United States
Environmental Protection
Agency
EPA/600/R-15/268
October 2015
www2.epa.gov/water-research
Demonstration of a Graywater
Management Project at a Community
Level on the Island of Puerto Rico
Office of Research and Development
Water Supply and Water Resources Division
-------
Demonstration of a Graywater Management Project
at a Community Level on the Island of Puerto Rico
by
Evelyn Huertas1, Exel Colon2, Thomas P. O'Connor3
'United States Environmental Protection Agency (USEPA), Region 2
San Juan, Puerto Rico 00968
2Integrated Global Solutions LLC
Guaynabo, Puerto Rico 00968
3United States Environmental Protection Agency (USEPA)
Office of Research and Development
Urban Watershed Management Branch
Edison, New Jersey 08837
Contract No. EP-1 l-C-000217
Project Officer
Thomas P. O'Connor
REGIONAL APPLIED RESEARCH EFFORT
OFFICE OF RESEARCH AND DEVELOPMENT
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
EDISON, NEW JERSEY 08837
-------
Notice
The United States of America Environmental Protection Agency (USEPA), through its Office of Research and
Development (ORD), funded and managed, or partially funded and collaborated in, the research described herein. It has
been subjected to the Agency's peer and administrative review and has been approved for publication. Any opinions
expressed in this report are those of the authors and do not necessarily reflect the views of the Agency, therefore, no
official endorsement should be inferred. Any mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
-------
Abstract
On-site sewage disposal for residents of many rural Puerto Rican communities were typically undersized due to small
lot sizes and have other operational difficulties. To reduce hydraulic loadings to on-site systems residents release
graywater to the nearest stormwater system or receiving stream. Graywater gardens are a low-cost/-maintenance
alternative to conventional sewer projects. A community graywater garden was constructed and monitored in the Maria
Jimenez community of the municipality of Gurabo. The garden infiltrated and evapotranspirated graywater from four
households.
Water quality analysis during an 18-month monitoring period showed no statistically consistent removal across the
graywater garden system. Some parameters had events indicative of removal as effluent concentrations were lower than
influent concentrations; however, there were also events when influent concentrations were lower than effluent
concentrations, potentially indicating a lagging response (e.g., total Kjeldahl nitrogen and chemical oxygen demand).
Some parameters had no lagging response (e.g. magnesium, potassium and total coliform). A statistical increase in
effluent iron most likely indicated the system was anaerobic and iron leached from the soil. Treatment within the
graywater garden was greatly influenced by system maintenance and periods of high precipitation and humidity which
limited evapotranspiration, infiltration and evaporation. Pretreatment through two 200-L tanks to control oil and grease
was insufficient. A larger (1000L) tank, installed six months before the end of the project, resulted in lower
concentrations of oil and grease in the garden.
Water quality monitoring indicated that the graywater garden behaved like an anaerobic drainage field for a septic
system. Despite limited pollutant reduction within the graywater garden system, graywater discharges from these
households were effectively eliminated or dramatically reduced. Recommendations for improving the graywater garden
system and other design modifications were made.
in
-------
Foreword
The U.S. Environmental Protection Agency (EPA) is charged by Congress with protecting the Nation's land,
air, and water resources. Under a mandate of national environmental laws, the Agency strives to formulate and
implement actions leading to a compatible balance between human activities and the ability of natural systems to
support and nurture life. To meet this mandate, EPA's research program is providing data and technical support for
solving environmental problems today and building a science knowledge base necessary to manage our ecological
resources wisely, understand how pollutants affect our health, and prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory (NRMRL) is the Agency's center for investigation of
technological and management approaches for preventing and reducing risks from pollution that threaten human health
and the environment. The focus of the Laboratory's research program is on methods and their cost-effectiveness for
prevention and control of pollution to air, land, water, and subsurface resources; protection of water quality in public
water systems; remediation of contaminated sites, sediments and ground water; prevention and control of indoor air
pollution; and restoration of ecosystems. NRMRL collaborates with both public and private sector partners to foster
technologies that reduce the cost of compliance and to anticipate emerging problems. NRMRL's research provides
solutions to environmental problems by: developing and promoting technologies that protect and improve the
environment; advancing scientific and engineering information to support regulatory and policy decisions; and
providing the technical support and information transfer to ensure implementation of environmental regulations and
strategies at the national, state, and community levels.
This publication has been produced as part of the Laboratory's strategic long-term research plan. It is published
and made available by EPA's Office of Research and Development to assist the user community and to link researchers
with their clients.
Cynthia Sonich-Mullin, Director
National Risk Management Research Laboratory
IV
-------
Table of Contents
Notice ii
Abstract iii
Foreword iv
Contents v
List of Figures vii
Acronyms and Abbreviations ix
Acknowledgements x
Executive Summary 1
Chapter 1 Introduction 3
Overview 3
Chapter 2 Background 6
Objective 6
Location 6
Chapter 3 Materials and Methods 10
Gray Garden Design and Construction 10
Changes after Construction and Implementation 13
Monitoring and Analyses 14
Statistical Analysis 15
Chapter 4 Analysis of Results 17
Weather Observations 17
Sampling Results and Statistical Analysis 18
Chapters Discussion and Recommendations 27
Chapter 6 Conclusions 30
Chapter 7 References 31
Appendix A Graphs of Data 33
Appendix B As-built Drawings 49
Appendix C Project Images 55
Appendix D Sampling Documentation 70
Appendix E Correlation Analysis 76
-------
List of Figures
Figure 1-1 Example of graywater collection for discharge (white PVC piping) for a typical residence 4
Figure 1-2 Typical graywater discharge to stormwater drainage system (street gutter) 4
Figure 1-3 Graywater discharge causing foam in stormwater drainage system 5
Figure 2-1 Aerial photograph of the Maria Jimenez community and proj ect site (PRPB, 2010) 7
Figure 2-2 Location of the site within Puerto Rico (PRPB, 2010) 7
Figure 2-3 Topographic map of site and surrounding area (USGS, 1982) 8
Figure 2-4 Temperature and precipitation averages in the Gurabo substation, 1981-2010 (NOAA-NWS, 2015) 9
Figure 3-1 Plan and cross-section view of piping 12
Figure 3-2 Alternative cross section view showing piping with freeboard 13
Figure 4-1 Historical precipitation maximum, minimum and average accumulation behavior for Gurabo compared to
that of 2013 17
Figure 4-2 Median iron concentration and non-outlier range for influent and effluent sampling locations 23
Figure 4-3 Box plot of sulfide concentration before and after change in pretreatment system 24
Figure 4-4 Box plot of phosphorous concentration before and after change in pretreatment system 25
Figure 4-5 Effluent pH and sulfide concentration 26
Figure A-l Parameter concentration profiles for aluminum 33
Figure A-2 Parameter concentration profiles for barium with limit of detection 34
Figure A-3 Parameter concentration profiles for calcium 34
Figure A-4 Parameter concentration profiles for chromium with limit of detection 35
Figure A-5 Parameter concentration profiles for copper 35
Figure A-6 Parameter concentration profiles for iron 36
Figure A-7 Parameter concentration profiles for magnesium 36
Figure A-8 Parameter concentration profiles for manganese 37
Figure A-9 Parameter concentration profiles for silver with limit of detection 37
Figure A-10 Parameter concentration profiles for zinc 38
Figure A-ll Parameter concentration profiles for ammonia 38
Figure A-12 Parameter concentration profiles for nitrate and nitrite with limit of detection 39
Figure A-13 Parameter concentration profiles for total Kjeldahl nitrogen 39
Figure A-14 Parameter concentration profiles for potassium 40
Figure A-15 Parameter concentration profiles for phosphorous 40
Figure A-16 Parameter concentration profiles for chemical oxygen demand 41
Figure A-17 Parameter concentration profiles for oil and grease in log scale (due to large initial concentration) 41
Figure A-18 Parameter concentration profiles for total organic carbon 42
Figure A-19 Parameter concentration profiles for total dissolved solids 42
Figure A-20 Parameter concentration profiles for total suspended solids 43
Figure A-21 Parameter concentration profiles for chloride 43
Figure A-22 Parameter concentration profiles for sodium 44
Figure A-23 Parameter concentration profiles for fluoride 44
Figure A-24 Parameter concentration profiles for sulfide 45
Figure A-25 Most probable number profiles for fecal coliform 45
Figure A-26 Most probable number profiles for total coliform 46
VI
-------
Figure A-27 Most probable number profiles for Escherichia coll 46
Figure A-28 Profiles for pH 47
Figure A-29 Profiles for conductivity 47
Figure A-30 Parameter concentration profiles for surfactant 48
Figure B-l As-built drawings for the graywater garden, sheet one 50
Figure B-2 As-built drawings for the graywater garden, sheet two 51
Figure B-3 As-built drawings for the graywater garden, sheet three 52
Figure B-4 As-built drawings for the graywater garden, sheet four 53
Figure B-5 As-built drawings for the graywater garden, sheet five 54
Figure C-l Maria Jimenez graywater garden site before excavation 56
Figure C-2 Piping scheme in the excavation and gravel in the bottom 56
Figure C-3 Filling in the excavation with dirt 57
Figure C-4 Pipe installation of the graywater garden 57
Figure C-5 Back view of the recently-filled graywater garden 58
Figure C-6 Front view of the recently-filled graywater garden 58
Figure C-7 Side view of the recently-filled graywater garden with germinating plants 59
Figure C-8 Side view of the recently-filled graywater garden with germinating plants 59
Figure C-9 View of the garden from street lamp post 60
Figure C-10 Garden with some developed vegetation 60
Figure C-ll Pre-treatment units of the graywater garden 61
Figure C-12 Vegetation grown in the garden 61
Figure C-13 Vegetation grown in the garden, alternative angle 62
Figure C-14 Plantain trees and tannia plants growing 62
Figure C-15 High view of the garden with several plantain trees grown 63
Figure C-16 Local street view leading to the site 63
Figure C-17 view of the garden before cutting grass 64
Figure C-18 View of the garden after cutting grass 64
Figure C-19 Accumulation of water downstream of the system 65
Figure C-20 Entrance of the site and stormwater pipe underthe street 65
Figure C-21 Surface ponding in the garden 66
Figure C-22 Installation of French drain 66
Figure C-23 Covering up French drain 67
Figure C24 Covering up French drain with stone 67
Figure C-25 View of the graywater garden with copious vegetation 68
Figure C-26 View of the graywater garden with grown Musa spp. and Xanthosoma sp. plants 68
Figure C-27 View of the graywater garden with remaining plantain (Musa spp.) plants 69
Figure D-l Example of a Laboratory Analysis Request form used for USEPA Region 2 laboratory in Edison, New
Jersey 71
Figure D-2 Example of a Chain of Custody form used for the local laboratory 72
Figure D-3 Example of the second page of the Sampling and Analysis Plan presented two weeks before each sampling
event 73
Figure D-4 Example of the third page of the Sampling and Analysis Plan presented two weeks before each sampling
event 74
Figure D-5 Example of the fourth and last page of the Sampling and Analysis Plan presented two weeks before each
sampling event 75
vn
-------
List of Tables
Table 3-1 Soil characteristics at the site 10
Table 3-3 Parameter Testing method, sample volume, preservation and holding time 15
Table 4-1 Sampling results and statistics for metals 19
Table 4-2 Sampling results and statistics for nutrients 20
Table 4-3 Sampling results and statistics organics, solids, salts, flouride and sulfide 21
Table 4-4 Sampling results and statistics pathogenic indicators and other water quality parameters 22
Vlll
-------
Acronyms and Abbreviations
AASHTO American Association of State Highway and Transportation Officials
ASTM American Society for Testing and Materials
COD Chemical oxygen demand
CV Coefficient of variance
E. coll Escherichia coll
FAO Food and Agriculture Organization
IBC Intermediate bulk container
FIDPE High-density polyethylene
LOD Limit of detection
MBAS Methylene blue active substance
MPN Most probable number
NA Not applicable.
NH3 Ammonia
NOAA National Oceanic and Atmospheric Administration
NRMRL National Risk Management Research Laboratory
ORD Office of Research and Development
PRASA Puerto Rico Aqueduct and Sewer Authority
PRDNER Puerto Rico Department of Natural and Environmental Resources
PRDOH Puerto Rico Department of Health
PREQB Puerto Rico Environmental Quality Board
QAPP Quality Assurance Project Plan
PDCA Phosphate Detergents Control Act
PVC Polyvinyl chloride
RARE Regional Applied Research Effort
SAP Sampling and Analysis Plan
SC Specific Conductivity
SE Standard Error
SM Standard Methods
TKN Total Kjeldahl nitrogen
USCS Universal Soil Classification System
USEPA United States Environmental Protection Agency
USGS United States Geological Survey
WSP Watershed Stewardship Program
IX
-------
Acknowledgements
A project of this type requires the dedication and cooperation of a team. The technical direction and coordination for
this project was provided by the technical project team of the Urban Watershed Management Branch (UWMB), under
the direction of Thomas P. O'Connor, Project Officer and Evelyn Huertas, Technical Lead for Region 2 and member
of Puerto Rico's Watershed Stewardship Program. Jim Ferretti and staff performed water quality analysis on the
samples at EPA's Region 2 Laboratory in Edison, New Jersey. We also want to acknowledge the help of Ramon Garcia-
Caraballo, former Mayor of Gurabo and resident of the Maria Jimenez community, for his continuous help to complete
this project; Ms. Maria Garcia for providing access to her property for constructing the graywater garden; and the
neighbors of Maria Jimenez community who have agreed to participate in the project. Personnel from the Municipality
of Gurabo Department of Public Works should also be acknowledged for their assistance in the project, specifically
Conception Cruz, Dennys Torres, and Juan Calderon. Robert Goo of the EPA's Office of Water and Daniel Murray of
ORD provided peer reviews. Josephine Gardiner of the UWMB provided a technical edit.
-------
Executive Summary
Past construction practices have created a legacy of communities in rural areas of Puerto Rico that were constructed
without properly planned residential sewage disposal infrastructure. Many of these communities were built with on-site
wastewater disposal systems even though the conditions were often not conducive to the proper operation of these
systems due to a variety of conditions. In an effort to reduce overflows from these onsite systems, homeowners have
generally adopted the practice of discharging graywater to the property surrounding the house or when available, to a
storm drain, street gutter or a nearby water body or course. This discharge practice has elevated pathogenic indicator
microorganisms in the local streams and lakes, and has increased phosphorous and other pollutant concentrations in
downstream reservoirs.
The Puerto Rico Watershed Stewardship Program (WSP), a collaborative effort comprised of federal and local agencies,
has sought various ways to improve receiving water quality and protect reservoirs. This EPA Regional Applied
Research Effort (RARE) project funded the design, construction and monitoring of a community graywater garden
system in the rural community of Maria Jimenez in the municipality of Gurabo, a small town 35 km south of San Juan.
The graywater garden system received graywater from four residences comprised of 13 persons in all. The period of
monitoring was from May 2013 to September 2014 during which samples were collected for standard water quality
parameters (e.g. solids, organics, nutrients, metals and indictor microorganisms) for eight monitoring events.
The original design assumed there would not be graywater from the kitchen sink. Initially, a simple 400 L pretreatment
system was designed to capture oil and grease from kitchen water but this was later replaced with a larger 1000 L
pretreatment system which captured oil and grease better and provided more storage which reduced surges. Other
operational changes were made during the demonstration project including construction of a French drain, addition of
material to surface and plant harvesting in an effort to eliminate observed surface ponding during the rainy season.
Influent concentrations were similar to literature values of septic system discharges to drainage fields. While differences
were observed between influent and effluent concentrations within the graywater system for individual events, the long
term analysis showed no statistical difference, except for an increase in iron.
Effluent concentrations for several parameters changed during the course of the project and this was due to operational
and maintenance changes. There was an increase in sulfide concentration which correlated with a drop in pH; this
change potentially indicated that the 1000 L tank led to more stable anaerobic conditions. A quick look through the
figures in Appendix A indicated that many of the lowest concentrations for both the influent and effluent came during
the last two sampling events. This also points to better operation and maintenance by the larger pretreatment system;
oil and grease capture was crucial to reducing all pollutant concentrations in the graywater garden. Phosphorous effluent
concentration increased after plant harvesting, indicating that the plants had been effective in reducing nutrient
concentrations.
Recommendations for improved design were made, and many of these design recommendations were derived from
recommended practices for septic system drainage fields.
Overall, the graywater garden demonstration project reduced flows of graywater that would have alternatively
discharged directly to storm drains and receiving water. The effluent sampling point was a submerged sampling point;
most of the pollutant discharge from the graywater garden was to the soils surrounding the garden. As such, the use of
1
-------
graywater gardens if practiced elsewhere on Puerto Rico and in sufficient numbers may reduce discharges to receiving
streams and downstream reservoirs.
-------
Chapter 1 Introduction
This report discusses the findings and results of a project investigating graywater management through the use of a
small-scale graywater garden as a way to improve water resource management and reduce waste streams in rural areas
of Puerto Rico. The chemical analysis conducted, including measurements of metals and bacterial levels, provides an
assessment of the influent/effluent characteristics of the graywater in the garden system.
Overview
Throughout Puerto Rico, housing development practices in the past have often not been closely monitored by regulatory
agencies. Past construction practices have often created a legacy of rural communities constructed without properly
planned sewage disposal infrastructure. Many of these communities were built with on-site wastewater disposal systems
though the conditions in these rural areas of Puerto Rico are often not conducive to the proper operation of these systems
due to a variety of conditions such as high groundwater, steep slopes, clay soils and small lot sizes with improperly
sized drainage fields. The combination of these factors with a less-than-required regulatory presence has resulted in a
situation where homeowners often find it difficult or impossible to properly dispose of all of the wastewater emanating
from their residence. In an effort to reduce overflows from the septic systems, homeowners have generally adopted the
practice of discharging graywater to property surrounding the house or when available, a storm drain, street gutter or a
nearby receiving water (WSP, 2011). Figure 1-1 is an example of the typical PVC piping used to collect and discharge
household graywater.
The graywater being released, which includes laundry, sinks, kitchen/cooking and bathing water, can eventually end up
in the drinking water reservoirs of Puerto Rico (Tetra Tech Inc., 2011). For example, in Puerto Rico's Rio Grande de
Loiza and La Plata watersheds, where 57% of the population is unsewered (PREQB, 2007), eutrophication has been
identified as a major water quality problem that impacts its reservoir (Quinones, 1980). Figure 1-2 shows development
of green bio-mat in a storm drain as a result of the nutrient rich graywater discharge, while Figure 1-3 shows foaming
agents in a pool of graywater ponding in the stormwater conveyance system.
-------
Figure 1-1 Example of graywater collection for discharge (white PVC piping) for a typical residence
Figure 1-2 Typical graywater discharge to stormwater drainage system (street gutter).
-------
Figure 1-3 Graywater discharge causing foam in stormwater drainage system.
The streams and lakes that discharge to Puerto Rico's reservoirs are known to contain high levels of pathogenic bacterial
indicator microorganisms and phosphorous (Caspe, 2008). The U.S. Environmental Protection Agency (EPA), Puerto
Rico Department of Health (PRDOH), Puerto Rico Environmental Quality Board (PREQB), Puerto Rico Aqueduct and
Sewer Authority (PRASA) and Puerto Rico Department of Natural and Environmental Resources (PRDNER) created
the Watershed Stewardship Program (WSP); this collaborative effort seeks to develop and implement replicable and
affordable pollution control strategies to protect the watersheds that drain to the reservoirs. These strategies included
elimination of phosphate detergents (Puerto Rico Environmental Quality Board, 1983; Puerto Rico Environmental
Quality Board, 2007; Quinones, 1980), extension of sewers and provision of affordable septic system cleanout services.
This program has also targeted the rural mountain communities in the Rio Grande de Loiza and La Plata watersheds
because the drainage from these areas lead to the island's drinking water reservoirs for approximately 40% of the
population of Puerto Rico.
The Puerto Rico Phosphate Detergents Control Act (PDCA), which was adopted in 2009 and became effective on
January 1, 2010, has successfully reduced nutrient enrichment in the reservoirs (WSP, 2011). However, despite these
pollution control strategies like the PDCA, residential graywater discharge will most likely continue to be practiced in
the watersheds. Graywater gardens are potentially a way to reduce the impact of these discharges. Graywater gardens
should be effective in Puerto Rico's tropical climate as the absence of freezing temperatures and corresponding plant
dormancy should increase annual effectiveness of these planted systems. Infiltration, evaporation and
evapotranspirative losses are effective mechanisms for planted systems to manage and treat water. The treatment of
household graywater composed of the nutrient-rich effluent from showers, washing machines and sinks (not used for
disposal of hazardous, toxic materials, food preparation, or food disposal) (McGovern, 2010) has been shown to reduce
energy and costs needed for wastewater transport, treatment and disposal.
This report describes the design, construction, and monitoring of a small-scale, community graywater garden system.
Influent/effluent water samples were analyzed for temperature, pH, specific conductivity (SC), metals, chemical oxygen
demand (COD), ammonia (NHs), chloride (Cl~), fluoride (F~), nitrate (NOs) and nitrite (NC>2), total Kjeldahl nitrogen
(TKN), oil and grease, organic carbon, phosphorus (P~), sulfide (82"), filterable residue, non-filterable residue, fecal
coliforms, total coliforms, and Escherichia coll (E. coll). These water quality parameters were collected quarterly in
order to evaluate the overall efficiency of the graywater garden treatment system over time, as well as to characterize
incoming water from the connected residences.
-------
Chapter 2 Background
In 1983, the Commonwealth of Puerto Rico enacted the Underground Injection Control Rules (PREQB Resolution No.
R-83-23-1) to protect surface and ground water resources throughout the island, as well as help prevent the pollution of
potable water sources. The enacted regulation required owners of subsurface wastewater discharge structures to meet
rigorous operational compliance criteria; however, single-family dwellings were excluded from the promulgated rules.
This exclusion has contributed to communities with inappropriate onsite wastewater disposal systems throughout the
island.
As noted, Puerto Rico's WSP collaborative effort is developing and implementing pollution control strategies (e.g.
elimination of phosphate detergents, sewer construction and affordable septic system cleanout services) to protect
receiving water quality of watersheds that drain to the drinking water reservoirs, focusing particularly on the rural,
unsewered, mountain communities that have inadequate on-site treatment systems. In lieu of the traditional, expensive
and disruptive sewershed build-out to rectify Puerto Rico's poorly-serviced sewershed areas, developing and
demonstrating ways to treat graywater separately from residential blackwater could also help to decrease downstream
impact to the reservoirs. As such, graywater garden treatment systems could support the proper operation and
performance of existing onsite wastewater disposal systems by helping to maintain appropriate inflow and wastewater
effluent characteristics.
USEPA's Office of Research Development (ORD) Urban Watershed Management Branch (UWMB) develops
innovative urban technologies to assist municipalities and utilities in the selection of watershed approaches to control
polluted urban discharges. Research by the UWMB on "graywater gardens" will help identify if this approach is a
potential means of supplemental treatment in rural Puerto Rico to remedy the current practice of discharging graywater
to the nearest water body or drainage system. Funding for this project was provided through the Regional Applied
Research Effort (RARE) in which Region 2 works with the ORD to prioritize research project needs. The RARE
Program is administered by the ORD's Regional Science Program.
Objective
This Region 2 RARE project demonstrated a graywater management system in a rural area of Puerto Rico. Household
graywater was treated in a community graywater garden that was designed, constructed and monitored for this project.
This goals of this project were to determine whether such gardens could improve water resource management and
reduce wastewater streams, and to ascertain whether the technology could be adopted elsewhere on the island.
Location
This project was located in the Maria Jimenez community (18.266575 N, 65.937861 W) within the municipality of
Gurabo (Figure 2-1). The residents of this small rural community were discharging graywater to property surroundings,
nearby water bodies or storm drains. The municipality of Gurabo is located at the central-eastern part of Puerto Rico
(Figure 2-2). It has a land area of 72.4 km2 and is comprised of nine rural neighborhoods and one urban zone. Gurabo
-------
borders with Trujillo Alto to the north, San Lorenzo to the south, Carolina to the north-east, Juncos to the east and
Caguas to the west. The population of Gurabo, based on the 2010 census, was 45,373 inhabitants. Although population
across Puerto Rico has declined in recent years, Gurabo is one of several municipalities with population growth as it
has become a satellite suburb of the metropolitan area of San Juan due in part to highway access. Although most of the
population lives in the rural area and works outside the municipality, the main economic activities of Gurabo include
the manufacturing of metals, paper, plastics, chemicals, pharmaceutical products, textiles, machinery and electrical
equipment. There is also some minor livestock and fruit crops activity (Fundacion Puertorriquena de las Humanidades,
2009).
, •
Figure 2-1 Aerial photograph of the Maria Jimenez community and project site (PRPB, 2010)
Figure 2-2 Location of the site within Puerto Rico (PRPB, 2010)
-------
Gurabo is comprised of three different geological regions: the southern section of Gurabo belongs to Puerto Rico's
eastern mountainous zone, the north is part of the Puerto Rico northern wet, and the central section belongs to the
Caguas Valley. The municipality's surface is composed of alluvial deposits and volcanic and plutonic rocks.
Specifically, the soils are mainly composed of the Mucara and Caguabo series which are respectively identified as a
shallow, well-drained soil, formed from gravelly residuum from basic volcanic rocks and a moderately deep, slightly
acid, well-drained soil, formed from weathered residual material from volcanic rock. Both of these series are
characterized as moderately permeable soils (Boccheciamp, 1978). Gurabo is located within the Rio Grande de Loiza
watershed and is crossed by the Gurabo River, the Valenciano River, and some minor creeks. A topographical
representation of the site location is provided in Figure 2-3
••- - '/
Figure 2-3 Topographic map of site and surrounding
The climate of Gurabo is classified as semi-tropical (Grupo Editorial EPR, 2009) with two temperature zones, a tropical
zone in the plains and temperate zone in the mountains. The average temperature is approximately 25 °C though the
average maximum daily temperatures increases to over 32 °C from July to September. Precipitation is abundant, with
an annual average of 1700 mm of rain (NOAA, 2015); the mean monthly rainfall is provided in Figure 2-4. During
hurricane season (June 1 - October 31) winds blow from east to west.
-------
35
400
30
25
S 20
*•>
-------
Chapter 3 Materials and Methods
This section describes criteria for garden design, construction details and monitoring procedures.
Gray Garden Design and Construction
Soils for the site were characterized (Table 3-1) by standard testing methods (ASTM methods D 3282 and D 2487)
based on the American Association of State Highway and Transportation Officials (AASHTO) Soil Classification
System and Universal Soil Classification System (USCS), respectively. Soil percolation tests results yielded a rate of
8.4 cm/h for the Maria Jimenez community soil using a procedure of digging a 0.6 m x 0.6 m x 0.6 m hole into the
ground which was then filled with water; water depth was measured every 15 minutes until all the water infiltrated
(Integrated Global Solutions, 2012).
Table 3-1 Soil characteristics at the site
Test
Method
ASTM
D3282
ASTM
D2487
ASTM
D3282
ASTM
D2487
ASTM
D3282
ASTM
D2487
Classification
System
AASHTO
USCS
AASHTO
USCS
AASHTO
USCS
Sample
Depth
(cm)
20
20
41
41
61
61
Gravel
(%)
26
26
15.2
15.2
8.5
8.5
Sand
(%)
23.1
23.1
25.8
25.8
30.5
30.5
Fines
(%)
50.9
50.9
59
59
61
61
Plastic
Limit
31
31
30
30
25
25
Liquid
Limit
47
47
43
43
42
42
Classification
A-7-5(6)
ML
A-7-5 (6)
ML
A-7-6 (9)
CL
Description
Clayey Soil
Gravelly Silt
with Sand
Clayey Soil
Sandy Silt
with Gravel
Clayey Soil
Sandy Lean
Clay
Based on field observations and interviews, it was anticipated there would be graywater from 13 residents among the
four households to be connected to the garden. Daily graywater input to the graywater garden was determined using
calculations presented in the Quality Assurance Project Plan (QAPP) (Integrated Global Solutions, 2012), which
assumed 1 m2 of garden area could manage graywater from three residents assuming 110 L/d per capita. This 110 L/d
per capita estimate was based on typical washing machine, shower and a fraction of sink usage (Mayer et al. 1999),
resulting in projected graywater flows for all 13 residents of 1430 L/day. Peak direct rainfall infiltration requirements
in the (Integrated Global Solutions, 2012) were 500 mm of rainfall in a period of 24 h, which is 21 mm/hr. Adding a
safety factor, the gray garden's surface area was increased to 11 m2. This resulted in a required 0.54 cm/hr infiltration
rate with peak infiltration of 2.6 cm/hr to handle peak rainfall conditions as well, both of which were well below the
8.4 cm/hr measured rate.
10
-------
Water level and sampling wells were placed at three different locations within the garden. These wells had depths of
250 cm, 175 cm and 50 cm measured from the existing elevation prior to excavation. Sampling wells were placed and
angled in a manner that allowed sampler to stay out of the garden while taking a sample. Although it was recommended
to perform excavation, construction and fill of the garden with hand instruments (e.g., shovels, pickaxes or
wheelbarrows) to avoid compaction (which could negatively affect exfiltration from the rain garden), excavation and
construction was performed with heavy machinery with the assistance of the Gurabo municipality. The site was
excavated to a depth of 130 cm and a filter fabric was placed at the bottom to prevent fill migration. Over the filter
fabric, a 10-cm gravel layer was added for improved percolation at the bottom of the garden. The greywater distribution
pipes were designed to manage the discharge flows and spread the graywater throughout the planted area of the garden.
These pipes had a 3% slope, producing an approximate velocity of 70 cm/s. Perforated 5-cm diameter PVC pipes were
also installed to act as system drains and improve flow. Cleanouts were placed at various locations to provide access
for maintenance in case of clogging.
The pre-project design assumed that kitchen sink water would not be treated in the graywater garden. With the inclusion
of kitchen water, the expected per capita discharge volume would increase. Based on Mayer et al. (1999) per capita
usage rates for washing machine (57 L/d), shower (44 L/d), faucet (41 L/d) and dishwasher (4 L/d), the discharge of
graywater was estimated to be 144 L/d/capita. The addition of kitchen water to the graywater stream only increased the
infiltration rates to 0.71 cm/hr and of 2.8 cm/hr for peak rate, both rates still well below design capacity. As long as the
system did not clog, there appeared to be adequate infiltration capacity.
The four households were connected to the graywater garden with the assistance of the municipality of Gurabo, through
a series of 5-cm and 10-cm diameter PVC pipes. Graywater was collected in the two 200-L pretreatment tanks. The
pipe from the residences discharged to the bottom of the first tank allowing the grease to float and remain in the drum
while the rest of the water continued to the second tank. In the second, screening tank, graywater passed through a 19-
L plastic perforated bucket; water then discharged from the bottom of the screen tank into the graywater garden. At this
point, flow was controlled through a pair of battery-powered timer valves to modulate graywater input to the garden.
The screen tank also had an overflow pipe located approximately 30 cm above the bottom pipe of the tank. Water
entering the garden passed through the 5-cm diameter PVC perforated irrigation pipes to diffuse into the soil. These
perforations were the same diameter as in the bucket in the screen tank, so that anything passing through the screen
tank, would also pass through the irrigation pipe holes without clogging them. There was also a bypass for both tanks
to allow for maintenance. The grease trap and screen tank were designed for ease of construction to allow graywater
gardens to be installed in other rural communities without need of extensive technical expertise.
As a safety factor, a graywater garden overflow pipe was designed to manage any excess water coming either from
heavy rainfall or excessive graywater influent. Figure 3-1 shows the plan and cross section of the piping incorporated
into the graywater garden system design. The complete as-built drawings with legends are located in Appendix B.
The graywater garden was initially planted with plantains (Musa sp.) and arrowleaf elephant ear (Xanthosoma
sagittifolium), locally known as malanga. These species provided large vapor exchange area in their leaves and, in the
case of plantains, their stems could store considerable amounts of water (Carr, 2009).
11
-------
© '
t ortu* c» p*g c-oj
"=^x
-------
Figure 3-2 Alternative cross section view showing piping with freeboard
Grease trap cleaning was scheduled every two or three weeks with the collaboration of the Public Works Department
personnel of Gurabo. Landscaping maintenance was also performed every two to four weeks, i.e., cutting grass and
keeping the garden as clean as possible (refer to Appendix C for pictures). Frequency of the landscaping depended on
recent precipitation and inspection of the growth of grass.
Changes after Construction and Implementation
Surface ponding started to occur in the graywater garden coinciding with the start of rainy (hurricane) season around
August or September 2013. Neighbors were concerned by strong odors coming from the garden and believed mosquito
breeding could present a health risk to anyone nearby. Several corrective actions to eliminate surface ponding were
performed: the construction of a French drain (refer to figure Appendix B-5 and Appendix C-23 and -24 for details);
reduction of the number of the plants inside the garden; scarification of the garden's surface; topping of the garden with
approximately 5 cm of crushed stone; and later, providing an additional 5-cm of sand to the garden surface. These
measures were effective in eliminating the surface ponding. While it was thought that this situation was caused by
clogging of the pipes, flow through the pipes was verified twice by flushing with water and no clogs were detected.
Due to the nearby location of two tall mango trees which kept the garden shaded (demonstrated by numerous
photographs in Appendix C), many of the initial plantings were removed so that air and sunlight could directly enhance
evapotranspiration and evaporation. Even though there was wind (not quantified), shading limited the phenomenon of
evapotranspiration and evaporation, which accounts for part of the removal of water in these systems (Maidment et al.
, 1988). Excess rain along with high humidity during the hurricane season of 2013 caused surface ponding in the
graywater garden. Some plantains were left in the garden as their leaves provided large surface area for
evapotranspiration and also stored water in their stems and trunks (FAO, 2013). As a replacement to the initial plantings,
a species of small shrub, brisselet (Erythroxylum brevipes) (Francis, 2004), was planted in various parts of the graywater
garden, however, the brisselet grew very slowly. During the remainder study, a number of the brisselet were cut along
with grass during landscaping maintenance.
13
-------
Finally, the initial pretreatment configuration comprised of two 200-L plastic drums to trap grease and screen influent
was replaced with a 1000-L intermediate bulk container (IBC) to facilitate maintenance and increase storage capacity
to reduce impacts due to surges of graywater. The timer valves installed at the lower exit of the screening 200-L plastic
drum to regulate the entrance of graywater to the garden were not well coordinated with the times and duration of the
influent water, therefore, the screen tank operated most of the time in overflow mode and the desired control of the
valves was overridden. Table 3-2 shows the timeline of construction and all modifications as well as the dates of the
sampling events.
Table 3-2 Timeline of construction, modifications and sampling events
Task Number
1
2
o
J
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Task
Percolation test was performed
Excavation was performed and pipe system was installed
Pretreatment grease trap and screen tank (200 L) tanks were
installed
Construction completed and houses connected
Plantains and malanga were planted
First quarterly sampling event
Second quarterly sampling event
Water ponding on the garden was observed
Third quarterly sampling event
French drain was installed
All malanga and some plantains plantings were removed
from the garden and alternative plantings, brisselet,
incorporated
Fourth quarterly sampling event
Larger pretreatment (IBC 1000 L) tank was installed
Fifth quarterly sampling event
Sixth sampling event
Seventh sampling event
Eighth (final) sampling event
Date
December 13,2011
September 6, 2012
September 8, 2012
February 22, 2013
April 13, 2013
May 15, 2013
August 13, 2013
September 12, 2013
December 3, 20 13
December 2 1,20 13
January 21, 2014
February 11,2014
March 12, 2014
June 3, 2014
Augusts, 2014
August 26, 2014
September 23, 20 14
Monitoring and Analyses
Quarterly sampling was initiated on May 15, 2013. The sampling portion of the project was terminated in October 2014,
due to a request by the property owners who were looking to sell the property. Until June 2014, quarterly samples ran
as scheduled, then the last three were performed at an accelerated rate resulting in eight sampling events overall (Table
3-2).
Prior to the scheduled sampling event, documentation and notification for the sampling events were prepared (normally
two weeks in advance). An Analysis Request Form for the USEPA Region 2 laboratory identified which tests would be
performed on the samples. Respective chains of custody were also completed. The Sampling and Analysis Plan (SAP)
presented an overview of the sampling event, type of analyses to be performed on the samples and all information
specifying quantity, collection and preservation of the samples. A copy of the SAP was sent to the local laboratory as
well so that they could prepare the bottle order, which was usually picked up the previous afternoon. All bottles
contained the preservatives for their respective contents, Region 2 having provided bottles at the onset of the sampling
portion of the project. Appendix D provides example documentation from the third sampling event.
The day before a sampling event, all equipment was checked. Materials for the sampling event included: sample bottles,
nitrile gloves, a 19-L bucket, two manual pumps, two 45-L coolers, six bags of ice, a marker, a pen, a pH and
14
-------
temperature meter, resealable plastic bags, bubble wrap, a 500-mL plastic cup and tape. Several resealable plastic bags
were filled with ice the day before and stored in a freezer to increase efficiency during sampling. Approximately 15 L
of sample were collected using a dedicated manual pump from the sampling point S2, located at the exit end of the
garden. Each sampling bottle was carefully filled and put into one of the coolers with the ice bags. Similarly, a 15-L
sample was manually pumped at the S-l located at the grease trap.
Holding times for the parameters are provided in Table 3-3. Sample bottles with parameters that had short holding
times, i.e., mainly bacteriological and some sanitary, were loaded into one cooler and were transported to a local
laboratory. A second cooler was filled with sample bottles with longer holding times; this cooler was shipped to EPA's
Region 2 laboratory in Edison New Jersey for analysis of the remainder of the parameters. Similarly preserved
parameters were shipped in the same bottle, so there were only five bottles per influent and effluent. However, there
was increased preparation time to prevent spillage during shipping as exemplified by the glass bottles in this second
cooler which were also bubble-wrapped and placed in a plastic resealable bags. Sampling was usually scheduled for the
morning hours, between 08:00 and 10:00, so that the cooler going to EPA Region 2's laboratory could be sent out later
that afternoon.
Table 3-3 Parameter Testing method, sample volume, preservation and holding time
Parameter
Total suspended solid
Total dissolved solids
Flouride
Chloride
Conductivity
Oil and grease
Sulfide
Chemical oxygen demand
Total organic carbon
Nitrate + Nitrite [as N]
Nitrogen, Total Kjeldahl
Ammonia [as N]
Phosphorous
Metals
pH
Fecal Coliform
Total Coliform
Escherichia coli
Surfactants
Method
SM1 2540D
SM 2540C
EPA 300.0
EPA 300.0
SM2510 A
EPA 1664A
SM 4500 S2 D
EPA 4 10.4
SM5310
EPA 353.2
EPA 351.2
EPA 350.1
EPA 365.4
EPA 200.7
EPA 150.1
SM9221E
SM9221C
SM9221F
SM 5540C
Container (volume)
1 L HOPE (400 ml)
1L HOPE (100 ml)
1 L HOPE (50 ml)
1L HOPE (100 ml)
1 L Amber glass (3 L)
250 mL HOPE
500 mL HOPE (50 mL)
500 mL HOPE (50 mL)
500 mL HOPE (100 mL)
500 mL HOPE (100 mL)
500 mL HOPE (100 mL)
500 mL HOPE (50 mL)
250 mL HOPE
Field measurement
125 mL HOPE
125 mL HOPE
125 mL HOPE
500 mL HOPE
Preservative
Ice, 4 °C
Ice, 4 °C
Not required
Not required
Not required
HC1
ZN acetate +NAOH pH>9
pH < 2 H2SO4
pH < 2 H2SO4
pH < 2 H2SO4
pH < 2 H2SO4
pH < 2 H2SO4
pH < 2 H2SO4
HNO3
Field measurement
0.008% Na2S3,4°C
0.008% Na2S3,4°C
0.008% Na2S3, 4°C
Not required
Holding
Time
7d
28 d
28 d
28 d
28 d
28 d
28 d
28 d
28 d
6 m
Oh
6h
6h
6h
48 h
1 Standards Methods (1998).
2 HDPE - High-density polyethylene
Statistical Analysis
Rudimentary statistical analyses were performed on the influent and effluent concentrations of the graywater garden
sampling results. These analyses, i.e., median, normality, mean, standard of deviation and coefficient of variance (CV)
along with t-statistic or Mann-Whitney U test are presented in Table 4-1 (metals), Table 4-2 (nutrients), Table 4-3
(organics and solids) and Table 4-4 (other parameters). Normality of data was tested the Shapiro-Wilk W test (StatSoft,
2011).
15
-------
When samples were not detected, one-half the detection limit was used for calculation of statistics for a frequency of
detection at or above 85%; from 85% to 50% detection, statistics were calculated using Aitchison's method (EPA,
2000). These calculations were made with Microsoft Office Excel (2013). For less than 50% detection, only frequency
of detections, if any were reported.
Change in operational procedures and effect on effluent concentrations was tested by t-statistic or Mann-Whitney U
test, as applicable. Correlation analyses were performed on effluent concentrations by the Spearman rank R correlation
test (StatSoft, Inc., 2011). A high correlation between specific parameters for all sampling dates may imply that there
were similar cause and effect for concentration changes, especially if these results increase and decrease in tandem.
Where applicable, only values above the one-half detection limit were used for the normality and non-parametric
correlation analysis.
Standard error (SE) was used for error bars and box plots.
16
-------
Chapter 4 Analysis of Results
Weather Observations
Annual evapotranspiration in Puerto Rico is 1140 mm/yr (45 in/yr) (Hanson, 1991). Evapotranspiration rates are
dependent on several variables, such as the surface area of the leaves, air temperature and the relative humidity. A high
relative humidity corresponds to a high water content in the air which would reduce the gradient that allows the transfer
of water from plants to the surrounding atmosphere.
An analysis of the data provided for precipitation in the area showed that precipitation was higher than normal from
July to December 2013 (which leads to high humidity conditions) and corresponded with observations of surface
ponding during this period of the project. Figure 4-1 shows a comparative plot of how different precipitation behaved
from the historic average during the period of the study where surface ponding started, approximately August or
September 2013. Figure 2-4 indicates that this is also the warmest period of the year.
3000
2500
1?
g2000
.2 1500
1955 (Lowest)
Normal
1970 [Highest)
Cumulative 2013
Date
Figure 4-1 Historical precipitation maximum, minimum and average accumulation behavior for Gurabo compared to that of 2013.
17
-------
Sampling Results and Statistical Analysis
Time series plots (Appendix A) were produced for parameters that had sufficient quantities of data so that statistical
analysis could be performed. Due to the small sampling size (8 events maximum), at least 50% detection (along with
normality, as tested by the Shapiro-Wilk W test) was required to calculate a mean concentration for the influent or
effluent. Several metals, i.e., antimony, beryllium, cobalt, selenium and thallium, were not detected during sampling;
arsenic, cadmium, mercury, and nickel, had only one detection while lead and vanadium had several detections but
insufficient numbers of detections of either the influent or effluent to perform any rudimentary statistics. Data for
parameters, as previously explained, are presented in Table 4-1 (metals), Table 4-2 (nutrients), Table 4-3 (organics and
solids) and Table 4-4 (other parameters).
A general observation of the behavior of all parameters was the near absence of sustained trend of removal as a function
of time, especially in the metals. Observation of the plots in Appendix A show that some parameters have some events
which indicate removal as effluent concentrations are lower than influent concentrations removals; however, there are
also sampling events when influent concentrations are lower than effluent concentrations, indicating a lagging in
response (e.g., TKN and COD). Some parameters have no lagging response (e.g. magnesium, potassium and total
coliform).
18
-------
Table 4-1 Sampling results and statistics for metals
Parameter
Aluminum
Barium
Calcium
Chromium
Copper
Iron
Magnesium
Manganese
Silver
Zinc
Location
Influent
Effluent
Effluent
Influent
Effluent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Effluent
Influent
Effluent
Limit of
detection
(ug/L)
100
100
100
500
500
5
10
10
50
50
500
500
5
5
5
20
20
Median
(ug/L)
430
2150
100
37000
33500
11
90
140
570
2500
12000
10200
53
235
5
180
455
Lower
quartile
(ug/L)
150
370
100
22000
21000
5
72
76
380
1100
7300
7500
46
140
5
120
170
Upper
quartile
(ug/L)
3400
7800
230
65000
43000
27
420
180
1900
4300
13000
15000
320
430
18
1400
690
Normality
(detections
/events)
No (7/7)
Yes (6/6)
Yes (3/6)
No (7/7)
No (6/6)
No (4/6)
No (7/7)
No (6/6)
No (7/7)
No (6/6)
Yes (7/7)
Yes (6/6)
No (7/7)
Yes (6/6)
Yes (3/6)
Yes (7/7)
No (6/6)
Mean
(ug/L)
—
3700
^OD1
—
—
—
—
—
—
—
10600
11000
—
303
61
650
-
Standard
of
deviation
(ug/L)
~
4000
~
~
~
~
~
~
~
~
2900
3800
~
230
5
690
~
Coefficient
of variation
~
1.07
~
~
~
~
~
~
~
~
0.27
0.35
~
0.76
0.8
1.1
~
Statistical
difference
No
NA2
No
NA
No
Yes
No
No
NA
No
1 Due to non-detects, mean and other normal parameters calculated using Atchison's method.
2 NA - not applicable.
19
-------
Table 4-2 Sampling results and statistics for nutrients
Parameter
Ammonia
Nitrate and
Nitrite
Total Kjeldalh
Nitrogen
Phosphorous
Potassium
Location
Influent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Limit of
detection
(mg/L)
0.51
0.51
0.05
0.05
I1
I1
0.51
0.51
0.5
0.5
Median
(mg/L)
4.2
5.7
0.06
0.06
12
17
3.6
4.6
9.4
8.9
Lower
quartile
(mg/L)
3.1
4.0
0.05
0.05
11
13
2.8
2.1
8.2
7.8
Upper
quartile
(mg/L)
5.7
9.2
0.21
0.08
18
27
5.3
5.4
12.0
13.0
Normality
(detections
/events)
Yes (8/8)
No (8/8)
Yes (5/8)
No (5/8)
No (7/7)
No (7/7)
Yes (8/8)
Yes (8/8)
Yes (7/7)
Yes (6/6)
Mean
(mg/L)
4.7
—
0.122
—
—
—
4.0
4.1
9.7
9.8
Standard
of
deviation
(mg/L)
3.1
~
0.15
~
~
~
1.6
1.9
1.9
2.6
Coefficient
of variation
0.65
~
1.2
~
~
~
0.39
0.45
0.2
0.3
Statistical
difference
No
No
No
No
No
1 Multiple detection limits; this is maximum detection limit.
2 Due to non-detects, mean and other normal parameters calculated using Atchison's method.
20
-------
Table 4-3 Sampling results and statistics organics, solids, salts, flouride and sulfide
Parameter
Chemical
oxygen demand
Oil and grease
Total organic
carbon
Total dissolved
solids
Total suspended
solids
Chloride
Sodium
Flouride
Sulfide
Location
Influent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Limit of
detection
(mg/L)
10001
4001
5
5
101
101
10
10
10
10
101
51
1
1
0.05
0.05
O.I1
O.I1
Median
(mg/L)
750
870
70
54
150
160
480
480
110
260
59
58
54
58
0.08
0.10
3.4
1.6
Lower
quartile
(mg/L)
620
550
50
35
130
130
430
440
60
50
45
53
43
53
0.03
0.03
0.03
0.05
Upper
quartile
(mg/L)
1900
1900
180
150
220
290
580
630
350
390
140
86
73
77
0.09
0.14
4.3
4.6
Normality
(detections
/events)
No (8/8)
Yes (8/8)
No (8/8)
No (8/8)
No (7/7)
Yes (7/7)
No (7/7)
No (7/7)
No (8/8)
No (7/7)
Yes (7/7)
Yes (7/7)
No (8/8)
Yes (7/7)
Yes (7/7)
Yes (7/7)
Yes (7/7)
Yes (6/7)
Mean
(mg/L)
—
1290
—
—
—
190
—
—
—
—
100
66
—
67
0.07
0.09
2.5
2.22
Standard
of
deviation
(mg/L)
~
1040
~
~
~
95
~
~
~
~
84
17
~
16
0.04
0.04
2.3
2.2
Coefficient
of variation
~
0.81
~
~
~
0.49
~
~
~
~
0.83
0.26
~
0.24
0.54
0.38
0.95
1.0
Statistical
Difference
No
No
No
No
No
No
No
No
No
1 Multiple detection
2 Due to non-detect,
limits; this is maximum
mean and other normal
detection limit.
parameters calculated using substitution of 1A detection limit.
21
-------
Table 4-4 Sampling results and statistics pathogenic indicators and other water quality parameters
Parameter
(units)
Fecal coliform
(MPNper 100/ml)
Total coliform
(MPNper 100/ml)
E. coll
(MPNper 100/ml)
pH
Conductivity
((iS/cm)
Surfactant (mg/L
as LAS, MW 320)
Location
Influent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Influent
Effluent
Limit of
detection
1.8
1.8
1.8
1.8
2.0
2.0
0.1
0.1
0.1
0.1
1.25B
2.5B
Median
2.2xl06
5.9xl05
1.6xl07
l.SxlO7
2.2xl06
160
6.4
6.3
600
545
19
20
Lower
quartile
7.2xl05
5.9xl04
3.6xl06
1.7xl06
7.2xl05
130
5.4
5.7
450
500
13
13
Upper
quartile
4.4xl06
4.5xl06
1.6xl07A
1.6xl07
3.9X106
4.5x106
7.0
6.7
720
600
30
29
Normality
(detections
/events)
No (8/8)
No (8/8)
No (8/8)
No (8/8)
No (8/8)
No (8/8)
Yes (7/7)
Yes (7/7)
No (6/6)
Yes (6/6)
Yes (8/8)
Yes (8/8)
Mean
—
—
—
—
—
—
6.3
6.4
—
547
22
22
Standard
of
deviation
~
~
~
~
~
~
0.7
0.7
~
84
9
13
Coefficient
of variation
~
~
~
~
~
~
0.11
0.11
~
0.15
0.43
0.55
Statisti
cal
Differe
nee
No
No
No
No
No
No
A Maximum count, values recorder as > 1.6x107.
B Multiple detection limits; this is maximum detection limit.
22
-------
T-test (when data was normal) or Mann-Whitney U test (when data was not normal) indicated there was no consistent
statistical differences from the influent sampling point to the effluent sampling point, except for an increase in iron in
the effluent as demonstrated in Figure 4-2. There was more manganese as well, but this is considered to not be
statistically different although the calculated p-values, i.e., p = 0.063, was very close to p < 0.05. High levels of organic
waste can lead to higher concentrations of iron and manganese in groundwater, especially under anaerobic conditions
(Sawyer and McCarthy, 1978).
4000
3500
3000
_j
~O)
~ 2500
o
IJ3
n
c 2000
o
o
0 1500
1000
500
n
• Median
31 Non-Outlier Range
1
1
Inflluent
Effluent
Sample
Figure 4-2 Median iron concentration and non-outlier range for influent and effluent sampling locations
Several operational changes during the course of the study were identified in Table 3-2. Probably the most important
change was the increase in the size of the pretreatment system to capture oil and grease, facilitate maintenance and
reduce impacts due to surges. A testing of the observed effluent concentrations before and after the installation of the
1000-L IBC resulted in three statistically relevant t-test results, i.e. reduced chloride and sodium concentration and
increased sulfide concentration. Figure 4-3 is a box plot of the sulfide. Similar changes in the sulfide concentration were
noted in the influent concentration as well which are indicative of the larger pretreatment system driving the water to
anaerobic conditions before being discharged to the graywater garden.
23
-------
O)
o
'-4-1
TO
O
§ 2
O
• Mean
Q| Mean±SE
lMean±1.96*SE
Small
Large
Pretreatment Tank- Effluent
Figure 4-3 Box plot of sulfide concentration before and after change in pretreatment system
The phophorous had an unusually large effluent concentration for the first event (see Figure A-154). It was speculated
that the plating of the garden just a month before may have contributed to this large effluent concentration particularly
since the influent concentration was much lower in comparison. It is suspected that the general backfill soils contained
nutrients. The plantains and malangas were purchased as bare roots from nearby garden store.
A retesting of the observed nutrient concentrations excluding first effluent event did not affect normality except for
NHs. A subsequent t-test did result in a statistical difference for phosphorous effluent before and after the change in the
pretreatment (p-values < 0.05) as depicted in Figure 4-4. As noted in Table 3-2, many of the original plantings were
removed from the garden and replaced with brisselets. The brisselets were provided by the PRDNER in pots and soil
from the pots was incorporated into the garden. Additionally, more fill material was incorporated into the garden to
reduce surface ponding. This addition of soil and fill along with replacement of original plantings with slow growing
brisselet would appear to have negatively affected the phosphorous concentration in the effluent samples.
24
-------
ra
o
13
0)
o
c
o
O
Mean
Mean±SE
Mean±1.96*SE
Small Large
Pretreatment Tank {w/o first effluent event)
Figure 4-4 Box plot of phosphorous concentration before and after change in pretreatment system
A nonparametric Spearman Rank Order Correlation analyses was performed on the observed effluent concentrations
only and is presented in Appendix E (StatSoft 2011). Correlations values in red are considered significant at p-value
<0.05. Many of the observed significant correlations were an expected result. For example, sodium correlates with
chloride and specific conductance; iron, aluminum and manganese correlate with each other; ammonia and TKN
correlate; sulfide negatively correlates with pH (see Figure 4-5); and COD, TOC and oil and grease all correlate.
Additionally, TDS correlated with calcium, magnesium and conductance; the water supply for this area of Puerto Rico
has hardness between 160-180 mg/1 Ca/Mg.
There are other correlations that demonstrate the impact of the high organic loading to graywater garden. COD
correlates with aluminum, copper, calcium, magnesium and zinc. Similarly, so does oil and grease; however oil and
grease also correlated with pH.
25
-------
o
7.5
7
6
5.5
5
3/2?
•-.
A
1
/ \
/ \ \
.•"'*' \ /
-*"" X
\
»_. —
•
\"*
>/13 7/7/13 10/15/13 1/23/14 5/3/14 8/11/14 11/1
o
5!
4 I
1
a
2 j
V
13
<*-
1 ~5
0
9/14
Date
--•-- pH --•-- Sulfide
Figure 4-5 Effluent pH and sulfide concentration
26
-------
Chapter 5 Discussion and Recommendations
Overall observed concentrations to the influent tank ware similar to literature values for septic tank effluent. TDS and
chloride influent concentrations were similar to septic tank effluent (EPA, 2000), with TDS median of 480 mg/L
compared to a mean of 497 mg/L, and chloride median of 59 compared to mean of 70, respectively. Fecal coliform was
an order of magnitude larger, while nutrients, as observed for TKN and phosphorous were slightly lower in the
graywater garden influent, with median of TKN at 12mg/l and a mean of phosphorous of 3.6mg/l as compared to mean
of 44.2 and 8.6 mg/1, respectively. However, organic loadings were much higher, with graywater influent median TOC
and COD concentrations of 150 mg/L and 750 mg/L, respectively, compared to septic tank mean TOC and biochemical
oxygen demand (BOD) effluent of 47.4 and 93.5 mg/L, respectively. The BOD to COD ratio for sanitary loading
typically is 0.4 to 0.8 (Tchobanoglous and Burton, 1991) which indicates the observed COD in the influent is much
higher than the comparative literature value. This larger organic loading in the observed graywater influent
concentration appears to be driven by the addition of kitchen water as literature values for kitchen sink water can be as
high as 880 mg/L for TOC and 1460 mg/1 for BOD (EPA, 1992).
Observed concentrations for February 2014 showed a sharp increase in many of the parameters (only phosphorous had
a statistical increase in effluent concentration). This followed several operational changes including construction of
French drain, removal of plants and change to larger pretreatment system, with these latter two being the most important
actions influencing observed changes in concentrations. The May 2014 data points showed a marked decrease in fecal
coliform and E. coll populations, which potentially signals that the system was entering a new operational phase. A
look through the parameters in Appendix A indicate that many of the lowest concentrations for both the influent and
effluent came during the last two sampling events. While this potentially indicates that the plantings were providing
more benefit than originally anticipated, the decrease in both influent and effluent concentration for the final two
sampling events points to better operation and maintenance of the pretreatment and oil and grease capture as being
crucial to reducing concentrations in the graywater garden. Potassium is one of the principal requirements for bacterial
growth, therefore decreases in potassium concentrations for the influent and effluent may also indicate reduced
opportunity for bacteriological growth (Leslie Grady Jr et al., 2011). Sulfide levels increased for the same period (a
statistical increase in effluent concentration), confirming anaerobic or septic conditions. The pH range which was 6-7
at the start of the study decreased to slightly acidic conditions between 5.5 and 6 nearing the end of the sampling period.
There was statistically more iron in the effluent from graywater garden than the observed influent concentration.
Continuous discharge leads to an anaerobic environment (Tchobanoglous and Burton, 1991), and soluble forms of iron
and manganese appear during anaerobic conditions (Sawyer and McCarty, 1978). The high values of the iron and
manganese (Smol, 2008) indicate a reducing environment, as both metals are insoluble in an oxidizing environment.
Iron, magnesium and manganese levels may be explained by the natural content of these minerals in the soil of the area,
and generally, soils in the Rio Grande de Loiza watershed which are hydrothermally-altered rocks (Seiders, 1971). Iron,
aluminum, mangesium and calcium silicates comprise a significant portion of the common rock forming minerals of
27
-------
the earth's crust (Klein and Hurlbut, 1985). The decrease in iron and manganese effluent, especially during the last two
sampling events, corresponds to increased sulfide production and most likely indicates that iron and manganese oxides
are potentially precipitating as ferrous and manganese sulfide. In a subsurface drainage field, mineral precipitates, i.e.
ferrous sulfide, aluminum, iron and calcium phosphate complexes can form and then be observed in leachate (Laak
1986). None of these metals are identified as limiting nutrients needed for bacterial growth in considerable quantities
(Leslie Grady Jr et al., 2011).
Levels of organic and inorganic content in the influent and effluent sampling points may have been influenced by the
fact that initial pretreatment tanks were not cleaned out consistently. An extended accumulation of oil and grease in the
tanks, with the addition of surges flows, transferred these pollutants to the graywater garden. Significant oil and grease
content were observed in the screen tank of the original pre-treatment configuration, and as previously noted, there was
prolonged use of the overflow pipe in the screen tank. Valve use and timing were not efficiently coordinated with peak
flows from the households, which may have caused the graywater garden to receive graywater directly as it came into
the system. Accumulation of oil and grease inside the graywater garden piping may have also occurred, contributing to
periods of surface ponding and the upwelling by the influent discharge into the graywater garden. This may have created
short circuiting routing for water to traverse through the graywater garden and, as a consequence, distributing unevenly
throughout the piped sections. As such, assuming that the specific point for the effluent sampling was representative of
the entire graywater garden system performance may not be entirely accurate. It is believed that the influence of factors
such as system maintenance, surges, rainfall run-on and the heterogeneity of conditions inside the garden contributed
to the observations of high effluent concentration within the graywater garden system.
During the course of the study and following strong periods of rain, surface ponding occurred due to surges in the
graywater influent, rainfall run-on from surrounding area or poor percolation of the water into the surrounding soils as
rainfall induced saturation occurs. Water sources other than graywater, i.e. water from rain gutters, basement sump
pump discharges or surface runoff, should be routed away from any onsite wastewater treatment system (EPA, 2002).
Measures to prevent rainfall runoff (Laak, 1986) from entering the graywater garden should be performed due to the
potential to overload the infiltration capacity of the system and lead to failure.
Even though the initial testing of the infiltration capacity appeared to be sufficient, the graywater garden may not have
been big enough due to the potential to clog when accepting higher organic loads associated with water from the kitchen
sink. In septic systems, release of greases and oils to the septic disposal field can lead to reductions in infiltration
capacity (Tchobanoglous and Burton, 1991). Laak (1986) recommended a large anaerobic pretreatment for graywater
septic systems due to large concentrations of grease in graywater. Septic system disposal fields typically develop a
biomat (Tchobanoglous and Burton, 1991 or clogging layer (Laak, 1986). Future sites should be tested more thoroughly
for infiltration capacity of the disposal field as per guidance for septic disposal fields, i.e. saturated coefficient of
permeability (Tchobanoglous and Burton, 1991). The initial infiltration testing performed was more appropriate for
intermittent infiltration not continuous infiltration with high organic content. The long-term acceptance rate (LTAR)
for onsite disposal systems is more typically 0.05-0.08 cm/hr (Tchobanoglous and Burton, 1991, which would have
required an 8 by 8 m to 12 by 12m field size for approximately 150 L/ day; instead of approximately 1 m2 per capita
designed for this project, approximately 10m2 may be required to meet the aforementioned LTAR.
As demonstrated by the hilly terrain in Figure 2-3 and potential for large rain falls as demonstrated in Figure 4-1, this
site may have benefited from an upslope curtain drain (EPA, 2002) to maintain unsaturated conditions in the soils
surrounding the graywater garden. Even when remedial measures were taken, i.e. French drain, water kept surfacing,
although not ponding; however, despite periods of surface ponding and remedial action, the majority of the water
influent to the graywater system either exfiltrated or evapotranspirated during the period of this demonstration project.
Besides maintaining the grease trap, additional design and operational changes like designing the system to have more
than one garden for acceptance of graywater may help functionality. As previously noted, the surrounding soils of the
graywater garden were identified to contain clay. Alternating subsurface drainage fields helps prevent clogging of on-
site systems (EPA, 1996). This resting of the field allows for cracking in the biomat and reaggregation of clay particles
which improves infiltration capacity (Tchobanoglous and Burton, 1991).
28
-------
As previously noted, the site was subject to shading. Only a few plantain plants were left until the end of the project
(see figure Appendix C-27), while the rest of the existing vegetation was removed. The replacement plants did not
provide an immediate benefit to the study. Siting future graywater gardens in sunnier locations and providing long-term
maintenance of plants in the garden with periods of harvesting and replanting would contribute to greater
evapotranspiration and pollution removal. A graywater demonstration project performed in Colorado (EPA, 2012), had
such a high period of evapotranspiration during the summer months that there was no observed effluent from this two
stage wetland system. This graywater wetland was lined, due to strict water usage laws in Colorado, so that effluent
concentrations were returned the sanitary sewer. The Colorado graywater system had better removals overall, but had
intermittent flows and was subject to much less annual precipitation and drier, less humid conditions. During winter
periods, there was a rise in the pathogen indicators of the Colorado graywater system that would have necessitated
further treatment if the purpose was to reuse the graywater. Any graywater garden or wetland system will be subject to
climatic conditions and the interaction of vegetation with climatic conditions. For future projects in Puerto Rico, it is
recommended to use more of both Musa sp. and Xanthosoma sp. plants in sunnier locations to better evaluate the full
evapotranspiration potential.
Recommendations for improved implementation are:
• Where possible, only graywater from washing machines and shower water should be applied to the garden
systems.
• When kitchen/cooking water is to be included, an adequately sized (large) septic pretreatment with a grease
trap is required with a maintenance agreement in place.
• A larger graywater garden is required based on a LTAR (especially if kitchen wastewater is included)
• Greater distance between influent discharge points and effluent sampling points may result in longer treatment
times and lessen observed short circuiting.
• At least two gardens should be constructed so that flows can be alternated between gardens.
• Graywater garden design should take into account more features to reduce rainfall induced surcharging as
direct discharge from the garden only occurred during the rainy season.
• A greater mix of plants could be tested.
• Graywater gardens should use automated valves after first determining optimal detention times and volumes
for storage and later discharge.
• Use of upslope curtain drains to improve unsaturated conditions in surrounding soils around graywater gardens.
29
-------
Chapter 6 Conclusions
Results from the quarterly analysis of the graywater garden constructed in the Maria Jimenez community of the
municipality of Gurabo in Puerto Rico did not demonstrate statistically consistent removals of monitored parameters.
The statistical increase in observed effluent iron was most likely an indicator of excessive organic loading to the system.
Operational changes had a statistically relevant effect on the effluent. The increased size of the pretreatment system,
from two 200-L plastic drums to a 1000-LIBC, facilitated maintenance and increased surge capacity. Higher values of
sulfide were observed indicating a more stable anaerobic environment; while aluminum, iron and manganese values
were initially large, higher sulfide concentration observed after this change potentially indicated the formation of metal
sulfide precipitates.
The plants were shown to have an effect on sampled effluent concentration as there was a statistical effect on the uptake
of phosphorous by reducing effluent concentration.
Alternate designs of graywater gardens and targeting all aspects of graywater flow (e.g. larger pretreatment systems
alternating discharge to parallel gardens to avoid clogging) or targeting specific graywater flows (e.g. washing machine
water only to reduce nutrient discharges) should be pursued.
This project tested a community graywater garden as a potential remedy to the practice by rural residents of Puerto Rico
of discharging graywater directly to storm drains or the nearest receiving water. The performed monitoring revealed
some aspects of the inner workings of the graywater garden system. Surface ponding required some modifications to
the system (e.g., installation of a French drain to address rainfall induced ponding and stormwater run-on, installation
of larger pretreatment system); however, other than the rainy season there was no surface discharge, which implies
there was a reduction in release of graywater to receiving waters by the participants in this study. As such, similarly
introduced graywater gardens may have a net benefit through the reduction of direct graywater discharges to receiving
waters.
30
-------
Chapter 7 References
Boccheciamp, R. A. (1978). Soil Survey of San Juan Area of Puerto Rico. San Juan: United States Department of
Agriculture Soil Conservation Service.
Carr, M. (2009). The Water Relations and Irrigation Requirements of Banana (Musa spp.). Expl Agric, 45, 333-371.
Caspe, R. L. (2008). An Overview of Water Quality Improvement Strategies for Puerto Rico. Dimension Ingenieria y
Agrimensura, Revista del Colegio de Ingenieros y Agimensores de Puerto Rico, Ano 22, Vol. 2, pp. 7-8.
EPA (1992). Assessment of On-Site Graywater and Combined Wastewater Treatment and Recycling Systems, EPA
Report No. 832R92900, http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=200046NF.txt.
EPA (1996). Wastewater Treatment: Alternatives to Septic Systems Guidance Document. U.S. Environmental
Protection Agency, Region 9 Drinking water Program EPA Report No. 909-K-96-001
EPA (2000). Guidance for Data Quality Assessment: Practical Methods for Data Analysis. Office of Environmental
Information, U.S. Environmental Protection Agency, Washington, DC, Report No. 600/R-96/084.
EPA (2002). Onsite Wastewater Treatment Systems Manual Revised 2002. Rep. No. EPA/625/R-00-008, pp. 369,
http://nepis.epa.gov/Exe/ZyPURL.cgi?Dockev=30004GXI.txt.
EPA (2012). Graywater Treatment Using Constructed Wetlands. EPA, Office of Research and Development,
Cincinnati, Ohio, EPA/600/R-12/684, http://nepis.epa.gov/Exe/ZvPURL.cgi?Dockev=P100FSPJ.txt.
Food and Agriculture Organization (FAO) (2013). Crop Water information: Banana. Retrieved 12/1/2014, from FAO
Water: http://www.fao.org/nr/water/cropinfo banana.html
Francis, J. K. (2004). Wildland Shrubs of the United States and Its Territories: Thamnic Descriptions. San Juan:
United States Department of Agriculture Forest Service. Retrieved from
http://www.fs.fed.us/rm/pubs other/iitf gtr026.pdf on 10/1/2014.
Fundacion Puertorriquena de las Humanidades. (2009). Gurabo General Information. Retrieved 2/1/2012, from
Encyclopedia of Puerto Rico: http://www.enciclopediapr.org/ing/article.cfm?ref=09021902
Grupo Editorial EPRL (2009). Fundacion Puertorriquena de las Humanidades. Retrieved 10/22/2009 from
Encyclopedia of Puerto Rico: http://www.enciclopediapr.org/esp/print_version.cfm?ref=09021902.
Hanson, R.L., (1991). Evapotranspiration and Droughts. Retrieved from:
http://geochange.er.usgs.gov/sw/changes/natural/et/
Integrated Global Solutions. (2012). Quality Assurance Project Plan for the Implementation of a Greywater
Management Project at a Community Level on the Island of Puerto Rico. Office of Research and Development,
USEPA.
Klein, C. and C. S. Hurlbut (1985). Manual of Mineralogy (20th Edition). John Wiley and Sons, Inc.
Laak, R. (1986) Wastewater Engineering Design for Unsewered Areas (2nd edition) Technomic Publishing Co.,
Lancaster, PA.
Leslie Grady Jr, C. P, G. T. Daigger, N. G. Love and C. D. Filipe (2011). Biological Wastewater Treatment (3rd
edition). CRC Press, Boca Raton, FL.
Maidment, D. R, L. W. Mays, and V. Chow (1988). Applied Hydrology (1st edition ed.). McGraw-Hill, New York,
31
-------
NY.
Mayer, P., W. DeOreo, E. M. Opitz, J. C. Kiefer, W. Y. Davis, B. Dziegielewski and J. O. Nelson (1999).
"Residential End-uses of Water" Published by the AWWA Research Foundation and American Waterworks
Association. http://www.waterrf.org/PublicReportLibrary/RFR90781 1999 241A.pdf
Microsoft® Excel®, 2013.
McGovern, C. (2010). Residential Graywater Reuse. Innovative Energy Management Workshop (p. 16). Oahu:
USEPA Region 9. http://www.epa.gov/region9/waterinfrastructure/training/energy-
workshop/docs/2010/ResidentialGravwaterReuseHiMauiOahuKauai .pdf
National Oceanic and Atmospheric Administration, National Weather Service (NOAA-NWS) (2015). Gurabo
Normals [online image]. Retrieved from http://www.srh.noaa.gov/sju/?n=climo_gurabo#GuraboAES
Puerto Rico Environmental Quality Board (1983). Puerto Rico Underground Injection Control Rules R-83-23-1. San
Juan: Puerto Rico Environmental Quality Board.
Puerto Rico Environmental Quality Board. (2007). Total Maximum Daily Loads (TMSL), Rio Grande de Loiza
Watershed: Fecal Coliform, Copper, Biochemical, Oxygen Demand, Ammonia. Water Quality Area. San Juan:
Puerto Rico Environmental Quality Board.
Puerto Rico Planning Board (PRPB). (2010). Puerto Rico Aerial Orthophotography, October 31, 2009 through
January 27, 2010 (Web Map Service). Retrieved from
http://gis.otg.pr .gov/ArcGIS/services/Ortofotos/Orthophoto2009_10/mapserver/WMSServer.
Quinones, F. (1980). Limnology of Lago Loiza: Puerto Rico (Water Resources Investigation Report No. 79-97).
United States Geological Survey, Water Resources Division, San Juan.
Sawyer, C. N. and P. L. McCarty (1978). Chemistry for Environmental Engineers. MacGraw-Hill Publishing Co. 3rd
Ed.
Seiders, V. M. (1971). Geologic Map of the Gurabo Quadrangle, Puerto Rico. United States Geological Survey.
SM (1998). Standard Methods for the Examination of Water and Wastewater. Edited by A. D. Eaton, L. S. Clesceri,
and A. E. Greenburg. Published Jointly by American Public Health association, American Water Works
Association and Water Environment Federation, 20th Edition.
StatSoft, Inc. (2011). STATISTICA (data analysis software system), version 10. www.statsoft.com.
Tchobanoglous, G. and F. Burton (1991). Wastewater Engineering: Treatment, Disposal and Reuse. McGraw-Hill,
Inc., 3rd Ed.
Tetra Tech Inc. (2011). Fecal Coliform Bacteria Total Daily Maximum Loads for Assessment Units in the
Commonwealth of Puerto Rico. Environmental Planning and Protection, USEPA Region 2. San Juan: Puerto Rico
Environmental Quality Board.
United States Geological Survey (USGS) (1982). Gurabo Quadrangle 7.5 minute series [map], (ca. 1:20,000.)
Retrieved from
http://gis.otg.pr.gov/ArcGIS/services/Geodata services/AmbientalTopografia/mapserverAVMSServer
Watershed Stewardship Program (WSP) (2011). Rio La Plata and Rio Grande De Loiza Watersheds Final Report,
Phase III, Puerto Rico Watershed Stewardship Program, November, 2011.
32
-------
Appendix A Graphs of Data
The following figures are presented in the same order of Tables 4-1 through 4-4 and show the raw data for parameters
that had detected values for both the influent and effluent sampling locations. Error bars, where applicable, are derived
from the standard error and are presented for data exhibiting normality. Only detected values are presented with the
exception of sulfide, i.e., one non-detect which used 1A detection was used to complete the graph.
— •— Influent
Effluent
16000
14000
< 12000
M
-JT 10000
o
'•£ 8000
u
O
6000
4000
2000
Q
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14
Date
Figure A-1 Parameter concentration profiles for aluminum
9/13/14
33
-------
1000
900
800
1 70°
~ 600
'•§ 500
4^
£ 400
o
O 300
200
100
Influent
Effluent
LOD(ug/L)
-•••- Influent
•Effluent
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-2 Parameter concentration profiles for barium with limit of detection
450000
400000
350000
~S 300000
§ 250000
+; 200000
c 150000
o
u
100000
50000
5/1/13 8/9/13 11/17/13 2/25/14
Date
Figure A-3 Parameter concentration profiles for calcium
6/5/14 9/13/14
34
-------
45
40
35
30
I 25
*-«
(0
-g 20
10
:
0
Influent
•Effluent
- ••- Influent
Effluent
800
700
< 600
•^r SOD
'•§ 400
£ 300
V
o 200
100
•LOD(ng/L)
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-4 Parameter concentration profiles for chromium with limit of detection
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-5 Parameter concentration profiles for copper
35
-------
- ••- Influent
Effluent (ng/L)
12000
union
DO
3. 8000
c
o
£ 6000
I
S 4000
2000
5/1/13 8/9/13 11/17/13 2/25/14
Date
Figure A-6 Parameter concentration profiles for iron
6/5/14 9/13/14
- •- Influent
Effluent
.M 11)011
18000
~ 16000
IjS 14000
~c 12000
o
ts 10000
i_
8000
6000
4000
2000
0
gj
u
c
o
LJ
5/1/13 8/9/13 11/17/13 2/25/14
Date
Figure A-7 Parameter concentration profiles for magnesium
6/5/14 9/13/14
36
-------
- •- Influent
Effluent
900
800
•3 7oo
^
3. 600
0 500
s
10
£ 400
c
u 300
3 200
100
0
5/1/13
8/9/13
11/17/13
2/25/14
Date
6/5/14
9/13/14
Figure A-8 Parameter concentration profiles for manganese
15
. •
§ 20
*^
2
£ 15
J 10
—•-- Influent
Effluent
LOD (ng/L)
'
i
'
/
i
i
i
i
i
i
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-9 Parameter concentration profiles for silver with limit of detection
37
-------
- •- Influent
Effluent
3
3500
3000
2500
2000
•
~ 1500
§ 1000
U
500
I _
""** — ....
J- ~ — — ..
^
0
5/1/13
8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-10 Parameter concentration profiles for zinc
- ••- Influent
Effluent
11/17/13 2/25/14
Date
Figure A-11 Parameter concentration profiles for ammonia
6/5/14 9/13/14
38
-------
0.6
0.5
0.4
c
o
°-3
LOD (ug/L)
0>
0.2
0.1
0
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-12 Parameter concentration profiles for nitrate and nitrite with limit of detection
- •- Influent
Effluent
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-13 Parameter concentration profiles for total Kjeldahl nitrogen
39
-------
- • - Influent
Effluent
16000
14000
< 12000
— 10000
c
o
'•«= 8000
§ 6000
u
o 4000
u
2000
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14
Date
9/13/14
Figure A-14 Parameter concentration profiles for potassium
- •- Influent
Effluent
B
/
f 5
O
I4
u
0
5/1/13
8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-15 Parameter concentration profiles for phosphorous
40
-------
- ••- Influent
Effluent
4000
3500 \
3000
— 2500
o
«= 2000
x_
1500
u
O 1000
u
500
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14
Date
Figure A-16 Parameter concentration profiles for chemical oxygen demand
9/13/14
- •- Influent
Effluent
10000
< 1000
M
E
100
u
o 10
o
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-17 Parameter concentration profiles for oil and grease in log scale (due to large initial concentration)
41
-------
— •• - Influent
Effluent
400
350
< 300
M
§250
1 200
§ 150
u
0 100
U
50
5/1/13 8/9/13 11/17/13 2/25/14
Date
6/5/14
9/13/14
Figure A-18 Parameter concentration profiles for total organic carbon
- •- Influent
Effluent
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14
Date
Figure A-19 Parameter concentration profiles for total dissolved solids
9/13/14
42
-------
- ••- Influent
Effluent
3000
^ 2500
_i
^ 2000
c
'•g 1500
S 1000
c
o
U
500
0
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-20 Parameter concentration profiles for total suspended solids
— •- Influent
Effluent
350
300
250
200
j= 150
4)
c 100
5
50
0
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-21 Parameter concentration profiles for chloride
43
-------
- •- Influent
Effluent
300
250
E, 20°
c
O
*= 150
I
g 100
c
o
u
50
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-22 Parameter concentration profiles for sodium
- •- Influent
Effluent
0.18
0,16
CT 0.14
^5
£ 0.12
§ 0-1
| 0.08
c
8 0.06
c
u 0.04
0.02
5/1/13
8/9/13 11/17/13 2/25/14
Date
6/5/14 9/13/14
Figure A-23 Parameter concentration profiles for fluoride
44
-------
- •- Influent
Effluent
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-24 Parameter concentration profiles for sulfide
- •- Influent
Effluent
l.OE+08
l.OE+07
l.OE+06
E
O1 l.OE+05
o
Jj l.OE+04
Q.
Z 1.0Et03
Q.
:
l.OE+01
l.OE+00
5/1/13
8/9/13 11/17/13 2/25/14
Date
6/5/14 9/13/14
Figure A-25 Most probable number profiles for fecal coliform
45
-------
- • - Influent
Effluent
l.OE+08
l.OE+07
l.OE+06
E
e> l.OE+05
o
fc l.OE+04
a,
Z l.OE+03
: l.OE+02
l.OE+01
l.OE+00
5/1
^^^^^^^-l ' " ' " **
/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-26 Most probable number profiles for total coliform
— •— Influent
Effluent
l.OE+08
l.OE+07
l.OE+06
E
o" l.OE+05
O
j; l.OE+04
Q.
Z l.OE+03
a.
'' l.OE+02
l.OE+01
l.OE+00
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14
Date
9/13/14
Figure A-27 Most probable number profiles for Escherichia coli
46
-------
- ••— Influent
Effluent
5/1/13 8/9/13 11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-28 Profiles for pH
1600
1400
,u 1200
;T 1000
.£" 800
- ••- Influent
Effluent
•q
O 400
u
200
0
5/1/13
8/9/13
11/17/13 2/25/14 6/5/14 9/13/14
Date
Figure A-29 Profiles for conductivity
47
-------
- ••- Influent
Effluent
8/9/13
11/17/13 2/25/14
Date
6/5/14 9/13/14
Figure A-30 Parameter concentration profiles for surfactant
48
-------
Appendix B As-built Drawings
49
-------
IMPLEMENTATION OF A GREYWATER MANAGEMENT
PROJECT AT A COMMUNITY LEVEL IN THE ISLAND
OF PUERTO RICO
AS-BUILT DRAWINGS
EP-11-C-000217
GURABO, PUERTO RICO
PRESENTED TO: U.S. ENVIRONMENTAL PROTECTION AGENC
DATE: JUNE 2014
fe INTteMAtxX>OlottAI.
INDEX OF DRAWINGS
Sheet No. Description Drawing No.
INFORMATIVE DRAWWOS
TOPOGRAPWC DRAMNQS
S1E DRAWINGS
Figure B-1 As-built drawings for the graywater garden, sheet one
50
-------
•
1 iMUMlt U
•rJwMwn
.0™,
EOHATt D GUOSA!. SOLUTIONS
DESIGN. CONSTRUCTION AND MONITORING OF A COMMUNITY BASED GRAY GARDEN
FLOW DIAGRAM
Figure B-2 As-built drawings for the graywater garden, sheet two
51
-------
CECRE"
• lit' i
DESIGN CONSTRUCTION AND MONITORING Of A COMMUNITY BASED GRAY GARDEN
TOPOGRAPHIC SURVEY
Figure B-3 As-built drawings for the graywater garden, sheet three
52
-------
SMS,.
CECRE"
i ii i ,'.< ;;.,-,. ] |<
DESIGN, CONSTRUCTION AND MONITORING OF A COMMUNITY 6ASED GRAY GARDEN
SITE PIPING LAYOUT
Figure B-4 As-built drawings for the graywater garden, sheet four
53
-------
,
,
GREASE TRAP DETAIL
s> -CI0
> "-^
' i,™
i uir*i.
"
IRRI
IRRIGATION PlPt DCTAi.
DRAINAGE PIPE DETAIL
•
•
INTEGRATED Ot-OBAL SOLUTIONS
DESIGN CONSTRUCTION AND MONfTORINGOf A COMMUNITY BASED GRAY GARDEN
SITE DETAILS
Figure B-5 As-built drawings for the graywater garden, sheet five
54
-------
Appendix C Project Images
55
-------
Figure C-1 Maria Jimenez graywater garden site before excavation
Figure C-2 Piping scheme in the excavation and gravel in the bottom
56
-------
Figure C-3 Filling in the excavation with dirt
Figure C-4 Pipe installation of the graywater garden
57
-------
' .""'•" ' "' ':•'.--„** .
Figure C-5 Back view of the recently-filled graywater garden
Figure C-6 Front view of the recently-filled graywater garden
58
-------
Figure C-7 Side view of the recently-filled graywater garden with germinating plants
*'; -''' '*
f -
Figure C-8 Side view of the recently-filled graywater garden with germinating plants
59
-------
Figure C-9 View of the garden from street lamp post
Figure C-10 Garden with some developed vegetation
60
-------
Figure C-11 Pre-treatment units of the graywater garden
Figure C-12 Vegetation grown in the garden
61
-------
Figure C-13 Vegetation grown in the garden, alternative angle
Figure C-14 Plantain trees and tannia plants growing
62
-------
Figure C-15 High view of the garden with several plantain trees grown
Figure C-16 Local street view leading to the site
63
-------
Figure C-17 view of the garden before cutting grass
Figure C-18 View of the garden after cutting grass
64
-------
Figure C-19 Accumulation of water downstream of the system
Figure C-20 Entrance of the site and stormwater pipe under the street
65
-------
Figure C-21 Surface ponding in the garden
Figure C-22 Installation of French drain
66
-------
Figure C-23 Covering up French drain
Figure C24 Covering up French drain with stone
67
-------
Figure C-25 View of the graywater garden with copious vegetation
Figure C-26 View of the graywater garden with grown Musa spp. and Xanthosoma sp. plants
68
-------
Figure C-27 View of the graywater garden with remaining plantain (Musa spp.) plants
69
-------
Appendix D Sampling Documentation
70
-------
u s
EPA REGION 2 LABORATORY A N A 1_ Y S
REQUEST FORM
SUR\1 ¥ NAME: Implementation of Crewvater Management at the C ommunitv L e-j-el in PR DATE OF RE QUE SI: February 07,2014
RE QUE SI OR: Evelyn Hnerlas AFFILIATION": CEPD PROGRAM: Water
E-MAIL ADDRE SSFOR FINAL ELECTRONIC REPORT: _huerta5.erelyng epa.gov
ADDRE SS FOR FINAL HARD COPY REPORT: City Vig.v Plaza 3-Suite 7000, 48 RD. J65 Km. 12, Ouavnabo. P.R. i»%S-9069
SAMPLING DATE S: February 07, 2014 ; ARRIVAL DATE S: ; ARRIVAL TIME:
ME T H OD OF SHIPM E NT: UPS
_; RE QUE STED TURNAROUND TIME:
#
Samp
Analvte
Mitrix
SANITARY
1
t
1
1
•1
ACIDITY
ALKALINITY.
TOTAL
AMMONIA
ASPHALTENES
*BOD,SDAY
*CBOD,SDAY
CHLORIDE
COD
* COL OR
CORROaVJTY
CYANIDE
CYANIDE.
AMEN. TO CL
CY'ANTDE,
WAD iFREE]
*DO(disolfed
FLUORIDE
*HES.
CHRO\DUM
IGNITAB1L1TY
*^fflAS
(surfactants)
*NJ IRATE
NITRATE+
NITRITE
* NITRITE
A
A
A
A
A
Samp
2
a
a
a
^
t
1
^
Anal\te
O1L& GREASE
ORTHO-
PHOSPHATE
PETROLEUM
HYDROC.
TOTAL
PHENOLICS
PHOSPHORUS
SOLIDS, m
105degC
a>Ecinc
CONDI" CTANCE
SULFATE
SULHDE
SCLF1DE.
UNIONIZED
SL'LFUR
TDS
TKN
TOC
TSS
*TURBID1TY
\TSCOSLTY
JAte:
A
A
A
A
A
A
A
A
MICROBIOLOGY
*CLOSIREOR"M
PERFRINGENS
* CRYPT Qt
GIARD1A
*E-COLI
*ENTERO-
COCCUS.MF
*EdAL REQUE SIS (turnaround time, additional anahles, etc):
Samp
Analyte
*ENTERO-
COCCUS.MPN
*F-COLIFORM.
MF
*F-COLIF.MPN
*HPC
* SALMONELLA
*T-COLIF.
COLILERT
*T-COLIF.MF
*T-COLIF.MPN
Nfctris
METALS
T"
Recife Metals:
list under Special
Requests
LEAD (inDW)
MEI.U.S-
aUDGE
H.1RDNESS
MET.US
HM9IING
MEI.U.STAL
CDW leveb)
METAL STAL
MET.USTCLP
A
ORGANICS
HALOACIUC
ACIDS
HERfflQDES
NVOA TCI
M'OA TCLP
Samp
Analvte ^
PAHs
PCBsTCL
PCBs TSCA
PESTICIDES
ICL
PESTICIDES
TCLJ
TRIBAL O-
ME THANES
V«V TCL
VOA TCLP
VOA TRACE
(DW lev els)
^trix
BIOLOGY
Y«
Mm
Y/N
Y/N
Y«
#EFFLUENT
ToxicrrY.icuiE
#EFFLUENT
TOXICITY.
CHRONIC
#SD.TC&FREgi
WATER
tfmics.
MiREtWATER
GRAIN
SIZE:
PIPE I
METHOD S
(PLUMB 1981)
HYDROMETER
METHOD S
ASTM-I12D-63
WTOTAL
SAKD?
5 SAND
FR-iCTIONS?
WTOTAL
HNES?
(sflt+clay)
WTOTAL
aLi:
« TOTAL
CLAY;
RE PORTING REQUIRE ME NTS: (attach separate sheet, if more room
needed):
RE QUE SI APPROVED BY:
DATE APPROVED:
Figure D-1 Example of a Laboratory Analysis Request form used for USEPA Region 2 laboratory in Edison, New Jersey
71
-------
d
i
iTl
[^
fS
m
m ii
H m
UJ -J
zi a
0 LJJ
g
8 t-
f • M
a 1 G
pi BJ
. |_ • ui 0
i § 1 5 ^
10 u i
UJ ,. «
3 It c*
g 0 g
1 I 1 . 3
o < S
g I 8 ..
> «8 S * H
•- Q. ~ g d
i j i s s
3 ro j
3 fc £
< of •
£ S 1
£ § 5
Z III tn ° «:
g D § S S
| uj E B f
* -> * d £
m 1 I
3i
s „
•* A
i i
e
3
d
a
1
*i
ui
i
§
S P
a
LJJ
h-
t/>
LJJ
Zl
0
LJJ
E
05
S
1
4
«
P
V3
UJ
t-
D
UJ
O
t-
£
RINFOR
CONTAINE
0
<
K
O
u.
z
UJ
Q.
1
Z
g
i s
'r 1
S 1"
UJ
2
d
n:
O
8.: S.
TYPE
PHESEftVAfl
O
J LJJ
if
•
* UJ
si s K
jig
i
§
9
I
1
u
3| si
LU
Q.
t
UJ ,
LJJ
Q.'
; S!
,,
]
^
if P
5 i£ o '
*
y x o
ID S OT
*
0
f\
S,
S
!
O
Z
o
P
u
3
£
i
_l
o_
w
UJ
Z
1
UJ
£
£
VJ
D
g
(J
LJJ
ce
>-
Q
O
H
W
3
O
|
•ir
pi:
i
O
™
•-S
J
S
u.
J^
—
s
1
u
L-l
J?
•a
ii
o:
J3
»J
O
LJJ
S
1
1
a
K
d
Ijj
s
i
i
c>
ac
"c
p
D
(J
O
T3
W
S
ti
O
^
- ' g)
g £
M 0
c
a
^
I
w
ii
2 fQ
ll
,ffl cn
_ «
fl] !b
> to
t cf
< UJ
• "5
Ii C
f" C
^ 2
« £
II
-** .«
It
IL -J
d o
Figure D-2 Example of a Chain of Custody form used for the local laboratory
72
-------
The third sampling event (SE3) for the greywater management project will be performed on Tuesday,
December 03, 2013, in the Maria Jimenez community in Hato Nuevo, Gurabo. The SE3 will follow the
sampling as established in Addendum 1 of the Quality Assurance Project Plan (QAPP) for this project.
Samples will be taken in two places, the system's influent (S1) and the effluent (S2). The influent
sample is taken inside the first drum, whereas the effluent is taken from one of the garden's sampling
pipes, to the end of the system. Two samples will be taken in each point, one to send to the EPA
Region 2 laboratory and another one to send to a local laboratory for those parameters that have short
holding times. These will be divided into five (5) bottles for certain parameters to be analyzed, as seen
in the following tables.
TABLE 1. Bottle summary for samples S1-EPA and S2-EPA to be sent to EPA's Region 2 laboratory.
NAME
PARAMETERS
SAMPLE BOTTLE TYPE PRESERVATIVE
Bottle 1
Bottle 2
Bottle 3
Bottle 4
Bottle 5
Total suspended solids (TSS),
Filterable residue (TDS),
Chloride, Fluoride. Conductivity
Oil & grease
Sulfide
Chemical oxygen demand (COD),
Total organic carbon (TOC),
Total Kjeldahl nitrogen (TKN)
Nilrate+Nitrile, Ammonia
Total phosphorus
Metals - TAL
1000 ml
3000 ml
250 mL
500 mL
250 mL
1000-rnLHDPE
1000-mL Amber
250-mL HOPE
500-rnL HOPE
250-mL HOPE
Ice
Hydrochloric acid
Zinc acetate /
Sodium hydroxide
Sulfuric acid
Nitric acid
Following is a table with a detail of the various parameters that will be measured for the event and their
respective reporting limits for EPA Region 2
Figure D-3 Example of the second page of the Sampling and Analysis Plan presented two weeks before each sampling event
73
-------
TABLE 2 . Parameters and respective reporting limits for samples S1-EPA and S2-EPA to be sent to
ERA's Region 2 laboratory.
PARAMETER
r RESE R VAT IV E
VOLUME, mL BOTTLE TTPE REGION Z LIMITS, mg/L
TSS
TDS
Fluoride
Chloride
Conductivity
Oil & grease
SulfidB
Ice
Ice
Ice
Ice
Ice
Hydrochloric acid
Zn acetate + Na hydroxide
COD Su If uric acid
Nitrate + Nitrite
TKN
Su If uric acid
Sulluric acid
Ammonia Sulturicacid
Total
phosphorus
TOC
Metals -TAL
Sulfuricacid
Su If uric acid
Nitric acid
400
-
100
50
100
3000
250
50
100
100
100
50
50
250
HOPE
HOPE
HOPE
HOPE
HOPE
1-L amber glass
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HOPE
HD 'E
10
10
0.1
1.0
0.1 uS/CM
5.0
.01
20.0
.05
0.1
.05
.05
1.0
Varies
Figure D-4 Example of the third page of the Sampling and Analysis Plan presented two weeks before each sampling event
74
-------
TABLE 3 . Bottle summary for samples S1 and S2 to be sent to EQLab's laboratory.
NAME
Bottle 1
Bottle 2
Bottle 3
Bottle 4
PARAMETERS SAMPLE BOTTLE TYPE
Filterable residue (TDS),
Chloride, Fluoride, Surfactants
Nitrate+Nitrite, Total phosphorus,
Total Kjeldahl nitrogen (TKN)
Total coliform (MTF-WW), Eco/i' (MTF-
WW), Fecal coliform (MTF-WW)
Total organic carbon (TOC)
1 500 ml_
500 mL
300 ml
40 ml
2000-mL HOPE
500-mLHDPE
300-mL HOPE
Amber vial
PRESERVATIVE
Ice
Sulfuric acid
Sodium
thiosulfate
Sulfuric acid
TABLE 4 . Parameters and respective reporting limits for samples S1 and S2 to be sent to EQLab's
laboratory.
PARAMETER
TDS
PRESERVATIVE
.' •:..
BOTTLE TYPE
HOPE
MDL LIMITS
5 mg/L
E Fluoride
Chloride
Surfactants
Nilrate+Nitrite
Tola! Phosphorus
Total Kjeldahl Nitrogen
Total Coliform
E. coli
Fecal Coliform
TOC
Ice
Ice
Ice
Sulfuric acid
Sulluric acid
Sulfuric acid
Sodium thiosulfate
Sodium thiosulfate
Sodium thiosulfate
Sulfuric acid
HOPE
HOPE
HOPE
HOPE
hL'i:h
HOPE
HOPE
HOPE
HOPE
Amber vial
0.01 mg/L
3 mg/L
0.5 mg/L as LAS
0.01 mg/L
0.25 mg/L
0.2 mg/L
1.8MPN/1QOmL
2.0MPN/100mL
1.8MPN/100mL
0.5 rng/L
Figure D-5 Example of the fourth and last page of the Sampling and Analysis Plan presented two weeks before each sampling event
75
-------
Appendix E Results of Spearman Rank Order Correlation
76
-------
A Spearman rank R correlation test was performed in Statistica 10 (StatSoft, Inc., 2011) on the effluent concentrations. A high correlation value
between specific parameters for all sampling dates may imply that there were similar cause and effect for concentration changes, if positively
correlated. If negative correlation are large, results may also be linked. Marked correlations or red values imply statistically significant results.
Variables on the left were tested against each column (e.g. COD is significantly, positively correlated with TKN, O&G, TOC and TSS).
Variable
NH4 as N
COD
Chloride (Cl)
Floride (F)
TKN
Oil&Grease (O&G)
TOC
TDS
TSS
Sodium (Na)
Specific conductance (SC)
Sulfide (S)
FCol
TCol
£ co/;
Iron (Fe)
Aluminum (Al)
Calcium (Ca)
Copper (Cu)
Magnesium (Mg)
Manganese (Mn)
Phosphorous (P)
Potassium (K)
Zinc (Z)
Ni+Ni
PH
Sodium (Na)
Surfactants (Sur)
Spearman Rank Order Correlations (Nutrients.sta)
MD pairwise deleted
Marked correlations are significant at p <. 05000
NH4
1.00
0.29
-0.18
0.77
0.58
0.21
0.28
-0.08
0.26
-0.00
-0.00
0.19
-0.65
0.27
-0.65
0.43
0.26
0.20
0.28
0.55
0.45
0.60
0.16
0.36
-0.09
-0.07
COD
0.29
1.00
0.28
0.28
0.78
0.87
0.85
0.42
0.80
0.07
0.38
0.07
-0.27
0.15
-0.27
0.46
0.69
0.87
0.92
0.74
0.36
0.40
0.19
0.74
0.39
0.38
Cl
-0.18
0.28
1.00
-0.02
0.19
0.21
0.06
0.35
0.41
0.91
0.76
-0.41
0.51
-0.08
0.52
-0.08
0.36
0.36
0.17
-0.00
0.02
0.05
-0.21
0.26
-0.15
0.32
-0.00 0.07| 0.91
-0.47
0.25
F
0.77
0.28
0.02
1.00
0.67
0.39
0.45
0.17
0.21
0.17
-0.00
0.13
-0.58
-0.15
-0.57
0.36
0.22
0.20
0.56
0.45
0.41
0.28
0.05
0.38
0.04
0.02
0.17
TKN
0.58
0.78
0.19
0.67
1.00
0.65
0.85
0.17
0.54
0.28
0.13
0.37
-0.32
0.10
-0.31
0.42
0.39
0.48
0.88
0.58
0.28
0.63
0.41
0.44
0.28
0.13
0.28
O&G
0.21
0.87
0.21
0.39
0.65
1.00
0.73
0.52
0.81
-0.00
0.43
-0.11
-0.30
0.01
-0.31
0.29
0.60
0.86
0.81
0.76
0.27
0.22
0.11
0.76
0.43
0.54
-0.00
TOC
0.28
0.85
0.06
0.45
0.85
0.73
1.00
0.39
0.54
0.14
0.14
0.37
-0.22
0.14
-0.21
0.28
0.32
0.52
0.78
0.65
0.08
0.37
0.52
0.32
0.42
0.43
0.14
0.29 -0.32 0.10 0.42 0.32
TDS
-0.08
0.42
0.35
0.17
0.17
0.52
0.39
1.00
0.34
0.39
0.68
-0.48
-0.09
-0.22
-0.09
0.14
0.44
0.72
0.29
0.78
0.14
-0.08
0.54
0.26
-0.09
0.62
0.39
0.41
TSS
0.26
0.80
0.41
0.21
0.54
0.81
0.54
0.34
1.00
0.30
0.54
-0.29
-0.14
0.12
-0.12
0.67
0.89
0.80
0.77
0.69
0.62
0.14
-0.07
0.85
0.23
0.62
0.30
Na
-0.00
0.07
0.91
0.17
0.28
-0.00
0.14
0.39
0.30
1.00
0.74
-0.43
0.33
-0.07
0.35
0.13
0.26
0.06
0.03
-0.02
0.24
0.11
-0.04
0.07
-0.23
0.42
1.00
SC
S
-0.00 0.19
0.38 0.07
0.76 -0.41
-0.00 0.13
0.13 0.37
0.43
0.14
0.68
0.54
0.74
1.00
-0.84
0.25
-0.29
0.27
0.07
0.60
0.65
0.14
0.50
0.25
0.17
-0.04
0.45
-0.05
0.75
0.74
0.43 0.20 0.47
-0.11
0.37
-0.48
-0.29
-0.43
-0.84
1.00
-0.07
0.51
-0.08
-0.25
-0.61
-0.44
0.05
-0.28
-0.55
0.39
0.35
FCol
-0.65
-027
0.51
-0.58
-0.32
-0.30
-0.22
-0.09
-0.14
0.33
0.25
-0.07
1.00
0.08
1.00
-0.26
-0.11
-0.18
-0.46
-0.33
-0.28
-0.42
-0.22
TCol
0.27
0.15
-0.08
-0.15
0.10
0.01
0.14
-0.22
0.12
-0.07
-0.29
0.51
0.08
1.00
0.07
0.13
-0.02
-0.00
-0.09
0.27
-0.17
0.28
0.15
E coli
-0.65
-0.27
0.52
-0.57
-0.31
-0.31
-0.21
-0.09
-0.12
0.35
0.27
-0.08
1.00
0.07
1.00
-0.26
-0.09
-0.17
-0.44
-0.34
-0.27
-0.42
-0.24
-0.44 -0.44| -0.24| -0.42|
0.24
-0.62
-0.43
-0.26
-0.11
0.06
-0.25
-0.18
0.33 -0.07
0.29
-0.11
0.06
0.35
-0.02 0.28
77
-------
Spearman Rank Order Correlations (Nutrients.sta) MD pain/vise deleted Marked correlations are significant at p <.05000
Fe
Al
Ca Cu
Mg
Mn
K
Na+Ni pH
Na Sur
NH4 as N
|0.43 |0.26 |0.20 |0.28 |0.55 |0.45 |0.60 |0.16 |0.36 |-0.09 |-0.07 |-0.00 |-0.47
COD
|0.46 |0.69 |0.87 |0.92 |0.74 |0.36 |0.40 |0.19 |0.74
|0.38 |0.07 |0.25
Chloride (Cl)
I-0.08 |0.36 |0.36 |0.17 |-0.00 |0.02 |0.05
|0.26
|0.32 |0.91 |0.29
Floride (F)
|0.36 |0.22 |0.20 |0.56 |0.45 |0.41 |0.28 |0.05 |0.38 |0.04 |0.02 |0.17 |-0.32
TKN
|0.42 |0.39 |0.48 |0.88 |0.58 |0.28 |0.63 |0.41 |0.44 |0.28 |0.13 |0.28 |0.10
Oil&Grease (O&G)
|0.29 |0.60 |0.86 |0.81 |0.76 |0.27 |0.22
|0.76 |0.43 |0.54 |-0.00 |0.42
TOC
|0.28 |0.32 |0.52 |0.78 |0.65 |0.08 |0.37 |0.52 |0.32 |0.42 |0.43 |0.14 |0.32
IDS
|0.14 |0.44 |0.72 |0.29 |0.78 |0.14 |-0.08 |0.54 |0.26 |-0.09 |0.62 |0.39 |0.41
TSS
|0.67 |0.89 |0.80 |0.77 |0.69 |0.62 |0.14 |-0.07 |0.85 |0.23 |0.62 |0.30 |0.43
Sodium (Na)
|0.13 |0.26 |0.06 |0.03 |-0.02 |0.24
|-0.04 |0.07
-0.23 |0.42 |1.00 |0.20
|Specific conductance (SC)) |0.07 |0.60 |0.65 |0.14 |0.50 |0.25 |0.17 |-0.04 |0.45
-0.05 |0.75 |0.74 |0.47
Sulfide (S)
|-0.25 |-0.61 |-0.44 |0.05 |-0.28 |-0.55 |0.39 |0.35 |-0.44 |0.24 |-0.62 |-0.43 |-0.26
FCol
|-0.26 |-
|-0.46 |-0.33 |-0.28 |-0.42 |-0.22 |-0.44
-0.11 |0.06 |0.33 |0.29
TCol
|0.13 |-0.02 I-O.OO |-0.09 |0.27
|0.28 |0.15 |-0.24
-0.25 |-0.18 |-0.07 |-0.02
Ecoli
|-0.26 I-0.09 |-0.17 |-0.44 |-0.34 |-0.27 |-0.42 |-0.24 |-0.42
-0.11 |0.06 |0.35 |0.28
Iron (Fe)
|1.00 |0.79 |0.38 |0.52 |0.50 |0.88
|-0.09 |0.47
-0.23 |0.45 |0.13 |0.14
Aluminum (Al)
|0.79 |1.00 |0.77 |0.65 |0.63 |0.78
|-0.22 |0.78
-0.19 |0.68 |0.26 |0.41
Calcium (Ca)
|0.38 |0.77 |1.00 |0.68 |0.79 |0.38 |0.14 |0.16 |0.80
-0.04 |0.48 |0.06 |0.47
|Copper (Cu)
|0.52 |0.65 |0.68 |1.00 |0.52 |0.41 |0.31 |0.04 |0.74 |0.42 |0.28 |0.03 |0.32
Magnesium (Mg)
|0.50 |0.63 |0.79 |0.52 |1.00 |0.38 |0.29 |0.37 |0.48
0.17 |0.70 |-0.02 |0.31
Manganese (Mn)
|0.88 |0.78 |0.38 |0.41 |0.38 |1.00 |-0.09 |-0.29 |0.57
0.12 |0.45 |0.24 |0.18
Phosphorous (P)
|0.14 |0.31 |0.29 |-0.09 |1.00 |0.52 |0.16 |0.02 |-0.26
|-0.25
Potassium (K)
|-0.09 |-0.22 |0.16 |0.04 |0.37 |-0.29 |0.52 |1.00
-0.07 |-0.09 |-0.04 |0.13
Zinc (Z)
|0.47 |0.78 |0.80 |0.74 |0.48 |0.57 |0.16
1.00
0.13 |0.36 |0.07 |0.38
Ni+Ni
I-0.23 |-0.19 |-0.04 |0.42 |0.17 |0.12 |0.02 |-0.07 |0.13 |1.00 |0.65 |-0.23 |0.42
|pH
|0.45 |0.68 |0.48 |0.28 |0.70 |0.45 |-0.26 |-0.09 |0.36
0.65 |1.00 |0.42 |0.47
Sodium (Na)
|0.13 |0.26 |0.06 |0.03 |-0.02 |0.24
|-0.04 |0.07
-0.23 |0.42 |1.00 |0.20
Surfactants (Sur)
|0.14 |0.41 |0.47 |0.32 |0.31 |0.18 |-0.25 |0.13 |0.38 |0.42 |0.47 |0.20 |1.00
78
------- |