamy Sand typically consists of
approximately 80% Sand, 15% Silt and 5% Clay (refer to Figure 1). AS 1726-1993 classifies
Sand as between 2.36 mm (2360 µm) and 0.075 mm (75 µm) in dia with Silts and Clays <
0.075 mm (75 µm) (refer to Figure 2). However, colloidal Clays are generally < 0.002 mm (2
µm) in dia. Consequently a Loamy Sand could contain a significant proportion of < 125 µm
dia soil particles. This has been confirmed by the recent GeoTesting Express ASTM D1140
sieve analysis, of the sediment retained within the system, which determined that in excess of
80% of the sediment was < 75 µm (refer to Figure 3), and provides a separate body of
evidence for the determination of a Suspended Solid Concentration (SSC < 500 µm)
Efficiency Ratio of 93.4% for the system (Contech Engineered Solutions, 2012).
Regardless, of the fact that a PSD was not undertaken initially, and that an ER for the
removal of the < 125 µm Suspended Solids component of the loads was not determined, the
low concentrations of Suspended Solids in the influent (10.3 mg/L to 98.2 mg/L) suggests that
an ER of at least 90% is possible when even higher concentrations of Suspended Solid, such
as those reported within the Australian literature, are encountered. Further, retention of the
solids in a dry to semi-dry condition limits the opportunity for anaerobic decomposition and re-
suspension of nutrients to occur.
With Compliments,
Hunter Environmental Services
HES – StormFilter® System Peer Review Page 6
Page 87
REFERENCES Brodie & Rosewell (2008) “Using soil loss models to estimate suspended solids concentrations in stormwater runoff from pre-urban areas”. Australian Journal of Water Resources V.12 No.1, Institution of Engineers Australia, 2008. Contech Engineered Solutions (2012) “North Carolina Department of Environment and Natural Resources Division of Water Quality Preliminary Evaluation Period Program Field Evaluation: Stormwater Management StormFilter® Stormwater Treatment System Mitchell Community College Mooresville, NC.” Contech Engineering Solutions (22 Aug 2012) EPASA (2007) “EPA Guidelines: Regulatory monitoring and testing Water and wastewater sampling.” Environment Protection Authority of South Australia (June 2007). Fletcher T, Duncan H, Poelsma P & Lloyd S (2004). “Stormwater Flow and Quality and the Effectiveness of Non-Proprietary Stormwater Treatment Measure – A Review and Gap Analysis Technical Report 04/8” Cooperative Research Centre for Catchment Hydrology (December 2004). Hunt and Giles (1992) “Farm Monitoring Handbook”. University of Western Australia Publishing, Nedlands WA. Hunter G (2008) “Predicting the Waterway Impacts of Urbanisation: modelling considerations pre, during & post urban development.” In proceedings of Urbanisation and Waterway Health: Policies and Processes for Management of Impacts, Sunshine Coast Regional Council, Qld, 17 June 2008. Kayhanian M, Young T, Stenstrom M (2005) “Limitations of Current Solids Measurements in Stormwater Runoff: a discussion of sampling, classifying and quantifying suspended solids.” In “Stormwater: The Journal for Surface Water Quality Professionals (July/August 2005)”. Morison P (2001) “Rainfall Simulation Experiment at St Martins Shopping Village” (accessed 12 August 2014) https://www.google.com.au/search?sourceid=navclient&aq=&oq=constituents+in+car+park+r unoff&ie=UTF- 8&rlz=1T4TSAU_enAU383AU383&q=peter+morison+mt+druitt+car+park&gs_l=hp..3.41l18.0 .0.4.186958...........0.CxJawMDyNU8 NWQMS (2000) “Australian Guidelines for Water Quality Monitoring and Reporting- Summary (2000)” Australia and New Zealand Environment and Conservation Council. SCCWRP (2001) “Characteristics of Parking Lot Runoff Produced by Simulated Rainfall.” Southern California Coastal Water Research Project (accessed 12 August 2014) ftp://ftp.sccwrp.org/pub/download/DOCUMENTS/TechnicalReports/343_characteristics_of_p arkinglot_runoff.pdf Siber A, Bar-Yosef B, Suryano S & Levkovitch I (2011) “Zinc adsorption by perlite: Effects of pH, ionic strength, temperature and pre-use as growth substrate.” Geoderma 170, Elsevier Press, pp. 159-167. (accessed 12 August 2014) www.researchgate.net/...adsorption_by_perlite.../00463519c642fc50b90 Urban Stormwater Preliminary Data Summary (2006). “Section 4.0 Environmental Assessment”. (accessed 12 August 2014) http://water.epa.gov/scitech/wastetech/guide/stormwater/upload/2006_10_31_guide_stormwa ter_usw_b.pdf HES – StormFilter® System Peer Review Page 7
Page 88
Figure 1 Soil Texture Triangle (Adapted from Hunt and Giles, 1992)
Figure 2 AS 1726-1993 Soil Classes
HES – StormFilter® System Peer Review Page 8
Page 89
Figure 3 75 µm Sieve Analysis of Sediment Captured by the System.
Figure 4 Sediment and Organic Matter retained in dry within the StormFilter®
(Source: John Pedrick (CPSWQ) Project Manager Contech Engineered Solutions LLC)
HES – StormFilter® System Peer Review Page 9
Page 90
The Stormwater Management StormFilter® Performance for
Removal of Total Phosphorus
Phosphorus in the Urban Environment
Phosphorus loading to freshwater can promote algal blooms and eutrophication that
threaten ecosystems by lowering dissolved oxygen levels. As shown in Figure 1, phosphorus
cycles through the environment in forms organic, inorganic and soluble forms.
Figure 1. Inorganic and Organic Cycle (RiverWatch, 2001).
Total phosphorus (TP), expressed in milligrams/liter is the sum of inorganic phosphate,
organic phosphate, and soluble phosphorus (Ortho-P). Organic phosphates are a part of plants
and animals, their wastes or decomposing rem ains. Inorganic phosphorus originates from
decomposing natural materials and man-made products.
Non-point source runoff (stormwater) increases phosphorus concentrations in lakes and
streams by transporting sediment and organic matter (bud shatter, leaves, lawn clippings, etc.)
from impervious surfaces. Additional phosphorus sources in stormwater are misapplied
fertilizers, some detergents, and animal waste from birds and domestic pets.
Phosphorus in urban runoff is typically measured as TP and sometimes Ortho-P is
measured as well. The non-soluble portion of the TP is commonly associated with the total
suspended solids (TSS). Of this form, the phosphorus can be in an organic or inorganic form.
TP concentrations in stormwater are variable but range from 0.01 to 7.3 mg/L (Minton, 2002).
Concentrations of Ortho-P in urban runoff are frequently in concentrations ranging from 0.05 to
0.2 mg/L (Wigginton, 1999). USEPA guidelines indicate that Ortho-P concentrations in stream
in excess of 0.10 mg/L can trigger algae blooms in fresh water lakes.
Removal of phosphorus can be accomplished by three mechanisms. The first is
removal of organic and inorganic P associated with solids. The second is removal by biological
uptake by plants or bacteria. The third is through chemical precipitation such as the reaction of
10/29/03
Page 91
Ortho-P with iron to form iron phosphate in aerobic conditions. Depending on the type of
treatment system, organic phosphorus can transform to Ortho-P and be released later. For
example, leaves trapped in a sump can decompose or fall senescence of wetland plant can
release Ortho-P.
Results
Performance data for removal of total phosphorus were summarized from ongoing field
evaluations. These field evaluations are a combination of first and third party investigations.
Data were collected from 9 sites located in different geographic locations (primarily from the
West Coast (WA, OR, CA) and a single Midwest site) and configured with different media types
at different flow rates. Available reports are listed in the reference section.
This performance summary focuses on Total Phosphorus removal only. The following
information presented in Figure 3 contains data collected since 2001, mostly during the late
spring, summer, and fall for total phosphorus removal by the Stormwater Management
StormFilter®. Fifty-five data points are presented in Figure 3. The mean removal efficiency
using linear regression was 62% with 95% confidence limits of 53% and 78% (lower and upper
limits, respectively). Sixteen data points that were included in the analysis did not have a
positive removal. Overall these systems demonstrated statistically significant removal (P<0.001;
99.9% probability of net removal) of Total Phosphorus.
Figure 3. Total phosphorus removal performance summary collected from 9 sites, in multiple geographic
locations, with different media. The linear regression produced an equation of y = 0.38x + 0.065, which
translates to a 62% removal with a 95% confidence interval of 53% and 78% (lower and upper limits,
respectively). Data was statistically significant with a P < 0.001. Data was current as of July 2003.
2
Page 92
Table 1. General Site Description
Site Description WQ Flow Unit Size Media No. of Location
Rate (cfs) Cartridges
Shopping Center 0.503 8 x 16 ZPG 23 Vancouver, WA
Carwash 0.070 CBSF CSF 2 Vancouver, WA
Hotel 0.165 6x8 CSF 5 Vancouver, WA
Mixed Use 1.600 8 x 16 (2) Perlite/Zeolite 48 Sammamish, WA
Shopping Center 0.033 CBSF Perlite 1 Vancouver, WA
Commercial Office 0.594 8 x 16 (2) Perlite/CSF 24,30 Olympia, WA
School 0.297 8 x 16 Perlite/Zeolite 14 Redmond, WA
Resort 1.650 CIP Perlite/Zeolite 50 California
Roadway 0.300 6 x 12 ZPG 9 Midwest
References
Delaware State Department of Natural Resources and Environmental Control. (no date). Urban
Stormwater Fact Sheet prepared for Inland Bays Watershed. Dover, DE. Retrieved 11/11/03 from
www.dnrec.state.de.us/water2000/Sections/Watershed/ws/fact_ib_stormwater.pdf
Minton, Gary. (2002). Stormwater Treatment: Biological, Chemical, & Engineering Principles. Resource
Planning Associates. Seattle, WA.
RiverWatch. (2001). Beyond Books Institute of Alberta. (2001). Retrieved on 11/11/03 at
www.riverwatch.ab.ca/how_to_monitor/p_info-types.cfm
Symons, James, Lee Bradley, Jr., Theodore Cleveland. (2000). The Drinking Water Dictionary. American
Water Works Association. McGraw & Hill. New York, NY.
Stormwater Management Inc. (SMI). 2002. Heritage Marketplace Field Evaluation: Stormwater
Management StormFilter with CSF Leaf Media. Author. Portland, OR.
Stormwater Management Inc. (SMI). 2003. University Place Field Evaluation: Stormwater Management
StormFilter with Perlite Media. Author. Portland, OR.
Stormwater Management Inc. (SMI). 2003. Overlake School Field Evaluation: Stormwater Management
StormFilter with Perlite/Zeolite Media. Author. Portland, OR.
Stormwater Management Inc. (SMI). 2003. Salmon Creek Plaza Field Evaluation: Stormwater
Management CatchBasin StormFilter™ with Coarse Perlite Media. Author. Portland, OR.
Stormwater Management Inc. (SMI). 2003. University Inn at Salmon Creek Field Evaluation: Stormwater
Management StormFilter with CSF Leaf Media. Author. Portland, OR.
Stormwater Management Inc. (SMI). 2003. Larry’s Carwash: Stormwater Management StormFilter with
CSF Leaf Media. Author. Portland, OR.
Stormwater Management Inc. (SMI). 2003. Saffron Village Field Evaluation: Stormwater Management
StormFilter with Perlite/Zeolite Media. Author. Portland, OR.
Wigginton, Byran O. James Lenhart. (2000). Using Iron-Infused Media and StormFilter technology for the
rd
removal of dissolved phosphorus from stormwater discharges. Water Environment Federation – 73
Annual Conference and Exposition. Anaheim, CA.
Data from Two third party investigations were used in this analysis, reports shall be available in fall of
2004 and spring of 2005 (Midwest & California site).
3
Page 93
February 29, 2012 John H. Pedrick Contech Construction Products, Inc. 11835 NE Glenn Widing Drive Portland, OR 97220 John, This letter provides a review of my involvement in the evaluation of the Stormfilter Stormwater Treatment System at Mitchell Community College in Mooresville, North Carolina. I was approached by Contech in October of 2010 to serve as a Technical Advisor (TA) for the project. As a TA, I provided support and guidance for the stormwater monitoring and evaluation of the Stormfilter from start to finish. This included review of monitoring methods and procedures as well as a site visit to observe and verify field monitoring setups. I am well qualified to serve as technical advisor for this project. I earned my M.S. degree in 2009 at North Carolina State University by completing a stormwater monitoring project. I have been intensively involved in field monitoring of more than 30 stormwater BMPs in North Carolina, including bioretention, permeable pavement, cisterns, filter strips, swales, and innovative wet ponds. Apart from field data collection, I have been involved in the production of 10 peer‐reviewed journal articles and numerous final reports. Therefore, I have a solid background in statistics and stormwater data analysis. From the start, I have reviewed and approved all reports produced on the data collected on the Mooresville Stormfilter. This included the Quality Assurance Project Plan (QAPP) in October 2010, the first quarterly report in May 2011, and the second quarterly report in August 2011. In short, the methods that have been used to monitor the Stormfilter are state‐of‐the‐science, and provide an excellent indication of how the product performed over the eight month evaluation period. The final report (February 2012) for the removal of solid pollutants provides NCDENR with the requisite monitoring data to show that the Stormfilter removed 85% of TSS during the 10 storm events that were sampled. A second report will be produced on the other analytes that were sampled, including nutrients and heavy metals. Should you have any questions, please do not hesitate to contact me. Thank you, Ryan J. Winston, M.S., E.I.
Page 94
1
Technical Papers
SOLID AND NUTRIENT POLLUTANT
REMOVAL BY AN ENGINEERED
STORMWATER FILTRATION MEDIA
Field evaluation of a radial cartridge media filter
M Wicks, J Lenhart, J Pedrick
Abstract (Wicks et al., 2011) using a granular suspended solids, together with
This paper provides a summary of perlite, zeolite and carbon filtration both soluble and particulate fractions
the results associated with a 20-month media blends. Similar to blended of both phosphorus and nitrogen
filtration media, activated alumina at mean concentrations that can be
field study conducted at the Mitchell
media employs both physical as well compared to the Australia context,
Community College test site located in
as chemical filtration characteristics to to validate the performance of the
the town of Mooresville, North Carolina,
promote adsorption of pollutants such activated alumina in the RFCS.
USA. The study was conducted to
as dissolved phosphorus (Ma, 2011). The
demonstrate the effectiveness of a radial The RCFS, as seen in Figure 1,
legislated removal requirements of both
cartridge filtration system (RCFS) using is typically comprised of a vault that
phosphorus and nitrogen have become
an activated alumina media, treating houses rechargeable, media-filled filter
commonplace in Australia, for example
stormwater runoff with respect to the cartridges. Stormwater entering the
the Queensland Planning Policy (DSDIP,
removal of solid and nutrient pollutants. system percolates horizontally through
2014) for the treatment of stormwater
these media-filled cartridges, where
Testing of the RCFS was conducted from development sites.
pollutant removal processes occur. Once
for a suite of pollutants, including Total
Although the stormwater treatment filtered through the media, the treated
Suspended Solids (TSS), Total Phosphorus
objectives for TSS, TP and TN have stormwater is directed to a collection pipe
(TP) and Total Nitrogen (TN) in accordance
been commonplace in Australia, reliable and/or discharged to the receiving water.
with an approved Project Plan.
and transferrable data from robust
The Mitchell Community College
Results from the study indicated field assessments in other regions
testing site is located in the town of
that the RCFS, operating at 0.5L/s can seldom be accomplished due to
Mooresville, North Carolina; it is owned
per cartridge, successfully treated variations in climatic conditions, particle
and operated by Mitchell Community
stormwater runoff with significant size distribution of solids and soluble
College and is used for parking. The site
reductions for solid and nutrient fractions of nutrients. For example,
is 68% impervious and the total drainage
pollutants that were observed between Wong and Walker (2009) found that
area for the site is 4,370m2. A view of the
influent and effluent sampling locations the particle size distributions range
finished parking lot located on site can
using the Efficiency Ratio (ER) calculation on Australian roadways were between
be seen in Figure 2. Stormwater runoff
(TSS 90%, TP 86%, and TN 56%) and approximately 2 and 500 microns.
from the contributing drainage area
Summation of Load (SOL) efficiency
This research provides information for is directed to the RCFS for treatment
calculation methods (TSS 91%, TP 87%,
a range of particle size distributions for before eventually discharging.
and TN 50%). These results demonstrate
that the radial cartridge filter system was OVERFLOW RISER AND HOOD
able to successfully meet the current INLET ENERGY DISSIPATER
load-based objectives from the NSW
DECC (2007) and QLD Single State
Planning Policy (2014) for all relevant
pollutants, including TSS, TP and TN.
Keywords: BMP, stormwater, TP, TN,
TSS, activated alumina, media filter
cartridge.
Introduction
This field trial complements previous
studies undertaken on the RCFS
utilising alternative filtration media
in both North America and Australia. OUTLET SUMP
RCFS have previously demonstrated INLET PIPE CONDUIT
significant removal of both phosphorus INLET SUMP
and nitrogen (about 44%) on a SOL basis Figure 1. Diagram of an RFCS system.
SEPTEMBER 2014 water
Page 95
2
Technical Papers
on a predictive determined to be identical with respect
probabilistic assessment to physical characteristics and therefore
of the sediment load the media should be considered
entrained within the equivalent with respect to expected
stormwater runoff solids removal performance.
from the site. Although
the cartridges were
Method
maintained on a yearly The Mitchell Community College
basis, a qualitative RCFS installation was evaluated over
assessment undertaken a 20-month period following system
at the time of the maintenance in November 2010.
cartridge exchange did Independent oversight of all aspects
not indicate that any of the project was provided by Ryan
bypassing of treatment Winston, MS, Extension Associate
occurred due to filter Engineer in the Department of Biological
occlusion. The yearly and Agricultural Engineering at North
maintenance frequency Carolina State University.
is unremarkable, given
Sample handling services (sample
the detention tank
retrieval, system reset and sample
upstream and the
RCFS’s passive surface submittal) were provided by Pace
cleaning mechanism, Analytical Services (Pace) and laboratory
which activates at work was conducted by Pace and Test
least once during America. Precipitation was measured with
Figure 2. Aerial view of the testing site. every runoff event a tipping bucket-type rain gauge. Influent
and deposits the and effluent water quality samples
The RCFS was designed as a capture- were collected by portable automated
waste on the floor of the cartridge bay.
and-treat system. The storage component samplers simultaneously collecting flow,
of the system (tank) is a 750mm diameter The RCFS system contained a total precipitation and water quality samples.
corrugated metal pipe (CMP) network of eight, 460mm high, media-filled Each automatic sampler was connected
designed to capture 75% of the calculated filter cartridges operating at a flow to the cellular network for remote
water quality volume (i.e. the runoff rate of 0.5L/s per cartridge. Each of communication and data access.
associated with a 25mm event). the filter cartridges was filled with an
activated alumina media. The media The influent sampler was equipped
The treatment component (StormFilter) used for this study was a granular perlite with an area velocity flow module with
was designed on a mass-loading basis coated with activated alumina; this was low profile sensor for flow analysis
required to meet the annual pollutant done to aid in the attenuation and/ and influent sample pacing. The effluent
loading requirements of the site with a or capture of nutrient pollutants by sampler was equipped with a bubbler
minimum estimated interval between cation exchange and adsorption. With flow module used in conjunction with
maintenance of one year. The estimation the exception of the surface coating, a weir for flow analysis and effluent
of the yearly maintenance was based coated and uncoated perlite media were sample pacing. Sample strainers and flow
water SEPTEMBER 2014
Page 96
3
Technical Papers
measurement equipment were secured efficiency as the average event mean Results and Discussion
to the invert of the influent and effluent concentration of pollutants over some Monitoring 13 storm events over
pipes using stainless steel spring rings. time period. a 20-month period resulted in the
Following a precipitation event, collection of cumulative rainfall
composite samples were submitted representing 704mm. Comparison of
for analysis according to accepted, The ER method assumes: 1) The weight the measured rainfall intensities is not
relevant EPA, ASTM (Suspended of all storm events is equal, regardless of wholly representative of that found in
Sediment Concentration) and SM254D the relative magnitude of the storm event; Australia due to varying climatic regions.
(TSS) methods. The field monitoring and 2) that if all storm events at the site The presence, however, of upstream
methods used for this study represent had been monitored, the average inlet storage attenuates flows from all storms
the current state-of-the-art practice and and outlet EMCs would be similar to those such that differences in rainfall intensities
are similar to those used by researchers that were monitored (URS/ EPA, 1999). ER are not a significant factor on treatment
in North Carolina to evaluate vegetated results for each parameter in the 13 events performance and pollutant wash-off. Full
stormwater treatment systems. sampled are summarised in Table 1. treatment flow through each RCFS is not
achieved until the water level inside the
To obtain a better understanding of Performance was also calculated using
chamber reaches a height of 460mm. At
RCFS performance with respect to solids, the Summation of Loads (SOL) efficiency
this water elevation a siphon is activated
the portion of SSC (Suspended Sediment calculation method. The SOL method
and the cartridge throughput of 0.5L/s
Concentration) particles smaller than defines the efficiency as a percentage
per cartridge is achieved, regardless of
500µm and 2000µm was also determined. based on the ratio of the summation of
the rainfall intensity.
all influent loads to the summation of all
For each of the 13 qualifying storm effluent loads. Non-parametric statistical methods
events sampled between November were used to evaluate correlations and
of 2010 and June of 2012: differences between non-transformed
1. The total rainfall was greater than
influent and effluent event mean
2.5mm for each event sampled; concentrations (EMCs), since influent
The SOL method assumes: 1)
and effluent EMCs were generally not
2. The minimum inter-event period was
Monitoring data accurately represents
from the same statistical distribution
greater than six hours for all storm the actual entire total loads in and out
due the complex nature and variability
events sampled; of the BMP for a period long enough
of stormwater monitoring. To test for
to overshadow any temporary storage
3. The minimum number of influent and
positive correlations between influent
or export of pollutants; and 2) Any
and effluent EMCs, the Spearman Rank
effluent aliquots collected per storm significant storm events that were not
Order Correlation test was used (USGS,
event was five; monitored had a ratio of inlet to effluent
1991). To evaluate the significance of
loads similar to the storms events that
4. Seven influent flow-weighted differences between influent and effluent
were monitored (URS/EPA, 1999). In an
composite samples covered ≥ 50% of EMCs, the Mann-Whitney Rank Sum Test
effort to eliminate the introduction of
the total storm flow, while six covered was used (USGS, 1991).
potential bias associated with observed
between 34% and 49%; effluent flow-
discrepancies between influent and For the Mann-Whitney Rank Sum
weighted composite samples covered
effluent measured volumes, it was Test the null hypothesis was that the
≥ 50% of the total storm flow for all
assumed that the influent volume was two samples were not drawn from
storm events sampled.
equal to the effluent volume. Measured populations with different medians. A
Performance was calculated using the effluent volume was used to calculate significant difference between influent
Efficiency Ratio (ER) efficiency calculation loads for both the influent and effluent and effluent EMCs was concluded
method. The ER method defines the sample locations. when P<0.05. Based on the use of the
Table 1. Summary of results.
Range of Median Mean Range of Median Mean Mean Removal Efficiency
no. of
Analyte Influent EMCs Influent Influent Effluent EMCs Effluent Effluent Efficiency Ratio
events
(mg/L) EMC (mg/L) EMC (mg/L) (mg/L) EMC (mg/L) EMC (mg/L) (Sum of Loads) (ER)
SSC < 2000
12 17.7 - 2080.0 53.4 231 1.9 - 7.2 3.4 3.9 98.3% 98.3%
micron
SSC < 500
12 9.0 - 393.0 28.6 66.1 1.7 - 10.0 2.8 4.4 93.7% 93.4%
micron
TP 11 0.065 - 0.9 0.14 0.223 0.025 - 0.058 0.025 0.031 87.1% 86.1%
PP 9 0.019 - 0.225 0.06 0.083 0.0 - 0.033 0.0 0.007 96.4% 91.3%
DP 9 0.025 - 0.850 0.054 0.155 0.025 - 0.160 0.025 0.04 67.3% 74.2%
TN 10 0.035 - 2.95 0.85 1.00 0.35 - 0.82 0.35 0.44 50.2% 55.9%
TKN 11 0.25 - 2.70 0.72 0.94 0.25 - 0.58 0.25 0.28 60.9% 70.2%
NH3-N 11 0.05 - 0.72 0.21 0.27 0.05 - 0.24 0.05 0.10 60.0% 62.8%
NOx 10 0.10 - 0.35 0.16 0.17 0.10 - 0.35 0.10 0.16 11.2% 9.8%
SEPTEMBER 2014 water
Page 97
4
Technical Papers
Spearman Rank Order correlation test, phosphorus, but are ineffective at aluminium oxide powder is mixed with
positive correlations (P<0.05) were removing soluble phosphorus. The clay to form a slurry, which is coated and
determined between influent and effluent absorptive filtration properties of the then baked onto the perlite media. In
EMCs for Ortho-P and NH3+. Based activated alumina media provides further operational conditions at low cartridge
on the use of the Mann-Whitney Rank removal mechanisms of total phosphorus flow rates, the RCFS also has the ability
Sum test, the difference in the median in stormwater through the synergistic to physically remove clay particles in
values between the influent and effluent effects of precipitation, adsorption and addition to the clay found within the
EMCs is greater than would be expected filtration (Ma, 2009). media from the manufacturing process.
by chance. Therefore, a statistically Substitution of silica by aluminum in
These results not only demonstrate
significant difference (P<0.05) was soil clay particles causes clays to have
that the system was able to provide
observed for TSS, SSC (<2000µm), SSC a negative charge (Cornell University,
substantial and consistent removal,
(<500µm), TP, PP, TKN, TN, and ON. 2007). Because of this negative charge
given the high solubility of phosphorus
at sites on or within the media structure,
Suspended Solids and at low mean influent concentration
we expect that the media would provide
Parameters for the study of 0.22mg/L, but was
some sorptive capacity and affinity for
Under Australian conditions Walker and able to attenuate TP captured by the
ammonium. The characteristics of the
Wong (1999) found that most suspended system over the entire course of the
media would also result in an increase in
solids in stormwater runoff is smaller than study. The soluble fraction appears to
pH. This increase in pH would cause the
500µm. In this study the closest parameter be higher than found by Vaze and Chew
ammonium to convert to ammonia and
was SSC<500µm. Influent EMCs for SSC (2004) under Australian conditions;
then volatilise into the atmosphere.
<500µm ranged from 9mg/L to 393mg/L they estimated that 20 to 30% of the
with a median of 29mg/L and a mean of phosphorus is soluble. It would be Fletcher et al. (2004) measured
66mg/L. Corresponding effluent EMCs expected, given the primary removal nitrogen concentrations of at least 2mg/L
ranged from 2mg/L to 10mg/L with a mechanism of the RCFS is physical for most urban land uses within Australia,
median of 3mg/L and a mean of 4mg/L, filtration, that a higher particulate while Duncan (1999) determined
resulting in an ER of 93% and a SOL fraction of phosphorus would yield median concentrations for roads and
efficiency of 94%. a similar result. urban catchments of 2.2 to 2.5mg/L.
In Melbourne, Taylor et al. found that,
This result needs to be explained Fletcher et al. (2004) measured mean
for TN in stormwater, 49% consisted of
in context with lower than expected TP concentrations of between 0.25
dissolved inorganic nitrogen (NH3+ and
influent concentrations. High pollutant and 0.50mg/L for most land uses while
NOx), which compares well with the 44%
concentrations lead to high percentage the BMP Database (2010) suggests
dissolved inorganic nitrogen load from
reductions and tend to over-estimate a typical TP range 0.11 to 0.38mg/L,
the RCFS study. The soluble component
the removal efficiency of pollutants from across a variety of land uses. Clearly
of nitrogen in stormwater found by
stormwater treatment systems (CSIRO, the mean influent TP concentration of
Wicks et al. (2011) and Miguntanna et al.
2010). A literature review by Duncan 0.22mg/L for this study correlates well
(2010) from roadways and commercial
(1999) showed that the median influent with the published data from Australia.
areas in Queensland was 50 to 60%. Vaze
concentration for total suspended In addition, the removal processes of
and Chew (2004), also under Australian
solids ranged from the lowest 41mg/L physical straining and adsorption are
conditions, found that the soluble portion
for roofs to 232mg/L for urban roads. independent of location and solely a
of nitrogen in stormwater can be up to
Fletcher (2004) found that mean influent function of water chemistry, filtration
50%. In the RCFS study, soluble organic
concentrations for roofs to roadways media and associated hydraulic
nitrogen was not measured; however,
ranged from 20mg/L to 270mg/L. conductivity. Hence our results should
assuming a modest allowance for soluble
apply to Australian stormwater.
Phosphorus Parameters organic nitrogen the soluble component
Nitrogen Parameters of nitrogen in stormwater for the RCFS
Given that the phosphorus removal
Given that the nitrogen removal target in study would be expected to be similar to
target in the Australian context is based
the Australian context is based solely on results from Australian studies.
solely on TP load removal efficiency, the
review of additional data was required TN load removal efficiency, the review of Conclusions
to further demonstrate the significance additional data was required to further
Between April 2011 and June 2012, 13
of the RCFS TP removal efficiencies demonstrate the significance of the
storm events were monitored and were
obtained. In an effort to isolate RCFS TN removal efficiencies obtained in
determined to meet the relevant USA
phosphorus removal efficiency based on this study. In an effort to further isolate
storm data collection requirements.
solubility of TP, dissolved phosphorus nitrogen removal efficiency NH3+ was
Significant reductions in sediment and
was also measured. Removal efficiencies also measured. Removal efficiencies
nutrient pollutant concentrations were
for TP and dissolved P were 86% and based on NH3+, led to overall removal
measured between influent and effluent
74%, respectively, using the ER method efficiency of 63% based on the ER and
sampling locations using the Efficiency
cf 87% and 67%, using the SOL efficiency 60% using the SOL efficiency calculation
Ratio (ER) calculation method (TSS 90%,
calculation method. method.
TP 86%, and TN 56%) and Summation
The removal of dissolved phosphorus There are several pathways in which of Load (SOL) efficiency calculation
is unsurprising but warrants further ammonium reduction may be occurring method (TSS 91%, TP 87%, and TN 50%).
discussion. Traditional forms of treatment in the RCFS system that require further These results demonstrate that the
such as settling and inert filtration discussion. During the manufacturing radial cartridge filter system was able
are able to remove particulate bound process of the activated alumina media, to successfully meet the current load-
water SEPTEMBER 2014
Page 98
5
Technical Papers
based objectives from the NSW DECC Urban Water Resources Research Council. Ma J, Lenhart J, Pedrick J & Tracy K (2010):
(2007) and QLD Single State Planning Jim holds a BS in Plant Sciences, a BS in Stormwater Phosphorus Removal Using
Innovative Filtration Media. American Society
Policy (2014) for all relevant pollutants, Agricultural Engineering and MS in Water
of Civil Engineers (ASCE), World Environmental
including TSS, TP and TN. Resources Engineering. He has authored
and Water Resources Congress: Challenges of
over 45 papers on the subject of water Change, Providence, Rhode Island.
Results from the 20-month study, that
quality and stormwater treatment.
represented 704mm of precipitation, Miguntanna NP, Goonetilleke A & Egodowatta
show that the RCFS tested was effective John Pedrick is Project P (2010): Understanding Nutrient Build-up on
in removing solid and nutrient pollutants Manager of Contech Urban Road Surfaces. Journal of Environmental
from the stormwater runoff. This study Engineered Solutions, Sciences, 22, 6, pp 806–812.
was completed using the recommended Portland Oregon and has Neumann L & Sharma A (2010): Literature Review
design criteria based on an individual 14 years of stormwater on Performance Testing Approaches for Gross
cartridge flow rate of 0.5L/s, activated management experience, Pollutant Traps. CSIRO, Water for a Healthy
alumina media, and a volume-based including managing numerous Country National Research Flagship.
design methodology. The RCFS was stormwater BMP monitoring programs NSW Department of Environment and
designed to capture and treat 75% of the in the US. He currently serves on the Climate Change (DECC, 2007): Managing
calculated water quality volume (i.e. the CPSWQ Council as the representative Urban Stormwater: Environmental Targets.
runoff volume associated with a 25mm for The Northwest Region and holds a Consultation Draft – October 2007,
Department of Environment and Climate
event). The RCFS was also designed on dual degree in Agricultural and Extension
Change NSW, p 4.
a mass-loading basis to meet the annual Education (AEE) and Environmental
pollutant loading requirements of the Resource Management (ERM) from The Rutgers, The State University of New Jersey
site with a minimum expected interval Pennsylvania State University College Department of Civil and Environmental
Engineering, New Jersey Department of
between maintenance of one year. of Agricultural Sciences.
Environmental Protection Division of Science,
The fraction of soluble nitrogen found References Research and Technology. (Rutgers/NJDEP)
in this study is in good agreement (2006): Correlation of Total Suspended Solids
BMP Database (2010): International Stormwater
(TSS) and Suspended Sediment Concentration
with the Australian data. The influent Best Management Practices (BMP) Database.
(SSC) Test Methods.: Trenton, Oregon:
concentrations, however, are at the lower
Pollutant Category Summary: Nutrients. Qizhong (George) Guo.
end of the Australian data. CSIRO (2010) Prepared by Geosyntec Consultants Inc. and
found that higher influent concentrations Taylor GD, Fletcher TD, Wong THF, Breen PF &
Wright Water Engineers Inc. (available from
Duncan HP (2005): Nitrogen Composition in
lead to higher percentage removals and, www.bmpdatabase.org).
Urban Runoff – Implications for Stormwater
in this context, we would expect results
Chiew FHS & Vaze J (2004): Nutrient Loads Management. Water Research, 39, pp
of the RCFS study to be conservative Associated with Different Sediment Sizes in 1982–1989.
when applied under Australian Urban Stormwater and Surface Pollutants.
URS Greiner Woodward Clyde, Urban Drainage
conditions. Journal of Environmental Engineering, 130,
and Flood Control District, Urban Water
4, pp 391–396. American Society of Civil
The Authors Engineers, April 2004.
Resources Research Council (UWRRC) of ASCE,
Office of Water, US Environmental Protection
Michael Wicks (email: Contech Stormwater Solutions Inc. (Contech) Agency (URS/EPA) (1999): Development of
michaelw@stormwater360. (2002): Evaluation of Lifecycle Loading Performance Measures Task 3.1 – Technical
com.au) is Technical Director Characteristics of The Stormwater Memorandum Determining Urban Stormwater
of Stormwater360 Australia. Management StormFilter® Cartridge Portland, Best Management Practice (BMP) Removal
Michael is a Civil Engineer Oregon. Efficiencies. Washington DC: Author.
with 15 years’ experience Cornell University Cooperative Extension US Geological Survey (USGS) (1980): Water
and an array of skills including hydraulics, (Cornell) (2007): Cation Exchange Capacity Resources Division by Office of Water Quality
hydrology, stormwater quality modelling, (CEC), Agronomy Fact Sheet Series, Fact (OWQ) Technical Memorandum No 80, 17.
structural design, project management, Sheet 22.
US Geological Survey (USGS) (1991): US
research and development. Michael has Duncan HP (1999): Urban Stormwater Quality: A Geological Survey, Techniques of Water-
designed, constructed, implemented, and Statistical Overview, Report 99/3, Cooperative Resources Investigations Reston, Virginia. DR
monitored numerous stormwater quality Research Centre for Catchment Hydrology, Helsel & RM Hirsch.
treatment facilities throughout the east Melbourne, Australia.
Walker TA & Wong THF (1999): Effectiveness
coast of Australia. Fletcher T, Duncan H, Poelsma P & Lloyd S, of Street Sweeping for Stormwater Pollution
(2004): Stormwater Flow and Quality, and the Control. Cooperative Research Centre for
Jim Lenhart is the founder of
Effectiveness of Non-Proprietary Stormwater Catchment Hydrology, Technical Report 99/8.
Stormwater Management Inc Treatment Measures – A Review and Gap
and former Chief Technology Washington State Department of Ecology
Analysis. Cooperative Research Centre for
Officer for CONTECH (WADOE) (2003): Guidance for Evaluating
Catchment Hydrology, Technical Report 04/8.
Emerging Stormwater Treatment Technologies:
Stormwater Solutions. He
Ma J, Lenhart J & Tracy K (2009): Phosphorus Technology Assessment Protocol – Ecology
is currently the owner of Removal in Urban Runoff Using Adsorptive (TAPE) (Publication Number 02-10-037).
Stormwater Northwest and consults with Filtration Media. Contech Stormwater Olympia, WA: Author. Available online: www.
companies that provide products and Solutions Inc. ecy.wa.gov/pubs/0210037.pdf
services in the stormwater related markets.
Ma J, Lenhart J & Tracy K (2011): Ortho-phosphate Wicks M, Vigar N & Hannah M (2011): Nutrients
Jim is currently an active member of the Adsorption Equilibrium and Breakthrough and Solids Removal by an Engineered
Water Environment Federation, the Water on Filtration Media for Stormwater Runoff Treatment Train – Field Evaluation of a Gully
Environment Research Foundation, ASCE Treatment. Journal of Irrigation and Drainage, Pit Insert and Cartridge Media Filter. Water
EWRI, and serves as Vice Chair for the April, 2011, pp 244–250. Journal, 38, 6, pp 83–88.
SEPTEMBER 2014 water
Page 99
HUNTER ENVIRONMENTAL SERVICES
Stormwater and Catchment Management
Geoffrey J. Hunter (MIEANZ); Churchill Fellow 1996 — Soil & Water Management
Jillian M. Hunter; Post Graduate Diploma Adult & Community Education
52 Centre Crescent Blaxland NSW Australia 2774
Phone (mobile): 0437 502 952 E-mail: [email protected]
A.B.N. : 90 264 168 968
Our Ref.: 1410 15 September 2014
Technical Director
Stormwater 360 Australia
PO Box 792
BOTANY NSW 1455
Attention: Mr Michael Wicks
RE: Peer Review of Product Evaluation Report
Stormwater Management StormFilter® Stormwater Treatment System
Mitchell Community College Mooresville, NC
Thank you for commissioning Hunter Environmental Services to peer review the field study of
the StormFilter® technology undertaken by Contech Engineered Solutions within the parking
lot of the Mitchell Community College in Mooresville North Carolina USA.
The primary goal of the field study and subsequent report “was to collect performance data in
order to support the approval of the StormFilter system and its potential inclusion into the
North Carolina Stormwater BMP Manual. A major condition of the approval for continued
participation in the PEP [Preliminary Evaluation Period] program was the execution of an
additional field evaluation in accordance with the conditions outlined in the PEP program.”
Consequently monitoring and evaluation of the performance of the StormFilter® system,
installed in the Mitchell Community College carpark, was carried out to demonstrate its
effectiveness at removing Total Suspended Solids (TSS) and nutrients (Phosphorus and
Nitrogen) from runoff, as required by the PEP program.
Our review has concentrated on answering the key issues raised in your commissioning letter
of 5 August 2014, namely:
“Review of treatment system, sampling setup and equipment and sampling methodology.
Confirmation of independent sampling chain of custody external to Contech Engineered
Solutions.
Review of Report including calculations, discussions and conclusions. (refer to Discussion
and Conclusion sections within this Peer Review).
Comment on the comparisons of EMC’s from the study compared with local data ie
Duncan & Fletcher + others (if applicable) to help provide transferability of results.
Comment on the comparison of nutrient species or the composition of soluble species
compared with local data.
HES – StormFilter® System Peer Review Page 1
Page 100
REVIEW
1. Review of StormFilter®Treatment System, Sampling and Methodology
The StormFilter® system described in the report (Contech 2012) is made up of an eight (8)
cartridge StormFilter® preceded by a detention tank designed to capture 75% of the runoff
volume from the 1.0 inch (25.4 mm) event. The catchment to the system consists of a 1.08
acre (0.437 ha) carpark with an equivalent impervious area of 0.73 acres (0.297 ha)
representative of a catchment that is 68% impervious.
The layout provides for detention of the runoff with a controlled discharge to the StormFilter®
treatment system at a flow rate determined by the hydraulic capacity of eight (8) StormFilter®
cartridges. Each cartridge was filled with an activated alumina coated perlite media
(PhosphoSorb).
The layout of this treatment system allows the catchment hydrology to be modified based on
the hydraulics of the tank cartridges and media rather than the runoff characteristics of the
catchment. This system provides a generic layout suitable for applications where there are
major constraints on the available site area for treatment options. It could be expected that
the alumina coating on the perlite media would enhance the adsorption of pollutants onto the
exterior of the perlite media, while the perlite itself has a demonstrated ability to absorb
pollutants (Siber A et al, 2011)
Influent and effluent samples were collected using twelve (12), 1 litre sample bottles.
Collection of flow proportional samples commenced once the flow rate exceeded 5 g(US)pm
(0.32 L/s). The sample strainers intakes and flow measurement equipment were secured to
the invert of the influent and effluent pipes, which biases the samples towards the larger
fraction of the suspended solids. Collection and testing of the samples was carried out by
Pace and TestAmerica independent of Contech and oversight of the project was provided by
Ryan Winston, MS, extension Associate in the Department of Biological and Agricultural
Engineering at North Carolina State University.
Rinsate blank and replicate samples were collected, which is consistent with the sampling
guidelines detailed in NWQMS (2000) and demonstrates scientific rigour within the sampling
methodology as well as a good understanding of the vagaries of sampling rainfall/runoff
events.
2. Sampling “chain of custody”
Personnel from Pace and TestAmerica collected the samples and delivered them to the
laboratory for testing. No record of the use of “chain of custody” procedures for the samples
is referred to in the Porject Evaluation, submitted for this peer review. However,
correspondence with Contech Engineered Solutions has identified that “chain of custody”
procedures, similar to those outlined in Sections 4 and 5 (NWQMS, 2000) and Section 6
(EPASA, 2007), were employed as part of the sampling Project Plan. Further, Pace and
TestAmerica are accredited laboratories in accordance with the National Environmental
Laboratory Accreditation Conference (NELAC) and all samples were tested in accordance
with their nationally approved testing protocols (John Pedrick pers com).
HES – StormFilter® System Peer Review Page 2
Page 101
3. EMC Comparison with Local Data
Very little published data exists on pollutants in runoff from carparks worldwide, and in
particular in Australia. The very nature of the carpark and its use can vary both the amount
and type of pollutants present in stormwater and the high impervious area and small
catchment area generally favours a “first flush” effect, especially in regions with high rainfall
intensities and defined wet and dry seasons. Further, maintenance/cleaning procedures e.g.
mechanical sweeping, can artificially alter the pollutant species and their concentrations, in
runoff from carparks.
Morison P (2001) reported on a study that categorised pollutant loads from a commercial
carpark in western Sydney using a rainfall simulator calibrated for a 1 in six month storm of 15
minutes duration. The results of the study demonstrate a marked “first flush” effect for 10
minutes with an approximate Event Mean Concentrations (EMCs) for a duration of 15
minutes of:
Suspended Solids 95 mg/L;
Total Phosphorus 0.15 mg/L;
Total Nitrogen 1.85 mg/L.
The EMCs, determined by this study, approximate Typical Wet Weather Concentrations for
Roads provided in Tables 2.43, 2.44 and 2.45, by Fletcher T et al (2004), which are generally
accepted as typical EMCs for various land uses in Australia:
Suspended Solids range 90 to 800 with a typical value of 270 mg/L;
Total Phosphorus range 0.15 to 1.5 with a typical value of 0.5 mg/L;
Total Nitrogen range 1 to 5 with a typical value of 2.2 mg/L.
The EMCs reported in the StormFilter® Product Evaluation carried out over twenty (20)
months within the Mitchell Community College carpark are:
Suspended Solids range 10.3 to 98.2 with a mean value of 34.6 mg/L;
Total Phosphorus range 0.07 to 0.90 with a mean value of 0.22 mg/L;
Total Nitrogen range 0.35 to 2.95 with a mean value of 1.00 mg/L.
Comparison of Event Mean Concentration (mg/L) for Various Locations
Variable St Martins Carpark CRCCH TR04/8 Mitchell Community
NSW (Australia) College NC
Suspended Solids 95 (approx) 90 to 800 10.3 to 98.2
typical 270 mean 34.6
Total; Phosphorus 0.15 (approx) 0.15 to 1.5 0.07 to 0.90
typical 0.5 mean 0.22
Total Nitrogen 1.85 (approx) 1 to 5 0.35 to 2.95
typical 2.2 mean 1.00
The above comparison identifies significantly lower EMCs measured in runoff from the
Mitchell Community College, Mooresville NC than are generally accepted as typical for car
parks in NSW and Australia. With respect to the determination of an Efficiency Ratio (ER) for
a stormwater treatment system, the higher the EMC in the influent the easier it is to achieve a
particular removal efficiency. Consequently the ER determined for the Mitchell Community
College system could be expected to be much higher if the influent EMCs approached those
reported in the Australian literature.
HES – StormFilter® System Peer Review Page 3
Page 102
4. Nutrient Species Comparison with Local Data
There is limited information available relating to the categorisation of nutrients, into their
constituent species, within runoff from carparks and roads in Australia. However, a literature
search did reveal a range of EMCs for Ammonia which could be used for comparative
purpose. These are referenced below.
Morison P (2001) reported on the Ammonia EMC in runoff from the St Martins carpark which
was in the order of 1 mg/L.
The recommended typical EMC values for nutrients in runoff from various land uses, provided
in Fletcher T et al (2004), do not include a species breakdown for either Total Phosphorus or
Total Nitrogen. Other data, referencing Ammonia EMCs in runoff from urban catchments in
Sydney and the Illawarra, suggest a mean range of between 0.02 and 0.54 mg/L while similar
EMCs measured in Hornsby returned mean Ammonia EMC ranges of between 0.01 and 0.11
mg/L. Similarly, Ammonia EMCs in runoff from urban catchments in Grafton were found to be
between 0.02 and 0.41 mg/L (Fletcher T et al, 2004).
The Contech Engineered Solutions (2012) product evaluation identified an EMC range of
between 0.05 mg/L and 0.72 mg/L with a mean of 0.27 mg/L for the Ammonia in runoff from
the Mitchell Community College carpark.
Consequently it would appear that the Ammonia EMCs measured in runoff from the Mitchell
Community College carpark are similar to those reported in runoff from the St Martins carpark
and other urban catchments in the Sydney, Illawarra and Grafton regions of NSW as reported
in Fletcher T et al (2004).
DISCUSSION
Based on the information provided in Contech Engineered Solutions (2012) product
evaluation of the StormFilter® installed, within the Mitchell Community College carpark, it is
our opinion that the sampling and monitoring of the system has been undertaken with
sufficient scientific rigour to meet the objectives of the project. Further, the sampling
methodology and testing procedures used by Contech Engineered Solutions are consistent
with the Australian and New Zealand Guidelines, NQWMS (2000).
With regards to compatibility with Australian conditions:
Pollutant concentrations in the influent appear lower than those that may be expected
under Australian conditions. Consequently the Efficiency Ratios determined may be
expected to be higher if the same StormFilter® treatment system were installed on a
similar carpark in Australia;
The use of a storage tank, upstream of the treatment system, allows the required
treatment volume of runoff to be released to the treatment system at a flow rate
comparable to the Treatable Flow Rate of the cartridge filter system. Consequently
comparisons between rainfall intensities is not thought necessary as the storage
volume within the detention tank and the size of the outlet orifice control the flow rate
HES – StormFilter® System Peer Review Page 4
The text continues on page 103.