oÉëÉ~êÅÜ=êÉéçêí Sloping Sand Filters for On-Site Wastewater Treatment External Research Program

oÉëÉ~êÅÜ=êÉéçêí Sloping Sand Filters for On-Site Wastewater Treatment External Research Program
oÉëÉ~êÅÜ=êÉéçêí
External Research Program
Sloping Sand Filters for On-Site Wastewater
Treatment
`je`ÔeçãÉ=íç
`~å~Çá~åë
Canada Mortgage and Housing Corporation (CMHC) has
been Canada’s national housing agency for more than 60 years.
Together with other housing stakeholders, we help ensure
that Canada maintains one of the best housing systems in the
world. We are committed to helping Canadians access a wide
choice of quality, affordable homes, while making vibrant,
healthy communities and cities a reality across the country.
For more information, visit our website at www.cmhc.ca
You can also reach us by phone at 1-800-668-2642
or by fax at 1-800-245-9274.
Outside Canada call 613-748-2003 or fax to 613-748-2016.
Canada Mortgage and Housing Corporation supports
the Government of Canada policy on access to
information for people with disabilities. If you wish to
obtain this publication in alternative formats,
call 1-800-668-2642.
SLOPING SAND FILTERS FOR ON-SITE WASTEWATER
TREATMENT
Report to:
Canada Mortgage and Housing Corporation
External Research Program
Submitted by:
Rob Jamieson, Department of Process Engineering and Applied Science, Dalhousie University
Janice Wilson, Department of Process Engineering and Applied Science, Dalhousie University
Peter Havard, Engineering Department, NS Agricultural College
June, 2009
This project was funded by Canada Mortgage and Housing Corporation (CMHC) under the terms of the
External Research Program, but the views expressed are the personal views of the authors and do not
represent the official views of CMHC.
research highlight
December 2009
Technical Series 09-112
Sloping Sand Filters for On-Site Wastewater
Treatment
introduction
PROJECT OBJECTIVES
Approximately 50% of Nova Scotia residents utilize
on-site septic systems for wastewater treatment and disposal,
however conventional treatment systems may be at greater
risk of premature failure due to geologic conditions
commonly found in the province, such as low permeability
soils, shallow bedrock, and high water tables. To address
these issues, sloping sand filters (SSF), or lateral flow sand
filters (LFSF) as presented in figure 1.1, have been approved
for use as remedial technology to replace failed conventional
systems. While laboratory studies have indicated satisfactory
performance very little field analysis has been undertaken.
This report presents findings from a two-year study
examining the hydraulics and treatment performance of
sloping sand filters (SSF), or lateral flow sand filters (LFSF),
for on-site residential wastewater treatment. This study
utilized, and expanded upon, an existing set of field scale
SSFs that were installed at the Bio-environmental
Engineering Centre (BEEC) in Truro, NS for experimental
purposes. The overall objective of this research project
was to assess current design guidelines for SSFs, and make
recommendations for the expanded use, and optimization,
of these types of systems. Specific objectives included:
(i) assessing the hydraulic behaviour and performance of
conventionally designed SSFs at loading rates greater than
those allowed within NS technical guidelines, (ii) assessing
the hydraulic behaviour and performance of SSFs that were
Figure 1
Schematic of a typical Lateral Flow Sand Filter used in Nova Scotia.
Research Highlight
Sloping sand filters for on-site wastewater treatment
substantially shorter than those required within NS technical
guidelines, (iii) determining if current hydraulic models of
SSF systems are appropriate.
■
In general, sand grain size had the greatest influence
on residence time characteristics, as opposed to slope
and wastewater loading rate. As expected, the fine
grained filters have higher residence times compared
to medium- and coarse-grained filters, presumably due
to the smaller pore space for the wastewater to travel
through. Residence times from medium and coarse
grained filters were comparable.
■
Tracer studies indicated that the residence time
characteristics of the filters did not change after the
wastewater loading rate was increased. This provided
further evidence that biomat hydraulics have a large
influence on the speed of wastewater movement
through the filters.
■
Comparison of measured mean residence times to
theoretical residence times computed assuming saturated
flow confirmed that saturated Darcy flow did not occur
within the SSFs.
■
In general the SSFs performed well over the monitoring
period, producing average effluent five-day biochemical
oxygen demand (BOD5) concentrations below 10 mg/L,
and average total suspended solids (TSS) concentrations
below 15 mg/L.
■
Median outlet E. coli levels were all below 10 CFU / 100 mL,
indicating that the sand filters were generally effective
at removing enteric bacteria. However, on several
occasions levels of E. coli in filter effluent exceeded
100 CFU / 100 mL, illustrating variability in treatment
performance. This indicated that effluent from
SSF systems could not be surface discharged without
additional disinfection.
■
SSFs that were loaded at elevated rates, as compared
to NS technical guidelines, performed well, providing
a level of treatment that was similar to that observed
at conventional loading rates.
Methodology
A total of eight pilot-scale SSF systems were installed
at the BEEC, and continuously dosed with septic tank
effluent. Filter construction encompassed the full range
of sand sizes and slopes allowed within NS technical
guidelines. Six of the SSFs were constructed according
to NS technical guidelines but loaded at approximately
double the recommended linear loading rate. Two replicate
SSFs were constructed with downgradient lengths that
were 50% of the length recommended in the NS technical
guidelines, but were loaded according to the recommended
linear loading rates. The hydrology and treatment
performance of the systems were monitored over
a 16-20 month period. A series of tracer studies were
conducted within all eight filters over the course of the
study. Tracer study data was fit to analytical residence time
distribution models to assess the hydraulic functioning
of the filters. Measured hydraulic characteristics were then
compared to those predicted using theoretical models
of porous media flow.
KEY FINDINGS
■
External hydrologic factors can have a large influence
on the flow characteristics of SSF systems. Outflows
from the SSFs varied greatly, increasing by a factor
of 4 during some precipitation events. However, the
influence of precipitation on SSF hydrology is complex,
and a strong function of antecedent moisture conditions.
■
Observations from piezometers installed at several
locations within each filter indicated that the filters
were not fully saturated, even after the wastewater
loading rate was increased.
■
2
Observations from monitoring wells installed in
the distribution trench of each filter indicated that
the biomat within each filter was stable. Minor increases
in ponding depths were observed in the distribution
trench after wastewater loading rates were increased.
Canada Mortgage and Housing Corporation
Research Highlight
Sloping sand filters for on-site wastewater treatment
■
The shortened filters performed at the same level
as a regular length filter under similar loading rate
conditions. These results suggest that the majority
of treatment occurs within the first part of the LFSF,
and that treatment processes occurring within the biomat
control the treatment efficiency of the system.
■
Removal of Total Nitrogen (TN) was not affected by
the increase in wastewater loading rate. The SSFs
consistently removed between 40 – 50% of TN.
■
A progressive reduction in phosphorus (P) removal
was observed in each filter. After 3 years of effluent
loading, several of the original six filters appear to be
saturated with P. As expected, the coarse-grained filters
saturate more quickly than fine-grained filters, as they
would possess less surface area for adsorption. These
results indicate that conventional sand-based disposal
systems have virtually no long-term phosphorus
removal capacity.
IMPLICATIONS FOR THE HOUSiNG
INDUSTRY
This study has shown that current design loading rates
for SSFs in NS are conservative, and provide an inherent
safety factor. The SSF system was shown to be a reliable,
relatively robust treatment system. However, sand-based
systems such as these indicate variable E. coli removal and
poor long-term phosphorous removal. Accordingly,
consideration should be given to additional treatment
and disinfection when utilized for surface discharge or
for discharge into sensitive receiving bodies.
Canada Mortgage and Housing Corporation
3
Research Highlight
Sloping sand filters for on-site wastewater treatment
CMHC Project Manager: Cate Soroczan
Consultants: Dr. Rob Jamieson, Janice Wilson, Department
of Process Engineering and Applied Science, Dalhousie University
Peter Harvard, Engineering Department, Nova Scotia
Agricultural College
This study was funded (or partially funded) by Canada Mortgage
and Housing Corporation (CMHC) under the terms of its External
Research Program. However, the views expressed are the personal
views of the author and do not necessarily reflect the views of
CMHC. CMHC’s financial contribution to this study does not
constitute an endorsement of its contents. For more information
on the ERP, please visit the CMHC website at www.cmhc.ca or
contact the Project Officer, Responsive Programs by e-mail [email protected]
cmhc-schl.gc.ca, or by regular mail: Project Officer, Responsive
Programs, External Research Program, Policy and Research Division,
Canada Mortgage and Housing Corporation, 700 Montreal Road,
Ottawa ON K1A 0P7.
To find more Research Highlights plus a wide variety of
information products, visit our website at
www.cmhc.ca
or contact:
Canada Mortgage and Housing Corporation
700 Montreal Road
Ottawa, Ontario
K1A 0P7
Phone:
Fax:
1-800-668-2642
1-800-245-9274
66702
©2009, Canada Mortgage and Housing Corporation
Printed in Canada
Produced by CMHC
09-12-09
Although this information product reflects housing experts’ current knowledge, it is provided for general information purposes only. Any reliance
or action taken based on the information, materials and techniques described are the responsibility of the user. Readers are advised to consult
appropriate professional resources to determine what is safe and suitable in their particular case. Canada Mortgage and Housing Corporation
assumes no responsibility for any consequence arising from use of the information, materials and techniques described.
Le point en recherche
Décembre 2009
Série technique 09-112
Filtres à sable en pente servant d’installation
d’assainissement sur place des eaux usées
INTRODUCTION
OBJECTIFS
Environ 50 % des résidents de la Nouvelle-Écosse utilisent
une installation septique sur place pour l’assainissement et
l’élimination des eaux usées. Les installations classiques,
toutefois, présentent un plus grand risque de défaillance
prématurée en raison des conditions géologiques que l’on
trouve communément dans la province, comme un sol à
faible perméabilité, une roche de fond à faible profondeur
ou une nappe d’eau élevée. Pour endiguer ces problèmes,
des filtres à sable en pente (ou filtres à sable à écoulement
latéral), illustrés à la figure 1.1, ont été approuvés à titre
de technologie de rechange pour remplacer les installations
classiques défectueuses. Bien que des essais en laboratoire
aient donné des résultats satisfaisants, très peu d’analyses sur
le terrain ont été entreprises.
Le rapport dont il est ici question fait état des constatations
d’une étude qui s’est déroulée sur deux ans et qui a porté sur
l’examen des caractéristiques hydrauliques et de la performance
en traitement des filtres à sable en pente, ou filtres à sable à
écoulement latéral, servant d’installation d’assainissement sur
place pour les habitations. L’étude a tiré parti d’une série
d’installations de filtres à sable en pente qui ont été mises en
place au Bioenvironmental Engineering Centre (BEEC), à
Truro (N.-É.), à des fins expérimentales. L’objectif principal
de la recherche consistait à évaluer les directives de conception
actuelles des filtres à sable en pente et à formuler des
recommandations quant à l’utilisation plus étendue de ce
genre d’installation et leur optimisation. Les travaux de
recherche avaient pour objectifs particuliers (i) d’évaluer le
comportement hydraulique et la performance des filtres à
Figure 1
Schéma d’un filtre à sable à écoulement latéral type utilisé en Nouvelle-Écosse.
AU CŒUR DE L’HABITATION
Le Point en recherche
Filtres à sable en pente servant d’installation d’assainissement sur place des eaux usées
sable en pente de conception courante fonctionnant à des
taux de charge plus importants que ceux autorisés par les
lignes de conduite de la N.-É., (ii) d’évaluer le
comportement hydraulique et la performance des filtres à
sable en pente qui étaient beaucoup plus courts que ceux
exigés par les lignes de conduite techniques de la N.-É. et
(iii) de déterminer dans quelle mesure les modèles
hydrauliques actuels visant les installations de filtres
à sable en pente sont adéquats.
■
Des observations tirées de piézomètres installés à divers
endroits à l’intérieur de chaque filtre révèlent que les
filtres ne sont pas entièrement saturés, même après
augmentation du taux de charge en eaux usées.
■
Des lectures tirées de puits d’observation installés dans la
tranchée de distribution de chaque filtre indiquent que le
film biologique à l’intérieur de chaque filtre était stable.
On a observé un faible accroissement de la profondeur
des accumulations de liquides dans la tranchée de
distribution après avoir augmenté les taux de charge
en eaux usées.
■
En règle générale, c’est la taille des grains de sable qui
influait davantage sur les caractéristiques de temps de
séjour, par rapport à la pente ou au taux de charge
en eaux usées. Comme prévu, les filtres à grain fin
présentent des temps de séjour plus longs
comparativement aux filtres à sable à grains moyens
ou grossiers, sans doute à cause du plus faible espace
interstitiel à travers duquel les eaux usées se déplacent.
Les temps de séjour des filtres à grains moyens ou
grossiers étaient du même ordre.
■
Les études sur traceur révèlent que les caractéristiques de
temps de séjour dans les filtres demeuraient inchangées
même si le taux de charge en eaux usées était augmenté.
Cette constatation constitue une autre preuve que
les caractéristiques hydrauliques des films biologiques
influent largement sur la vitesse d’écoulement des eaux
usées à travers les filtres à sable.
■
Une comparaison établie entre les temps moyens de
séjour et les temps de séjour théoriques calculés, en
supposant un écoulement saturé, a confirmé que
l’écoulement dans les installations étudiées ne se faisait
pas selon l’écoulement saturé de Darcy.
■
En général, les filtres à sable en pente ont affiché une
performance satisfaisante au cours de la période de suivi,
produisant un effluent dont les concentrations moyennes
en demande biochimique d’oxygène sur cinq jours
(BOD5) étaient inférieures à 10 mg/L, et dont les
concentrations moyennes totales de solides en suspension
(TSS) étaient inférieures à 15 mg/L.
MÉTHODE
Au total, on a aménagé huit installations pilotes de filtres
à sable en pente au BEEC, lesquelles ont été alimentées en
continu d’effluent de fosse septique. Pour la construction
des filtres, on a fait appel à toute la gamme de granulométrie
de sables et de pentes autorisées dans les lignes de conduite
techniques de la Nouvelle-Écosse. Six des installations ont été
construites selon les lignes de conduite techniques de la
N.-É., mais à une charge hydraulique d’environ le double de
la charge linéaire recommandée. Deux installations identiques
de filtres à sable en pente ont été mises en place avec des
longueurs de l’amont vers l’aval à 50 % de la longueur
recommandée dans les lignes de conduite techniques, tout
en recevant une charge conforme au taux de charge linéaire
recommandé. Les aspects de l’hydrologie et de la
performance en traitement des installations ont été suivis
pendant 16 à 20 mois. Au cours des travaux, une série
d’études par traceur a été menée dans les huit filtres à sable.
Les données des études par traceur ont été comparées à des
modèles de distribution analytiques en temps de séjour aux
fins d’évaluation du fonctionnement des filtres à sable.
Les caractéristiques hydrauliques mesurées ont par la suite
été comparées à celles prévues par les modèles théoriques
d’écoulement en milieux poreux.
CONSTATATIONS CLÉS
■
2
Des facteurs hydrologiques externes peuvent avoir un
impact important sur les caractéristiques d’écoulement
des installations de filtres à sable en pente. Le débit de
sortie a varié grandement, quadruplant lors de certains
épisodes de précipitations. L’effet des précipitations sur
l’hydrologie des filtres à sable en pente est toutefois
complexe et dépend en grande partie des conditions
antérieures d’humidité.
Société canadienne d’hypothèques et de logement
Le Point en recherche
Filtres à sable en pente servant d’installation d’assainissement sur place des eaux usées
■
Les niveaux moyens de colibacilles dans l’effluent étaient
tous inférieurs à 10 CFU/100 ml, ce qui indique que les
filtres à sable étaient généralement efficaces en matière
d’élimination d’entébactéries. À plusieurs occasions,
toutefois, les niveaux de colibacilles dans l’effluent des
filtres ont excédé 100 CFU/100 ml, illustrant ainsi la
variabilité de la performance en traitement. Cette
situation indique qu’il faut éviter d’acheminer l’effluent
des filtres à sable en pente vers les surfaces de
ruissellement sans désinfection préalable.
■
Les filtres à sable en pente qui ont traité une charge
élevée d’effluent comparativement aux lignes de conduite
techniques de la Nouvelle-Écosse ont présenté une
performance satisfaisante, c’est-à-dire qu’ils ont fourni un
niveau de traitement semblable à celui observé à des taux
de charge traditionnels.
■
Les filtres à sable raccourcis ont affiché la même
performance en traitement que les filtres de longueur
courante dans des conditions de charge semblables.
Ces résultats suggèrent que la majorité du traitement
s’effectue dans la première partie du filtre à sable
à écoulement latéral et que ce sont les processus de
traitement qui ont cours dans le film biologique qui
déterminent l’efficacité de l’installation.
■
L’augmentation du taux de charge en eaux usées n’a pas
influé sur l’élimination de l’azote total. Les filtres à sable
en pente ont éliminé de façon constante entre 40 et
50 % de l’azote total.
■
On a observé une réduction progressive du taux
d’élimination du phosphore dans chacun des filtres.
Après trois années de traitement de charge en effluent,
plusieurs des six filtres d’origine semblent être saturés en
phosphore. Comme on s’y attendait, les filtres à sable à
grains grossiers se sont saturés plus rapidement que les
filtres à grains fins, puisque les premiers présentent une
surface d’adsorption moins étendue. Ces résultats
révèlent que les installations d’assainissement courantes
à base de sable ne possèdent aucune capacité à long
terme d’élimination du phosphore.
CONSÉQUENCES POUR
LE SECTEUR DE L’HABITATION
Cette étude a montré que les taux de charge actuellement
utilisés lors de la conception de filtres à sable en pente en
Nouvelle-Écosse sont prudents et comportent donc un
facteur de sécurité inhérent. Il a été montré que le filtre
à sable en pente constitue une installation d’assainissement
fiable et relativement robuste. Cela dit, les installations
à base de sable de ce type affichent des taux variables
d’élimination des colibacilles et une piètre élimination
à long terme du phosphore. C’est pourquoi il faudrait
considérer la mise en œuvre de techniques de désinfection
et de traitement additionnelles lorsque l’effluent de sortie est
acheminé en surface du sol ou vers un milieu récepteur sensible.
Société canadienne d’hypothèques et de logement
3
Le Point en recherche
Filtres à sable en pente servant d’installation d’assainissement sur place des eaux usées
Directeur de projet à la SCHL : Cate Soroczan
Consultants : Rob Jamieson, Janice Wilson, départment
de génie des procédés opérationnels et des sciences appliquées,
Université Dalhousie
Peter Harvard, départment de génie, Collège agricole de la
Nouvelle-Écosse
Cette étude a été financée (ou financée en partie) par la Société
canadienne d’hypothèques et de logement (SCHL) dans le cadre du
Programme de subventions de recherche (PSR), mais les opinions
exprimées dans l’étude sont celles de l’auteur et ne reflètent pas
nécessairement les opinions de la SCHL. La contribution financière
de la SCHL à cette étude ne constitue nullement une approbation
de son contenu. Pour en savoir plus sur ce programme, visitez le site
Web de la SCHL à www.schl.ca ou communiquez avec l’agent de projets,
Recherche d’initiative privée, par courriel, à [email protected], ou par
la poste à : Agent de projets, Recherche d’initiative privée, Programme
de subventions de recherche, Division de la recherche et des politiques,
Société canadienne d’hypothèques et de logement, 700 chemin de
Montréal, Ottawa (Ontario) K1A 0P7.
Pour consulter d’autres feuillets Le Point en recherche et pour
prendre connaissance d’un large éventail de produits d’information,
visitez notre site Web au
www.schl.ca
ou communiquez avec la
Société canadienne d’hypothèques et de logement
700, chemin de Montréal
Ottawa (Ontario)
K1A 0P7
Téléphone : 1-800-668-2642
Télécopieur : 1-800-245-9274
66703
©2009, Société canadienne d’hypothèques et de logement
Imprimé au Canada
Réalisation : SCHL
09-12-09
Bien que ce produit d’information se fonde sur les connaissances actuelles des experts en habitation, il n’a pour but que d’offrir des
renseignements d’ordre général. Les lecteurs assument la responsabilité des mesures ou décisions prises sur la foi des renseignements contenus
dans le présent ouvrage. Il revient aux lecteurs de consulter les ressources documentaires pertinentes et les spécialistes du domaine concerné
afin de déterminer si, dans leur cas, les renseignements, les matériaux et les techniques sont sécuritaires et conviennent à leurs besoins.
La Société canadienne d’hypothèques et de logement se dégage de toute responsabilité relativement aux conséquences résultant de l’utilisation
des renseignements, des matériaux et des techniques contenus dans le présent ouvrage.
Sloping Sand Filters for On‐Site Wastewater Treatment
2009 EXECUTIVE SUMMARY
This report presents findings from a two year study examining the hydraulics and treatment performance
of sloping sand filters (SSF), or lateral flow sand filters (LFSF), for on-site residential wastewater
treatment. Lateral flow sand filters have been used in Nova Scotia (NS) for several years as a remedial
option for failed soil absorption fields. The use of these types of treatment systems could be greatly
expanded in NS, and across Canada, however, the performance, and hydraulic behavior, of these types
of systems is poorly understood. This study utilized, and expanded upon, an existing set of field scale
SSFs that were installed at the Bioenvironmental Engineering Centre (BEEC) in Truro, NS for
experimental purposes. The overall objective of this research project was to assess current design
guidelines for SSFs, and make recommendations for the expanded use, and optimization, of these types
of systems. Specific objectives included: (i) assessing the hydraulic behavior and performance of
conventionally designed SSFs at loading rates greater than those allowed within NS technical guidelines,
(ii) assessing the hydraulic behavior and performance of SSFs that were substantially shorter than those
required within NS technical guidelines, (iii) determining if current hydraulic models of SSF systems are
appropriate.
A total of eight pilot scale SSF systems were installed at the BEEC, and continuously dosed with septic
tank effluent. Six of the SSFs were constructed according to NS technical guidelines but loaded at
approximately double the recommended linear loading rate. Two replicate SSFs were constructed with
downgradient lengths that were 50% of the length recommended in the NS technical guidelines, but
were loaded according to the recommended linear loading rates. The hydrology and treatment
performance of the systems were monitored over a 16-20 month period. A series of tracer studies were
conducted within all eight filters over the course the study. Tracer study data was fit to analytical
residence time distribution models to assess the hydraulic functioning of the filters. Measured hydraulic
characteristics were then compared to those predicted using theoretical models of porous media flow.
In general, the results of the study provide further evidence that SSF systems are a relatively reliable,
and robust, form of on-site wastewater treatment. Water level measurements within the SSF systems
loaded at double the recommended linear rate revealed that filters largely remained unsaturated, and
biomats hydraulics remained stable. Mean residence times did not appear to increase with increased
loading rates. The treatment performance of the SSF systems loaded at double the recommended rate
was not statistically different from those reported previously for these systems when loaded at the
recommended rate. The treatment performance of the two shortened SSF systems was also similar to
that reported for a comparable SSF system that possessed a sand toe that was twice as long. These
findings confirm that existing technical guidelines for SSF design are conservative, and that the
hydraulic behavior of these types of systems is not well represented as saturated darcy flow. An
additional interesting observation from this study was the progressive reduction in phosphorus removal
within all the monitored SSFs. Results from this study suggest that sand-based disposal fields become
saturated with phosphorus within 3-5 years.
ii National Office
700 Montreal Road
Ottawa ON K1A 0P7
Telephone: (613) 748-2000
Bureau national
700 chemin de Montréal
Ottawa ON K1A 0P7
Téléphone : (613) 748-2000
Puisqu’on prévoit une demande restreinte pour ce document de
recherche, seul le résumé a été traduit.
La SCHL fera traduire le document si la demande le justifie.
Pour nous aider à déterminer si la demande justifie que ce rapport soit
traduit en français, veuillez remplir la partie ci-dessous et la retourner à
l’adresse suivante :
Centre canadien de documentation sur l’habitation
Société canadienne d’hypothèques et de logement
700, chemin Montréal, bureau C1-200
Ottawa (Ontario)
K1A 0P7
Titre du rapport: _______________________________________
_______________________________________
Je préférerais que ce rapport soit disponible en français.
NOM _____________________________________________
ADRESSE___________________________________________
rue
App.
___________________________________________________________
ville
province
Code postal
No de téléphone (
) ____________
Sloping Sand Filters for On‐Site Wastewater Treatment
2009 TABLE OF CONTENTS
EXECUTIVE SUMMARY……………………………………………………………….……………ii
TABLE OF CONTENTS………………………………………………………………………….…...iii
LIST OF TABLES………………………………………………………………………...……….…...iv
LIST OF FIGURES……………………………………………………………………….……………v
1.0 BACKGROUND AND PROJECT OBJECTIVES……………………………………………….1
1.1 Project Objectives…………………………………………………………………………..3
2.0 METHODOLOGY…………………………………………………………………………………4
2.1 SSF Construction and Loading Regimes……………..………………………………...…..4
2.2 Hydrologic Monitoring…………………………………………………………………...…4
2.3 Hydraulic Assessment and Modeling…………………….…………………………………6
2.4 Performance Monitoring…………………………………………………………………….7
3.0 RESULTS AND DISCUSSION……………………………………………………………….......8
3.1 Flow Characteristics…………………………………………………………………….…..8
3.2 Hydraulic Assessment and Modeling…………………………………………….……..….16
3.3 Treatment Performance………………………………………………………….……….…29
3.3.1. Statistical Summary……………………………………………………………….29
3.3.2. Temporal Trends in Treatment Performance……………………………….……..35
4.0 SUMMARY AND CONCLUSIONS…………………………………………………….…......…44
5.0 REFERENCES………………………………………………………………………………...….,.45
APPENDIX A: Photographs………………………………………………………………………..…..46
iii Sloping Sand Filters for On‐Site Wastewater Treatment
2009 LIST OF TABLES
Table 2.1. Physical characteristics of the filter sands……………………………………….…………...4
Table 3.1. Descriptive Statistics for SSF1-6, including Average Daily Flow, Standard Deviation,
Minimum and Maximum Flows………………………………………………………..…………..……..9
Table 3.2. Descriptive Statistics for SSF7-8, including Average Daily Flow, Standard Deviation,
Minimum and Maximum……………………………………………………………………………........9
Table 3.3. Hydraulic parameters generated from tracer studies on SSF1-8……………………………..16
Table 3.4. Theoretical residence times (tT, hours) for each SSF…………………………………….....….28
Table 3.5. Statistical summary of filter performance (SSF 1 - 6) from November 2004 to December
2008……………………………………………………………………………………………………...30
Table 3.6. Summary of treatment performance during the low loading rate period for filters 1-6……...31
Table 3.7. Summary of treatment performance during the high loading rate period for filters 1-6…......32
Table 3.8. Summary of p-values comparing low to high loading performance results for filters 1-6…..33
Table 3.9. Statistical summary comparison between SSF2, a regular length filter during the low loading
(~100 L/day) period (January 2005-November 2006) and shortened filters, SSF7-8, from November
2007 to December 2008………………………………………………………………………………….34
Table 3.10. Summary of p-values comparing performance results from SSF2 from January 2005 to
November 2006 to SSF7 to SSF8 from November 2007 to December 2008……………………………35
iv Sloping Sand Filters for On‐Site Wastewater Treatment
2009 LIST OF FIGURES
Figure 1.1. Schematic of a typical Lateral Flow Sand Filter used in Nova Scotia……………………….2
Figure 2.1. Layout of the SSF field experimental facility at the Bio-environmental Engineering Centre
in Truro, NS. The site consists of six 8 m long SSFs (Filters 1-6) and two 5.5 m SSFs (Filters 7 and 8).
Filters 7 and 8 were constructed with a medium grained sand and are therefore comparable Filter
2……………………………………………………………………………………………………………5
Figure 3.1. Total Precipitation and Daily Flow for SSF1 (a), SSF2 (b) & SSF3 (c) from February 2006 to
December 2008………………………………………………………………………………………..…10
Figure 3.2. Total Precipitation and Daily Flow from SSF4 (a), SSF5 (b) & SSF6 (c) from February 2006
to December 2008……………………………………………………………………………………..…11
Figure 3.3. Total Precipitation and Daily Flow for SSF7 (a) & SSF8 (b) from August 2007 to December
2008………………………………………………………………………………………………….…..12
Figure 3.4. Total Precipitation and Daily Flow for SSF1 (a), SSF2 (b) & SSF3 (c) from July 15, 2008 to
September 30, 2008…………………………………………………………………………………...…13
Figure 3.5. Total Precipitation and Daily Flow for SSF4 (a), SSF5 (b) & SSF6 (c) from July 15, 2008 to
September 30, 2008…………………………………………………………………………………..….14
Figure 3.6. Total Precipitation and Daily Flow for SSF7 (a) & SSF8 (b) from July 15, 2008 to
September 30, 2008………………………………………………………………………………...……15
Figure 3.7. Residence times from tracer studies from 2005-2008 on SSF1-8……………………...……17
Figure 3.8. Rhodamine tracer study concentration series for SSF1-3 from summer 2007 (a) and Flow
and precipitation data during tracer study, June 11-19, 2007 (b)……………………………………..…19
Figure 3.9. Rhodamine tracer study concentration series for SSF4-6 during summer 2007 (a) and Flow
and precipitation data during tracer study, June 11-19, 2007 (b)…………………………………..……20
Figure 3.10. Rhodamine tracer study concentration series for SSF1-3 during summer 2008 (a) and Flow
and precipitation data during tracer study, July 3-11, 2008 (b)…………………………………….…....21
Figure 3.11. Rhodamine tracer study concentration series for SSF4-6 during summer 2008 (a) and Flow
and precipitation data during tracer study, July 3-11, 2008 (b)……………………………………….…22
Figure 3.12. Rhodamine tracer study concentration series for SSF7-8 during summer 2008 (a) and Flow
and precipitation data during tracer study, July 3-11, 2008 (b)………………………………………….23
Figure 3.13. Rhodamine tracer study concentration series for SSF7-8 from summer 2008 with adjusted
tm…………………………………………………………………………………….…………………...24
Figure 3.14. Rhodamine tracer study concentration series for SSF1-3 during autumn 2008 (a) and Flow
and precipitation data during tracer study, October 14-22, 2008 (b)…………………………………….25
Figure 3.15. Rhodamine tracer study concentration series for SSF4-6 during autumn 2008 (a) and Flow
and precipitation data during tracer study, October 14-22 2008 (b)……………………………………..26
v Sloping Sand Filters for On‐Site Wastewater Treatment
2009 LIST OF FIGURES (cont’d)
Figure 3.16. Rhodamine tracer study concentration series for SSF7-8 from autumn 2008 (a) and Flow
and precipitation data from October 14-22, 2008 (b)……………………………………………………27
Figure 3.17. TSS inlet concentrations from November 2004 to December 2008…………………….….36
Figure 3.18. BOD5 inlet concentrations from November 2004 to December 2008……………….……..36
Figure 3.19. Outlet TSS concentrations for SSF1-3 (a) and SSF4-6 (b) from November 2004 to
December 2008 and for SSF7-8 (c) from October 2007 to December 2008…………..………………...37
Figure 3.20. Outlet BOD5 concentrations for SSF1-3 (a) and SSF4-6 (b) from November 2004 to
December 2008 and SSF7-8 (c) from October 2007 to December 2008…………..……………...…….38
Figure 3.21. Inlet TC and E. coli concentrations from November 2004 to December 2008 in CFU/100
mL…………………………………………………………………………………………………...…...39
Figure 3.22. Outlet TC concentrations (a) and E. coli concentrations (b) for SSF1-3 from November
2004 to December 2008 in CFU/100 ……………………………………………………………..……..40
Figure 3.23. Outlet TC concentrations (a) and E. coli concentrations (b) for SSF4-6 from November
2004 to December 2008 in CFU/100 mL..………………………………………………………………41
Figure 3.24. Outlet TC concentrations (a) and E. coli concentrations (b) for SSF7-8 from October 2007
to December 2008 in CFU/100 mL……...………………………………………………………………42
Figure 3.25. Removal % of TP for SSF1-3 (a) and SSF4-6 from November 2004 to December 2008 and
SSF7-8 (c) from October 2007 to December 2008………………………………………………………43
vi Sloping Sand Filters for On‐Site Wastewater Treatment
2009 1.0 BACKGROUND AND PROJECT OBJECTIVES
Approximately 50% of the population of Nova Scotia relies on an on-site system for wastewater
treatment and disposal (Check 1992). A conventional system in Nova Scotia consists of a septic tank
and absorption trench, constructed within native soil materials. Many of these systems are
malfunctioning, resulting in contamination of surface waters, groundwater, shellfish areas and
recreational areas. A septic system malfunction may include failure of the absorption field caused by
saturated soil conditions or clogging due to sewage solids, effluent breaking through the ground surface,
and sewage backups into toilets, tubs, or sinks. The associated health risks, environmental effects and
costs to shellfish and tourist industries, property owners and taxpayers are substantial.
System failures can be due to a number of reasons; however, the installation of conventional leaching
bed systems in areas where the soil and geologic conditions are not appropriate is a common occurrence.
The presence of low permeability soils, shallow bedrock and high water tables are the most common site
constraints in Nova Scotia. If sufficient depth of permeable soil is not present there is an increased risk
of system malfunction and contamination of ground or surface water. Improper septic tank maintenance
may also cause system malfunction.
Sand filtration has been used as a method for treating both drinking water and wastewater for many
years. Typically, systems are operated as either single pass or recirculating vertical flow filters
(Kristiansen 1981a; Pell and Nyberg 1989a; Harrison et al. 2000). Various studies have shown that
single pass sand filters are effective in removing organic material, suspended solids, enteric
microorganisms, and ammonia-nitrogen (NH4-N) from domestic wastewater (Brandes 1980; Pell and
Nyberg 1989a; Harrison et al. 2000; Rodgers et al. 2004). In these studies the studied sand filters were
operating under unsaturated flow conditions, thus allowing for almost complete nitrification of all
influent NH4-N to nitrate-nitrogen (NO3-N). Reported treatment efficiencies for phosphorus (P) within
sand filters receiving domestic wastewater are variable, ranging from less than 50% (Pell and Nyberg
1989a) to greater than 80% (Rodgers et al. 2004).
In Nova Scotia, sand filters are approved for use as a remedial technology to replace a failed soil
absorption field described as a malfunction (NSDEL 2005). The design of a typical sand filter currently
used in Nova Scotia is provided in Fig. 1.1. The system is termed a Lateral Flow Sand-Filter (LFSF), or
sloping sand filter (SSF), and is a variation of the commonly reported vertical flow sand filter design.
Septic tank effluent enters a gravel filled distribution trench and flows vertically down into the sand
medium (Fig. 1.1) until it reaches an impermeable base (bedrock or impermeable soil). It then flows
laterally downslope through the sand filter. The length of the distribution trench is based on the design
loading rate, the slope of the filter, and the hydraulic conductivity of the sand. The width of the gravel
trench (length of sand filter in the downslope direction) must be at least 5 m and the areal loading rate
for system must not exceed 33 L/m2.
Although SSF systems have been in use in Nova Scotia for more than 20 years, very little data have
been collected from field systems to evaluate their performance, and to determine if current design
criteria are appropriate. The LFSF concept was studied by Check et al. (1994) using laboratory models.
Within this laboratory study three 0.178 m wide x 5 m long filters were constructed. Each filter was
constructed using a different sand medium, encompassing the range of permeability and sizes
recommended in the Nova Scotia Department of Environment technical guidelines.
1 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Figure 1.1. Schematic of a typical Lateral Flow Sand Filter used in Nova Scotia.
They found that treatment efficiencies for SSF systems were comparable to vertical flow filters. The
SSF models provided sufficient removal of biochemical oxygen demand (BOD), total suspended solids
(TSS), NH4-N and enteric microorganisms. It was observed that P removal was satisfactory at the start
of the experiment and then declined as the experiment progressed. The authors also recommended use
of the coarsest sand size as the treatment performance of this filter was comparable to the two filters
constructed with finer sand materials, and the coarser sand would be less likely to clog.
Although the work conducted by Check et al. (1994) showed that the SSF is capable of providing
adequate treatment of septic effluent, data related to the performance of these systems under field
conditions was not available. Also, the experiments were relatively short-term (6 months) and
conducted under laboratory conditions. The influence of temperature and atmospheric processes
(precipitation/evapotranspiration) could not be examined. The influence of other key design criteria,
such as sand characteristics and slope, on system performance also required further study. To address
these information gaps, an experimental facility was constructed in the summer of 2004 to evaluate the
performance of SSFs under field conditions. The original facility consisted of six pilot scale SSFs
designed in accordance with specifications for SSFs in the Nova Scotia On-site Technical Guidelines.
The systems were constructed adjacent to the Bio-environmental Engineering Centre in Truro, Nova
Scotia, Canada. Three filters, containing media that met specifications for mortar, concrete, and silica
sands, were installed at a slope of 5%. Three others, containing the same sands, were installed at a 30%
slope. These filters differ in the particle size of the sand, with coarse, medium or fine sands providing a
range of hydraulic conductivity.
Each sand filter is 8 m long, 1.2 m wide and consists of a gravel bed at the head to evenly distribute the
effluent across the top of the sand filter, and then a layer of sand tapering from approximately 1 m at the
gravel to 0.45 m at the toe of the filter (Fig. 1.1). The filter bed is covered with 0.6 to 1 m of soil which
is separated from the sand by geotextile material. Piezometers and a sampling well were installed to
provide access for monitoring water levels and sampling effluent in the filter bed (Fig. 1.1). The
sampling well was installed to a depth below the filter bottom to act as a sump for effluent collection.
The sand filters were constructed within open plywood boxes (i.e. the top of the filter was open) lined
2 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 with plastic. Effluent was collected in a 100 mm perforated pipe running the width of the filter at the
downslope end and directed into the outflow sampling hut through a solid 100 mm PVC pipe. The
septic effluent for the sand filters is supplied by a PLC controlled pump to dose the sand filters once per
hour to ensure a loading rate of 100 L d-1 (375 mm mth-1) to each filter. The results from the first year of
monitoring were reported in Havard et al. (2008). Average removal efficiencies for all SSFs met NSE
requirements: biological oxygen demand (>98.5%), total suspended solids (>95.5%), and E. coli (>5.4
log reduction). Phosphorus removal ranged from 98% in the fine sand to 71.2% in the coarse sand filter.
Nitrification was favored because the filters were operating under aerobic and unsaturated conditions.
Therefore, denitrification was limited causing elevated nitrate effluent concentrations. Total nitrogen
removal ranged from 60 to 66 %. The SSF provided consistent year round treatment and did not appear
to be impacted greatly by slope, temperature or external hydrologic influences.
However, the hydraulic assumption on which the design of these systems is based, that flow occurs
through a saturated media at the base of the system, apparently does not apply. Moisture measurements
have indicated that the media in all of these systems is tension saturated, and that lateral flow occurs to
some degree through much of the media depth. Therefore, it is likely that current provincial guidelines
are resulting in over-designed systems.
1.1 Project Objectives
As stated previously, initial monitoring of pilot scale SSFs in the field produced promising results. The
initial results also provided evidence that current design SSFs were conservative, and that more costeffective designs could be developed. In addition, initial results suggested that the treatment efficiency
of certain parameters, specifically phosphorus, could change as the filters mature. The principal
objectives of this research were to:
1. Assess the hydraulic behavior and performance of conventionally designed SSFs at loading rates
greater than those allowed within NSE technical guidelines
2. Assess the hydraulic behavior and performance of SSFs that were substantially shorter lengths than
those required within NSE technical guidelines
3. Determine if current hydraulic models of SSF systems are appropriate
3 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 2.0 METHODOLOGY
2.1 LFSF Construction and Loading Regimes
The research utilized the existing infrastructure at the BEEC. To assess the performance of conventional
SSFs at increased loading rates the hydraulic loading rates applied to the original six SSF systems (Fig.
2.1) was approximately doubled. The flow rates to these filters (Filters 1 -6) was increased in February
of 2007. In order to address Project Objective 2, the project team had originally planned to install
suction lysimeters at varying distances throughout the existing SSF systems. This was attempted during
the summer of 2007, however, due to the low moisture contents, and coarse grained nature of the sand
material within the filters, it was not possible to obtain a representative water quality samples using
suction lysimeters. To address this issue the project team constructed an additional two SSF systems
(Filters 7 and 8) which possessed sand toes which were reduced to 3 m in length (Fig. 2.1). The effluent
leaving the filters was collected in the same manner as for the original six SSFs and directed to the
heated sampling hut. The two new shortened SSFs were both constructed on a 5% slope with a medium
grained sand (Table 2.1). Therefore the two new filters were identical to Filter 2, except they possessed
a shorter sand toe. The two new filters were loaded at the recommended rate of 100 L d-1. Construction
of these filters was completed in July, 2007 and they began receiving wastewater in August, 2007.
Table 2.1. Physical characteristics of the filter sands.
Sand Type
d10
(mm)
Uniformity Coefficient
(d60/d10)
Hydraulic Conductivity
(m s-1)
Fine
0.15
8
1.5 x 10-4
Medium
0.17
5.6
5.0 x 10-4
Coarse
0.30
1.8
1.0 x 10-3
2.2 Hydrologic Monitoring
Flow data from each sand filter was measured using calibrated tipping buckets and recorded every 10
minutes on a CR510 data logger (Campbell Scientific, Edmonton, AB). Meteorological conditions
(precipitation, solar radiation, air temperature, and relative humidity) were continuously monitored and
recorded using a Campbell Scientific CR10 datalogger (Edmonton, AB). The data were sampled every
60 s and 60 min averages were recorded. Precipitation was measured using a heated rain gauge
(Campbell Scientific, Edmonton, AB). Temperatures within the filters were measured using copper
constantan thermocouples.
4 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Figure 2.1. Layout of the SSF field experimental facility at the Bio-environmental Engineering Centre
in Truro, NS. The site consists of six 8 m long SSFs (Filters 1-6) and two 5.5 m SSFs (Filters 7 and 8).
Filters 7 and 8 were constructed with a medium grained sand and are therefore comparable Filter 2.
5 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 2.3 Hydraulic Assessment and Modeling
A series of tracer studies were conducted within the 8 filters several times during the course of the study.
Rhodamine tracer studies were conducted within the eight SSFs a total of 4 times and included both
warm and cool season climatic conditions. Within each tracer experiment 10 mL of rhodamine dye (20
wt/wt) was injected into the inlet pipe of each filter. Grab samples were collected at the outlet of each
filter at time intervals ranging from 2 – 48 h over a 28 day period. Concentrations of rhodamine dye
within effluent samples were determined on a UV/VIS spectrophotometer. Data obtained from the
rhodamine tracer studies were fit to analytical models which were used to determine the mean
residence time, and longitudinal dispersion coefficient, for each sand filter (Fogler 1992). The function
E(t) represents the residence time distribution (RTD) function which describes the amount of time that a
particular fluid element spends in the system (Fogler 1992):
(1)
C (t )
E (t ) =
∫ C (t )dt
Where C(t) represents the effluent tracer concentration at time, t, and the integral in the denominator is
the area under the C(t) curve. The mean residence time, tm, can then be calculated by taking the first
moment of the RTD function, as follows:
(2)
t m = ∫ tE (t ) dt
Variance (σ2), or the square of the standard deviation, was then calculated to obtain a measure of the
degree of “spread” of the data distribution. Variance was determined by taking the second moment
about the mean residence time (Fogler 1992):
(3)
σ 2 = ∫ (t − t m ) E (t ) dt
Variance was calculated for each sand filter and was then used to calculate a dispersion coefficient.
Variance as calculated above is in units of time; this must be converted to variance in space before
calculating dispersion (Apello and Postma 1993):
(4)
σ x2 =
σ 2x2
tm
2
where x is the filter length (8 or 5.5 m). Variance in space was then used to calculate a longitudinal
dispersion coefficient for each sand filter (Apello and Postma1993):
(5)
DL =
σ x2
2t m
6 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 The dispersivity, α L, for each sand filter was then computed as:
(6)
αL =
DL
v
where; v is velocity (filter length /mean residence time). The theoretical time of travel within a SSF,
assuming saturated flow can be modeled using Darcy’s Law. The true horizontal velocity, vh, can be
computed as:
(7)
vh =
K dH
n dL
Where K is the hydraulic conductivity of the sand (m/d), n is the porosity of the sand, and dH/dL is the
hydraulic gradient within the sand filter. When designing a SSF it is assumed that the hydraulic gradient
is equal to the slope of the filter. The porosity of the sand was estimated to be 0.30 (Freeze and Cherry,
1979). The average theoretical travel time, tT of a conservative particle within the filter could then be
computed as:
(8)
tT =
dL
vh
Water levels in mini-piezometers installed within both the distribution trench and filter sand in all filters
were also measured on a monthly basis to assess level of saturation.
2.4 Performance Monitoring
The influent wastewater, and effluent from each filter, was sampled on monthly basis for the duration of
the study. Filters 7 and 8 were allowed to mature for 2 months before sampling was initiated. During
each monthly sampling event autosamplers were deployed for a 24 h period to collect a composite
sample of influent wastewater. Two ISCO 6712 autosamplers (ISCO Inc., Montreal, PQ), with
multiplexers, were used to collect time weighted (100 mL/h) composite samples over a 24 h period from
the outlet of each filter system. These composite samples were analyzed at the Nova Scotia Agricultural
College’s Department of Environmental Sciences, Water Quality Research Laboratory. Analysis
included total suspended solids (TSS) (Std. Method 2540D), five-day biological demand (BOD5) (Std.
Method 5210B), total phosphorus (TP) (Std. Method 4500-P, Phosphorus - 1999 revision and ascorbic
acid method), total kjeldahl nitrogen (TKN) (Std. Method 4500-Norg B) and NH4-N (Std. Method 4500NH3). The anions phosphate-P (SRP) and NO3-N were quantitated according to Std. Method 4110
(2000 version). All procedures are outlined in Standard Methods for the Examination of Water and
Wastewater (Clesceri et al. 1998). Individual grab samples of the influent wastewater and effluent from
each filter were collected and analyzed for Escherichia coli using the membrane filtration technique in
conjunction with m-Coliblue24 culture media (HACH 1999). The performance of the eight filters was
evaluated in several manners. Percent removal, on a concentration basis, was computed for each water
quality variable for each month. Total influent and effluent nitrogen concentrations were determined on
a monthly basis by adding the TKN and NO3-N concentrations. T-tests were used to assess differences
in treatment performance.
7 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 3.0 RESULTS AND DISCUSSION
3.1 Flow Characteristics
Hydrology plays a significant role in the performance of SSFs. To evaluate this influence, flow
characteristics such as daily flow from the low and high loading rate periods, the relationship between
external hydrological events, and seasonal trends were examined. Filters 1 through 6 were installed in
the summer of 2004 and began receiving wastewater in September 2004. Each filter’s daily loadin rate
capacity was to be 100 L/day, as determined by the current provincial guidelines. Preliminary findings
indicated little to no ponding of water in the filters, an indication of saturation, so daily wastewater
loading was increased to 200 L/day in January 2007. It took approximately three months to calibrate
loading rates and address for pump failures and mechanical problems. Shorter filters 7 and 8 were
constructed in July 2007 and started receiving 100 L/day in August 2007.
Table 3.1 provides average daily outflow for SSF1-6. From February 2006 to December 2006, SSF1-5
averaged approximately 100 L/day and SSF6 averaged 55.6 L/day. Flow ranges from 0 L/day to 380.8
L/day. Values of 0 L/day were recorded during periods of pump failures. High loading rate data, taken
from April 2007 to December 2008, ranges from 0 L/day to 1092.7 L/day. Average daily flows are
under 200 L/day. The range in flows is indicative of real events where the SSF would be in high usage
and/or during heavy rainfall events and spring thaws. Since their operation in August 2007, SSF7-8
achieved daily average outflows of 90.4 L/day and 65.9 L/day, shown in Table 3.2. Filter 8 underwent
some clogging issues with the flow adapter which explains the lower daily outflow average. Flows range
from 672.6 L/day to 0 L/day.
External hydrological events such as rainfall and snowfall (total precipitation) have an impact on flow.
The relationship is a multi-variable one; flow is a function of the history of precipitation events and
antecedent moisture conditions. When the filters are more unsaturated, response to external hydrological
events will not be as great. The filters are not steady state hydrological systems and are subject to
variable conditions. Figures 3.1, 3.2, and 3.3 represent this graphically where peaks in total
precipitation correspond with peaks in daily flow from the filters.
To better illustrate how the pre-existing moisture status of the soil and precipitation events influence the
flow responses of the filters, the period of mid-July 2008 to the end of September 2008 was examined
(Figures 3.4 – 3.6). There were few precipitation events occurring during the last two weeks of July; all
were comparatively small. As such, the filter soil was dry. Subsequently the flow responses occurring
after the precipitation events in early August are dampened. However, precipitation events in early
September trigger a greater response in the filters due to moisture in the soil from early August.
Water level measurements conducted in Filters 1-6 after the loading rate was increased indicated that
ponding of water in the distribution trench was occurring, however no water was detected in any
piezometers located within the sand bed. The hydraulic characteristics of the biomat appeared stable, as
ponded water levels did not increase significantly during the experiment.
8 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Table 3.1. Descriptive Statistics for SSF1-6, including Average Daily Flow, Standard Deviation,
Minimum and Maximum Flows.
Parameters
SSF1
Average
Daily Flow
(ADF)
Standard
Deviation
(STD)
Minimum
(MIN)
Maximum
(MAX)
ADF
STD
MIN
MAX
SSF4
SSF5
SSF6
101.7
SSSF2
SSF3
Low Flow (L/day)
99.1
104.7
103.5
101.0
55.6
48.9
50.0
51.6
47.8
52.0
51.4
2.5
1.9
1.1
1.4
0.9
0.0
280.5
311.1
297.9
239.9
380.8
237.4
176.8
81.3
0.0
974.6
High Flow (L/day)
181.9
170.4
76.2
54.5
0.0
0.0
1077.4
549.9
180.4
79.4
0.0
1002.5
175.1
96.8
0.0
1092.2
164.3
58.5
2.8
709.7
Table 3.2. Descriptive Statistics for SSF7-8, including Average Daily Flow, Standard Deviation,
Minimum and Maximum.
Parameters
ADF (L/day)
STD (L/day)
MIN (L/day)
MAX (L/day)
SSF7
90.4
59.5
1.2
672.6
SSF8
65.9
44.5
0.0
623.7
9 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) Flow (L/day)
1400
10
1200
20
1000
30
800
600
40
400
50
200
0
Feb-06
Rainfall (mm/day)
0
1600
60
Aug-06
Mar-07
Sep-07
Apr-08
Nov-08
Date
Total Precipitation (mm)
SSF1
(b) 1600
10
Flow (L/day)
1200
20
1000
800
30
600
40
400
50
200
0
Feb-06
Rainfall (mm/day)
0
1400
60
Aug-06
Mar-07
Sep-07
Apr-08
Oct-08
Date
Total Precipitation (mm)
SSF2
(c) Flow (L/day)
1400
10
1200
20
1000
30
800
600
40
400
50
200
0
Feb-06
Rainfall (mm/day)
0
1600
60
Aug-06
Mar-07
Sep-07
Apr-08
Nov-08
Date
Total Precipitation (mm)
SSF3
Figure 3.1. Total Precipitation and Daily Flow for SSF1 (a), SSF2 (b) & SSF3 (c) from February 2006 to
December 2008
10 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) 1600
Flow (L/day)
10
1200
20
1000
800
30
600
40
400
50
200
0
Feb-06
Rainfall (mm/day)
0
1400
60
Aug-06
Mar-07
Sep-07
Apr-08
Nov-08
Date
Total Precipitation (mm)
SSF4
1200
0
1000
10
800
20
600
30
400
40
200
50
0
Feb-06
Rainfall (mm/day)
Flow (L/day)
(b) 60
Aug-06
Mar-07
Sep-07
Apr-08
Nov-08
Date
Total Precipitation (mm)
SSF5
(c) 1600
0
1400
10
Flow (L/day)
1200
20
1000
800
30
600
40
400
50
200
0
Feb-06
60
Aug-06
Mar-07
Sep-07
Apr-08
Nov-08
Date
Total Precipitation (mm)
SSF6
Figure 3.2. Total Precipitation and Daily Flow from SSF4 (a), SSF5 (b) & SSF6 (c) from February 2006 to
December 2008
11 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) 800
700
10
Flow (L/day)
600
20
500
400
30
300
40
200
50
100
0
Aug-07 Sep-07 Nov-07 Jan-08 Feb-08 Apr-08 Jun-08 Jul-08 Sep-08 Nov-08
Rainfall (mm/day)
0
60
Date
Total Precipitation (mm)
SSF7
(b) 800
700
10
Flow (L/day)
600
20
500
400
30
300
40
200
50
100
0
Aug-07 Sep-07 Nov-07 Jan-08 Feb-08 Apr-08 Jun-08 Jul-08 Sep-08 Nov-08
Rainfall (mm/day)
0
60
Date
Total Precipitation (mm)
SSF8
Figure 3.3. Total Precipitation and Daily Flow for SSF7 (a) & SSF8 (b) from August 2007 to December
2008
12 (a) 1200
Flow (L/day)
0
5
10
15
20
25
30
35
40
45
1000
800
600
400
200
0
15-Jul
4-Aug
24-Aug
Date
2009 Rainfall (mm/day)
Sloping Sand Filters for On‐Site Wastewater Treatment
13-Sep
Total Precipitation (mm)
SSF1
1200
0
1000
5
10
15
20
800
600
25
30
400
35
40
200
0
15-Jul
Rainfall (mm/day)
Flow (L/day)
(b) 45
25-Jul
4-Aug
14-Aug
24-Aug
3-Sep
13-Sep
23-Sep
Date
Total Precipitation (mm)
SSF2
1200
0
1000
5
10
800
15
20
600
25
30
400
35
40
200
0
15-Jul
Rainfall (mm/day)
Flow (L/day)
(c) 45
25-Jul
4-Aug
14-Aug
24-Aug
3-Sep
13-Sep
23-Sep
Date
Total Precipitation (mm)
SSF3
Figure 3.4. Total Precipitation and Daily Flow for SSF1 (a), SSF2 (b) & SSF3 (c) from July 15, 2008 to
September 30, 2008
13 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Flow (L/day)
(a)
1200
0
1000
5
10
15
20
800
600
25
30
400
35
40
200
0
15-Jul
45
25-Jul
4-Aug
14-Aug
24-Aug
3-Sep
13-Sep
23-Sep
Date
Total Precipitation (mm)
SSF4
1200
0
1000
5
10
800
15
20
600
25
30
400
35
40
200
0
15-Jul
Rainfall (mm/day)
Flow (L/day)
(b) 45
25-Jul
4-Aug
14-Aug
24-Aug
3-Sep
13-Sep
23-Sep
Date
Total Precipitation (mm)
SSF5
1200
0
1000
5
10
800
15
20
600
25
30
400
35
40
200
0
2-Aug-08
Rainfall (mm/day)
Flow (L/day)
(c) 45
12-Aug-08
22-Aug-08
1-Sep-08
11-Sep-08
21-Sep-08
Date
Total Precipitation (mm)
SSF6
Figure 3.5. Total Precipitation and Daily Flow for SSF4 (a), SSF5 (b) & SSF6 (c) from July 15, 2008 to
September 30, 2008
14 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 1200
0
1000
5
10
15
20
800
600
25
30
400
35
40
200
0
15-Jul
Rainfall (mm/day)
Flow (L/day)
(a) 45
25-Jul
4-Aug
14-Aug
24-Aug
3-Sep
13-Sep
23-Sep
Date
Total Precipitation (mm)
SSF7
1200
0
1000
5
10
15
20
800
600
25
30
400
35
40
200
0
15-Jul
Rainfall (mm/day)
Flow (L/day)
(b) 45
25-Jul
4-Aug
14-Aug
24-Aug
3-Sep
13-Sep
23-Sep
Date
Total Precipitation (mm)
SSF8
Figure 3.6. Total Precipitation and Daily Flow for SSF7 (a) & SSF8 (b) from July 15, 2008 to September 30,
2008.
15 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 3.2 Hydraulic Assessment and Modeling
Tracer studies were conducted within each filter to determine hydraulic characteristics and residence
times within each system. These results explore the relationship between the hydraulic responses of
individual filters and variables including filter length, slope, grain size, biomat development, loading
rate and precipitation. A comparison of previous tm results from 2005 to subsequent tracer studies
performed in 2007 and 2008 are shown in Table 3.3.
Table 3.3. Hydraulic parameters generated from tracer studies on SSF1-8.
SSF1
SSF2
SSF3
SSF4
SSF5
SSF6
SSF7
SSF8
Slope
5%
5%
5%
30%
30%
30%
5%
5%
Grain Size
Fine
Medium
Coarse
Fine
Medium Coarse Medium Medium
Summer 2005
tm (hrs)
150.2
97.5
91.4
153.5 94.4
62.0
-
-
σ (hours)
104.2
97.6
86.4
94.6
80.9
66.5
‐ ‐ D (cm2/s)
0.29
0.91
0.87
0.22
0.69
1.65
‐ ‐ Summer 2007
tm (hrs)
138.8
51.8
106.2
99.2
66.4
48.2
-
-
σ (hours)
52.1
258.9
43.0
27.7
32.4
31.9
-
-
D (cm2/s)
0.090
42.9
0.137
0.070 0.319
0.805
-
-
Summer 2008
tm (hrs)
187.6
76.6
66.7
115.1 90.8
64.3
210.5
91.3
σ (hours)
49.1
223.5
35.5
37.4
22.1
52.3
44.8
D (cm2/s)
0.033
9.903
0.377
0.082 0.090
0.163
0.012
0.111
27.6
Autumn 2008
tm (hrs)
201.4
65.2
55.0
116.4 110.4
57.0
56.5
81.8
σ (hours)
57.2
218.5
26.4
49.3
25.4
29.1
23.5
D (cm2/s)
0.036
15.3
0.373
0.137 0.151
0.309
0.196
0.040
47.4
16 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Mean residence times vary between filters, ranging from 48.2 hours (SSF6, summer 2007) to 210.5
hours (SSF7, summer 2008). Temporal changes in tm are shown in Figure 3.7. Generally, tm decreases
from 2005 to 2007, possibly due to the doubling of the loading rate from 100 L/day to 200 L/day. The
exception to this is SSF3; however the increase in tm is small, from 91 hours to 106 hours. Fine grained
filters have higher residence times compared to medium and coarse grained filters, presumably due to
the smaller pore space for the wastewater to travel through. Residence times from medium and coarse
grained filters are comparable; SSF6 has the fastest tm, which is due to its coarse grain size and 30%
slope.
Residence Time (hours)
250.0
200.0
150.0
100.0
50.0
0.0
SSF1
SSF2
SSF3
SSF4
SSF5
SSF6
SSF7
SSF8
Filters
Autumn 2005
Summer 2007
Summer 2008
Autumn 2008
Figure 3.7. Residence times from tracer studies from 2005-2008 on SSF1-8.
Filter 2 appeared to be short circuiting during the tracer study in summer 2007 because tm decreases by
half. Figure 3.8 (a) shows SSF2 reaching a concentration peak earlier than SSF1, which is unusual
because SSF3 is coarse grained and should therefore should peak first. An analysis of the flow and
precipitation data in Figure 3.8 (b) indicates that excess flow and precipitation to SSF2 were not the
cause of this occurrence. Therefore, the short circuiting in SSF2 may be due to preferential flow patterns
within the filter during that tracer study. Subsequent tracer studies show that this was an isolated
incident as SSF2 does not short circuit again.
From Figure 3.9 (a) we note that SSF5-6 achieve higher concentration peaks than SSF2-3 (Figure 3.8
(a)). This may be due to the difference in slopes. However, the magnitude of the concentration peak for
SSF4 is comparable to SSF1. Both filters are fine grained, leaving less pore space for the wastewater to
travel through. Another cause for lower concentration peaks and greater residence times for the fine
grained filters (SSF1, SSF4) may be due to the development of the biomat, which is known to control
the rate of infiltration of wastewater into the filters (Crites & Tchobanoglous, 1998). From this we can
infer that slope does not play as significant a role as filter grain size in hydraulic characteristics of SSFs.
The concentration peaks are lower in summer 2008 than in summer 2007, shown in Figure 3.10 (a).
This may be due to the lack of precipitation events that occurred during this study, as seen in Figure 4
17 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (b). Note that SSF2 is not short circuiting anymore as it peaks in between SSF1 and SSF3. These filters
behave as expected.
Filters 5 and 6 achieved lower concentration peaks than in previous studies (Figure 3.11 (a)). This may
be due to the lack of precipitation events that occurred during this study. Concentration peaks for SSF6
are comparable to previous studies. It was expected that results from the first Rhodamine tracer study
from SSF7-8 would be similar. However, SSF8 reaches a higher concentration peak earlier than SSF7,
shown in Figure 3.12 (a). Filter 7 has a tm of over 210.5 hours compared to that of SSF8 at 64.3 hours.
Corresponding flow and precipitation data shown in Figure 3.12 (b) during this period indicate the filter
received little to no wastewater loading during the first 150 hours of the study. This is due to clogging
of the orifice plate of the filter. Notably, at approximately the 210 hour point, SSF7 begins receiving
more flow and SSF8 receives none. This indicates a problem with the proper distribution of flow
between the filters. An adjusted tm from this tracer study for SSF7 was calculated using the start time of
the tracer study at t = 150 hours. The adjusted tm is 61.5 hours (Figure 3.13). After the adjustment is
made for tm for SSF7 in summer 2008, in general, mean residence times from SSF7-8 are comparable to
those of SSF5, a medium-grained filter located on a 30% slope. This suggests that slope and filter
length play lesser roles in the hydraulic characteristics of SSFs.
Results are as expected for SSF1-6 for the Rhodamine tracer study conducted in autumn 2008 (Figures
3.14-3.15). There is more rainfall than in previous studies, and variations in flow, which produced little
influence on the hydraulic response of the filters. Flow data from Figure 3.15 (b) shows that SSF4
received over 200 L/day for the first 250 hours of the study, yet there is little impact on the residence
time or the concentration peak. Notably, SSF1 has a very low concentration peak and the greatest tm of
any previous tracer study. Therefore, we can infer that SSF1-6 have well established biomats,
specifically SSF1 and SSF4 (both fine grained filters). External precipitation events and variations in
flow do not appear to influence hydraulic response from filters with well-established biomats. In
comparison, results from the autumn 2008 tracer studies on SSF7-8, filters with less established biomats,
showed the highest magnitude of concentration peaks in any tracer study to date (Figure 3.16 (a)). Flow
and precipitation data shown in Figure 3.16 (b) indicate that SSF8 received reduced flows during the
first hours of the study. This is due to clogging and can explain the slight delay in concentration peak
and greater tm compared to SSF7.
18 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) (b) Figure 3.8. Rhodamine tracer study concentration series for SSF1-3 from summer 2007 (a) and Flow and
precipitation data during tracer study, June 11-19, 2007 (b).
19 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) (b) Figure 7.9. Rhodamine tracer study concentration series for SSF4-6 during summer 2007 (a) and Flow and
precipitation data during tracer study, June 11-19, 2007 (b).
20 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) (b) Figure 3.10. Rhodamine tracer study concentration series for SSF1-3 during summer 2008 (a) and Flow
and precipitation data during tracer study, July 3-11, 2008 (b).
21 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) (b) Figure 3.11. Rhodamine tracer study concentration series for SSF4-6 during summer 2008 (a) and Flow
and precipitation data during tracer study, July 3-11, 2008 (b).
22 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) (b) Figure 3.12. Rhodamine tracer study concentration series for SSF7-8 during summer 2008 (a) and Flow
and precipitation data during tracer study, July 3-11, 2008 (b).
23 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Figure 3.13. Rhodamine tracer study concentration series for SSF7-8 from summer 2008 with adjusted tm.
24 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) (b) Figure 3.14. Rhodamine tracer study concentration series for SSF1-3 during autumn 2008 (a) and Flow and
precipitation data during tracer study, October 14-22, 2008 (b).
25 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) (b) Figure 3.15. Rhodamine tracer study concentration series for SSF4-6 during autumn 2008 (a) and Flow and
precipitation data during tracer study, October 14-22 2008 (b).
26 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) (b) Figure 3.16. Rhodamine tracer study concentration series for SSF7-8 from autumn 2008 (a) and Flow and
precipitation data from October 14-22, 2008 (b).
27 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 From these tracer studies, it has been shown that a number of factors influence the hydraulic
characteristics of individual filters. There is variability in residence times between each filter. Through
the course of four tracer studies conducted from 2005 to 2008, there appears to be no difference in
response when wastewater loading was doubled. Grain size has an influence; fine grained filters had
increased residence times which also may be due to greater biomat development. However, there was
very little difference between medium grained and coarse grained filters. Filters 7 and 8, which do not
have as well established a biomat as others, saw higher concentration peaks than other filters, which may
be due to the short length of the filter and a less mature biomat. However, residence times from these
filters, composed of medium grained sand and located on a 5% slope, are comparable to SSF5, a
medium-grained filter located on a 30% slope. This suggests that filter slope and length does not play a
large role in hydraulic behavior of filters. Precipitation played an insignificant role in filter hydraulic
response because there were few rainfall events that occurred during the studies (the greatest rainfall
event was 17.2 mm during the last hours of the summer 2008 study).
Theoretical hydraulic retention times were simulated using Darcy’s Law and are provided in Table 3.4.
The theoretical retention times were much smaller than those measured in the tracer studies (Table 3.3).
This result, in combination with the water level measurements, confirm that saturated flow is not
occurring with the LFSF systems. The larger measured retention times indicate that the biomat may be
controlling residence time characteristics within the filters. The fact that a saturated zone does not exist
at the bottom of the filters also suggests unsaturated flow may be occurring within a larger component of
the sand than previously assumed.
Table 3.4. Theoretical residence times (tT, hours) for each LFSF.
SSF1
SSF2
SSF3
SSF4
SSF5
SSF6
SSF7
SSF8
Slope
5%
5%
5%
30%
30%
30%
5%
5%
Grain Size
Fine
Medium
Coarse
Fine
90
27
13
15
tT (hrs)
Medium Coarse Medium Medium
5
2.5
13
13
28 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 3.3 Treatment Performance
3.3.1 Statistical Summary: The treatment performance of key parameters such as E. coli, TC, BOD5,
TSS and nutrients over time provided essential information about the long term performance of SSFs.
In this study, primary comparisons were made between low and high wastewater loading rate periods,
and between regular length and shortened filters. As well, the influence of external hydrological factors
on treatment performance was also assessed. The low loading rate (~100 L/day) period extends from
January 2005 to November 2006 for filters 1-6. The filters were given three months to acclimate to the
doubling of the loading rate. The high loading rate period occurs from March 2007 to December 2008.
In Table 3.5, a statistical summary of the average monthly influent and effluent concentrations for all
filters from 2004 to 2008 is presented. In general the six original filters performed well over this four
year period, producing average effluent BOD5 concentrations below 10 mg/L, and average TSS
concentrations below 15 mg/L. Some variability in effluent concentrations of BOD5 and TSS were
observed, as evidenced by the maximum concentrations shown in Table 3.5. As will be shown later,
these spikes in BOD5 and TSS can be attributed to excessively high solids and organic loading to the
filters which occurred in the fall of 2006. This was due to solids accumulation in the septic tank, which
was remediated in January, 2007. Median E. coli levels were all below 10 CFU / 100 mL, indicating
that the sand filters were generally effective at removing enteric bacteria. However, on several
occasions levels of E. coli in filter effluent exceeded 100 CFU / 100 mL, illustrating variability in
treatment performance.
Average removal efficiencies and effluent concentrations for each of the six filters for the low and high
loading rate periods are provided in Table 3.6 and 3.7, respectively. A summary of statistical testing,
comparing effluent concentrations from low and high loading rate periods for each water quality
parameter in each filter is provided in Table 3.8. Results from t-tests comparing low to high loading rate
performance results indicate that there were no difference in treatment of TN, NO3-N and TSS. Other
parameters have mixed results. With respect to BOD5, the results indicated that effluent concentrations
were significantly higher during the low loading rate period. However, this is largely attributable to the
elevated organic loading observed during the end of this period. With these values removed there is no
significant difference in effluent BOD5 concentrations between the high and low loading rate periods.
Two of the filters were statistical different with respect to E. coli effluent concentrations, but these
differences were small. The statistical test results for TP show that TP concentrations were statistically
higher in filters 1, 4, and 4 during the high loading rate period. However, these results should be viewed
with caution as the TP removal efficiency of the sand filters is progressively reduced with time due to
saturation of adsorption sites. In general the increase in loading rate did not appear to affect the
treatment efficiency of the SSFs.
29 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Table 3.5. Statistical summary of filter performance (SSF1-6) from November 2004 to December 2008
Parameter
E. coli
(CFU/100
mL)
Influent
2.2E+06
1.6E+06
1.7E+07
1.0E+04
2.8E+06
1
6.9E+01
3.0E+00
1.2E+03
0.0E+00
2.3E+02
2
4.5E+01
7.0E+00
6.0E+02
0.0E+00
1.1E+02
3
2.5E+01
3.5E+00
3.0E+02
0.0E+00
6.5E+01
4
1.1E+01
2.0E+00
3.0E+02
0.0E+00
4.3E+01
5
1.9E+01
2.0E+00
3.0E+02
0.0E+00
6.0E+01
6
5.8E+01
4.0E+00
9.0E+02
0.0E+00
1.7E+02
Mean
Median
Maximum
Minimum
Stdev
TSS
(mg/L)
522.3
52.4
5200.0
6.5
1198.6
14.3
4.8
114.0
2.0
21.5
10.4
3.2
95.0
2.0
18.4
9.2
5.3
59.8
2.0
11.2
7.7
2.6
30.6
2.0
7.7
5.5
2.2
51.6
2.0
7.8
7.1
2.0
78.4
2.0
13.7
Mean
Median
Maximum
Minimum
Stdev
BOD5
(mg/L)
278.0
95.7
1878.0
29.6
457.6
2.7
2.4
5.2
2.4
0.7
2.9
2.4
10.7
2.4
1.3
3.6
2.4
35.7
2.1
4.9
2.7
2.4
5.0
2.4
0.6
2.7
2.4
4.7
2.4
0.7
3.0
2.4
7.9
2.4
1.2
Mean
Median
Maximum
Minimum
Stdev
TP (mg/L)
8.3
3.1
55.8
0.8
10.6
0.9
0.7
3.8
0.1
0.9
1.3
1.0
5.2
0.1
1.0
2.4
2.2
6.2
0.1
1.3
1.0
0.7
2.9
0.1
0.8
1.1
1.0
3.0
0.1
0.7
2.3
2.2
5.1
0.1
1.1
Mean
Median
Maximum
Minimum
Stdev
NH4-N
(mg/L)
20.1
17.5
69.5
1.7
13.1
0.4
0.1
2.6
0.1
0.6
0.3
0.1
2.0
0.1
0.4
0.3
0.1
1.8
0.1
0.4
0.2
0.1
1.0
0.1
0.2
0.2
0.1
2.2
0.1
0.4
0.4
0.1
4.8
0.1
0.8
Mean
Median
Maximum
Minimum
Stdev
NO3-N
(mg/L)
0.9
0.2
20.0
0.0
3.0
14.9
14.4
48.8
0.3
7.9
16.4
15.0
45.6
5.5
7.7
16.5
15.4
37.2
4.2
7.7
15.6
14.1
33.9
3.8
7.2
14.5
13.0
38.7
0.2
8.0
14.5
13.1
32.8
3.0
7.1
43.1
29.3
236.8
8.4
41.7
16.8
51.5
2.2
2.2
8.8
17.9
16.2
48.3
6.6
8.3
18.7
16.4
46.6
5.9
9.5
17.0
15.5
34.4
3.9
15.5
16.0
14.0
40.3
0.6
8.6
15.5
14.0
33.3
2.1
8.4
Mean
Median
Maximum
Minimum
Stdev
Mean TN (mg/L)
Median
Maximum
Minimum
Stdev
30 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Table 3.6. Summary of treatment performance during the low loading rate period for filters 1-6.
Filter
Parameter
Influent
Fine
Medium
Coarse
Fine
Medium
Coarse
(5%)
(5%)
(5%)
(30%)
(30%)
(30%)
SSF1
SSF2
SSF3
SSF4
SSF5
SSF6
E. coli
(CFU/100
1.5E+06
Outlet
Log removal
3.2E+00
5.58
4.4E+01
5.05
2.7E+01
5.21
1.5E+01
5.28
2.0E+01
5.29
4.7E+01
5.25
TSS
(mg/L)
910.1
Outlet
% removal
21.0
91.9%
13.2
93.1%
13.4
93.2%
9.9
94.0%
7.9
93.3%
10.0
94.6%
BOD5
(mg/L)
426.8
Outlet
% removal
2.8
97.8%
3.0
97.8%
4.3
97.6%
2.8
97.7%
2.8
97.6%
3.1
97.4%
TP (mg/L)
9.1
Outlet
% removal
0.6
87.0%
1.1
74.9%
2.2
56.9%
0.6
85.8%
0.8
82.7%
2.1
59.1%
NH4-N
(mg/L)
22.4
Outlet
% removal
0.3
97.3%
0.3
97.2%
0.3
97.2%
0.2
98.0%
0.2
97.9%
0.6
96.8%
TKN
(mg/L)
52.8
Outlet
% removal
2.3
93.8%
2.3
93.6%
2.6
92.9%
1.8
94.7%
1.8
95.5%
1.9
95.1%
NO3-N
1.5
Outlet
14.6
16.7
16.0
15.3
14.2
14.0
Outlet
% removal
16.9
55.2%
18.9
47.5%
18.6
48.1%
17.1
50.7%
15.9
56.8%
15.4
54.8%
TN (mg/L) 54.3
31 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Table 3.7. Summary of treatment performance during the high loading rate period for filters 1-6.
Filter
Parameter Influent
Fine
Medium
Coarse
Fine
Medium
Coarse
(5%)
(5%)
(5%)
(30%)
(30%)
(30%)
SSF1
SSF2
SSF3
SSF4
SSF5
SSF6
3.1E+06
E. coli
(CFU/100
Outlet
Log removal
1.5E+02 4.3E+01 2.4E+01 6.6E+00 1.9E+01 5.0E+01
4.87
5.16
5.28
5.59
5.52
5.10
TSS
(mg/L)
65.5
Outlet
% removal
6.9
89.2%
8.1
92.4%
4.3
90.0%
4.5
91.2%
2.8
92.4%
4.3
91.8%
BOD5
(mg/L)
91.2
Outlet
% removal
2.7
95.7%
2.9
95.9%
2.5
96.6%
2.5
96.4%
2.5
96.3%
2.6
96.2%
TP
(mg/L)
4.3
Outlet
% removal
1.3
53.9%
1.6
42.8%
2.8
2.1%
1.5
47.6%
1.4
48.9%
2.6
9.5%
NH4-N
(mg/L)
17.9
Outlet
% removal
0.4
97.2%
0.1
99.2%
0.1
99.1%
0.1
99.2%
0.2
98.8%
0.2
98.9%
TKN
(mg/L)
30.5
Outlet
% removal
1.2
95.5%
0.8
96.7%
0.4
98.4%
1.0
96.1%
0.6
97.8%
1.0
95.9%
NO3-N
0.3
Outlet
15.6
16.2
17.3
15.8
14.7
14.9
TN
(mg/L)
30.7
Outlet
% removal
16.6
40.4%
16.6
40.7%
17.5
36.8%
16.9
39.0%
15.7
43.4%
14.9
45.1%
32 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Table 3.8 Summary of p-values comparing low to high loading rate performance results for filters 1-6.
Filter
E.Coli
BOD5
TSS
SSF1
NS
0.021
SSF2
0.000
NS
NS
SSF3
0.001
0.003
NS
SSF4
NS
SSF5
SSF6
TP
NS 0.031
NH4-N
TKN
NO3-N
TN
NS
0.014
NS
NS
NS
0.016
0.006
NS
NS
NS
0.031
0.030
NS
NS
0.009
NS 0.000
NS
NS
NS
NS
NS
0.022
NS 0.004
NS
0.003
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
Where NS = Not significant (p > 0.05)
A statistical summary comparing performance results from SSF2 during the low loading rate period and
the shortened filters (SSF7-8) are presented in Table 3.9. This is followed by Table 3.10 which lists the
results from t-tests comparing performance results between the filters. There are no significant
differences between results from SSF7 and SSF8. The only parameters which indicate differences
between the levels of treatment provided are TKN and NH4-N (only in SSF8). Thus, the shortened
filters are performing at the same level as a regular length filter under similar loading rate conditions.
These results suggest that the majority of treatment occurs within the first part of the LFSF, and that
treatment processes occurring within the biomat control the treatment efficiency of the system.
33 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Table 3.9. Statistical summary comparison between SSF2, a regular length filter during the low loading
(~100 L/day) period (January 2005-November 2006) and shortened filters, SSF7-8, from November
2007 to December 2008.
Parameter
Influent
concentration
Medium
(5%, 8 m)
(Jan ’05 – Nov
’06)
Mean
Median
Maximum
Minimum
Stdev
Mean
Median
Maximum
Minimum
Stdev
Mean
Median
Maximum
Minimum
Stdev
Mean
Median
Maximum
Minimum
Stdev
Mean
Median
Maximum
Minimum
Stdev
Mean
Median
Maximum
Minimum
Stdev
Mean
Median
Maximum
Minimum
Stdev
E. coli
(CFU/100
mL)
TSS (mg/L)
BOD5
(mg/L)
TP (mg/L)
NH4-N
(mg/L)
NO3-N
(mg/L)
TN (mg/L)
1.5E+06
1.4E+06
3.5E+06
3.0E+04
9.5E+05
910.1
273.0
5200.0
6.5
1532.2
426.8
156.0
1878.0
38.5
574.6
9.1
6.2
33.1
0.8
8.9
22.4
17.6
69.5
1.7
15.9
1.5
0.4
20.0
0.0
4.0
54.3
41.5
236.8
8.4
52.6
SSF2
4.44E+01
4.00E+00
6.00E+02
0.00E+00
1.30E+02
13.2
4.9
63.5
2.0
17.4
3.0
2.6
5.0
2.4
0.8
1.1
0.7
5.2
0.2
1.1
0.3
0.1
1.9
0.1
0.5
16.7
13.8
45.6
5.5
9.3
18.9
16.3
48.3
6.6
9.8
Influent
concentration
(Nov ’07 –
Dec ’08)
Medium
Medium
(5%, 5.5 m)
(5%, 5.5 m)
SSF7
SSF8
3.7E+06
1.5E+06
1.7E+07
1.0E+04
5.0E+06
49.5
34.1
197.0
21.4
49.9
66.6
68.4
113.0
29.6
27.4
3.2
2.9
9.2
1.4
2.1
16.4
17.3
29.6
5.8
8.1
0.05
0.04
0.08
0.04
0.02
30.5
26.7
90.1
15.2
20.6
4.8E+01
8.0E+00
2.8E+02
0.0E+00
9.3E+01
11.0
7.0
36.8
2.0
9.9
2.7
2.4
6.2
2.4
1.0
1.0
1.0
1.4
0.5
0.2
0.2
0.1
0.9
0.1
0.2
14.0
14.2
25.6
7.2
5.9
14.8
14.5
25.7
7.3
5.7
6.6E+01
1.4E+01
6.0E+02
2.0E+00
1.6E+02
13.7
13.6
41.4
2.0
9.9
2.7
2.4
5.7
2.4
0.9
0.9
0.9
1.7
0.1
0.4
0.1
0.1
0.1
0.1
0.0
17.8
15.2
28.2
9.3
6.3
18.6
16.6
28.8
9.7
6.5
34 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Table 3.10. Summary of p-values comparing performance results from SSF2 from January 2005 to
November 2006 to SSF7 to SSF8 from November 2007 to December 2008.
Parameter
SSF2 vs. SSF7
SSF2 vs. SSF8
SSF7 vs. SSF8
E. Coli (CFU/100 mL)
NS
NS
NS
TSS (mg/L)
NS
NS
NS
BOD5 (mg/L)
NS
NS
NS
TP (mg/L)
NS
NS
NS
NH4-N (mg/L)
NS
0.015
NS
0.035
0.003
NS
NO3-N (mg/L)
NS
NS
NS
TN (mg/L)
NS
NS
NS
TKN (mg/L)
Where NS = Not significant (p > 0.05) 3.3.2. Temporal Trends in Treatment Performance: Influent concentrations appear to vary when
comparing high to low loading period averages in Tables 3.6 and 3.7 (910.1 mg/L to 65.5 mg/L,
respectively). However, peak values shown in Figure 3.17 occurring in February, June, August and
September 2006 due to mechanical issues with the pump are an explanation for the inflated low loading
period averages. This is evident when comparing inlet concentrations (Figure 3.17) to outlet
concentrations for SSF1-6 (Figure 3.19, a-b): spikes in influent are matched by increased outlet
concentrations, specifically in late summer 2006. However, it is important to note that increased influent
concentrations did not affect TSS removal %. Statistically, each filter is achieving over an 89% TSS
removal rate. Average outlet concentrations for the high loading rate period range from 2.8 mg/L (SSF5)
to 8.1 mg/L (SSF2). Each filter performed very well with respect to TSS removal.
35 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 6000
TSS (mg/L)
5000
4000
3000
2000
1000
0
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
Figure 3.17. TSS inlet concentrations from November 2004 to December 2008.
Results for BOD5 treatment are similar to TSS removal. Filters demonstrate an ability to dampen out
variability in influent concentrations. Outlet concentrations from the high loading rate period range from
2.5 mg/L (SSF3, SSF4) to 2.9 mg/L (SSF2). Temporal trends of effluent BOD5 concentrations show that
the systems consistently produce levels less than 10 mg/L.
BOD5 (mg/L)
2000
1500
1000
500
0
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
Figure 3.18. BOD5 inlet concentrations from November 2004 to December 2008.
36 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) 120
TSS (mg/L)
100
80
60
40
20
0
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
SSF1
SSF2
SSF3
(b) 120
TSS (mg/L)
100
80
60
40
20
0
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
SSF4
SSF5
SSF6
(c) 120
TSS (mg/L)
100
80
60
40
20
0
Sep-07 Nov-07 Jan-08 Feb-08 Apr-08
Jun-08
Jul-08
Sep-08 Oct-08 Dec-08 Feb-09
Sampling date
SSF7
SSF8
Figure 3.19. Outlet TSS concentrations for SSF1-3 (a) and SSF4-6 (b) from November 2004 to
December 2008 and for SSF7-8 (c) from October 2007 to December 2008.
37 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) 12
BOD5 (mg/L)
10
8
6
4
2
0
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
SSF1
SSF2
SSF3
(b) 12
BOD5 (mg/L)
10
8
6
4
2
0
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
SSF4
SSF5
SSF6
(c) 12
BOD5 (mg/L)
10
8
6
4
2
0
Sep-07
Dec-07
Mar-08
Jun-08
Oct-08
Jan-09
Sampling date
SSF7
SSF8
Figure 3.20. Outlet BOD5 concentrations for SSF1-3 (a) and SSF4-6 (b) from November 2004 to
December 2008 and SSF7-8 (c) from October 2007 to December 2008.
38 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Influent bacteria counts have remained constant over time from 2004 to 2008 with occasional spikes
which simulate realistic variability (Figure 3.21). With respect to E. coli, outlet concentrations did not
change from low to high loading periods, except for SSF1 which increased by a magnitude of 2.
Accordingly, log removal for that filter fell from 5.58 to 4.87. Except for SSF1, each filter achieves at
least a 5-log removal rate of E. coli. Temporal variations in bacteria outlet concentrations (Figure 3.223.24) represent a level of variability which may not be acceptable for surface discharge, due to the
possible health hazard associated with such levels.
1.0E+08
1.0E+07
CFU/100 mL
1.0E+06
1.0E+05
1.0E+04
1.0E+03
1.0E+02
1.0E+01
1.0E+00
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
Total coliform
E. coli
Figure 3.21. Inlet TC and E. coli concentrations from November 2004 to December 2008 in CFU/100
mL.
Temporal trends in TP removal efficiencies are illustrated in Figure 3.25. The TP removal efficiency of
each filter is progressively decreasing with time. All eight filters initially removed close to 100% of the
influent TP, but after a relatively short period of time (1 year) the treatment efficiency begins to
decrease. This reduction in treatment efficiency happened much quicker in filters 7 and 8 due to the
smaller flow path lengths. After 3 years of effluent loading, several of the original six filters appear to
be saturated with phosphorus. As expected, the coarse-grained filters saturate more quickly than finegrained filters, as they would possess less surface area for adsorption. These results indicate that
conventional sand-based disposal systems have virtually no long-term phosphorus removal capacity.
39 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Total coliform (CFU/100 mL)
(a) 1.0E+04
1.0E+03
1.0E+02
1.0E+01
1.0E+00
1.0E-01
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
SSF1
SSF2
SSF3
(b) E. coli (CFU/100 mL)
1.0E+04
1.0E+03
1.0E+02
1.0E+01
1.0E+00
1.0E-01
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
SSF1
SSF2
SSF3
Figure 3.22. Outlet TC concentrations (a) and E. coli concentrations (b) for SSF1-3 from November
2004 to December 2008 in CFU/100 mL.
40 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Total coliform (CFU/100 mL)
(a) 1.0E+04
1.0E+03
1.0E+02
1.0E+01
1.0E+00
1.0E-01
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
SSF4
SSF5
SSF6
(b) E. coli (CFU/100 mL)
1.0E+04
1.0E+03
1.0E+02
1.0E+01
1.0E+00
1.0E-01
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
SSF4
SSF5
SSF6
Figure 8.23. Outlet TC concentrations (a) and E. coli concentrations (b) for SSF4-6 from November
2004 to December 2008 in CFU/100 mL.
41 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Total coliform (CFU/100 mL)
(a) 1.0E+04
1.0E+03
1.0E+02
1.0E+01
1.0E+00
Sep-07
Dec-07
Mar-08
Jun-08
Oct-08
Jan-09
Sampling date
SSF7
SSF8
(b) E. coli (CFU/100 mL)
1.0E+04
1.0E+03
1.0E+02
1.0E+01
1.0E+00
Sep-07
Dec-07
Mar-08
Jun-08
Oct-08
Jan-09
Sampling date
SSF7
SSF8
Figure 3.24. Outlet TC concentrations (a) and E. coli concentrations (b) for SSF7-8 from October 2007
to December 2008 in CFU/100 mL.
42 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 (a) Removal % of TP
150%
100%
50%
0%
-50%
-100%
-150%
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
SSF1
SSF2
SSF3
(b) Removal % of TP
150%
100%
50%
0%
-50%
-100%
-150%
Nov-04
May-05
Dec-05
Jun-06
Jan-07
Jul-07
Feb-08
Sep-08
Mar-09
Sampling date
SSF4
SSF5
SSF6
(c) Removal % of TP
150%
100%
50%
0%
-50%
-100%
-150%
Sep-07 Nov-07 Jan-08 Feb-08 Apr-08 Jun-08 Jul-08 Sep-08 Oct-08 Dec-08 Feb-09
Sampling date
SSF7
SSF8
Figure 3.25. Removal % of TP for SSF1-3 (a) and SSF4-6 from November 2004 to December 2008 and
SSF7-8 (c) from October 2007 to December 2008.
43 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 4.0 SUMMARY AND CONCLUSIONS This report presents results from a detailed field scale analysis of the performance and hydraulic
functioning of sloping sand filter systems receiving septic tank effluent. A total of eight SSFs,
constructed using varying slopes, sand types, lengths, and loaded at varying rates, were intensively
monitored over a 20 month period. The SSF system was shown to be a reliable, relatively robust
treatment system. SSFs that were loaded at elevated rates, as compared to NS technical guidelines,
performed well, providing a level of treatment that was similar to that observed at conventional loading
rates. The treatment performance of two identical SSF systems that were substantially shorter than that
recommended in the NS technical guidelines also performed well. The two shortened SSFs produced
effluent concentrations of all primary water parameters that were similar to those achieved by a
conventional length SSF possessing similar sand and slope characteristics.
Several tracer studies, using rhodamine dye as a conservative tracer, were conducted on all filters during
the study period. Tracer study data was fit to analytical residence time distribution models to assess the
hydraulic functioning of the filters. Measured hydraulic characteristics were then compared to those
predicted using theoretical models of porous media flow. The increase in flow to the original six SSFs
did not cause a noticeable change in the mean residence time characteristics of the filters. Water level
measurements from piezometers installed throughout the sand bed in each filter confirmed that the filters
remained unsaturated after the loading rates were increased. Ponding of water in the distribution trench
was observed, but biomat hydraulics appeared to be stable. A comparison of tracer generated residence
times with those predicted using theoretical porous media flow models provide further evidence that
saturated darcy flow does not exist within the SSF systems, even at elevated loading rates. This study
has shown that current design loading rates for SSFs in NS are conservative, providing an inherent
safety factor
The treatment performance of most water quality parameters (BOD5, TSS, E. coli, TN) remained
consistent as the SSF systems aged. However, a progressive reduction in phosphorus removal was
observed in all filters, regardless of sand type or slope. Results from this study indicated that
conventional sand based disposal fields become saturated with phosphorus within 3- 5 years. The
incorporation of alternative materials for phosphorus adsorption within passive filters should be
examined in future studies.
44 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 5.0 REFERENCES
Apello, C.A.J. and Postma, D. (1993). “Geochemistry, groundwater and pollution”. A.A. Balkema
Publishers, Rotterdam, Netherlands.
Brandes, M. (1980). “Effect of precipitation and evapotranspiration of a septic tank-sand filter disposal
system.” J. Water Pollut. Control Fed., 52(1), 59-75.
Check, G. (1992). “Design and Evaluation of On-site Wastewater Treatment with Lateral Flow Sand
Filters”. MASc Thesis. Dalhousie University. Halifax, Nova Scotia, Canada.
Check, G., Waller, D., Lee, S., Pak, D., and Mooers, J. (1994). “The lateral-flow sand-filter system for
septic effluent treatment.” Water. Environ. Res., 66(7), 919-928.
Clesceri, L.S., A.E. Greenberg, and A.D. Eaton. (Eds). (1998). “Standards Methods for the
Examination of Water and Wastewater, 20th ed.” Am. Public Health. Assoc. (APHA), Am. Water
Works Assoc., Water Environ. Fed.; APHA, Washington, DC.
Fogler, H. S., (1992). “Elements of Chemical Reaction Engineering, Second Edition”. Prentice Hall
Inc., New Jersey, NJ.
Freeze, R. A., and J. A. Cherry. (1979). “Groundwater”. Prentice-Hall Inc. Englewoods Cliff, NJ.
HACH Company. (1999). “Products for analysis; microbiological testing kits and procedure”:
http://www.hach.com, (April 10, 2006).
Harrison, R., Turner, N., Hoyle, J., Krejsl, J., Tone, D., Henry, C., Isaken, P., and Xue, D. (2000).
“Treatment of septic tank effluent for fecal coliform and nitrogen in coarse textured soil only and sand
filter systems.” Water, Air Soil Poll., 124, 205-215.
Havard, P. Jamieson, R., Boutilier, L., and Cudmore, D. 2008. Treatment Performance and Hydraulic
Characteristics of Lateral Flow Sand Filters for On-site Wastewater Treatment. Journal of Hydrologic
Engineering. 13,720-728.
Kristiansen, R. (1981a). “Sand-filter trenches for purification of septic tank effluent: I. The clogging
mechanism and soil physical environment.” J. Environ. Qual., 10(3), 353-357.
Nova Scotia Department of Environment and Labour (NSDEL). (2005). “On-Site Sewage Disposal
Guidelines Technical Guidelines”.
Pell, M., and Nyberg, F. (1989a). “Infiltration of wastewater in a newly started pilot sand-filter system:
I. Reduction of organic matter and phosphorus.” J. Environ. Qual., 18, 451-457.
Rodgers, M., Clifford, E., Mulqueen, J., and Ballantyne, P. (2004). “Organic carbon and ammonium
nitrogen removal in a laboratory sand percolation filter.” J. Environ. Sci. Health. Part AToxic/Hazardous Substances and Environmental Engineering., A39, 2355-2368.
45 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 APPENDIX A: PHOTOGRAPHS
46 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Photograph of LFSFs located on the 5 % slope showing piezometers and sampling wells.
Photograph of the plywood containment boxes that the LFSFs were constructed within.
47 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Photograph of liner installation and placement of the filter sand.
Photograph of sampling hut showing tipping buckets and autosamplers.
48 Sloping Sand Filters for On‐Site Wastewater Treatment
2009 Photograph showing the construction of the 5.5 m LFSF systems and installation of piezometers.
49 
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement