A Deep Dive Into Natural Swimming Pool Filtration: Living Walls As Technical Wetland Filters

Utah State University DigitalCommons@USU

All Graduate Theses and Dissertations Graduate Studies

12-2020

A Deep Dive Into Natural Swimming Pool Filtration: Living Walls as Technical Wetland Filters

Anna Farb Utah State University

Follow this and additional works at: https://digitalcommons.usu.edu/etd

Part of the Environmental Design Commons, and the Landscape Architecture Commons

Recommended Citation Farb, Anna, "A Deep Dive Into Natural Swimming Pool Filtration: Living Walls as Technical Wetland Filters" (2020). All Graduate Theses and Dissertations. 7970. https://digitalcommons.usu.edu/etd/7970

This Thesis is brought to you for free and open access by the Graduate Studies at DigitalCommons@USU. It has been accepted for inclusion in All Graduate Theses and Dissertations by an authorized administrator of DigitalCommons@USU. For more information, please contact [email protected]. A DEEP DIVE INTO NATURAL SWIMMING POOL FILTRATION:

LIVING WALLS AS TECHNICAL WETLAND FILTERS

Anna Farb

A thesis submitted in partial fulfillment of the requirements for the degree

MASTER OF LANDSCAPE ARCHITECTURE

Approved:

______David Anderson, M.L.A. Phillip S. Waite, M.A.A. Major Professor Committee Member

______Wolfram Kircher, Ph.D. D. Richard Cutler, Ph.D. Committee Member Interim Vice Provost of Graduate Studies

UTAH STATE UNIVERSITY Logan, Utah

2020

iii

ABSTRACT

A deep dive into natural swimming pool filtration: Living walls as technical wetland filters

Anna Farb, Master of Landscape Architecture

Utah State University, 2020

Major Professor: David Anderson Department: Landscape Architecture and Environmental Planning

Vertical greenery systems such as living walls can function as air biofilters and wastewater treatment systems, in addition to cooling buildings, reducing noise, increasing urban biodiversity, providing food, and enhancing well-being. Natural swimming pools

(NSPs) are an ecologically sound alternative to chemically treated pools, but they have not reached their potential in the U.S. We investigated whether a living wall could be integrated into an NSP system as a technical wetland filter, given that the vertical filter would have to produce excellent water quality for human swimmers. This could be a novel landscape design, particularly in cases of steep contours, urbanized sites with limited space, or the retrofitting of old pools. We evaluated two filter media: limestone gravel, based on a lime fen ecosystem, and Sphagnum moss, based on an acidic bog ecosystem. We also tested the effects of vertical vegetation by planting half of the filter media variants and leaving the other half unplanted. We installed 12 systems (3 of each of the 4 variants), each comprised of a living wall module atop a basin with water pumped and recirculated through the living wall. Beginning in summer 2019 and ending iv after summer 2020, we analyzed the water quantity, water quality, and vegetation quality over two growing seasons. We found that all of the variants attained good water quality with minor exceptions. Also, most of the trial systems consistently met German NSP water quality standards. As expected, evapotranspiration losses occurred, especially with the Sphagnum moss acidic bog variants. The vegetation did not have a significant impact on water quality, but the benefits of vertical planting (including cooling, noise reduction, and air pollution reduction) could justify their use. Overall, this exploration of living walls as technical wetland filters verified their feasibility for further study and practice.

As the U.S. faces more water shortages, heat waves, and sprawling development, we could integrate pools into our water systems more ecologically, which NSPs and living walls could facilitate. Our study concludes with a hypothetical design concept proposal for an NSP with a vertical technical wetland filter.

(167 pages)

PUBLIC ABSTRACT

A deep dive into natural swimming pool filtration: Living walls as technical wetland filters

Anna Farb

Vertical gardens such as living walls can filter air and water, in addition to

cooling buildings, reducing noise, increasing urban biodiversity, providing food, and

enhancing well-being. Natural swimming pools (NSPs) are an ecologically sound

alternative to chemically treated pools, but they have not reached their potential in the

U.S. We investigated whether a living wall could be integrated into an NSP system for water filtration purposes, given that the vertical filter would have to produce excellent water quality for human swimmers. This could be a novel landscape design, particularly in the cases of steep contours, urbanized sites with limited space, or the retrofitting of old pools. We evaluated two filter media: limestone gravel, based on a lime fen ecosystem,

and Sphagnum moss, based on an acidic bog ecosystem. We also tested the effects of

vertical vegetation by planting half of the filter media variants and leaving the other half

unplanted. We installed 12 systems (3 of each of the 4 variants), each comprised of a

living wall frame atop a basin with water pumped and recirculated through the living

wall. Beginning in summer 2019 and ending after summer 2020, we analyzed the water

quantity, water quality, and vegetation quality over two growing seasons. We found that

all of the variants attained good water quality with minor exceptions. Also, most of the

trial systems consistently met German NSP water quality standards. As expected,

evapotranspiration losses occurred, especially with the Sphagnum moss acidic bog

variants. The vegetation did not have a significant impact on water quality, but the vi benefits of vertical planting (including cooling, noise reduction, and air pollution reduction) could justify their use. Overall, this exploration of living walls as water filters verified their feasibility for further study and practice. As the U.S. faces more water shortages, heat waves, and sprawling development, we could integrate pools into our water systems more ecologically, which NSPs and living walls could facilitate. Our study concludes with a hypothetical design concept proposal for an NSP with a vertical water filter.

vii

ACKNOWLEDGMENTS

I would like to thank everyone who helped care for the study trials, including

Ilka, Lucas, Marie, Zhangxin, Jessica, and Evgenii. Thank you to Tino Bräuchle,

Norbert Gäng, and everyone at Balena for helping us with the water measurements.

Also, Maximilian Colditz generously provided the water pumps out of interest for the project. Mick Hilleary and Troy Becker were extremely generous in sharing information about their companies. The LAEP department was very supportive of me completing my thesis overseas, and I really appreciate all the administrators and professors who answered my questions at all hours. Thanks to my family, Tonya,

Emily, Julie, and my pup Otter. I am so grateful to Dave Anderson for his calming

and guiding presence, and to Phil for working during retirement! Lastly, Wolfram

Kircher’s fascinating mind and ideas got us all here. Thanks so much to you all.

Anna Farb

viii

CONTENTS

Page

Abstract ...... iii

Public Abstract ...... v

Acknowledgments...... vii

List of Tables ...... x

List of Figures ...... xi

Chapter 1 Introduction ...... 1

1.1 Statement of the Problem ...... 1 1.2 Purpose of the Study ...... 1 1.3 Definition of Terms...... 2 1.4 Significance of the Research ...... 9

Chapter 2 Literature Review ...... 11

2.1 Introduction ...... 11 2.2 Living walls ...... 11

2.2.1 Classifications ...... 12 2.2.2 Plants ...... 17 2.2.3 Irrigation ...... 18 2.2.4 Maintenance ...... 20 2.2.5 Building Cooling and Energy Reduction ...... 22 2.2.6 Air Pollution Reduction ...... 25 2.2.7 Noise Reduction ...... 27 2.2.8 Water Management ...... 28 2.2.9 Biodiversity ...... 29 2.2.10 Food and Urban Farming ...... 30 2.2.11 Socioeconomic Aspects ...... 31 2.2.12 Case Study: Fashion Valley Mall, San Diego ...... 34 2.2.13 Conclusion ...... 35

2.3 Natural Swimming Pools ...... 40

2.3.1 Classifications ...... 45 2.3.2 Plants ...... 50 ix

2.3.3 Filter Media ...... 53 2.3.4 Water Quantity ...... 55 2.3.5 Water Quality ...... 57 2.3.6 Biodiversity ...... 62 2.3.7 Socioeconomic Aspects ...... 65 2.3.8 Case Study: Los Angeles Conversion ...... 68 2.3.9 Conclusion ...... 70

2.4 Added Value of Integrated System ...... 70

Chapter 3 Methodology ...... 74

3.1 Study Model ...... 74 3.2 Framework ...... 74 3.3 Filter Media Selection ...... 79 3.4 Vegetation Selection ...... 80 3.5 Water Quantity Analysis ...... 88 3.6 Water Quality Analysis ...... 88 3.7 Vegetation Analysis ...... 90 3.8 Strategies and Limitations...... 90

Chapter 4 Results ...... 92

Chapter 5 Discussion ...... 101

5.1 Key Findings and Overall Feasibility ...... 101 5.2 Limitations ...... 106 5.3 Further Research ...... 107 5.4 Conclusions ...... 108

Chapter 6 Hypothetical Design Study...... 109

References ...... 115

Appendices ...... 136

Appendix A. Filling Water Data ...... 137 Appendix B. Water Chemistry Data ...... 148 Appendix C. Vegetation Data ...... 153

LIST OF TABLES

1. Generalized VGS Classifications ...... 15

2. Review of Various Benefits of Living walls ...... 36

3. Comparison of Relevant Water Features in Landscape Design ...... 42

4. NSP Filter Classifications ...... 47

5. NSP Model Classifications ...... 49

6. Comparison of P-limiting and C-limiting System Specifications...... 50

7. Water Chemistry and Sanitary Microbiological Standards for NSP Water ...... 58

8. NSP Organism Classification ...... 64

9. Description of Variants ...... 76

10. Visual Representations of Variants ...... 77

11. Species Planted on Vegetated Elements ...... 81

12. Attributes of Selected Vegetation ...... 82

13. Schedule of Water Quality Measurements ...... 89

14. Scales for Assessing Vegetation Quality ...... 90

LIST OF FIGURES

1. VGS classification tree ...... 13

2. Green façade and modular living wall examples ...... 14

3. Vertical cross-section through modular panel living wall and geotextile/pocket

living wall ...... 16

4. Fashion Valley Mall living wall ...... 34

5. Modular construction with organic, sponge-like media block ...... 35

6. Commonly designed water features ...... 41

7. Examples of more organic and more formal NSPs in terms of style and form ...... 43

8. An example of NSP framework ...... 44

9. NSP classification system ...... 45

10. Biological filtration flow chart ...... 46

11. Processes by which nutrients enter and exit NSP systems ...... 59

12. Residential Los Angeles NSP by California Natural Pools ...... 69

13. Filtration and usage areas of the low-flow filter substrate NSP ...... 69

14. Added value of integrating living walls and NSPs for landscape architects ...... 71

15. Conceptual sketch of a living wall acting as the regeneration area for an NSP ...... 72

16. Summary of variants and replicates and spatial randomization of systems ...... 78

17. Setup of systems in the greenhouse ...... 79

18. Limestone gravel filter media; moistened Sphagnum moss block ...... 80

19. Irrigation of the living walls from recirculated water in the basins below ...... 88

20. Average monthly filling water needs by variant ...... 92 xii

21. Water chemistry parameter measurements by variant (nutrients) ...... 94

22. Water chemistry parameter measurements by variant ...... 96

23. Visual assessments of vegetated variants ...... 98

24. Formation of living Sphagnum (a); cranberries (b); moisture gradient of lime

fen variants (c) ...... 100

25. Commercial living wall system ...... 105

26. Conceptual sketch of a green roof acting as the regeneration area for an NSP ...... 106

27. Geographical context and slope ...... 110

28. Existing NSP with cascading regeneration areas ...... 111

29. Potential conceptual plan ...... 112

30. Water circulation and basic plumbing system ...... 113

CHAPTER 1

INTRODUCTION

1.1 Statement of the Problem

Natural swimming pools (NSPs) are adaptable design solutions for those who wish to swim without using harmful chemicals. Vertical greenery systems such as living walls can function as air biofilters and wastewater treatment systems, in addition to cooling buildings, reducing noise, increasing urban biodiversity, providing food, and enhancing well-being. However, NSPs have not reached their potential in the U.S. and could be utilized more if the ground space required were reduced. We investigated whether a living wall could be integrated into an NSP system as a technical wetland filter, given that the vertical filter would have to produce excellent water quality for human swimmers.

1.2 Purpose of the Study

This study aims to explore the feasibility and value of integrating natural swimming pools (NSPs) and living walls. Thus, we explored the capacity of the medium

and vegetation of the living wall to purify NSP water. This particular filtration method

has never been done before, so we took a broad, exploratory approach.

We aimed to answer four research questions: (1) Which living wall medium would filter water better and most efficiently when comparing limestone gravel, based on a lime fen ecosystem, and Sphagnum moss, based on an acidic bog ecosystem?; (2) How would vegetation affect water filtration capacity?; (3) Could the living wall filter the 2 water sufficiently to meet the NSP water quality guidelines developed by the German

FLL organization, with particular attention to the phosphorus limit?; (4) How and where could a vertical technical wetland filter be employed in a practical sense?

1.3 Definition of Terms

Aerobic. Having a high supply of oxygen (Kircher & Thon, 2016a).

Anaerobic. Having a low supply of oxygen (Kircher & Thon, 2016a).

Aquatic plant. Various wetland vegetation species which inhabit various water

levels.

Helophyte. A plant that grows in marshy ground with submerged

regeneration buds and shots emerging above the water level.

Hydrophyte. An aquatic plant with floating leaves or growing completely

submerged (Kircher & Thon, 2016a).

Biofilm. An aggregate of microorganisms embedded within a matrix covering submerged vascular plants, stones, substrates, wood, or the pond sealing itself. The accumulation of microbial bacteria, algae, fungi, viruses, protozoa, and tiny invertebrates secretes a mucus film that adheres to the surface or boundary layer (Kircher & Thon,

2016a).

Biocoenosis. Closely integrated diverse biological communities, including algae, can develop autonomously in water, substrates, and root zones and purify the water with the appropriate conditions. The film (of the biofilm) structurally encapsulates the biocoenosis (BioNova Natural Pools, 2019; FLL, 2011). 3

Biological Water Purification. The process of cleaning water using plant and animal organisms, microorganisms, and filtration mechanisms. The water is purified in the usage area (in situ) and in the purification area (ex situ).

Usage Area (in situ). Water purification in this area involves natural

filtration by zooplankton, reduction of microorganisms by sunlight, and

elimination of nutrients through sedimentation.

Purification Area (ex situ). Supplementary biotechnological filtering

processes can be added in order to meet the water parameter requirements for

natural swimming pools (FLL, 2011).

Conductivity. The measure of the salt content of water (Kircher & Thon, 2016a).

Dissolved Inorganic Carbon (DIC). Carbon compounds in water bodies that can be incorporated by plants. Forms include carbon dioxide (CO2), hydrogen carbonate or

- 2- bicarbonate (HCO3 ), and carbonate (CO3 ) (Kircher & Thon, 2016a).

Dissolved Organic Carbon (DOC). The sum of scarcely degradable carbon

compounds found in water bodies. In NSPs, this is often in the form of wildlife droppings or incompletely decomposed plant material (Kircher & Thon, 2016a).

Dissolved Oxygen (DO). The presence of oxygen in water. This occurs through

the dissolution of oxygen from the atmosphere, photosynthesis of submerged plants, the

reduction of particular chemical compounds, or the influx of oxygen-rich water. Large

amounts of organic matter and murky conditions decrease the oxygen concentration in

water (Kircher & Thon, 2016a).

Ethylene Propylene Diene Monomer (EPDM). A very elastic synthetic rubber

material for liners (Kircher & Thon, 2016a). 4

Evapotranspiration (ET). The sum of evaporation, water loss by the

vaporization from the soil or water surfaces, and transpiration, water loss from the

vegetation surface (Kircher & Thon, 2016a).

Filamentous Algae. Also known as string algae, they form clusters or blankets on

the water’s surface, mainly in shallow water. They can even live in oligotrophic water

(Kircher & Thon, 2016a).

German Landscaping and Landscape Development Research Society (FLL).

This independent non-profit organization was founded by eight professional

organizations to enhance “environmental conditions through the advancement and

dissemination of plant research and its planned applications.” They develop standards and

guidelines for several forms of green infrastructure technology (FLL, 2011).

Hardness. Lime content in water bodies, or more precisely, dissolved alkaline

earth metals. Carbonate hardness (CH) is the concentration of dissolved Ca2+ and Mg2+

- with an equivalent concentration of HCO3 ions; measured in °dH or mmol/l; non-

carbonate hardness (NCH) is the concentration of dissolved Ca2+ and Mg2+ with an

equivalent concentration of other ions than the ones responsible for carbonate hardness,

2- - - e.g. SO4 , Cl , NO3 . Total hardness (TH) is the sum of carbonate hardness and non-

carbonate hardness (Kircher & Thon, 2016a).

Mire. A peat-forming ecosystem.

Bog. A type of mire characterized by high water tables and low nutrient

availability, with low pH and alkalinity, in which only precipitation provides

nutrients (Gore, 1983; Aerts, Verhoeven, & Whigham, 1999). Sphagnum mosses 5

primarily, in addition to slow-growing evergreen and deciduous shrubs and trees,

dominate temperate bogs (Bridgham, Pastor, Janssens, Chapin, & Malterer,

1996).

Fen. A type of mire characterized by high water tables and low nutrient

availability, with high pH and alkalinity, in which precipitation, surface water,

and groundwater provide nutrients (Gore, 1983; Aerts, Verhoeven, & Whigham,

1999). Graminoids (mainly Carex and Cladium species) and deciduous shrubs

and trees dominate temperate fens (Bridgham, Pastor, Janssens, Chapin, &

Malterer, 1996).

Mycorrhiza. The symbiosis between specialized fungi and the roots of certain vascular plants wherein the fungi enhance the plant’s uptake of nutrients and water

(Kircher & Thon, 2016a).

Natural Swimming Pool (NSP). Artificially created bodies of water with defined requirements for water quality for the purpose of human recreational bathing. They are sealed against the subsoil and have no chemical disinfection treatment, instead cleansing the water biologically and/or physically. Ultraviolet radiation treatment is also unacceptable because it disables the desired biological activity (Casanovas-Massana &

Blanch, 2013; Giampaoli, Garrec, Donzé, Valeriani, Erdinger, & Romano Spica, 2014;

FLL, 2011). The various models utilize one or more of the systems listed below.

Biofilm-accumulating Substrate Filter Bed. Water is treated via vertical

percolation through filter beds without plants or with purely decorative planting

(Kircher & Thon, 2016a). 6

Hydrobotanical System. Water is treated via densely planted filter beds

in shallow water (Kircher & Thon, 2016a).

Technical Wetland. Water is treated via percolation through a planted

filter bed. Purification is largely done by the substrate plus microbes in the

rhizosphere (Kircher & Thon, 2016a).

Nitrogen (N). The basic element for building proteins. In order to grow, plants

+ - absorb nitrogen in the forms of ammonium (NH4 ) and nitrate (NO3 ) (Kircher & Thon,

2016a).

Nitrogen Use Efficiency. The ratio between aboveground biomass production

and nutrient loss in litterfall (Vitousek, 1982).

Nutrient Load. The amount of dissolved nutrients in a water body (particularly nitrogen, but also phosphorus and potassium) impacts the productivity of wetland vegetation and its species composition (Kircher, 2004). The trophic level (below) reflects the site’s productivity based on the nutrient load (Kircher & Thon, 2016a).

Eutrophic. A site condition with a high production rate of organic matter;

mostly water with a high nutrient level (Kircher & Thon, 2016a).

Hypertrophic. Site conditions that provide an excessively high

production rate of organic matter; water with an abnormally high nutrient level

(Kircher & Thon, 2016a).

Mesotrophic. Site conditions that provide a moderate production rate of

organic matter; water with a medium nutrient level (Kircher & Thon, 2016a).

Oligotrophic. Site conditions that provide a low production rate of

organic matter; nutrient-poor water (Kircher & Thon, 2016a). 7

Particulate Matter (PM). Fine dust particles.

pH Value. The measure of the acidity or alkalinity of water, defined as the

negative decimal logarithm of hydrogen ion (H+) activity. A pH of less than 7 indicates

an acidic pH; a pH of more than 7 indicates an alkaline pH; a pH of 7 indicates a neutral

pH (Kircher & Thon, 2016a).

Phosphorus (P). An essential element, known as the essential energy for all

known life forms, in DNA and RNA and in membranes. It is considered particularly

relevant in terms of hygiene and algae growth. Total Phosphorus (Ptot) is the sum of

phosphorus in all compounds (Kircher & Thon, 2016a).

Plankton. Microscopic organisms encompassing many diverse groups (e.g.

microscopic algae are considered zooplankton) (Kircher & Thon, 2016a).

Planktonic Green Algae. Also known as floating algae, they cause the water to become murky if present in high concentrations (Kircher & Thon, 2016a).

Rhizome. The stem axis (runner) of a plant that grows underground (Kircher &

Thon, 2016a).

Rhizosphere. The area of substrate surrounding plants’ roots (Kircher & Thon,

2016a).

Sphagnum Moss. The appearance of typical peat mosses (such as Sphagnum

magellanicum and S. rubellum) indicates development of a raised bog. Sphagnum mosses have special cells (Hyalocytes) that are able to exchange cations for hydrogen (H+),

causing further acidification. They carpet the bog’s surface, growing upward while their

lower parts die, gradually forming a peat bog (Kircher, 2004).

Sp. Abbreviation for a singular species. 8

Stone Wool. Also called mineral wool, an insulation product created by spinning

molten rock and minerals with steel slag to create a cotton-candy-like wool substance.

The company Rockwool produces stable, water-repellent, sound-absorbent, and fire- resistant stone wool with non-directional fiber orientation (This Old House, 2020).

Swimmer Equivalent Value. The standardized influx per user for microorganisms (E. coli) and plant nutrients (phosphate). Water purification specifications are based on the swimmer equivalent value (FLL, 2011).

USDA Hardiness Zone. A classification system of plants that denotes how well they can tolerate various levels of colder winter temperatures (Kircher & Thon, 2016a).

Vertical Greenery System (VGS). Different forms of vegetated wall surfaces with plant species spreading across vertical structures that may be fixed to an indoor wall or to a building façade (Francis & Lorimer, 2011).

Green Façade. Traditional vertical greenery systems in which the

vegetation cover is formed by climbing or hanging plants. They are rooted in the

ground or grown in containers attached to the walls or integrated onto balconies at

various heights (Fernández-Cañero, Pérez Urrestarazu, & Perini, 2018).

Living Wall. Contemporary vertical greenery systems in which a

supporting structure is adapted to each cultivation system (Francis & Lorimer,

2011). Active living walls involve pushing air through the living wall, out of

which comes biofiltered air (Darlington, Dat, & Dixon, 2000). The adiabatic

interchange between the air and plants also cools emitted air (Franco, Fernández-

Cañero, Pérez-Urrestarazu, & Valera, 2012; Pérez-Urrestarazu, Fernández-

Cañero, Franco, & Egea, 2016). Geotextile/Pocket systems use two layers of 9

synthetic fabric to create pockets holding the growing media and plants (Hopkins

& Goodwin, 2011). Modular living wall systems consist of panels that hold

growing media to support the plant material (Hopkins & Goodwin, 2011).

Water Cycle. The natural swimming pool water cycle involves several phases.

Filling Water. Water used for initial filling and refilling (e.g. to

compensate for water lost due to evaporation, swimmers, or cooling).

Pool Water. Water in the swimming usage area.

Pure Water. Water that has just gone through biological purification

before being introduced into the pool.

Surface Water. Water flowing off of surfaces (both sealed and unsealed)

that may contaminate the pool water.

Untreated Water. Water in the swimming usage area on its way to the

purification process (FLL, 2011).

Zeolite. Crystalline aluminosilicates with a microporous structure, some of which can fix special ions. The vast inner surfaces promote biofilm development (Kircher &

Thon, 2016a).

1.4 Significance of the Research

By using walls as productive, green spaces, we allow for ample possibilities on the ground plane. Natural swimming pools (NSPs) can already fit ecologically and aesthetically into most spaces, but the spatial element can be a challenge due to the requirements for regeneration areas. Thus, we explored whether a living wall could perform technical wetland filtration vertically for an NSP. This could open up pool 10

designers to more creative options. Moreover, people will and should be able to swim, especially in places facing water scarcity and heat waves. NSPs and living walls could facilitate the integration of our pools into our water systems more ecologically.

CHAPTER 2

LITERATURE REVIEW

2.1 Introduction

This study aims to explore the feasibility and value of integrating natural swimming pools (NSPs) and living walls. Scholarly literature covers these separate topics extensively. As such, this literature review will describe the relevant information on living walls and natural swimming pools separately, followed by the arguments for bringing them together.

2.2 Living Walls

The benefits of green infrastructure include increasing urban biodiversity, rainwater retention and decreased runoff, noise reduction (indoors and outdoors), cooling from shading and evaporation, prevention from overheating, air quality improvement

(oxygen production, air purification, binding particulate matter), placemaking, privacy, enhancing ambient design quality, and building preservation (Dettmar, Pfoser, & Sieber,

2016; Pérez-Urrestarazu, Fernández-Cañero, Franco-Salas, & Egea, 2015; Hopkins &

Goodwin, 2011). Specifically, recent research has shown that living walls offer a biophilic strategy for climate change mitigation by reducing greenhouse gas (carbon dioxide) emissions especially if used on a larger scale, lowering energy consumption, increasing the thermal performance of buildings, enhancing water sensitive urban design, and easing the urban heat island effect (Ottelé, 2015; Zhao, Zuo, Wu, & Huang, 2019; 12

Solera Jimenez, 2018; Pérez-Urrestarazu, Fernández-Cañero, Franco-Salas, & Egea,

2015; Loh, 2008; Roehr & Laurenz, 2008). Living walls are particularly suitable for cities because of the limited availability of land, as well as their potential to enrich dwellers’ living conditions and health, improving indoor air quality, mitigating noise pollution, and producing food (Sheweka & Mohamed, 2012; Pérez-Urrestarazu, Fernández-Cañero,

Franco-Salas, & Egea, 2015; Loh, 2008). Still, the benefits to air quality, reduction of noise, positive effects on hydrology, and visual benefits need much further empirical testing (Radić, Brković Dodig, & Auer, 2019).

2.2.1 Classifications. Vertical greenery systems (VGSs) are divided into two major groupings: green façades and living walls (Figure 1; Kontoleon & Eumorfopoulou,

2010; Hopkins & Goodwin, 2011). Classical façade greening uses perennial climbing plants, while the plants in more modern forms of VGSs do not have ground contact, instead rooting into various supporting structures (Figure 2; Hommel, Veres, & Röseler,

2012). These modular systems are recently more prevalent because of their efficient installation, quick surface coverage, and straightforward maintenance measures

(disassembly and replacement of each element), whereas geotextile systems are less widespread except for in more artistic applications (Table 1; Figure 3; Manso & Castro-

Gomes, 2015; Azkorra, Pérez, Coma, Cabeza, Bures, Álvaro, Erkoreka, & Urrestarazu,

2015). The latest ideas involve entire vertical façade “parks” planted with trees and shrubs (Hommel, Veres, & Röseler, 2012). This study focuses on outdoor modular living walls; thus, this literature review largely will not cover the other VGSs, such as green façades. Also, since active living walls are generally indoor systems, this study will not 13

address them further. Nevertheless, there could be potential to apply our findings to the

active living wall system.

Figure 1. VGS classification tree (Besir & Cuce, 2018; Radić, Brković Dodig, & Auer,

2019; Ottelé, 2011; Perini, Ottelé, Haas, & Raiteri, 2013). 14

Figure 2. Green façade (above; source: Carl Stahl) and modular living wall (below; source: Tournesol) examples.

Table 1. Generalized Living Wall Classifications (Fernández-Cañero, Pérez Urrestarazu,

& Perini, 2018; Hopkins & Goodwin, 2011).

Living wall Structural Location Growing method Vegetation type material Hydroponic cultures Active Mostly Nonwoven with porous indoor porous felt Epiphytes, inorganic substrates lithophytes, Geotextile/ Outdoor/ Textile or Hydroponic cultures/ bromeliads, Pocket Indoor nonwoven felt organic substrate ferns, with hanging succulents, pockets herbs/forbs, small shrubs, Modular Outdoor/ Galvanized Hydroponic climbing panel Indoor steel, cultures/ plants, polyethylene, or organic or vegetables recycled plastic inorganic substrate

Figure 3. Vertical cross-section through modular panel living wall (left) and geotextile/pocket living wall (right; Loh, 2008; drawing by M. Murray, 2008).

Although these are simple classification systems, VGSs can be much more complex; one wall can even host a hybrid of various types. Also, spontaneous VGSs can emerge, such as when wild vines climb up a wall, freestanding VGSs can be used as dividers, and moss walls can be art pieces (Fernández-Cañero, Pérez Urrestarazu, &

Perini, 2018; Hopkins & Goodwin, 2011; Radić, Brković Dodig, & Auer, 2019). Textile systems can use Sphagnum moss or topsoil as their organic substrate. Modular systems can also use organic substrate or inorganic compounds, including coconut fiber (coir) or

Sphagnum moss, or stone wool or polyurethane foam, respectively (Fernández-Cañero,

Pérez Urrestarazu, & Perini, 2018). They are constructed from elements such as modules, 17

panels, containers, vessels, planter tiles, leaf boxes, framed boxes, wire cages, perforated

boxes, and slanted cell boxes, depending on the construction and design (Radić, Brković

Dodig, & Auer, 2019).

2.2.2 Plants. Living wall planting depends on the microclimatic conditions

(temperature and wind), orientation (sun exposure), cultivation system, and nursery stock

availability. Living wall plant types include epiphytes, lithophytes, bromeliads, ferns,

succulents, herbs/forbs, small shrubs, climbing plants, and vegetables/fruits, such as

strawberries and lettuce. Appropriately chosen native plants can work as well. They can

be chosen based on adaptations to jungle understories, river and stream basins, and/or

rocky environments (Fernández-Cañero, Pérez Urrestarazu, & Perini, 2018). Plants with higher air pollution tolerance are favorable for living walls due to the higher pollution

levels of cities (Singh, Jain, Mathur, & Kumar, 2017).

Plant selection and characteristics affect air quality improvement, energy savings

(depending on the plants’ evaporation capacity), coverage of the wall, and particulate

matter collecting capacity (Pérez, Rincón, Vila, González, & Cabeza, 2011; Cameron,

Taylor, & Emmett, 2014; Perini, Magliocco, & Giulini, 2017; Ottelé, van Bohemen, &

Fraaij, 2010; Sternberg, Viles, Cathersides, & Edwards, 2010; Perini, Ottelé, Giulini,

Magliocco, & Roccotiello, 2017). Species selection depends on wind/air movement

because with more wind, the plant cannot absorb as much moisture and risks drying out.

They should have drought and high wind tolerance, cliff top/cliff face habituation, and

road edge or median strip adaptability. Furthermore, living walls have limited space for

the root zone, so plant success is aided by a fibrous root system, a strong stem to root 18

connection, an appropriate growth habit, and adequate orientation to sun or shade. Plants

that overhang or droop over the plants below are acceptable as long as the plants below

are shade tolerant (Hopkins & Goodwin, 2011). Accordingly, plants toward the top of the

wall should be more tolerant of sunlight and drought, plants in the middle should have

medium lighting requirements, and plants toward the bottom of the wall should be more

tolerant to shade and humidity (Fernández-Cañero, Pérez Urrestarazu, & Perini, 2018).

The growing medium and species composition choices are important for a

successful living wall, so planting combinations should be tested. The planting position

affects the plants’ ability to grow and utilize nutrients and water from different parts of

the growing medium. The study of several varied interactions indicated the complexities

in species composition and substrate choice, but generally plants toward the top were observed to grow better (Jørgensen, Thorup-Kristensen, & Dresbøll, 2018). Planting a

living wall involves cutting the felt or modules onsite or pre-planting into modules. Using

larger plants for the initial planting decreases the amount of time the wall takes to fill in

(Fernández-Cañero, Pérez Urrestarazu, & Perini, 2018).

2.2.3 Irrigation. Living wall irrigation generally involves horizontal branching

with drippers distributed per sector such that water flows down through the entire surface

by gravity and water losses are minimized (Hopkins & Goodwin, 2011). The timing and

duration depend on field observations for the particular site, but the substrate should

never be allowed to dry out. Accordingly, the drainage capacity of the substrate needs to

provide sufficient water retention while preventing root saturation (Pérez-Urrestarazu &

Urrestarazu, 2018). Irrigation requirements depend on sun exposure, temperature, 19

humidity, vertical greenery system (VGS) type, runoff, species selection (plant uptake),

and substrate selection (direct evaporation from the substrate; Pérez-Urrestarazu &

Urrestarazu, 2018; Loh, 2008). For recirculating systems, runoff can be collected below

the living wall, mixed with renewed water, filtered, and pumped back up (Pérez-

Urrestarazu & Urrestarazu, 2018; Hopkins & Goodwin, 2011). The circulating flow

depends on the number of emitters and their flow. The working pressure depends on the

emitter type, the height of the living wall, and the head losses. Additional equipment could include filters (for recirculating systems), valves, and automation/control systems

(Pérez-Urrestarazu & Urrestarazu, 2018).

Most living walls are fertilized via irrigation water (fertigation); again,

recirculating systems require less input because fewer nutrients leach out (Pérez-

Urrestarazu & Urrestarazu, 2018). Since inorganic substrate does not provide nutrients to

the plants, most living walls of this sort require fertigation (Fernández-Cañero, Pérez

Urrestarazu, & Perini, 2018). Hydraulic substrate properties, plant water requirements,

and the design and operation of irrigation systems determine water distribution; every

living wall can be considered a unique ecosystem (Segovia-Cardozo, Rodríguez-Sinobas,

& Zubelzu, 2019).

The combination of drip irrigation and percolation can result in two hydro zones:

one in the upper section with a deficit and the other at the base with a surplus and prone

to root asphyxia (Segovia-Cardozo, Rodríguez-Sinobas, & Zubelzu, 2019). In contrast,

the top levels exhibit significantly higher plant activity compared to the bottom levels,

with 4 times the water uptake (Prodanovic, Wang, & Deletic, 2019). No optimal drip line

spacing value exists, but irrigation lines that are closer together are more likely to achieve 20

water distribution uniformity. Systems that recirculate water should employ higher flows

(via high flow emitters) because of the resulting higher distribution uniformity, despite

the higher runoff losses (Pérez-Urrestarazu, Egea, Franco-Salas, & Fernández-Cañero,

2014). Pressure compensating emitters can be used (Pérez-Urrestarazu, & Urrestarazu,

2018; Urrestarazu and Burés, 2012).

Water consumption may discourage the use of living walls; water used/misused for plant transpiration and substrate evaporation is related to air temperature and humidity, incoming solar energy, speed of the air flow, vegetation type, and substrate characteristics (thickness, transfer surface, presence of vegetation, etc.; Pérez-

Urrestarazu, Fernández-Cañero, Franco-Salas, & Egea, 2015). However, living walls can

consume about half the amount of water of a standard in-ground garden, depending on the site. A modular container system consumes a maximum of 5 l/m2/day during the

summer and 1 l/m2/day during the winter (Prodanovic, Wang, & Deletic, 2019). Water

sensors can automatically turn off the irrigation system when the living wall is saturated

(Hopkins & Goodwin, 2011). There is a tradeoff between evaporative cooling and water use, so living wall design must balance energy savings and urban heat island effect reduction with efficient use of water for plant growth (Pérez & Coma, 2018). Eco- friendly water supply alternatives include rainwater, air conditioner condensation, greywater, and blackwater, with various requisite pre-treatments (Pérez-Urrestarazu &

Urrestarazu, 2018).

2.2.4 Maintenance. Achieving a successful living wall involves ensuring an appropriate support structure, maintaining the proper amount of water, oxygen, nutrients 21 and pH levels, choosing plants that can survive seasonal climatic changes, establishing the appropriate lighting conditions, and maximizing affordability, sustainability, and longevity (Riley, 2017). Accordingly, maintenance mainly involves sufficient watering during plant establishment (irrigation system adjustment), managing airborne seed germination, providing access to the wall for equipment, annual weeding and pruning, and inspecting irrigation and fertilization systems on a consistent basis (Hopkins &

Goodwin, 2011). In particular, the maintenance program should include regular inspections of the entire irrigation system (Pérez-Urrestarazu & Urrestarazu, 2018).

Sporadic water testing ensures that water and nutrient uptake are not causing major changes to the conductivity level (Fernández-Cañero, Pérez Urrestarazu, & Perini, 2018).

Additional aspects of maintenance include the plant-growing medium relationship, pruning, and pest and disease management. Specifically, growing medium, plant species, and planting position affect the spatial root growth of the plants. The roots generally do not stray far from planting position. A comparison of coir and stone wool as the growing media showed that plants in coir had stronger root growth in all parts of the medium than plants in stone wool. Slow growing plants require less maintenance, but slow initial root growth may cause difficulties in establishment (Jørgensen, Dresbøll, &

Thorup-Kristensen, 2014). Pruning needs depend on the type of living wall and the vigor of the vegetation. Pruning, weeding, and cleaning can be completed simultaneously

(Pérez-Urrestarazu & Urrestarazu, 2018). Lastly, pest and disease management depend on local climatic conditions, species selection, and concerns of the proprietor. Living wall pests found in temperate climates include aphids, spider mites, whiteflies, mealybugs, 22 and leaf miners. Extermination strategies involve chemical controls (pesticides) or biological controls (beneficial organisms) (Pérez-Urrestarazu & Urrestarazu, 2018).

Thus far, we have discussed how living walls are made. Although they are complex, living walls come with extensive potential benefits, as supported by the literature.

2.2.5 Building Cooling and Energy Reduction. Living walls can act as carbon sinks and reduce energy consumption for buildings, especially during summers

(Charoenkit & Yiemwattana, 2016; Singh, Jain, Mathur, & Kumar, 2017). They block light from penetrating the building façade, providing insulation capabilities. The layer of air trapped within the mass of vegetation decreases heat from reaching the building façade. Shade from vegetation, in addition to evapotranspiration, contributes to a decrease in ambient temperature, further reducing the heat load on the building façade.

Additionally, living walls provide a buffer against wind in winter (Hopkins & Goodwin,

2011). Solar radiation on a living wall acts as the engine of evaporation (from the soil) and transpiration (from the vegetation); consequently, evaporative cooling of air by the outer surface of the building components occurs, which allows a lower heat transfer inside (Ascione, 2017; Sheweka & Magdy, 2011; Safikhani, Abdullah, Ossen, &

Baharvand, 2014). This cooling energy reduction can be significant and offers a valuable solution for retrofitting existing buildings and for the urban heat island effect (Mazzali,

Peron, Romagnoni, Pulselli, & Bastianoni, 2013; Pulselli, Pulselli, Mazzali, Peron, &

Bastianoni, 2014; Ottelé & Perini, 2017). 23

Nevertheless, the literature is insufficient on the impacts of living walls on heating period (winter) thermal performance and the separate effects of insulation, cooling, and wind barrier capacity (Pérez, Coma, & Cabeza, 2018). With so much variability among living walls, the energy savings and urban heat island effect reduction depends on the system’s materials and layers (substrate typology and thickness), support elements, air cavity type (gap thickness and ventilation), species selection (foliage thickness/coverage), maintenance program, and climatic conditions such as solar radiation, temperature, relative humidity, rainfall, and wind. For example, geotextile felt living walls do not have an air gap, while modular panels do (Pérez, Rincón, Vila,

González, & Cabeza, 2011; Pérez, Coma, & Cabeza, 2018; Scarpa, Mazzali, & Peron,

2014).

The role of living wall vegetation cannot be understated regarding the thermal and energy performance of buildings. Plants intercepting the solar radiation upon the wall produce shade (due to their higher reflective capacity than typical building materials), plant and substrate evapotranspiration produce the cooling effect on the façade and surroundings, and plants and support structures affect wind resistance capacity (Pérez,

Rincón, Vila, González, & Cabeza, 2011; Taha, 1997; Sheweka & Magdy, 2011;

Sheweka & Mohamed, 2012; Feng & Hewage, 2014; Sudimac, Cukovic-Ignjatovic, &

Ignjatovic, 2018). Evaporative performances depend not only on the selected plants, but also on the whole system panel (water evaporates directly from the panel); foliage density affects building surface shading and consequently, summer thermal performance (Perini,

Magliocco, & Giulini, 2017). Although vegetation can reduce negative heat transfer 24 through the building façade, in colder climates, the energy savings may not be cost- effective in winter months (Feng & Hewage, 2014).

Several studies of physical living walls generally support these theories of the beneficial effects of living walls on building thermal performance and energy usage, at least in terms of external temperatures. When comparing studies, the external building wall surface temperature reduction should be used as the parameter of focus (Pérez,

Coma, Martorell, & Cabeza, 2014). Living walls in Chile reduced wall surface temperatures up to 30°C. Canopy shading and evapotranspiration could be responsible for the reduced wall surface temperature, depending on the vegetation species, vegetation covering, and the exposed wall structure of the living wall (Victorero, Vera, Bustamente,

Tori, Bonilla, Gironás, & Rojas, 2015).

In London, living walls can reduce the exterior surface temperature by up to

12°C, ambient air temperature between 0.5°C and 4.1°C, and wind speed by up to 0.7 m/s

(Solera Jimenez, 2018). Moreover, the maximum temperature difference between a bare wall and a living wall system in the Netherlands was 8.4°C, compared to the direct green façade system, which only had a maximum temperature difference of 1.7°C. Curiously, increasing the living wall width could actually lower the influence of evaporation, and mineral wool was the most effective substrate for summer cooling (Ottelé & Perini,

2017). Lastly, in warm temperate climates, living walls were able to reduce external surface temperatures of buildings by a range of 12°C (Mazzali, Peron, Romagnoni,

Pulselli, & Bastianoni, 2013) to 20.8°C (Chen, Li, & Liu, 2013). Accurate, comparable studies of thermal resistance are difficult to find, due to the variability of the timing of measurements both seasonally across the year and through the day. 25

Thermal behavior of living walls can also be mathematically modeled. Models of various types of living walls can simulate physically measured temperatures and reproduce water stress (Malys, Musy, & Inard, 2014). However, further research is needed to validate and improve the models of living wall carbon sequestration, interactions between living wall plants, and substrate and performance as an insulating layer and reducing CO2 (Charoenkit & Yiemwattana, 2016). Studies employing models have revealed that vertical greenery systems (VGSs) can provide energy savings, especially during the cooling period (summer) in warm temperate and arid climates, yielding 20-30% reductions most frequently. The limitations of these studies include the ambiguity of species selection and insufficient validation from real scale experimental assessments (Wong, Tan, Tan, & Wong, 2009; Kontoleon & Eumorfopoulou, 2010;

Pérez, Coma, & Cabeza, 2018).

2.2.6 Air Pollution Reduction. Living wall vegetation can remove a significant amount of particulate matter (PM) from polluted air; thus, they can contribute to air pollution mitigation, such as from traffic (Viecco, Vera, Jorquera, Bustamente, Gironás,

Dobbs, & Leiva, 2018; Weerakkody, Dover, Mitchell, & Reiling, 2017; Weerakkody

Dover, Mitchell, & Reiling, 2018). PM clings to plant surfaces, and plants absorb PM and uptake gaseous pollutants (CO2, NO2, and SO2; Ottelé, van Bohemen, & Fraaij, 2010;

Baik, Kwak, Park, & Ryu, 2012). In accordance, an Australian modular living wall study confirmed that vegetated filters outperformed filters with only substrate (Pettit, Irga,

Abdo, & Torpy, 2017). Plant vitality and morphological variation likely influence the ability of a living wall to improve air quality. In particular, foliage density, as measured 26

by leaf area index (i.e., leaf area per m2 of wall surface), and specific leaf characteristics

(length and shape of hair, depth and shape of ridges or grooves) are attributed to making

an impact (Perini & Roccotiello, 2018; Janhäll, 2015; Perini, Ottelé, Haas, & Raiteri,

2011; Perini, Ottelé, Giulini, Magliocco, & Roccotiello, 2017; Lin, Hagler, Baldauf,

Isakov, Lin, & Khlystov, 2016). Specifically, leaves with thick cuticles, cuticular waxes, and hairs remove PM2.5 from the air more effectively (Perini, Ottelé, Giulini, Magliocco,

& Roccotiello, 2017).

Also, in a couple of UK-based living wall studies, all plant species removed a

wide range of elements from the atmosphere, including potentially hazardous heavy

metals, but plants with smaller leaves, plants with hairy and waxy leaf surfaces, and

conifers, removed more PM (Weerakkody, Dover, Mitchell, & Reiling, 2017;

Weerakkody Dover, Mitchell, & Reiling, 2018). Different plant species have different

single pass PM removal efficiencies; fern species removing particles most efficiently in

the Australian study. This demonstrates the importance of plant choice for increased

particulate removal (Pettit, Irga, Abdo, & Torpy, 2017).

In terms of the living wall planting design as a whole, interspersing plants of

different heights throughout the living wall immobilizes more atmospheric PM than with

a more uniform planting design scheme (Weerakkody, Dover, Mitchell, & Reiling, 2019).

Zooming out further to long-term air pollution reduction, a model-based study

demonstrated that living wall vegetation can accumulate substantial amounts of carbon

dioxide with longevity, from the point of planting to harvest and composting (Marchi,

Pulselli, Marchettini, Pulselli, & Bastianoni, 2015). 27

2.2.7 Noise Reduction. Vertical greenery systems (VGSs) have the potential to reduce noise transmissions in the urban environment, such as from traffic, especially with deciduous and perennial planting (Hopkins & Goodwin, 2011; Pérez, Coma, Barreneche, de Gracia, Urrestarazu, Burés, & Cabeza, 2016; Pérez, Coma, & Cabeza, 2018).

Although there are insufficient studies confirming noise reduction by living walls at the city or building scale, factors such as the construction system (sealing joints between modules and support structure), species selection (morphology), wall shape and dimensions (thickness of substrate and vegetation layers), location in relation to the sound source, substrate selection (porosity and density), and operating methods affect

their ability in this respect (Pérez, Coma, & Cabeza, 2018; Horoshenkov, Khan, &

Benkreira, 2013; Van Renterghem, Hornikx, Forssen, & Botteldooren, 2013).

Nevertheless, an acoustic laboratory-based study showed that modular living walls have a similar or better acoustic absorption capacity compared to typical building materials, even performing better than some materials used particularly for sound-absorption. By

sealing the joints of the modules tightly, noise could be reduced further (Azkorra, Pérez,

Coma, Cabeza, Bures, Álvaro, Erkoreka, & Urrestarazu, 2015).

Compared to other VGSs, living walls with a substrate layer contribute more to

noise reduction in middle frequencies, whereas the scattering effect of greenery reduces

noise at high frequencies (Wong, Kwang Tan, Tan, Chiang, & Wong, 2010; Pérez, Coma,

Barreneche, de Gracia, Urrestarazu, Burés, & Cabeza, 2016). At the same time, modular

living walls installed at the ground level could absorb human voices effectively

(approximately 60 dB), which could be beneficial in public places such as restaurants

(Azkorra, Pérez, Coma, Cabeza, Bures, Álvaro, Erkoreka, & Urrestarazu, 2015). Lastly, 28 in terms of acoustic insulation, an outdoor living wall would only reduce noise inside the corresponding building with a physical separation between the living wall and the façade, thus preventing an acoustic bridge (Pérez, Coma, & Cabeza, 2018).

2.2.8 Water Management. Possible water sources for living walls include

(green) roof and greywater capture from the building with appropriate storage and filtration systems, in addition to stormwater runoff from impervious pavement stored underneath (Pérez & Coma, 2018; Hopkins & Goodwin, 2011). Living walls can also collect precipitation and buffer it somewhat for later evapotranspiration, as modular panels have higher evaporation levels than planter boxes (van de Wouw, Ros, &

Brouwers, 2017). For example, the living wall on Tooley Street in London is fed by rainwater stored behind the planting and absorbed by the plants via capillary action, rather than a pump requiring power. In addition, the Espai Tabacalera’s living wall in

Tarragona, Spain treats the rainwater and greywater coming off a neighboring park as a tertiary treatment filter. Lastly, the Los Cabos International Convention Center’s living wall in Mexico uses greywater for irrigation, which is then cycled back into the building for toilet flushing (Pérez & Coma, 2018). Creativity in designing for closed-loop systems is boundless; living walls could even form a cascade of greywater treatment and recycling from floor to floor (Hopkins & Goodwin, 2011).

As in the examples above, modular living walls can treat wastewater (e.g. household greywater) according to the established concepts of constructed wetlands, depending on plant types and media selection (Castellar Da Cunha, Arias, Carvalho,

Rysulova, Canals, Pérez, Gonzalez, & Morató, 2018; Rysulova, Kaposztasova, & 29

Vranayova, 2017; Pradhan, Al-Ghamdi, & Mackey, 2019). Plants treat wastewater,

especially that which is not heavily polluted, in vertical subsurface flow-constructed wetlands efficiently, wherein wastewater is fed from the top through a web of pipes, distributed over the whole basin surface, and collected at its bottom (Sakkas, 2013). With proficient design, living walls can even treat household greywater to a quality sufficient for drinking (Emeric, 2009). Moreover, evidence suggests that other types of wastewater are feasible; brewery wastewater has reduced turbidity and biological oxygen demand as a result of living wall filtration (Wolcott, 2015).

Academic studies of water chemistry parameters also support this concept.

Greywater-irrigated ornamental plants uptake nitrogen and phosphorus (Prodanovic,

McCarthy, Hatt, & Deletic, 2019). Furthermore, living walls as greywater treatment systems reduced 5-day biochemical oxygen demand, chemical oxygen demand, total suspended solids, total nitrogen, total phosphorus, and E. coli with a flow of 1000 l/m2/day (Svete, 2013). The effluent from living wall-treated greywater fulfilled the

Indian standards for reuse in irrigation for all the analyzed samples, but not all the samples fulfilled the standards for flushing toilets (Masi, Bresciani, Rizzo, Edathoot,

Patwardhan, Panse, & Langergraber, 2016). As infiltration rate increases, pollutant removal decreases, due to insufficient time for biological processes. Also, living wall media alone is limited in its ability to remove nutrients, once more highlighting the importance of vegetation (Prodanovic, Zhang, Hatt, McCarthy, & Deletic, 2018).

2.2.9 Biodiversity. Although there are insufficient studies on the use of living walls by fauna for habitat, they can host a wide variety of plant life forms in various 30 stages of development (Mayrand, Clergeau, Vergnes, & Madre, 2018). Living walls have the potential to tackle habitat fragmentation issues in cities; they could provide refuge for declining species and urban wildlife (Shimwell, 2009). Habitat function depends on container type and substrate quantity (Mayrand, Clergeau, Vergnes, & Madre, 2018).

Using native species would promote local biodiversity (Deguines, Juliard, de Flores, &

Fontaine, 2016).

Vertical greenery systems (VGSs) provide habitat for more abundant and diverse fauna compared to bare walls. Diverse plant cover and litter contribute to more prey and predator habitat, and consequently, more beetles and spiders (Chiquet, 2014; Chiquet,

Dover, & Mitchell, 2013; Madre, Clergeau, Machon, & Vergnes, 2015). Moreover, vegetated areas surrounding living walls positively affect the presence of low dispersal species (Madre, Clergeau, Machon, & Vergnes, 2015). Connecting a green roof to the ground via a living wall could enrich habitat diversity and accessibility, with the Acros

Building in Fukuoka, Japan serving as a prime example (Hopkins & Goodwin, 2011).

This is especially true for wingless species, such as certain beetles (Kyrö, Brenneisen,

Kotze, Szallies, Gerner, & Lehvävirta, 2018).

2.2.10 Food and Urban Farming. With sufficient water and nutrients, vegetables can grow on living walls (Demling, 2018; Suparwoko & Taufani, 2017). For example, chives (Allium schoenoprasum), minty herbs (Calamintha nepeta), and strawberries

(Fragaria vesca) are feasible in living wall systems (Mårtensson, Fransson, & Emilsson,

2016). Vertical farming on living walls can also be integrated into aquaponic systems

(Khandaker & Kotzen, 2018), which involve the hydroponic production of plants in 31

combination with fish farming in a closed cycle (Pérez & Coma, 2018). However, very

few academic studies exist on this specific use of living walls.

2.2.11 Socioeconomic Aspects. Living walls can add value to buildings by enhancing the aesthetic appeal and demonstrating ‘green’ initiative. They are highly visible in public, enhancing placemaking capacity. Moreover, they can help preserve the integrity and longevity of the building structure in terms of protecting the surface

finishing and augmenting the insulation (Hopkins & Goodwin, 2011). Green

infrastructure focusing on stormwater management (to which living walls can contribute) can increase real estate value if executed well (Urban Land Institute, 2017). Specifically,

Jialin (2013) observed that living walls could increase residential and commercial property values by 7% and 15%, respectively. However, no general agreement exists regarding the environmental and economic advantages of vertical greenery systems

(VGSs); further research is needed on plant properties, orientation, ventilation, materials, and substrates (Safikhani, Abdullah, Ossen, & Baharvand, 2014).

Although VGSs have high installation and maintenance costs, they also have several individual and social benefits, some of which can be measured economically

(Rosasco, 2018). Individual economic benefits include energy savings on heating/cooling, building façade preservation, and increase in real estate value

(McPherson, Scott, & Simpson, 1998; Mazzali, Peron, & Scarpa, 2012; Alexandri &

Jones, 2008; Dunnet & Kingsbury, 2008; Perini & Rosasco, 2013; Veisten, Smyrnova,

Klæboe, Hornikx, Mosslemi, & Kang, 2012; Hopkins & Goodwin, 2011). Social benefits include carbon reduction, urban water management (reduced runoff), biodiversity enhancement, habitat development, urban heat island effect reduction, added privacy, and 32

aesthetic value; however, these are more difficult to assess economically (Rosasco, 2018;

Perini & Rosasco, 2013). The most famous living walls have been designed by Patrick

Blanc for aesthetics over other benefits (Magliocco, 2018).

Some VGSs are economically sustainable over their lifecycle. Costs include

initial establishment, maintenance, and disposal (Perini & Rosasco, 2013). However,

budgeting is possible, especially with appropriate plant selection that can be grown from

seed in-situ and by avoiding the use of fertilizer (Riley, de Larrard, Malécot, Dubois-

Brugger, Lequay, & Lecomte, 2019). Modular systems are generally more expensive to

install than geotextile felt systems, costing approximately $850-$1350/m2 (Perini, Ottelé,

Haas, & Raiteri, 2011; Radić, Brković Dodig, & Auer, 2019).

Perceptions of living walls can impact real-life practice. Living walls have numerous perceived benefits such as clean air, enhanced aesthetics, biodiversity, and thermal comfort (Solera Jimenez, 2018). However, perceptions depend on the level of knowledge of the object. The city of Sydney, Australia completed a perception study on the implementation of green infrastructure, the results of which showed that people were concerned about maintenance costs and the functioning of the irrigation system but appreciated the potential environmental advantages and increase in biodiversity

(Magliocco, 2018). Similarly, in a UK-based study, willingness to pay was associated with green infrastructure that increases biodiversity. Specifically, living walls were associated with higher utility than green façades, the value exceeding the estimated costs

(Collins, Schaafsma, & Hudson, 2017). Living wall-specific perception studies are lacking in the U.S. In general, however, a higher presence of green areas in one’s living space brings about an improved perception of general health (Maas, Verheij, 33

Groenewegen, de Vries, & Spreeuwenberg, 2006), and urban sites with vegetation are

perceived to be more restorative than those without vegetation (Hernández & Hidalgo,

2005).

Numerous studies have found that urban greenery can positively impact societal

mental health. For example, window views of natural elements contributed to improved

satisfaction and well-being of residents, whereas window views of built elements only

affected satisfaction (Kaplan, 2001). Using high-resolution satellite imagery to identify greenness, schoolchildren were found to have benefitted from exposure to green space in terms of cognitive development. This could partially have resulted from the associated improved air quality of green space (Dadvand, Nieuwenhuijsen, Esnaola, Forns,

Basagaña, Alvarez-Pedrerol, Rivas, López-Vicente, De Castro Pascual, Su, Jerrett,

Querol, & Sunyer, 2015). Additionally, as the nature available in one’s surroundings

increases, the crime level of the community decreases (Weinstein, Balmford, DeHann,

Gladwell, Bradbury, & Amano, 2015).

It can be extrapolated that living walls have comparable social benefits to general

urban greenery and green roofs. Taking micro-breaks to observe green roofs can help

people give sustained attention to work tasks (Lee, Williams, Sargent, Williams, &

Johnson, 2015). Similarly, visual access to the Melbourne City Council CH2 building’s

living wall boosted productivity and decreased worker sick days (Chilla, 2004; Hopkins

& Goodwin, 2011). Living walls integrated well into the urban architecture could add

measurable value to human life (Shaikh, Gunjal, & Chaple, 2015). The limited literature

supports this extrapolation. 34

2.2.12 Case Study: Fashion Valley Mall, San Diego. The public living wall on the commercial property of Fashion Valley Mall (Figure 4), owned by Simon Properties,

is a good example of a robust modular system with a soilless media and gravity-fed low-

flow (drip) irrigation, as discussed in this thesis (Figure 5). In 2012, Rocco Campanozzi

from Mission Landscape Architecture studio designed the living wall with a system from

Tournesol Siteworks. At 74 m2, it is comprised of recycled plastic soil-based planting

modules attached to stainless steel hanging rails. The planting is formal, yet diverse, the

design concept evoking piano keys. Species include spider plants, lemon button ferns,

ribbon ferns, bugleweed, moneywort, and mondo grass (Greenroofs.com, 2018).

Figure 4. Fashion Valley Mall living wall (source: Tournesol).

Figure 5. Modular construction with organic, sponge-like media block (source:

Tournesol).

2.2.13 Conclusion. Living walls are an aspirational endeavor that bring about benefits such as heat, air, and noise pollution reduction, biodiversity, food production, health, and beauty. Researchers have been assessing whether these benefits are true in practice, lab settings, and computer models. Table 2 presents a review of modular living wall (and otherwise unspecified living wall and vertical greenery system) study results in terms of these variables (Radić, M., Brković Dodig, M., & Auer, 2019). So far, researchers are unclear as to whether living walls can achieve true economic, socially equitable, and environmental sustainability through lifecycle analysis. Some of the latest living wall research is less positive than proponents may wish, yet in order to become sustainable, the industry must not only sell the ‘wall,’ but an entire system, including rainwater storage tanks (Riley, 2017). 36

Table 2. Review of Various Benefits of Living Walls, Focusing on Modular Systems

When Possible (Radić, Brković Dodig, & Auer, 2019).

Variable Type† Result Source Cooling Modular vessels Cooling performance in Jim & He (2013)* performance humid subtropical climate Modular framed Heating and cooling Coma Arpón (2016)* boxes performance in Mediterranean warm/cool summer climate Modular framed Cooling performance in Wong, Tan, Chen, Sekar, boxes tropical rainforest climate Tan, Chan, Chiang, & Wong (2010)* Modular framed Cooling performance in Hui & Zhao (2013)* boxes unspecified climate Modular framed Cooling performance in Chen, Li, & Liu (2013)* boxes humid subtropical climate Modular framed Cooling performance in Cheng, Cheung, & Chu boxes Mediterranean hot (2010)* summer climate Modular framed Cooling performance in Solera Jimenez (2018)* boxes oceanic climate Modular wire Cooling performance in Wong, Tan, Chen, Sekar, cages tropical rainforest climate Tan, Chan, Chiang, & Wong (2010)* Modular wire Cooling performance in Davis, Ramírez, & Vallejo cages tropical savanna with dry (2015)* winter climate Air Living wall Significantly lower Timur & Karaca (2013) pollution concentrations of toxins in reduction the area surrounding a living wall Living wall Improved air quality Loh (2008) *: Empirical study †: The level of specificity varied by study. 37

Variable Type† Result Source Air Vertical Improved citywide air Hop & Hiemstra (2012) pollution greenery system quality, especially if reduction vegetation space is limited Canyon concentrations of Vertical Pugh, MackKenzie, NO2 and PM10 reduced by greenery system Whyatt, & Hewitt (2012)* 15% and 23%,

respectively Vertical Peak PM2.5 concentration Viecco, Vera, Jorquera, greenery system reduced by up to 45.3% Bustamante, Gironas, and 71% Dobbs, & Leiva (2018)* Noise Living wall Reduced noise Loh (2008) reduction Living wall Acoustic insulation up to Haggag (2010) 30 dB better than that of exposed wall Outside noise and Living wall Timur & Karaca (2013) vibration reduced by up to

40 dB Vertical 2-5 dB reduction Hop & Hiemstra (2012) greenery system Modular framed 2-3.9 dB insertion loss Wong, Tan, Tan, Chiang, boxes & Wong (2010)* Module wire 2-3.9 dB insertion loss Wong, Tan, Tan, Chiang, cages & Wong (2010)*

Living wall Significant influence of Bratičević, Ristanović, & plants Drinčić (2016)* Modular framed 15 dB weighted sound Azkorra, Pérez, Coma, boxes reduction index; 0.40 Cabeza, Bures, Álvaro, weighted sound Erkoreka, & Urrestarazu absorption coefficient (2015)* Modular framed 1 dB increase in sound Pérez, Coma, Barreneche, boxes insulation for traffic noise; Gracia, Urrestarazu, 2 dB increase in sound Burés, & Cabeza (2016)* insulation for pink noise *: Empirical study †: The level of specificity varied by study. 38

Variable Type† Result Source Noise Living wall ≤ 1 dB influence on the Van Renterghem, reduction averaged road traffic noise Hornikx, Forssen, & insertion loss over a Botteldooren (2013)* courtyard Water Living wall Enhanced urban Loh & Stav (2008) manage- stormwater management ment through recycled water or rainfall usage Modular framed Excess water collected Radić, Arsić, Džaleta, & boxes and returned to the Stevović (2012) irrigation system Living wall Water retained to control Sheweka & Magdy (2011) roof runoff Biodiversity Vertical Birds, butterflies, Timur & Karaca (2013) greenery system hoverflies, flycatchers, and moths attracted Places for insects and Vertical Hop & Hiemstra (2012) birds to hide and nest greenery system provided Vertical Potential to enhance Mayrand & Clergeau greenery system biodiversity in cities if (2018) integrated into wildlife corridors

Socio- Vertical Aesthetic value in urban Hui & Zhao (2013) economics greenery system environment; improved human health and well- being; enhanced public spaces and building identity Decreased stress; Vertical Sheweka & Magdy (2011) improved patient recovery greenery system rate; improved illness

resistance Vertical Crime reduction (reduced Jialin (2013) greenery system fear, incivilities, violent behavior) *: Empirical study †: The level of specificity varied by study. 39

Variable Type† Result Source Three categories of Socio- Vertical Sutton (2014) beauty: enjoyable, economics greenery system admirable, and

ecologically beautiful Living wall Much more creative and Perini & Ottelé (2014) aesthetic potential than green façade Hydroponic Great potential to be used Bakar, Mansor, & Harun

living wall as public art (2014)

Vertical Perceived by users as Meral, Başaran, greenery system beautiful, special, natural, Yalçınalp, Doğan, Ak, & memorable, relaxing, Eroğlu (2018)* colored, aesthetic, reliable, compatible, and functional

Vertical Increased value of real Hop & Hiemstra (2012) greenery system estate, especially if more outdoor living space created Living wall Residential and Jialin (2013) commercial property values increased by 7% and 15%, respectively Vertical Climatic stress on Johnston & Newton greenery system building façades reduced; (1996) prolonged life of buildings Life of building façades Vertical Haggag (2010) prolonged if construction greenery system had waterproof panels and

an air gap Vertical Life of building façades Wong, Tan, Chen, Sekar, greenery system prolonged by limiting Tan, Chan, Chiang, & diurnal fluctuation of wall Wong (2010) surface temperatures *: Empirical study †: The level of specificity varied by study. 40

If environmental resources are not overused and if complex benefits are taken into

account, living walls installed on a massive wall envelope can achieve sustainability in a

25-year lifetime. These qualitative and not easily measured benefits include ecosystem services, outdoor climate comfort, urban heat island mitigation, air quality, acoustic filtering, biological continuity, biodiversity, and landscape perception (Pulselli, Pulselli,

Mazzali, Peron, & Bastianoni, 2014). Living wall design should take into account integration with the building envelope, sustainable and locally produced material choices, environmental impact, and the symbiosis between the growing medium and the vegetation (Perini, Ottelé, Haas, & Raiteri, 2011). We should aspire to apply living walls in citywide ‘green’ and ‘blue’ networks.

Both living walls and natural swimming pools (NSPs) are extremely variable in their designs, types, and contexts. They can also both be compared to their more conventional counterparts, namely regularly constructed walls/façades and chemically treated swimming pools. Scholarly literature on living walls is much more extensive than that on NSPs. Although the research tends to be tempered in extolling their design features, the positive aspects of both are apparent.

2.3 Natural Swimming Pools

In terms of large, outdoor water features, a natural swimming pool (NSP) is an environmentally conscious alternative to a conventional, chemically treated pool (Figure

6; Table 3; Hoffman, 2013). The price and convenience of chlorine has made it the most widespread treatment for typical swimming pools (Olsen, 2007). Such high levels of chlorine can be detrimental to aquatic organisms. Moreover, among adolescents and 41 children, high exposure to chlorine from swimming pools has been associated with respiratory issues (Hoffman, 2013; Bernard, Nickmilder, Voisin, & Sardella, 2009).

Swimming pool chlorination forms disinfection byproducts, the risks of which have been discussed as a public health issue but need to be balanced against the risks of harmful microbiological outbreaks (Florentin, Hautemanière, & Hartemann, 2011; Chowdhury,

Alhooshani, & Karanfil, 2014).

Figure 6. Commonly designed water features.

Table 3. Comparison of Relevant Water Features in Landscape Design (adapted from

Kircher & Thon, 2016a).

Primary Water body Treatment Aesthetics purpose Chemically treated Swimming Chlorine, chlorine Primarily formal swimming pool dioxide, mineral salts, styles/forms organic biocides, ozone Natural Swimming Biological processes Variable styles/forms swimming pool (traditional to organic) Garden pond Aesthetics, Eutrophic standing Primarily organic habitat body styles/forms Constructed Water Biological processes Variable styles/forms treatment wetlands filtration (traditional to organic)

In response to health issues from chlorine byproduct exposure and pool water runoff, several European countries have popularized NSPs (Olsen, 2007). The goal of the chemical treatment of swimming pools is to interrupt natural, inherent processes, creating issues on both sides, whereas the goal of the NSP is to mobilize those natural processes in a self-regulating ecosystem (Hilleary & Gracy, 2014). Salt pools can also be designed to be a more sustainable alternative for traditional chlorine pools, but that topic is outside the scope of this study. Nevertheless, NSPs vary widely in their design; aesthetically, they can have formal or organic styles (Figure 7), and functionally, biological water purification can happen in a number of ways. 43

Figure 7. Examples of more organic (above; source: Total Habitat) and more formal

(below; source: Biotop) NSPs in terms of style and form.

NSPs are modeled after natural water systems in terms of chemistry, physics, biology, and the interactions thereof (Kircher & Thon, 2016a), applying natural limnological principles in order to foster biofilm growth and biocoenosis (BioNova 44

Natural Pools, 2019). NSPs involve a closed water and nutrient cycle except for evapotranspiration losses (Hoffman, 2013). Figure 8 describes an example of an NSP design framework. Natural wetlands provide a prototype for NSP vegetation types, habitats supporting wetland organisms, and the boggy aesthetic. The biotic community includes bacteria, zooplankton, and emergent and floating macrophytes. NSPs utilize filtration zones, similar to treatment wetlands, but they have different treatment goals, operations, aesthetics, and maintenance. Biological filtration systems such as NSPs provide wildlife habitat, year-round aesthetic interest, and plant diversity, while cultivating stewardship and reducing harmful chemical use (Hoffman, 2013).

Furthermore, they can enhance broader conservation measures, human connections to nature, and emotional health, especially with regard to experiencing small wild spaces on a daily basis (Manuel, 2003).

Figure 8. An example of NSP framework (adapted from Hilleary & Gracy, 2014).

2.3.1 Classifications. Natural swimming pools (NSPs) are classified by nutrient

limitation, filter type, and model. Figure 9 shows how these classifications relate to one

another, and Figure 10 shows the NSP biological filtration cycle. The filter types (Table

4) primarily depend on the percolation rate and vegetation. They can be employed

together or separately, which determines the model number (Table 6). This study focuses on phosphorus-limiting technical wetlands, which is part of the low-flow substrate filter system (Model 3).

Table 5

Table 6

Table 4

Figure 9. NSP classification system (adapted from Kircher & Thon, 2016a).

Figure 10. Biological filtration flow chart (adapted from FLL, 2011; Kircher & Thon,

2016a). 47

Table 4. NSP Filter Classifications (adapted from ÖNORM L, 2013; FLL, 2011; Kircher & Thon, 2016a).

Filter type Characteristics Filter Medium Limiting factor Percolation Purification Maintenance Hydrobotanical Densely planted Sand/gravel Either < 0.2 m/hr Plants, plankton Trim/harvest system filter beds in (soft water  C) (slow/standing) plants shallow water Technical Vertical or Relatively fine- Either < 0.3 m/hr Substrate, plants Trim/mow wetland horizontal grained gravel (alkaline (slow) (helophytes), plant percolation (1-8 mm; substrate  P; microorganisms sprouts, through filter limestone/ acidic substrate, adjacent to roots harvest after beds, moderately dolomite/ soft water  C) and rhizomes 10-15 years dense plantings silicate/peat

Biofilm- Vertical Highly Only P > 0.5 m/hr Biofilm Regular accumulating percolation permeable (permanent backwashing substrate through filter beds, gravel (quartz) water no planting/only with 30% movement) weak or shallow lime/dolomite rooted/floating chips (≥ 2 mm; island(s) low P products)

Filter type Characteristics Filter Medium Limiting factor Percolation Purification Maintenance Supplementary Dirt collection in N/A N/A N/A Mechanical Sieve or pre-filtering/ sieves cleaning treatment Sedimentation of N/A N/A N/A Sedimentation Sediment suspended matter chamber removal Binding N/A N/A N/A Chemical/ Backwashing phosphorus via physical binding and medium special granulates capacity exchange Aeration (oxygen N/A N/A N/A Aerobic Sporadic enrichment) organisms checks

Table 5. NSP Model Classifications (adapted from FLL, 2011; Kircher & Thon, 2016a).

Water flow Water movement Regeneration area (as a Model Filter type(s) Conditions speed method percent of pool surface) 1.Standing water N/A Natural circulation Hydrobotanical ≥ 65% densely planted P- or C-limitation (no equipment) system 2.Surface flow Slow Skimmer  pump  Hydrobotanical ≥ 50% densely planted P- or C-limitation inflow through system regeneration area 3.Low-flow Slow Skimmer  pump  Hydrobotanical ≥ 30-40% densely Primarily P- substrate filter (300-500 inflow into pool system + planted on percolated limitation, l/m2/hr) (through percolated/ technical substrate hard/soft water intermittent filter wetland regeneration area) 4.High-flow Fast Skimmer  pump  Biofilm- ≥ 25% (≥ 5-20% in P-limitation, hard substrate filter (> 500 continuously accumulating professional systems water, pH ≥ 8.4, l/m2/hr) percolated substrate substrate with supplementary high O content filters (no plants (others optional) treatment) necessary)

5.High-flow Fast Skimmer  pump  Biofilm- ≥ 20% (≥ 5-10% in P-limitation, hard technical unit (> 500 percolated substrate accumulating professional systems water, pH ≥ 8.4, (specialized l/m2/hr) filters (sand cartridge) substrate with supplementary high O content substrates and in a separate unit treatment) aeration) within a large metal vessel (no plants) 49

A nutrient is limiting when in its absence, no growth is possible. Accordingly,

adding said nutrient to the system instigates growth, and said nutrient is not limiting

when its addition to the system does not instigate growth (Gibson, 1971). In agriculture, identifying and adding the limiting nutrient back to the soil helps production, but in NSP systems, retaining the limiting nutrient keeps the algae down. Nitrogen and phosphorus are of particular importance for NSPs due to their connection to eutrophication and algae development (Hoffman, 2013). Limiting phosphorus or carbon can suppress algal growth

(Table 6). Phosphorus-limiting systems are generally easier to establish and more stable

than carbon-limiting systems. However, the choice depends largely on the available water

used to fill the NSP and the climatic conditions of the site (Kircher & Thon, 2016a).

Table 6. Comparison of P-limiting and C-limiting System Specifications (adapted from

Kircher & Thon, 2016a).

Limiting nutrient pH Hardness Climate Phosphorus pH ≈ 8 Moderate hardness Hot regions with low (P sedimentation) (P sedimentation) rainfall Carbon Low pH Low hardness Cool climates, high (reduced DIC) (reduced hydrogen rainfall (oceanic carbonate) climate)

2.3.2 Plants. Nutrient cycling is the key to successful natural swimming pools

(NSPs), and although abiotic factors affect nutrient cycling, plant-based effects often overrule them in wetland environments (Urban & Eisenreich, 1982; Hemond, 1983;

Morris, 1991; Koerselman & Verhoeven, 1992). In terms of nutrient removal, treatment 51

wetland filters using plants perform better than those without, especially when it comes to

nitrogen and phosphorus removal during the peak months of June to October (Hunter,

Combs, & George, 2001; Fraser, Carty, & Steer, 2004; Picard, Fraser, & Steer, 2005).

Plants impact nutrient availability through litter production, uptake in aboveground

biomass, decomposition leading to sedimentation, and denitrification via carbon

production in the root zone (Aerts, Verhoeven, & Whigham, 1999; Hoffman, 2013). This

supports the survival of the microbes and their filtration efforts (Lin, Jing, Wang, & Lee,

2002). Furthermore, plants can mediate the seasonal temperature changes that cause

fluctuations in nutrient filtration (Hunter, Combs, & George, 2001). Studies have

observed that diverse plantings can temper these fluctuations more effectively than

monocultures (Picard, Fraser, & Steer, 2005).

In some wetlands systems, plants do not contribute much to nutrient removal, but

in low load systems like NSPs, plants do take part in nutrient removal (Vymazal, 2007).

Although unplanted regeneration areas contribute to nutrient removal via microbial

processes, plants can further this process by enhancing microbial population richness

(Ottová, Balcarová, & Vymazal, 1997). Accordingly, planted laboratory-based NSP

variants performed better than unplanted variants in terms of water clarity (Thon,

Kircher, Pesch, Schmidt, & Thon, 2009). Also, plants can extract nutrients from moving

water more effectively than from standing water (Schwoerbel, 1987; Kircher, 2008).

For the most part, NSP plantings include helophytes from eutrophic areas, which

exhibit nutrient deficiency and consequently, weak growth, due to the strong filtration of

nutrients out of the system (Kircher, 2007; Kircher & Thon, 2016a; Kircher & Thon,

2016b; Thon, 2014). Inspiration from fens and bogs could offer a solution. Compared to 52 other ecosystems, fens and bogs have lower concentrations of nitrogen and phosphorus in mature leaves of vascular plants and higher nitrogen use efficiency on the leaf level.

Whereas nitrogen use efficiency and nutrient mineralization are comparable among fen and bog species, bog species have higher phosphorus use efficiency than fen species

(Aerts, Verhoeven, & Whigham, 1999).

Oligotrophic and mesotrophic species from bogs and fens work well in NSP regeneration areas atop submersed gravel filters (Kircher, 2007; Kircher & Thon, 2016b).

Since species from oligotrophic fens have low phosphorus requirements, plants from this habitat are well-adapted for the regeneration area of NSPs with P-limitation (Wassen,

Venterink, Lapshina, & Tanneberger, 2005; Van Duren & Pegtel, 2000; Kircher & Thon,

2016a). Acidic bogs also have depleted nutrient availability but with lower pH levels.

Sphagnum moss species form dense carpets in these habitats. Carbon-limiting NSPs could employ these and other related plant species, requiring slightly higher nutrient levels (Kircher & Thon, 2016a). In oligotrophic environments, nitrogen use efficiency is not the only beneficial factor—minimizing plant nutrient loss also contributes to adaptability. This is related to low tissue nutrient concentrations, long tissue lifespans, and night nutrient resorption from senescing tissues (Aerts & van der Peijl, 1993;

Berendse, 1994).

NSP planting designs should utilize native or climate-adapted species. Moreover, obligate wetland plants with high ground production, showy aesthetics, and high growth index measure, rather than rapid establishment, should perform well in NSP systems

(Hoffman, 2013). However, plants used in typical treatment wetlands may not be appropriate for NSPs, especially those with sharp, pointed rhizomes and those without 53 aesthetic qualities (Hoffman, 2013). Plants with aggressive rhizomes or roots can protrude through seams in an NSP liner when they get caught in a fold and cannot escape, but this situation is rare (Kircher & Thon, 2016a). Emergent plant species for NSPs include sedges, lesser cattails, and rushes; submergent species include common waterweed and hornwort; floating species include pondweeds (Buege & Uhland, 2002).

NSP planting design can also utilize naturalistic planting mixture concepts according to design function: dominant plants are tall specimens and should be sufficiently spaced out; companion plants are in small groupings; groundcover plants are low carpets, planted close together to unite the entire planting; scattered plants take up very little space and provide transient visual effects; filler plants are short-lived species that can provide visual interest while slower growers establish (Riedel, Kietsch, Heinrich,

Messer, & Kircher, 2007; Borchardt, 1996; Kircher & Thon, 2016a). NSP plantings should have a dense mix of 5-10% dominant plants, 20-40% companion plants, and at least 50% groundcover plants in order to hold back weed competition (Kircher, 2007).

This balanced plant biodiversity has impacts beyond aesthetic interest. Lime fen vegetation regeneration areas perform better than eutrophic fen vegetation regeneration areas in this respect, due to the invasion of more aggressive plants occurring on the latter

(Thon, Kircher, Pesch, Schmidt, & Thon, 2009).

2.3.3 Filter Media. Natural swimming pool (NSP) filter media should have high hydraulic conductivity, high phosphorus adsorption capacity, low phosphorus content, potential to support plant growth, long-term phosphorus saturation potential, reasonable pH, and air space (no clogging). Furthermore, it should be readily available and not 54

prohibitively expensive (Hoffman, 2013; Hatt, Fletcher, & Deletic, 2009). As has been

shown in treatment wetland systems, substrates could become saturated as the system

ages, making nutrient removal more difficult. However, since NSPs have low loads,

media saturation may not even occur until after the decline of other system elements

(Hoffman, 2013).

In order to reach a phosphorus level of ≤ 0.01 mg/l according to the FLL

guideline, vegetative filters could be augmented with filter media filtration (Hoffman,

2013). Applying reactive materials to regeneration areas can increase the proportion of area used for recreation because they remove phosphorus and protect against undesirable algae. Reactive material selection should be based on phosphorus-sorption properties and

local availability. Examples include clays (lightweight expanded clay, calcined clay),

iron-, manganese-, aluminum oxides and hydroxides, organic matter, zeolites, activated

carbon, electric air furnace slag oolites, silica-calcite sedimentary rock (opoka), and commercial products such as FerroSorp and Norlite (Bus & Karczmarczyk, 2015; Kircher

& Thon, 2016a; Vohla, Põldvere, Noorvee, Kuusemets, & Mander, 2005; Vohla, Alas,

Nurk, Baatz, & Mander, 2007; Erickson, Gulliver, & Weiss, 2007; White, Taylor,

Albano, Whitwell, & Klaine, 2011; Brix, Arias, & del Bubba, 2001; DeBusk, Dierberg,

& Reddy, 2001; Hsieh & Davis, 2005; Vohla, Kõiv, Bavor, Chazarenc, & Mander, 2011;

Hoffman, 2013).

Some of these media come with high pH, poor hydraulic conductivity, and rapid

phosphorus saturation (Hoffman, 2013). Hilleary & Gracy (2014) use expanded shale

because it has a high surface area and can absorb phosphorus. They use sandstone for the

sides of the pool, allowing the walls to act as filters. Alternatively, Thon, Kircher, Pesch, 55

Schmidt, & Thon (2009) tested wood fiber and coir as permeable regeneration area

substrates and found that the former performed better in terms of water clarity. This study

explored other filter media options relating to acidic bogs and lime fens.

2.3.4 Water Quantity. As outdoor water features, natural swimming pools

(NSPs) take part in the water cycle, taking in precipitation and surface runoff, and letting

out water in the form of evapotranspiration. NSPs are filled by tap or well water. Surface

runoff should be avoided as a source due to the added nutrients from lawn fertilizers,

pesticides, and organic matter. However, stormwater could be collected for use in NSPs

(Hoffman, 2013; Dunnett & Clayden, 2007). Given the need for a relatively stable water

level throughout, adding a separate regeneration pond downstream to handle the overflow

would enhance this design element (Kircher & Thon, 2016a). Filling water (from the tap

or well) can be a significant source of phosphorus into the NSP system, so its use should

be limited (Hoffman, 2013).

Users introduce excretions (sweat, urine, and saliva), pharmaceuticals, and

personal care products (sunscreen, cosmetics, hair products, and lotions) to the swimming

usage area. These convey microorganisms (E. coli) and plant nutrients (phosphate), on which the water purification specifications are based, through the swimmer equivalent value (FLL, 2011). The amalgamation of user inputs, recurrent additions of disinfectant, lack of freshwater inputs, and recirculating said water continuously can concentrate the resulting contaminants in the pool water (Carter & Joll, 2017). Whereas in chemically treated pools the disinfectants are employed in order to fight these and other inevitable inputs, NSPs aim to accept these nutrient loads to be incorporated as part of the natural 56

cycle (BioNova Natural Pools, 2019; Figure 10). Pollen, filling water, precipitation, and

user impurities are the main sources of phosphorus into the NSP system; precipitation

and user impurities are the main sources of nitrogen (Vymazal, 2007).

Whereas NSPs require partial refilling to replace evapotranspiration losses,

chemically treated pools require full refilling to discard imported contaminants

periodically. Evapotranspiration varies seasonally, peaking during the summer,

depending on wind, air temperature, anoxic conditions, and humidity (Cronk &

Fennessey, 2001). Evapotranspiration losses of more than 5 mm per day can occur on hot

summer days. Species with large leaves and additional water features such as streams,

waterfalls, and fountains can exacerbate the issue. However, running the water features

intermittently can curb the losses. Overall, reducing evapotranspiration can reduce

phosphorus influx, but maintaining the water level by refilling is necessary (Kircher &

Thon, 2016a). We can estimate evapotranspiration losses using NOAA evaporation pan

technical reports; open water and vegetated wetland evapotranspiration amounts are

comparable (Mitsch & Gosselink, 2007). Small plant species from oligotrophic fens are

well-adapted to phosphorus-limiting systems and exhibit relatively low

evapotranspiration (Thon, 2014), but Sphagnum moss species have high

evapotranspiration rates. Nevertheless, covering the swimming usage area in NSPs with

separate regeneration areas can decrease evapotranspiration losses.

With more evaporation comes more water replacement, which is not ideal due to

the added phosphorus from the filing water. Thus, plantings with minimal perspiration are beneficial. A comparison between a non-planted control filter, a conventionally planted filter, a lime fen planted filter, and an acidic bog planted filter showed that the 57

lime fen planted filter required the least filling water (Kircher & Thon, 2015a; Kircher &

Thon, 2016b). Calculating the operant number of gallons for a high-flow substrate filter

NSP involves calculating the amount needed for 4 full-volume daily water circulations in the warmest months (June-August). Since warmer water takes in less oxygen, this overestimation ensures sufficient oxygenation for the aerobic bacteria to filter the water

throughout the year (Hilleary & Gracy, 2014).

2.3.5 Water Quality. Natural swimming pool (NSP) users expect their pools not

to have murky water, undesirable weeds (especially string algae), and pathogenic germs.

Artificial water circulation through a natural system can reduce nutrients, preserving clear

water. Shade can also reduce algae, so locating at least part of the pool in the shade is

worthwhile (Kircher & Thon, 2016a; FLL, 2011). NSPs have to conform to specific

codes in terms of health and safety regulations; some countries like Germany have NSP-

specific standards. The German Landscape Research, Development, and Construction

Society (FLL) publishes NSP standards. In order to eradicate algae and maintain

aesthetics and health, the NSP water must remain within the standards (FLL, 2011; Table

7). Nutrients can enter the system through some obscure sources but can also be removed

(Figure 11). US-based companies also use the German standards, along with state-level

safety requirements for all swimming pools. Additionally, some set their own standards,

such as Total Habitat setting their enterococci level at 30 cfu/100 ml and fecal coliform at

200 cfu/100 ml (Hilleary & Gracy, 2014). Public pools require more regular water

chemistry testing, and the parameters can fluctuate between day and night (Kircher &

Thon, 2016a). 58

Table 7. Water Chemistry and Sanitary Microbiological Standards for NSP Water

(ÖNORM L, 2013; Frei, 2012; FLL, 2011; Baumhauer & Schmidt, 2008).

Parameter Pool water Filling water + Ammonium (NH4 ) ≤ 0.3 mg/l ≤ 0.5 mg/l Carbonate hardness (CH) ≥ 5.6°dH ≥ 5.6°dH Conductivity (at 25°C) 200-1000 µS/cm ≤ 1000 µS/cm Iron (Fe2+ / Fe3+) N/A ≤ 0.2 mg/l Manganese (Mg2+) N/A ≤ 0.05 mg/l - Nitrate (NO3 ) ≤ 30 mg/l ≤ 50 mg/l - Nitrite (NO2 ) 0.1 mg/l N/A Oxygen saturation 80-120% N/A pH 6.0-8.5 N/A 2- Sulphur (SO4 ) N/A ≤ 40 mg/l ≤ 25°C (long-term) Temperature N/A ≤ 28°C (5-day) Total hardness (TH) ≥ 1.0 mmol/l ≥ 1.0 mmol/l Total phosphorus (Ptot) < 0.01 mg/l < 0.35 mg/l Escherichia coli 100 cfu/100 ml N/A Enterococci 50 cfu/100 ml N/A Pseudomonas aeruginosa 10 cfu/100 ml N/A Legionella Below detection limit (100 ml) N/A

Figure 11. Processes by which nutrients enter and exit NSP systems (adapted from Van

Duren & Pegtel, 2000).

Because NSPs are kept in low-nutrient, closed systems, they come with distinct challenges. Algae can cause water clarity and visibility issues, endangering users.

Furthermore, algae compete with emergent macrophytes and aquatic organisms for nutrients, which can impact pH and dissolved oxygen values (Hoffman, 2013). To combat them, water purification is achieved by plants, plankton, biofilm, and technical equipment, both in-situ and ex-situ (Figure 11; Kircher & Thon, 2016a; FLL, 2011).

Skimming the usage area is the main form of mechanical filtration, skimmers collecting large debris (e.g. leaves, sticks, etc.) in the net or basket, and small debris (e.g. dust, dirt, hair, etc.) in the filter pad (Hilleary & Gracy, 2014). They filter particulate matter before it affixes to nutrients (Littlewood, 2005; von Berger, 2010). 60

Overall, biological filtration performance depends on the season; however,

phosphorus filtration varies less due to being related more to sediment adsorption rather

than biological processes (Spieles & Mitsch, 2000; Richardson, 1985). Also, organic

pollutant removal in artificial wetlands depends on the hydraulic load applied and

presence/absence of vegetation (Guardia-Puebla, Pérez-Quintero, Rodríguez-Pérez,

Sánchez-Girón, Llanes-Cedeño, Rocha-Hoyos, & Peralta-Zurita, 2019). Although plants can only uptake phosphorus to a certain point and do not represent a significant portion of its uptake (less than 10% to 20% of phosphorus inputs), they can store a significant amount of nitrogen (Lucas & Greenway, 2008; Kadlec & Wallace, 2009).

In NSPs, there is no consensus among researchers as to the degree to which vegetation impacts nutrient levels. Achieving the target phosphorus level is more difficult than the target nitrogen level, but biomass uptake can account for some nitrogen and phosphorus removal, depending largely on climate, plant age, plant health, growth stage, substrate, and nutrient loading rate. For instance, in her dissertation from Penn State,

Hoffman (2013) demonstrated that nitrogen removal significantly correlates with biomass production. Kadlec and Wallace (2009) assert that biomass removal is of greatest significance for low loading rates, corroborating Hoffman (2013). But overall, very few studies exist on nutrient removal performance of planted versus unplanted NSP biofilters.

NSPs will always have a small number of algae, even as C- and P-limitation suppresses algal emergence and growth (Jaksch, Wesner, & Fuchs, 2013). Only chemical disinfection can completely eradicate them (Graber, 2007). Algae grow in conditions with high temperatures, light intensity, nutrients, and pH levels (Kircher & Thon, 2016a).

Specifically, cyanobacteria thrive in warm, clear, still, eutrophic or hypertrophic water. 61

They can cause harm to humans via cyanobacterial toxins when they come to the surface

in a bloom (Hunter, 1998). Decreasing filling water needs and using substrates with low

organic contents can limit algal growth. The nutrient-limitation approaches depend on pH and hardness levels, C-limitation corresponding with low pH and soft water and P- limitation corresponding with high pH and harder water (Kircher & Thon, 2016a). A pH level of 8.4 optimizes biofilm development and establishes a well-buffered system

(Wesner, 2013), demonstrating P-limitation. Alternatively, filters planted with Sphagnum moss species can also control algae (Kircher & Schönfeld, 2008; Thon, 2014), which would perform well in a C-limiting system.

A comparison between a non-planted control filter, a conventionally planted filter, a lime fen vegetation filter, and an acidic bog vegetation filter showed that the acidic bog planted filter yielded the least filamentous algae (Kircher & Thon, 2015a). However, lime fen-modeled filters are still worth exploring because lime fen vegetation mat filters performed better than eutrophic fen vegetation mat filters in that they were able to derive sufficient nutrients to foster good vegetation coverage, aesthetic longevity, and diversity

(Thon, Kircher, Pesch, Schmidt, & Thon, 2009). Filter media also have an impact on filtration, as previously discussed. Oligotrophic substrates of sand, peat, and bark compost (acidic bog) and the same with added limestone chips (lime fen) yielded water with very low amounts of nitrate, ammonium, and phosphorus, achieving oligotrophic conditions (Kircher, 2007).

Phosphorus control in particular is critical for NSP maintenance. Algal blooms can occur with phosphorus levels down to 0.02 mg/l (Daniel, Sharpley, & Lemunyon,

1998), but regularly checking the filling water chemistry can ensure the proper measures 62

are taken to keep total phosphorus down (Kircher & Thon, 2016a). Also, periodic

sediment extraction removes adsorbed phosphorus, which could then be composted

(Hoffman, 2013). In order to stop nutrients from leaching back into the NSP, the

maintenance manager should remove aboveground vegetation directly after fall

dormancy (FLL, 2011), only harvesting up to 50% of the regeneration area vegetation at

one time to protect species biodiversity (Kircher & Thon, 2016a).

Whereas chlorine pool maintenance entails daily pH and chlorine checks, skimmer cleanouts, and shocking, NSPs only require water quality testing (nutrients and clarity) on a seasonal basis, sporadically removing leaves and debris from the skimmer, and harvesting plant material in the fall (Littlewood, 2005; von Berger, 2010; Kircher &

Thon, 2016a; Hoffman, 2013). NSP planting maintenance involves fertilization (as needed) and weeding. Urea (CH4N2O) can be dissolved in NSP water as a nitrogen

fertilizer for plants. This is often necessary due to the lack of nitrogen in NSP systems as

a result not only of water circulation driving nitrogen decomposition, but also of

denitrification in the anaerobic areas on the pool floor and in densely planted areas.

Vacuuming the mud off pool floor can address denitrification issues to some extent

(Kircher & Thon, 2016a). With respect to weeding, oligotrophic plantings should

eventually cover the substrate and keep weeds out; before this, weeding is necessary

(Kircher, 2007; Kircher & Thon, 2016a).

2.3.6 Biodiversity. Natural swimming pools (NSPs) can provide habitat for

beneficial animals such as dragonflies, damselflies, zooplankton (cyclops and water

fleas), water insects (pond skaters, great diving beetles, whirligig beetles), water snails, 63 frogs, toads, newts, and even mammals (mice, hedgehogs) visit sporadically to quench their thirst. Table 8 lists various species and their respective roles in NSP systems. 64

Table 8. NSP Organism Classification (adapted from Kircher & Thon, 2016a).

Organism group Desirability Benefits Challenges/Maintenance Stoneworts Yes Underwater mat Easily weeded by hand Filamentous algae No N/A Causes murky water, can thrive in oligotrophic water Phytoplankton No/little N/A Can cause murky water (green algae, blue- and toxic excretions only green algae/ in neglected NSPs cyanobacteria) Zooplankton Some Eat phytoplankton N/A Other planktonic Yes Water filtration Sedimented substances algae (diatoms, can be removed with bacteria and suction pond cleaners flagellates, amoeba, rotifers, etc.)

Microscopic Yes Water filtration N/A multicellular organisms (roundworms, small crustaceans, water mites, insect larvae) Biofilm Some Water filtration Moving water enhances (central component) uptake of nutrients, pollutants, and pathogens Mycorrhizal fungi Yes Purification Harvest plant material to (improve plants’ remove incorporated uptake of ions and substances from system contaminants) Plants Yes Purification Harvest plant material to (depending (phytoextraction of remove incorporated on design) heavy metals, substances from system suppress/compete with floating algae)

Various species of stoneworts and beneficial algae found in NSPs thrive in hard and soft water, in addition to oligotrophic conditions (Kircher & Thon, 2016a). Dragonfly 65 larvae, backswimmers, predatory water beetles, and newts control mosquito populations.

Nevertheless, users can avoid swimming with these organisms by designing their NSP with flowing water through multiple areas, the regeneration area separated from the usage area. Fish (mostly Japanese koi) are also a popular planned addition to the NSP system.

More harmful species can also appear, such as Lymnaea snails, backswimmer bugs, water scorpions, water snakes, and the parasitic trematode species Trichobilharzia ocellata, which can cause swimmer’s itch. Ducks also flock to NSPs, but their feces contaminate the water, along with other wildlife (Kircher & Thon, 2016a; Casanovas-Massana &

Blanch, 2013). In contrast, swimming pool runoff with chlorine concentrations as low as

0.011 ppm can cause harm to fish and other aquatic life when released into the environment (Olsen, 2007).

NSP customers of the design/build firm Total Habitat have reported seeing frogs

(especially in the tadpole phase), birds, dragonflies, snails, aquatic insects, snakes, deer, bears, racoons, and even a great blue heron (S. Elniff, personal communication, April 14,

2020). A 2-year-old NSP in Tennessee was home to 17 species of benthic (bottom- dwelling) macroinvertebrates, the most dominant of which were scuds and rams horn snails. These and other local species took part in the food web established by the NSP, such as leeches (mostly harmless), caddisflies, mayflies, damselflies, dragonflies, water striders, and minnows (Total Habitat, 2019).

2.3.7 Socioeconomic Aspects. Generally, people are inclined toward natural landscapes rather than built environments, in particular with elements of water and an open setting (Ulrich, 1993). As previously mentioned, natural swimming pools (NSPs) 66

can have a variety of styles, from formal to naturalist, which can be reflected in the

planting design as well. In contrast with more sterile chemically treated pools, NSPs can

highlight the blooming dynamics throughout the growing season and from year to year.

(Kircher & Thon, 2016a). Aesthetically pleasing plantings for low-nutrient NSP

environments depend on availability of suitable plant material from local nurseries, but

this is possible (Kircher, 2007). Limited space presents an issue for installing NSPs in

residential settings. Even though residents tend to enjoy the plantings around NSPs, an

emerging trend is to filter them biologically but without plants (Thon, Kircher, Pesch,

Schmidt, & Thon, 2009). This study presents a strategy to address this issue.

NSP filters do not need to be pressurized, unlike those of chemically treated

pools; thus, they require much less energy to run at the same flow rate. Furthermore, by

harnessing a renewable energy source such as with a solar array, the NSP system could

operate off the grid with a net zero carbon footprint (BioNova Natural Pools, 2019).

Therefore, the low-input, closed system philosophy of NSPs extends beyond the water

system.

Regarding the health impacts, the comparison between NSPs and chemically

treated swimming pools is most relevant. Chemically treated swimming pools must

balance the risk of adverse health effects from disinfectants with the risk of viral

outbreaks if disinfectant concentrations are too low (Bonadonna & La Rosa, 2019).

Health effects of acute exposure to chlorine can be serious, but are mostly minor, such as

dry hair and skin, fatigue, and eye irritation, redness, and watering. Corneal swelling and

erosion can occur, so wearing goggles is recommended (Olsen, 2007; Momas, Brette,

Spinasse, Squinazi, Dab, & Festy, 1993). Furthermore, swimming in chlorinated pools 67

can aggravate allergic diseases such as asthma among children and adolescents, likely

due to inadequate ventilation in situ and chlorine compounds in the air above the water

(Bernard, Nickmilder, Voisin, & Sardella, 2009; Bernard, Carbonnelle, de Burbure,

Michel, & Nickmilder, 2006; Nemery, Hoet, & Nowak, 2002). People perceive chlorine

treatment of pools to cause more health risks (eye/skin irritation, respiratory problems,

and skin dryness) than other types of treatments, such as ozone and UV, and salt

electrolysis (Fernández-Luna, Burillo, del Corral, García-Unanue, & Gallardo, 2016).

The relationship between health risks and fecal indicator concentrations is still

uncertain for NSPs. Chemically treated swimming pools have less of a sanitary risk or are

at least more controlled. They eliminate E. coli more effectively, but NSPs eliminate protozoans more effectively (Bruns & Peppler, 2019). Also, three of four tested NSPs by

Casanovas-Massana & Blanch (2013) exceeded the E. coli or enterococci limits, but

concentrations of other fecal bacteria were acceptable. Overall, however, the chance of

contracting infection from contaminated water has been vastly exaggerated (Kircher &

Thon, 2016a). The risk of infections such as Dermatophytes, Legionellae, Pseudomonas

aeruginosa, and Staphylococcus aureus is greater in chemically treated pools than in

NSPs (Müller, 2001).

Blue-green algae (cyanobacteria), known to be harmful to human health, are not

likely to be present in NSPs because they essentially only occur in eutrophic or

hypertrophic waters with low oxygen levels (Kircher & Thon, 2016a; Chorus, Deuckert,

Fastner, & Klein, 1992). Percolating filters such as technical wetlands, rock garden

filters, and biofilm-accumulating substrate filters play a large role in eradicating

pathogens, whereas densely planted regeneration areas associated with hydrobotanical 68 systems contribute in a less significant manner by promoting plant and plankton excretions (Kircher & Thon, 2016a).

2.3.8 Case Study: Los Angeles Conversion. California Natural Pools, a natural swimming pool (NSP) dealer and builder, retrofitted a damaged chemically treated pool into an NSP (Figure 12). The homeowners Robin Petersson and David Zucker also contributed to the design. The usage area covers 63.5 m², and the regeneration area covers 50.6 m². As a low-flow substrate filter, the regeneration area includes a hydrobotanical system and technical wetlands (Figure 13). The pure water flows directly into the usage area in the form of a small cascade. We chose this pool for our representative case study not only because it demonstrates the NSP model in our study, but also because it shows that conventional pools can be converted into NSPs. This pool was the first retrofit of its kind. They repaired the substantial crack in the original pool construction and resized the pool so as to balance the space between swimming and hosting guests. Furthermore, the plants in the regeneration area have brought pollinators to their garden ecosystem. This case study shows that defunct pools can gain new life as

NSPs (California Natural Pools, 2020). 69

Figure 12. Residential Los Angeles NSP by California Natural Pools.

Figure 13. Filtration and usage areas of the low-flow filter substrate NSP. 70

2.3.9 Conclusion. Natural swimming pools (NSPs) present different challenges depending on location in terms of physical issues (climate) and socioeconomic issues.

They are not very prevalent in the U.S., but if more people were aware of them as a feasible and comparably priced option, their demand would grow, and more landscape architects would be able to include them in their designs.

2.4 Added Value of Integrated System

Natural swimming pools (NSPs) with living wall filtration could be a valuable strategy for landscape architects. If living walls could function as NSP technical wetland filters, a new type of green infrastructure would be available to sites that formerly could not have accommodated an NSP, along with the compiled benefits of both elements together (Figure 14; Figure 15). Implementing living walls and natural pools in urban areas (smaller spaces) could prevent more suburban sprawl by enhancing livelihood in high density housing and/or urban infill. 71

Figure 14. Added value of integrating living walls and NSPs for landscape architects. 71

Figure 15. Conceptual sketch of a living wall acting as the regeneration area for an NSP

(adapted from Weinreich, 2005)

Vertically filtered NSPs could provide an alternative to high-flowing NSPs with added noise insulation, air purification, stormwater management, food production, and summer cooling. On steep sites, the living wall filter could connect to dynamic design

elements like streams and waterfalls. Furthermore, integration with green roofs would

offer a ground to roof connection, comprehensive water catchment systems, and multi-

layered green space. Lastly, since living walls can grow indoors, designers could

experiment with indoor pool design and indoor (active) living walls (Liem, 2014). The

potential applications are only limited by creativity. 73

In order to assess the feasibility of a living wall to function as a technical wetland

filter for an NSP, we developed a model-based pilot study of oligotrophic living wall systems to test the water against the NSP standards. We expected that the living walls could maintain oligotrophic water, but that transpiration losses would vary, and that vegetation would not have a significant impact on water quality. We also aimed to inspire new research and development by pushing against the conventional beliefs about swimming pools.

CHAPTER 3

METHODOLOGY

3.1 Study Model

Our study advances the natural swimming pool (NSP) trials from Kircher & Thon

(2016b) and Thon, Kircher, Pesch, Schmidt, & Thon (2009). Kircher & Thon (2016b) set up three different planting variants and three different water movement variants. They used liner-sealed containers filled with water, representing small NSPs, and inserted boxes filled with substrate as filters. Data collection and analysis consisted of measuring and comparing total phosphorus, nitrate, carbonate hardness, evapotranspiration, and algae growth. Similarly, Thon, Kircher, Pesch, Schmidt, & Thon (2009) studied permeable, shallow vegetation mats as oligotrophic regeneration areas for NSPs by pumping water from the containers into vegetated mats above, representing green roof filters. Our methodological framework stems from these practical experiments.

3.2 Framework

We evaluated four oligotrophic variants of model natural swimming pools (NSPs) with living wall filters. We established three replicates per variant, necessitating a total of

12 living wall elements, each made up of one module (Natur-Zaun modules; 37 cm wide x 72 cm long x 15 cm deep). We circulated water from a small basin below each living wall element, representing the NSP. This was closest to NSP Model 3, with the living wall as the technical wetland aspect, but without the hydrobotanical system. Regarding 75

the nutrient limitation of the systems, we assumed P-limitation and tested phosphorus, but

predicting this was not possible. The Sphagnum variants may have tended more toward

C-limitation, but we could not confirm this. Nevertheless, both P-limitation and C-

limitation can combat algae growth.

The variants encompassed two habitat types (lime fen with limestone gravel

substrate and acidic bog with Sphagnum moss substrate) and presence of vegetation

(vegetated and non-vegetated; Table 9; Table 10). Natur-Zaun also supplied the

Sphagnum, which was cultivated in Chile. This product is typically extracted

unsustainably, but the material we used is considered a renewable resource. A 0.5-inch

inner diameter irrigation pipe attached to a small pump (Oase AquariusUniversal

“Brunnenpumpe⁄600⁄ Neptun 600⁄10m” 7-watt pump) recirculated the water through each

living wall element into the hard-plastic aquatic plant crate (40 cm wide x 60 cm long x

28 cm high) below. These crates held about 40 liters of water. Initially, we used ventilated, stackable crates lined with a synthetic rubber liner, which we then replaced with watertight crates during the first growing season.

Table 9. Description of Variants.

Substrate Vegetation Substrate Species type(s) variant variant Lime fen Non-vegetated Limestone gravel N/A surrounded by Rockwool Lime fen Vegetated Limestone gravel Oligo- to mesotrophic surrounded by Rockwool Acidic bog Non-vegetated Sphagnum moss (block) N/A

Acidic bog Vegetated Sphagnum moss (block) Oligo- to mesotrophic

Table 10. Visual Representations of Variants.

Variant Section Photo

Pumped water

Lime fen; non- Module vegetated

Basin

Pumped water

Module Lime fen; vegetated

Basin

Pumped water

Acidic bog; Module non-vegetated

Basin

Pumped water

Module Acidic bog; vegetated

Basin

We applied simple random sampling without replacement in the field in order to

randomize the systems spatially (Figure 16; Figure 17). During April of 2019, we constructed the systems by filling the living wall elements with the filter media and planting the vegetated variants. We viewed each living wall element as a representation of a full living wall. Correspondingly, each basin represented an NSP. We discussed the results and formulated conclusions based on these assumptions. We monitored the systems for two growing seasons, until September 2020. In order to control the surrounding environmental conditions to an extent, and to ensure overwintering between the two growing seasons, we constructed the systems in a greenhouse on the campus of

Anhalt University in Bernburg, Germany.

Lime fen Acidic bog Non-       vegetated 1 7 12 2 4 8 Vegetated       6 9 10 3 5 11

           

Figure 16. Summary of variants and replicates (above) and spatial randomization of

systems (below).

Figure 17. Setup of systems in the greenhouse.

3.3 Filter Media Selection

Given that the systems used P-limitation, each filter medium had to have a low

phosphorus content. The lime fen filter medium consisted of prewashed 2-8mm limestone gravel (approximately 32 liters per element) surrounded by a layer of Rockwool to hold it

into the module grid construction. The acidic bog filters used compressed and moistened

Sphagnum moss blocks (one block per module; Figure 18). 80

Figure 18. Limestone gravel filter media (left); moistened Sphagnum moss block (right).

3.4 Vegetation Selection

We selected the plant species using prior knowledge of their functioning in each particular habitat type and based on availability. For the lime fen variant, we selected plants that could tolerate alkaline conditions; for the acidic bog variant, we selected plants that could tolerate acidic conditions (Table 11; Table 12). All the species needed to have low nutrient requirements. Moreover, we sought fast-growing herbaceous plants and grasses that were not overly erect and tall; overhanging plants are preferable for living walls. Plants with aerenchymatous (spongy) tissue provide more oxygen and keep the technical wetland filter open, due to the higher activity of microorganisms. Lastly, the acidic bog variant included two edible plants, Vaccinium macrocarpon (Large Cranberry) and Vaccinium oxycoccos (Small Cranberry), providing the social benefit of food production. We randomly placed each plant by hand, contributing to a mixed meadow effect.

Table 11. Species Planted on Vegetated Elements.

Plants per Variant Species Common name element Lime fen Allium suaveolens Fragrant Leek 2 Aster flaccidus Weak Violet Aster 1* Campanula cochlearifolia Fairy's Thimble 3* Carex capillaris Hair-like Sedge 3 Carex davalliana Davall’s Sedge 4* Carex pulicaris Flea Sedge 3 Carex serotina Little Green Sedge 4 Epipactis palustris Marsh Helleborine 2 Leontopodium sp. 2 Lysimachia nummularia Moneywort 3 Eastwood's Mimulus eastwoodiae 2 Monkeyflower Liverwort Mimulus jungermannioides 2 Monkeyflower Schoenus ferrugineus Brown Bog-rush 5-6; 5* Succisella inflexa ‘Frosted Pearls’ Devil’s Bit 3* Tofieldia calyculata German Asphodel 3 Trichophorum alpinum Alpine Bulrush 3 Acidic bog Calluna vulgaris Common Heathe 2 Hare's-tail Eriophorum vaginatum 3 Cottongrass Gaultheria mucronata Prickly Heath 2 Eastwood's Mimulus eastwoodiae 2 Monkeyflower Oclemena nemoralis Bog Aster 6 Pogonia ophioglossoides Snakemouth Orchid 4-5 Sarracenia purpurea subsp. Purple Pitcher Plant 2 purpurea Sphagnum sp. (living moss) 1 liter Trichophorum alpinum Alpine Bulrush 2 Vaccinium macrocarpon Large Cranberry 6 (on top) Vaccinium oxycoccos Small Cranberry 2 *Added after the first growing season

Table 12. Attributes of Selected Vegetation (adapted from Kircher & Thon, 2016a).

4 d

3 6

Species Sun/shade tolerance (cm) Height and Bloom month color hardness Water preference regime Water Nitrogen deman Estimated sociability Growth habit Plant type hardiness USDA zone Natural distribution Allium Companion/ 50 Aug-Sep 3  2 4a Eu suaveolens ++ scattered Companion/ Aster flaccidus 30 Jun-Jul 4 As scattered Campanula Companion/ 10 Apr-Jul 3  5 Eu cochlearifolia scattered

4Scale from 1 to 9 (1: oligotrophic; 3: oligo- to 1 6 : full sun : partial shade : full mesotrophic; 5: mesotrophic; 7: eutrophic; 9: : compact clumps/rosettes shade hypertrophic) 2 : wide clumps/tussocks 1: only thrives in soft water/acid soil 5 -- : short lifespan 2: prefers soft water - : solid presence followed by decline : moderate runners 3: accepts wide range ± : sustainable presence, not becoming overly 4: only thrives in hard water/alkaline soil competitive : strong runners (rhizomes or stolons) 3 : no connection to water/dry to damp + : moderately spreading, but rarely overly : wetland area/wet, but not flooded competitive : extremely rampant competitive : swamp area (5 cm under water to 5 cm ++ : overly competitive after a few years runners above water) +++ : outgrowing all weaker plants 82

5 1

2 4

Species Sun/shade Height (cm) Bloom Hardness Water Nitrogen Sociability Growth Type Zone distr. Nat. Companion/ Eu, Carex capillaris 40 May-Jul 3  2 scattered N-Am Companion/ Eu, Carex davalliana 20 May-Jul 3-4  2 3b ++ groundcover N-Am

Carex pulicaris 30 May-Jun 3  2 Eu Companion/ Eu, As, Carex serotina 20 May-Aug 3  2 3b + groundcover N-Am Epipactis 50 Jun-Jul 3-4  2 Companion 6a Eu, As palustris ++

5 1

2 4

Species Sun/shade Height (cm) Bloom Hardness Water Nitrogen Sociability Growth Type Zone distr. Nat. Companion/ Leontopodium sp. 15 Jun-Aug  scattered Lysimachia 5 Apr-May 3  5 Groundcover 3a Eu, As nummularia ++ Mimulus Companion/ 30 Jul-Sep 3  5 N-Am eastwoodiae scattered Mimulus Companion/ 15 Jun-Jul 3  7a N-Am jungermannioides scattered Schoenus 40 Mar-Dec 3  2 Companion 5b Eu ferrugineus ±

5 1

2 4

Species Sun/shade Height (cm) Bloom Hardness Water Nitrogen Sociability Growth Type Zone distr. Nat. Succisella inflexa Companion/ 75 Jun-Sep 2-3  5 Eu 'Frosted Pearls' scattered Tofieldia Companion/ 45 Jun-Jul 3-4  2 Eu calyculata scattered Trichophorum Eu, As, 25 May-Jul 2   2 Companion 2 alpinum ( ) ± N-Am Woody, Eu, Af Dominant/ Calluna vulgaris 60 Jul-Oct 1-2  1 multi- 4a As, ++ groundcover stem N-Am

5 1

2 4

Species Sun/shade Height (cm) Bloom Hardness Water Nitrogen Sociability Growth Type Zone distr. Nat. Eriophorum Eu, As, 60 May-Jul 1  1 Dominant 1b vaginatum + N-Am Gaultheria S-Am, 100  May-Jun 1-2  7a mucronata Au Mimulus Companion/ 30 Jul-Sep 3  5 N-Am eastwoodiae scattered Oclemena Companion/ 70 Jun-Sep 1  3a N-Am nemoralis scattered Pogonia Companion/ 70 May-Aug 1-2  3 N-Am ophioglossoides scattered

5 1

2 4

Species Sun/shade Height (cm) Bloom Hardness Water Nitrogen Sociability Growth Type Zone distr. Nat.

Sarracenia purpurea subsp. 25 Jun-Jul 1-2  3 + Companion 4b N-Am purpurea

Sphagnum sp. 10-15 N/A 1-2  2 + Groundcover 4 Wide Trichophorum Eu, As, 25 May-Jul 2   2 Companion 2 alpinum ( ) ± N-Am Vaccinium Aug-Feb; 15 1  6 ++ Groundcover 2b N-Am macrocarpon Apr-May Vaccinium 15 Jun-Aug 1  2 Groundcover 3a N-Am oxycoccos

4Scale from 1 to 9 (1: oligotrophic; 3: oligo- to 1 : full sun : partial shade : full mesotrophic; 5: mesotrophic; 7: eutrophic; 9: : wide clumps/tussocks shade hypertrophic) 2 : moderate runners 1: only thrives in soft water/acid soil 5 -- : short lifespan 2: prefers soft water - : solid presence followed by decline : strong runners (rhizomes or stolons) 3: accepts wide range ± : sustainable presence, not becoming overly 4: only thrives in hard water/alkaline soil competitive : extremely rampant competitive 3 : no connection to water/dry to damp + : moderately spreading, but rarely overly runners : wetland area/wet, but not flooded competitive : swamp area (5 cm under water to 5 cm ++ : overly competitive after a few years above water) +++ : outgrowing all weaker plants 6 : compact clumps/rosettes 87

3.5 Water Quantity Analysis

We set the pumps at about half-pressure, resulting in a percolation rate of about

300 l/m2/hr, which is considered low-flow for a natural swimming pool (NSP; Figure 19).

Our system represented a simplified vertical subsurface flow constructed wetland. In order to ensure the establishment of the plantings, we started with a continuous flow, but switched to alternating watering (starting at 6 am, ending at 7:15 pm; 15 minutes on, 45 minutes off) at the end of September 2019. Also, we applied a dimpled plastic membrane sheet on the backside of each living wall element for protection from the sun. Throughout the trial period, we measured the amount of filling water needed for each basin in order to assess evapotranspiration losses. We determined the separate effects of the filter media and vegetation on water quality using two-way repeated ANOVA.

Figure 19. Irrigation of the living walls from recirculated water in the basins below.

3.6 Water Quality Analysis

We assessed water quality, focusing on phosphorus elimination of water after circulation through the living wall-natural swimming pool (NSP) integrated systems. 89

Table 13 describes our assessment schedule for pH, conductivity, nitrate, ammonium,

total phosphorus, and total hardness, total phosphorus being the most important

parameter. We sent the water samples to the lab associated with the Balena NSP

company in Eppingen, Germany, pre-washing the sample bottles with the sample water

and collecting the samples before adding filling water to the system for that day. We used handheld meters onsite for the pH values. We also collected water chemistry data for the filling water from a soft water reservoir via the tap. We determined the separate effects of

the filter media and vegetation on water quality using two-way repeated ANOVA

excluding the baseline measurements from July 2019.

Table 13. Schedule of Water Quality Measurements.

Date Parameter(s) Purpose 22-Jul-2019 Conductivity, nitrate, Baseline assessment ammonium, total phosphorus, and total hardness 31-Jul-2019 pH Periodic assessment 2-Oct-2019 pH Periodic assessment 13-Nov-19 Conductivity, nitrate, Post-growing season ammonium, total phosphorus, assessment to check for and total hardness oligotrophic conditions 14-Nov-19 pH Periodic assessment 3-Apr-20 pH Periodic assessment 22-Jul-2020 Conductivity, nitrate, Second growing season ammonium, total phosphorus, assessment to check for and total hardness oligotrophic conditions 29-Jul-20 pH Periodic assessment

3.7 Vegetation Analysis

For the vegetated variants, we compared the quantitative and subjective aesthetic value of the lime fen and acidic bog vegetation by taking photographs of the modules each month, excluding the dormant season. From the photographs, the author assessed coverage and vitality of each module as a whole using Likert scales (Table 14) and estimated species richness, defined as the number of species clearly visible as a ratio of the total species planted for each module. We considered yellow leaves, indicating iron or nitrogen deficiency, for the vitality parameter. One limitation of the vegetation analysis was observation bias, but the evaluator was trained and consistent. We compared the lime fen and acidic bog vegetation using unpaired t-tests and Mann-Whitney tests.

Table 14. Scales for Assessing Vegetation Quality (Ramírez, Ayuga-Téllez, Gallego,

Fuentes, & García, 2011).

Parameter Numerical value and meaning (1) (2) (3) (4) (5) Coverage <5% 5-25% 25-50% 50-75% >75% Poor: Fair: Good: Very good: Excellent: Mostly Several Few dead/ Very few Vitality No dead/ dead/weak dead/weak weak plants dead/weak weak plants plants plants plants

3.8 Strategies and Limitations

Real world living walls and natural swimming pools (NSPs) require periodic maintenance. Thus, throughout the trials, we chose to address vegetation care and living wall construction issues through a few measures causing minimal disturbance. We 91 completed two rounds of plant additions, the first during the first growing season and the second before the second growing season. Furthermore, we fertilized Mimulus eastwoodiae (Eastwood’s Monkeyflower) and Lysimachia nummularia (Moneywort) of the lime fen variants by spraying their foliage with a 0.5% urea solution in April 2020.

We also removed weeds when they appeared in the vegetated living wall variants and set up a small oscillating fan facing the systems in order to address heat in the greenhouse during the summer of 2020. Occasionally, pieces of Sphagnum, limestone gravel, and plant debris fell into the basins when they came loose from the living wall modules.

Consequently, this could block the pump filters, so we cleaned them off as needed. We considered these disturbances when evaluating our results.

CHAPTER 4

RESULTS

Although the trends we observed were underpowered due to the low sample size, we can make some preliminary statements about the feasibility of living walls as natural swimming pool (NSP) filtration systems in terms of water losses, water chemistry parameters, and vegetation assessments. The results largely supported our hypotheses.

First, we expected the acidic bog variants to have higher water losses than the lime fen variants, due to the high moisture level and evaporation of Sphagnum mosses (Nichols &

Brown, 1980), which we found to be true (P=.005; Figure 20). Also, the switch from continuous to alternating irrigation in September 2019 decreased water losses. Filling water use decreased substantially during the winter months.

Figure 20. Average monthly filling water needs by variant.

Total phosphorus, the most important parameter that we tested, fell to sufficiently

low levels after the baseline measurement (Figure 21a) and did not significantly vary by filter media or vegetation. Most of the total phosphorus measurements (except from the non-vegetated lime fen variant) far exceeded the FLL standard on the first test. This could have happened due to an issue with the samples or the filter media needing to be

flushed out. Nitrate levels also did not vary significantly by filter media or vegetation and

remained below standard across the board (Figure 21b). Lastly, ammonium levels did not

vary significantly by filter media or vegetation, but the non-vegetated acidic bog variant

average exceeded the standard on the two test dates after the baseline (Figure 21c).

(a)

(b)

(c)

*Baseline measurements (July 22, 2019) were taken just after irrigation was turned on, before filtration could be achieved.

†Ptot values for the baseline acidic variants 2019 were omitted from the display due to being extremely high. Figure 21. Water chemistry parameter measurements by variant (nutrients). 95

Conductivity did not vary significantly by filter media or vegetation, and only the vegetated acidic bog variant average fell slightly below the standard range in the baseline assessment (Figure 22a). Total hardness climbed to sufficiently high levels after the baseline measurement, with the exception of the vegetated acidic bog variant average on the first test date after the baseline (Figure 22b). Filter media influenced total hardness, the lime fen variants having significantly harder water than the acidic bog variants, as expected (P=.001). Lastly, the pH varied significantly by filter media, as expected

(P<.001; Figure 22c). The acidic bog variants kept the pH at a low, steady level, which were all outside of the FLL standard range. The lime fen variants maintained a slightly alkaline pH within the FLL standard range. Despite the slightly less reliable water chemistry measurements of the acidic bog variants, oligotrophic conditions were achieved overall.

(a)

(b)

(c)

*Baseline measurements (July 22, 2019) were taken just after irrigation was turned on, before filtration could be achieved. Figure 22. Water chemistry parameter measurements by variant. 97

Anecdotally, we found more filamentous algae in the lime fen variants than in the

acidic bog variants. In general, we observed some floating and some filamentous algae,

but not a substantial amount. The visual water quality was good over the duration of the

study.

Vegetation did not negatively impact water quality, so the added visual benefits

were worth considering. We compared the lime fen and acidic bog vegetation in terms of

coverage, vitality, and species richness (Figure 23). Acidic bog vegetation scored

significantly higher than lime fen vegetation in terms of both coverage (P=.016 according to the unpaired t-test; P=.011 according to the Mann-Whitney test) and vitality (P<.0001 according to both tests). Also, the estimated species richness differed significantly between lime fen and acidic bog vegetation (P=.023). The urea solution foliage application to the acidic bog species on April 3, 2020 improved the blooms but had little effect on the paleness of the leaves. Mimulus (monkeyflower) species require more nutrients than were available. Throughout the experiment, the vegetated acidic bog variants performed better visually, but toward the end, the vegetated lime fen variants trended toward similar coverage, vitality and species richness.

Figure 23. Visual assessments of vegetated variants. 99

The acidic bog vegetation was more consistently attractive than the lime fen

vegetation throughout the experiment, covering the filter media well even though the

species richness decreased over time. Living Sphagnum orblike formations flourished

(Figure 24a), which could have made the water even more acidic, due to the specialized

+ + + cells that exchange cations (K , NH4 ) for H ions. Vaccinium macrocarpon (Large

Cranberry), an aggressive runner, outcompeted some of the other species planted.

Nevertheless, this plant had the added benefit of producing an edible product, cranberries

(Figure 24b). The Rockwool outer layer of lime fen living wall modules exhibited a

pronounced vertical moisture gradient, with heavy moisture at the bottom and dryness at

the top (Figure 24c), causing some issues with the planting. Some plants randomly placed

toward the bottom of the modules failed to establish and thrive, specifically the

Leontopodium sp. and Carex serotina (Little Green Sedge). As a result, the lime fen variants needed more plant addition and replacement.

100

(a) (b) (c)

Figure 24. Formation of living Sphagnum (a); cranberries (b); moisture gradient of lime fen variants (c).

Before the second growing season in 2020, we filled in bare spots of the lime fen variants by planting Schoenus ferrugineus (Brown Bog-rush), Carex davalliana (Davall’s

Sedge), Aster flaccidus (Weak Violet Aster), Succisella inflexa 'Frosted Pearls' (Devil’s

Bit), and Campanula cochlearifolia (Fairy’s Thimble; see Appendix B). Lime fen

variants performed slightly better in terms of water quantity and quality, whereas acidic

bog variants performed better in terms of vegetation. Therefore, applying one of these

systems would likely come with tradeoffs.

101

CHAPTER 5

DISCUSSION

5.1 Key Findings and Overall Feasibility

This work is wholly original to our knowledge, as living walls have never been studied academically as natural swimming pool (NSP) filters. The added value of the living wall to the NSP system, including cooling, noise reduction, air purification, food harvest, and rainwater catchment, could make this feature an attractive option for pool and green infrastructure designers. We found sufficient success in terms of the water quality and vegetation growth to state that living walls are feasible as technical wetland filters for NSPs.

We found that over time, all the variants produced good water quality for NSPs,

with the exception of the ultimately high ammonium levels of the non-vegetated acidic

bog variant. Total phosphorus levels specifically remained below maximum limit. This is

a good sign for NSP safety, as total phosphorus levels maintained below 0.03 mg/l can

prevent masses of cyanobacteria from forming (Chorus, Deuckert, Fastner, & Klein,

1992). This finding is consistent with previous studies on the use of Sphagnum moss and

limestone gravel for water filtration. Peat, the accumulation of dead Sphagnum, has a

high specific surface area, which contributes significantly to its purification capacity

(Babel & Kurniawan, 2003). Furthermore, peat can efficiently adsorb contaminants such

as heavy metals, cyanide, phosphate, and organic matter, due to its biological makeup

and chemical bonding capacity (Coupal & Lalancette, 1976; Ho & McKay, 2000). 102

Sphagnum extraction from places like Chile could threaten the existence of the resource

(Díaz, Tapia, Jiménez, & Bacigalupe, 2012), so it should be sourced responsibly.

Limestone gravel could be a more sustainable option as a filter medium but is not

regenerative. It is widely used to remove metals from wastewater and can also effectively remove ammonia, biological oxygen demand, chemical oxygen demand, turbidity, and

color (Aziz, Adlan, Zahari, & Alias, 2004; Affam & Adlan, 2013). Therefore, the acidic

bog and lime fen filter media performed predictably well in terms of water quality.

Neither the acidic bog nor the lime fen plants affected water quality, but we were

not surprised by this result because in practice, the primary function of NSP technical

wetland plants is to keep the filter body open with their roots, not necessarily to uptake

pollutants. Nevertheless, living wall vegetation can mitigate noise pollution, air pollution,

and building thermal effects, and it can enhance aesthetics and biodiversity (Table 2).

Thus, since the vegetation did not negatively impact the water quality, the benefits of the

presence of living wall plants could outweigh the costs. Several endangered plant species

persist under phosphorus-limiting conditions; thus phosphorus-limiting ecosystem

preservation and restoration is important for conservation of endangered species

(Wassen, Venterink, Lapshina, & Tanneberger, 2005). Although we used a specific plant

palette for our climatic conditions, it could be adapted to other climates by substituting

plants with similar tolerances for high/low pH levels and equivalent ecological roles and

sociability factors, keeping in mind local invasive species and those with the potential to

become invasive.

Since NSPs and living walls are perceived as more environmentally friendly to

conventional pools and walls, they should actually be applied in such a way as to limit 103 resource use. Thus, the high levels of evapotranspiration-causing water losses of the acidic bog variants, consistent with the findings of Kircher & Thon (2015a), could be a hindrance to their real-life application. However, this high evapotranspiration would increase the cooling effect of a building walled with such a vertical filter. The greenhouse conditions could have increased the evapotranspiration rate, and as long as dripping from the plants and substrate fell into the basin, water loss was minimal. Also, covering the basins would have decreased water losses, as pool coverings do (Lee & Heaney, 2008).

More formally designed NSPs commonly have receding coverings, and more organic forms can utilize solar sun rings, or “lily pad” pool coverings.

Evaluating the evapotranspiration losses of this study in relation to other studies, existing pools, or constructed wetlands presents difficulties. The water balance, or budget, for a swimming pool is defined as:

water volume / time = Inputs Outputs

InputsΔ = PrecipitationΔ + Filling −

Outputs = Evapo(transpi)ration + Backwash + Splash-out + Leaks

(Lee & Heaney, 2008; Hoffman, 2013; Mitsch & Gosselink, 2007).

All the variables depend on localized conditions, and evapotranspiration specifically can vary spatiotemporally due to the relation of the water table to the root zone and the timing of plant senescence. Thus, we cannot extrapolate our estimates to compare to a larger wetland complex (Lott & Hunt, 2001). Generally, though, along other NSP technical wetlands, managing evapotranspiration comes with striking balances. For example, a tradeoff exists between evapotranspiration rate and string algae growth (Kircher & Thon,

2016b). Although evapotranspiration can increase pollutant accumulation in media of 104

constructed wetlands, pollutant removal can decrease if evapotranspiration increases over

a certain threshold (Białowiec, Albuquerque, & Randerson, 2014). This determination is

site-specific, the more prudent choice depending on the alternatives, in terms of water

consumption.

Due to the issues we observed with the acidic bog system, its application is more

limited than the lime fen system. The acidic bog filter could be used in two ways. First, it

could promote a C-limiting, soft water system while also minimizing phosphorus levels.

This is recommended mainly in places with cool summers and high precipitation at high

altitudes (oceanic or mountainous climates). Second, it could be used for its filter medium only, to fix the roots of plants. Combining the Sphagnum moss with carbonate

gravel would then stabilize the pH, promoting P-limitation. Lime fen plants (or plants preferring a pH > 8) could be used. Sphagnum may not survive on limestone, but some species that could work include S. palustre, S. subsecundum, S. squarrosum, S. contortum, and S. platyphyllum. Overall though, the lime fen system is more applicable.

If a pool designer were to use our results, they should make some changes to this pilot study. First, combining the living wall technical wetland with a hydrobotanical system according to FLL standards, feasibly in the catchment basin, would complete the

NSP system (see Figure 15). Also, our living wall wire cage construction was not robust enough and could just barely take the weight of the wet filter media, even with extra wire reinforcement for the lime fen modules. A system with solid material walls with holes for plants would be a more appropriate construction. For example, Biotecture’s system could better support the filter media we used (Figure 25). Our results correspond with the conclusion that a green roof can function as the regeneration area for an NSP (Figure 26; 105

Thon, Kircher, Pesch, Schmidt, & Thon, 2009). Living wall filters could connect NSPs to a greater green infrastructure or wastewater treatment network, such as with green roofs.

Figure 25. Commercial living wall system (adapted from Biotecture).

106

Figure 26. Conceptual sketch of a green roof acting as the regeneration area for an NSP

(adapted from Weinreich, 2005).

5.2 Limitations

The fundamental limitation of this study was the small sample size. We approached it as a pilot study of feasibility with an emphasis on real-world applications.

Thus, we chose to execute disturbances to the systems during the experimental period according to the conditions of the systems at certain times of observation. Otherwise, the water chemistry samples could have been altered by plastic from the sample bottle, affecting phosphorus measurements especially. Also, the transport of the samples could have caused CO2 production from suspended algae, again affecting the lab results.

Furthermore, we selected plants based not only on the conditions of the systems, but also 107

on availability. Consequently, our plants were not entirely weed free, so we had to

monitor and remove weeds in the plant establishment phase. Weeds emerged that looked

like Carex (sedge) species, so it was difficult to pull them before their roots established.

However, most practical situations face this challenge and facilitators should be prepared

to weed as a part of the maintenance plan.

5.3 Further research

Subsequent studies would assess the nutrient reduction capacity specifically, and the longer-term and larger scale viability of the living walls as technical wetland filters.

By adding nutrients into the system in the form of a dissolved solid inorganic macronutrient fertilizer, we could determine if the systems could filter it out and reach

the FLL standards again. By expanding, researchers could assess the thresholds and

replacement needs of the filtration system. Both the flow rate and the age of the filter can impact the phosphorus retention capacity in mineral filters, low flow filters continuing to

decrease over time (Karczmarczyk, Bus, & Baryła, 2019). The study could also be

relocated in order to test the adaptability in other climates in terms of plant selection and

evapotranspiration issues. In order to assess the vegetation, independent observers

unaware of treatment groups could be surveyed. As we researched natural swimming

pools (NSPs) and living walls, we discussed the concept with various NSP experts. We

found that added values of living wall filters could be of interest to design-build

landscape design firms, so case studies of potential installations and retrofits would

inform them on the possibilities.

108

5.4 Conclusions

Living walls as technical wetland filters for natural swimming pools (NSPs) have

a sufficient amount of promise to justify further studies. Using oligotrophic filter media

and vegetation kept nutrient levels in the water low while growing aesthetically pleasing

living walls. Although the acidic bog variants grew superior vegetation, their

evapotranspiration levels were higher, and this system has fewer practical applications

when compared with the lime fen system. An NSP with a vertical filtration system could offer landscape architects and pool builders a new type of green infrastructure for sites that formerly could not have accommodated an NSP. In urbanized areas especially, the added benefits of living walls include cooling, tranquility, stormwater management, aesthetic value, and added privacy. Our findings fit with the general flexibility and adaptability of NSPs to various sites. As the U.S. faces more water shortages, heat waves, and sprawling development, we could integrate our pools into our water systems more ecologically, which NSPs and living walls could facilitate.

109

CHAPTER 6

HYPOTHETICAL DESIGN STUDY

This exploratory study concludes with a practical design concept for a living wall integrated natural swimming pool (NSP). In order to assess the feasibility and demand for living wall filters for NSPs, we met with associates from several NSP companies in the

US and Germany, visiting several sites both in progress and completed. One California- based NSP dealer, California Natural Pools, was particularly interested in the potential of vertical living wall filters. Thus, we inquired about any of their projects for which a living wall technical wetland filter would be appropriate, and they helped us identify a heavily sloped, urbanized site in Los Angeles with great potential. After touring and analyzing the property, we developed a conceptual plan addressing the filtration area, risk management, local codes for water use, and climatically appropriate vegetation.

The residence is located in the hilly neighborhood of Silver Lake in East Los

Angeles, built into the side of a large hill in the densely packed suburbs. From the house to the road, the land slopes downward, approximately 19% overall (Figure 27). The selected residence installed an NSP in 2018 (Figure 28) with a pool usage area of approximately 20 m2, an adjacent regeneration area of approximately 11 m2, and

cascading regeneration areas below of approximately 10 m2. The biologically purified

water overflows into several catchment basins at different elevations before it is pumped

back to the swimming pool at the top of the hill.

110

Figure 27. Geographical context and slope.

111

Figure 28. Existing NSP with cascading regeneration areas (top left source: BioNova,

2018; other photos taken by the author).

The homeowners could increase the size of their pool usage area while decreasing the size of their ground-plane regeneration area, instead using the available vertical space.

Figure 29 demonstrates how this could look and function with a more vertical low-flow substrate filter system. This design would increase the pool usage area by 10 m2, in

addition to increasing the available regeneration area with the use of a living wall as a

technical wetland filter. After percolating through the living wall and the hydrobotanical

system below, the pure water would be pumped back up into the usage area (Figure 30).

The vegetable garden is already located downflow from the pool and filter system, so the

overflow water could be used for irrigation. 112

Figure 29. Potential conceptual plan. 113

Figure 30. Water circulation and basic plumbing system.

Climate conditions determine plant selection; the site is located in USDA hardiness zone 10b, so plant species should be able to tolerate minimum temperatures of

35-40°F. The plant palette of the living wall could include such species as Acorus gramineus (Japanese Sweet Flag), Anemopsis californica (Yerba Mansa), Caltha palustris (Marsh Marigold), Eleocharis maculosa (Spikerush), Houttuynia cordata

(Houttuynia), Juncus effusus (Soft Rush), Schoenus ferrugineus (Brown Bog-rush), and

Schoenus nigricans (Black Bog-rush; Kircher & Thon, 2016a). The plants should not be too tall or erect and should be able to tolerate oligotrophic conditions. The water source for the site generally supplies hard water. This necessitates a P-limiting system with limestone gravel filter media. 114

Los Angeles has one of the highest pool concentrations in the US, and thus, the

potential retrofitting and installation opportunities are substantial (Forrest & Williams,

2010). The state of California in particular must routinely deal with water and

development issues in their zoning codes. The pool safety codes require some kind of

enclosure, cover and/or alarm system, which our site already has. As California faces

drought and water shortages, the water laws may change for outdoor water use as they

have for indoor water use, but currently, residents can still install swimming pools. With

proper maintenance and covering, their water use remains acceptable.

If living wall filters could enable more Los Angeles residents to be able to build

NSPs, we believe that this could be beneficial when compared to chemically treated pools or other outdoor water uses. Living walls could use less water than in-ground gardens,

and both together could be integrated into wastewater management systems, rescuing

water from being lost. Generally, living walls have so far remained outside of the scope

of local building code, so permitting would depend on the specific situation. However,

several constituencies across the country offer incentives for installing green or

stormwater infrastructure (Green Roofs for Healthy Cities, 2019), under which an NSP

with a living wall filter would fall. Therefore, this concept could be reproduced around

Los Angeles and beyond.

115

REFERENCES

Aerts, R., & van der Peijl, M. J. (1993). A simple model to explain the dominance of low- productive perennials in nutrient-poor habitats. Oikos, 66(1), 144.

Aerts, R., Verhoeven, J. T. A., & Whigham, D. F. (1999). Plant‐mediated controls on nutrient cycling in temperate fens and bogs. Ecology, 80(7), 2170-2181.

Affam, A. C., & Adlan, M. N. (2013). Operational performance of vertical upflow roughing filter for pre-treatment of leachate using limestone filter media. Journal of Urban & Environmental Engineering, 7(1), 117-125.

Alexandri, E., & Jones, P. (2008). Temperature decreases in an urban canyon due to green walls and green roofs in diverse climates. Building and Environment, 43(4), 480–493.

Ascione, F. (2017). Energy conservation and renewable technologies for buildings to face the impact of the climate change and minimize the use of cooling. Solar Energy, 154, 34–100.

Aziz, H. A., Adlan, M. N., Zahari, M. S. M., & Alias, S. (2004). Removal of ammoniacal nitrogen (N-NH3) from municipal solid waste leachate by using activated carbon and limestone. Waste Management & Research, 22(5), 371-375.

Azkorra, Z., Pérez, G., Coma, J., Cabeza, L. F., Bures, S., Álvaro, J. E., Erkoreka, A., & Urrestarazu, M. (2015). Evaluation of green walls as a passive acoustic insulation system for buildings. Applied Acoustics, 89, 46–56.

Babel, S., & Kurniawan, T. A. (2003). Low-cost adsorbents for heavy metals uptake from contaminated water: A review. Journal of Hazardous Materials, 97(1-3), 219- 243.

Baik, J. J., Kwak, K. H., Park, S. B., & Ryu, Y. H. (2012). Effects of building roof greening on air quality in street canyons. Atmospheric Environment, 61, 48–55.

Bakar, N. I. A., Mansor, M., & Harun, N. Z. (2014). Vertical greenery system as public art? Possibilities and challenges in Malaysian urban context. Procedia-Social & Behavioral Sciences, 153, 230-241.

Baumhauer, J., & Schmidt, C. (2008). Schwimmteichbau: Handbuch für Planung [Swimming pond construction: Manual for planning], Technik und Betrieb [Technology and Operation]. Berlin-Hannover, Germany: Patzer.

116

Berendse, F. (1994). Competition between plant populations at low and high nutrient supplies. Oikos, 71(2), 253.

Bernard, A., Carbonnelle, S., de Burbure, C., Michel, O., & Nickmilder, M. (2006). Chlorinated pool attendance, atopy, and the risk of asthma during childhood. Environmental Health Perspectives, 114(10), 1567–1573.

Bernard, A., Nickmilder, M., Voisin, C., & Sardella, A. (2009). Impact of chlorinated swimming pool attendance on the respiratory health of adolescents. Pediatrics, 124(4), 1110–1118.

Besir, A. B., & Cuce, E. (2018). Green roofs and facades: A comprehensive review. Renewable and Sustainable Energy Reviews, 82, 915–939.

Białowiec, A., Albuquerque, A., & Randerson, P. F. (2014). The influence of evapotranspiration on vertical flow subsurface constructed wetland performance. Ecological Engineering, 67, 89-94.

BioNova Natural Pools. (2019). The definitive guide to natural swimming pools and ponds. Unpublished manuscript.

Bonadonna, L., & La Rosa, G. (2019). A review and update on waterborne viral diseases associated with swimming pools. International Journal of Environmental Research and Public Health, 16(2), 166.

Borchardt, W. (1996). Pflanzenverwendung im Garten- und Landschaftsbau – Der Gärtner Vol. 6 [Plant use in gardening and landscaping – The Gardener Vol. 6]. Stuttgart, Germany: Ulmer.

Bratičević, J., Ristanović, I., & Drinčić, D. (2016, June). Apsorpciona Svojstva Razlicˇitih Tipova Zemljišta I Moguc ́nost Primene u Tehnologiji Vertikalnih Vrtova [Absorption properties of different soil types and possibility of application in vertical garden technology]. Paper presented at ETRAN 2016 - 60th Conference, Zlatibor, Serbia.

Brix, H., Arias, C. A., & del Bubba, M. (2001). Media selection for sustainable phosphorus removal in subsurface flow constructed wetlands. Water Science and Technology, 44(11–12), 47–54.

Bruns, S., & Peppler, C. (2019). Hygienic quality of public natural swimming pools (NSP). Water Supply, 19(2), 365–370.

Buege, D., & Uhland, V. (2002, August/September). How to build a natural swimming pool. Mother Earth News. https://www.motherearthnews.com/diy/natural- swimming-pool-zmaz02aszgoe 117

Bus, A., & Karczmarczyk, A. (2015). Kinetic and sorption equilibrium studies on phosphorus removal from natural swimming ponds by selected reactive materials. Fresenius Environmental Bulletin, 24, 2736-2741.

California Natural Pools. (2020). Our recent projects. California Natural Pools. https://californianaturalpools.com/recent-projects/

Cameron, R. W. F., Taylor, J. E., & Emmett, M. R. (2014). What’s ‘cool’ in the world of green façades? How plant choice influences the cooling properties of green walls. Building & Environment, 73, 198–207.

Carter, R. A. A., & Joll, C. A. (2017). Occurrence and formation of disinfection byproducts in the swimming pool environment: A critical review. Journal of Environmental Sciences, 58, 19–50.

Casanovas-Massana, A., & Blanch, A. R. (2013). Characterization of microbial populations associated with natural swimming pools. International Journal of Hygiene & Environmental Health, 216(2), 132–137.

Castellar Da Cunha, J. A., Arias, C. A., Carvalho, P., Rysulova, M., Canals, J. M., Perez, G., Gonzalez, M. B., & Morató, J. F. (2018). “WETWALL”—An innovative design concept for the treatment of wastewater at an urban scale. Desalination and Water Treatment, 109, 205–220.

Charoenkit, S., & Yiemwattana, S. (2016). Living walls and their contribution to improved thermal comfort and carbon emission reduction: A review. Building & Environment, 105, 82–94.

Chen, Q., Li, B., & Liu, X. (2013). An experimental evaluation of the living wall system in hot and humid climate. Energy & Buildings, 61, 298–307.

Cheng, C. Y., Cheung, K. K., & Chu, L. M. (2010). Thermal performance of a vegetated cladding system on I walls. Building & Environment, 45(8), 1779-1787.

Chilla, T. (2004). ‘Natur’-Elemente in der Stadtgestaltung [‘Elements of Nature’ and Urban Design]. Diskurs, Institutionalisierung und Umsetzungspraxis am Beispiel von Fassadenbegrünung [Discourse, Institutionalization and Implementation Practice Using the Example of Façade Greening].

Chiquet, C. (2014). The animal biodiversity of green walls in the urban environment (Unpublished doctoral dissertation). Staffordshire University, Staffordshire, England.

Chiquet, C., Dover, J. W., & Mitchell, P. (2013). Birds and the urban environment: The value of green walls. Urban Ecosystems, 16(3), 453–462. 118

Chorus, I., Deuckert, I., Fastner, J., & Klein, G. (1992). Toxine und Allergene aus Algen in Badegewässern [Toxins and allergens from algae in bathing water]. Bundesgesundheitsblatt, 35(8), 404-407.

Chowdhury, S., Alhooshani, K., & Karanfil, T. (2014). Disinfection byproducts in swimming pool: Occurrences, implications, and future needs. Water Research, 53, 68-109.

Collins, R., Schaafsma, M., & Hudson, M. D. (2017). The value of green walls to urban biodiversity. Land Use Policy, 64, 114–123.

Coma Arpón, J. (2016). Green roofs and vertical greenery systems as passive tools for energy efficiency in buildings (Unpublished doctoral dissertation). University of Lleida, Lleida, Spain.

Coupal, B., & Lalancette, J. M. (1976). The treatment of waste waters with peat moss. Water Research, 10(12), 1071-1076.

Cronk, J. K., & Fennessey, M. S. (2001). Wetland plants: Biology and ecology. Boca Raton, FL: Taylor & Francis Group.

Dadvand, P., Nieuwenhuijsen, M. J., Esnaola, M., Forns, J., Basagaña, X., Alvarez- Pedrerol, M., Rivas, I., López-Vicente, M., De Castro Pascual, M., Su, J., Jerrett, M., Querol, X., & Sunyer, J. (2015). Green spaces and cognitive development in primary schoolchildren. Proceedings of the National Academy of Sciences, 112(26), 7937–7942.

Daniel, T. C., Sharpley, A. N., & Lemunyon, J. L. (1998). Agricultural phosphorus and eutrophication: A symposium overview. Journal of Environmental Quality, 27(2), 251–257.

Darlington, A. B., Dat, J. F., & Dixon, M. A. (2001). The biofiltration of indoor air: Air flux and temperature influences the removal of toluene, ethylbenzene, and xylene. Environmental Science & Technology, 35(1), 240–246.

Davis, M. J. M., Ramirez, F., & Vallejo, A. L. (2015). Vertical gardens as swamp coolers. Procedia Engineering, 118, 145-159.

DeBusk, T. A., Dierberg, F. E., & Reddy, K. R. (2001). The use of macrophyte-based systems for phosphorus removal: An overview of 25 years of research and operational results in Florida. Water Science & Technology, 44(11–12), 39–46.

Deguines, N., Julliard, R., de Flores, M., & Fontaine, C. (2016). Functional homogenization of flower visitor communities with urbanization. Ecology & Evolution, 6(7), 1967–1976. 119

Demling, F. (2018). Food production on greening of roofs and façades. Acta Horticulturae, 1215, 171–174.

Dettmar, J., Pfoser, N., & Sieber, S. (2016). Gutachten Fassadenbegrünung [Expert opinion on façade greening]. Gutachten über quartiersorientierte Unterstützungsansätze von Fassadenbegrünung für das Ministerium für Klimaschutz, Umwelt, Landwirtschaft, Natur-und Verbraucherschutz (MKUNLV) NRW [Expert opinion on neighborhood-oriented support approaches for façade greening for the Ministry for Climate Protection, Environment, Agriculture, Natura, and Consumer Protection]. Darmstadt.

Díaz, M. F., Tapia, C., Jiménez, P., & Bacigalupe, L. (2012). Sphagnum magellanicum growth and productivity in Chilean anthropogenic peatlands. Revista Chilena de Historia Natural [Chilean Journal of Natural History], 85(4), 513-518.

Dunnett, N., & Clayden, A. (2007). Rain gardens: Sustainable rainwater management for the garden and designed landscape. Portland, OR: Timber Press.

Dunnett, N., & Kingsbury, N. (2008). Planting green roofs and living walls. Portland, OR: Timber Press.

Emeric, B. (2009) An innovative way to treat wastewater. Unpublished manuscript, Uppsala University, Uppsala, Sweden.

Erickson, A. J., Gulliver, J. S., & Weiss, P. T. (2007). Enhanced sand filtration for storm water phosphorus removal. Journal of Environmental Engineering, 133(5), 485- 497.

EU (ed.). (2006). Directive 2006/7/EC of the European Parliament and the Council of 15 February 2006 concerning the management of Bathing Water Quality and Repealing Directive 76/160/EEC. Official Journal of the European Union, L 64/37.

Feng, H., & Hewage, K. (2014). Energy saving performance of green vegetation on LEED certified buildings. Energy & Buildings, 75, 281–289.

Fernández-Cañero, R., Pérez Urrestarazu, L., & Perini, K. (2018). Vertical greening systems: Classifications, plant species, substrates. In G. Pérez & K. Perini (Eds.), Nature-based Strategies for Urban & Building Sustainability (pp. 45–54). Oxford, UK: Butterworth-Heinemann.

Fernández-Luna, Á., Burillo, P., Felipe, J. L., del Corral, J., García-Unanue, J., & Gallardo, L. (2016). Perceived health problems in swimmers according to the chemical treatment of water in swimming pools. European Journal of Sport Science, 16(2), 256–265. 120

FLL. (2011). Recommendations for planning, construction, servicing and operation of outdoor swimming pools with biological water purification (swimming and bathing ponds). Retrieved from https://shop.fll.de/de/english- publications/swimming-pools-with-biological-water-purification-405.html

Florentin, A., Hautemanière, A., & Hartemann, P. (2011). Health effects of disinfection by-products in chlorinated swimming pools. International Journal of Hygiene & Environmental Health, 214(6), 461–469.

Forrest, N., & Williams, E. (2010). Life cycle environmental implications of residential swimming pools. Environmental Science & Technology, 44(14), 5601-5607.

Franco, A., Fernández-Cañero, R., Pérez-Urrestarazu, L., & Valera, D. L. (2012). Wind tunnel analysis of artificial substrates used in active living walls for indoor environment conditioning in Mediterranean buildings. Building & Environment, 51, 370–378.

Fraser, L. H., Carty, S. M., & Steer, D. (2004). A test of four plant species to reduce total nitrogen and total phosphorus from soil leachate in subsurface wetland microcosms. Bioresource Technology, 94(2), 185–192.

Frei, M. (2012). Grundlagen der Bildung von Algen und Biofilm in Abhängigkeit von Nährstoffangebot und – anlieferung [Basics of the formation of algae and biofilm depending on the supply and supply of nutrients] (Unpublished master’s thesis). ZHAW School of Life Sciences and Facility Management, Wädenswil, Switzerland.

Gibson, C. E. (1971). Nutrient limitation. Journal (Water Pollution Control Federation), 2436-2440.

Gore, A. J. P. (1983). Ecosystems of the world. Mires: swamp, bog, fen and moor. Elsevier: Amsterdam, Netherlands.

Graber, A. (2007). Forschungsresultate aus der Schweiz [Research results from Switzerland]. Presentation to the International Congress for Natural Bathing Waters. Hannover, Germany.

Green Roofs for Healthy Cities. (2019). Green roof and wall policy in North America: Regulations, incentives, and best practices (2019). https://static1.squarespace.com/static/58e3eecf2994ca997dd56381/t/5d84dfc371cf 0822bdf7dc29/1568989140101/Green_Roof_and_Wall_Policy_in_North_Americ a.pdf

121

Greenroofs.com. (2018, September 11). Fashion Valley Mall living wall. Greenroofs.com: Connecting the planet + living architecture. https://www.greenroofs.com/projects/fashion-valley-mall-living-wall/

Guardia-Puebla, Y., Pérez-Quintero, F., Rodríguez-Pérez, S., Sánchez-Girón, V., Llanes- Cedeño, E., Rocha-Hoyos, J., & Peralta-Zurita, D. (2019). Effect of hydraulic loading rate and vegetation on phytoremediation with artificial wetlands associated to natural swimming pools. Journal of Water & Land Development, 40(1), 39–51.

Haggag, M. A. (2010). The use of green walls in sustainable urban context: With reference to Dubai, UAE. WIT Transactions on Ecology & the Environment, 128, 261-270.

Hatt, B. E., Fletcher, T. D., & Deletic, A. (2009). Hydrologic and pollutant removal performance of stormwater biofiltration systems at the field scale. Journal of Hydrology, 365(3–4), 310–321.

Hemond, H. F. (1983). The nitrogen budget of Thoreau’s bog. Ecology, 64(1), 99–109.

Hernández, B., & Hidalgo, M. C. (2005). Effect of urban vegetation on psychological restorativeness. Psychological Reports, 96(3), 1025.

Hilleary, M., & Gracy, D. (2014). Natural swimming pools & ponds: The total guide. Bonner Springs, KS: Habitat Books.

Ho, Y. S., & McKay, G. (2000). The kinetics of sorption of divalent metal ions onto sphagnum moss peat. Water Research, 34(3), 735-742.

Hoffman, M. C. (2013). Nutrient removal in natural swimming pools: A mass balance analysis (Unpublished doctoral dissertation). Pennsylvania State University: State College, Pennsylvania, USA.

Hommel, J., Veres, E., & Röseler, H. (2012). Bewertungsmethode zur Beurteilung der bauphysikalischen Auswirkung von Fassadenbegrünungen [Assessment method for assessing the physical impact of façade greening].

Hop, M. & Hiemstra, J.A. (2012). Contribution of green roofs and walls to ecosystem services of urban green. Acta Horticulturae, 990, 475–480.

Hopkins, G., & Goodwin, C. (2011). Living architecture: Green roofs and walls. Australia: Csiro Publishing.

122

Horoshenkov, K. V., Khan, A., & Benkreira, H. (2013). Acoustic properties of low growing plants. The Journal of the Acoustical Society of America, 133(5), 2554– 2565.

Hsieh, C., & Davis, A. P. (2005). Evaluation and optimization of bioretention media for treatment of urban storm water runoff. Journal of Environmental Engineering, 131(11), 1521–1531.

Hui, S. C. M., & Zhao, Z. (2013, November). Thermal regulation performance of green living walls in buildings. Paper presented at Joint Symposium on Innovation and Technology for Built Environment, Hong Kong, China.

Hunter, P. R. (1998). Cyanobacterial toxins and human health. Journal of Applied Microbiology, 84, 35-40.

Hunter, R. G., Combs, D. L., & George, D. B. (2001). Nitrogen, phosphorous, and organic carbon removal in simulated wetland treatment systems. Archives of Environmental Contamination and Toxicology, 41(3), 274-281.

Jaksch, H., Wesner, W., & Fuchs, A. (2013). ASC Formblätter Z0013-Z0020 – 10 Gebote Schwimmteichbau [ASC Forms Z0013-Z0020 – 10 commandments for swimming pond construction]. http://www.asceuropa.org/

Janhäll, S. (2015). Review on urban vegetation and particle air pollution – Deposition and dispersion. Atmospheric Environment, 105, 130–137.

Jialin, T. (2013). Living wall: Jungle to concrete, 1st ed. Design Media Publishing Limited: Hong Kong, China.

Jim, C. Y., & He, H. (2011). Estimating heat flux transmission of vertical greenery ecosystem. Ecological Engineering, 37(8), 1112-1122.

Johnston, J., & Newton, J. (1996). Building green: A guide for using plants on roofs, walls, and pavements, 1st ed. Greater London Authority: London, UK.

Jørgensen, L., Dresbøll, D. B., & Thorup-Kristensen, K. (2014). Spatial root distribution of plants growing in vertical media for use in living walls. Plant & Soil, 380(1–2), 231–248.

Jørgensen, L., Thorup-Kristensen, K., & Dresbøll, D. B. (2018). Against the wall—Root growth and competition in four perennial winter hardy plant species grown in living walls. Urban Forestry & Urban Greening, 29, 293–302.

Kadlec, R. H., & Wallace, S. (2009). Treatment wetlands. (2nd ed.). Boca Raton, FL: Taylor & Francis Group. 123

Kaplan, R. (2001). The nature of the view from home: Psychological benefits. Environment & Behavior, 33(4), 507–542.

Karczmarczyk, A., Bus, A., & Baryła, A. (2019). Influence of operation time, hydraulic load and drying on phosphate retention capacity of mineral filters treating natural swimming pool water. Ecological Engineering, 130, 176–183.

Khandaker, M., & Kotzen, B. (2018). The potential for combining living wall and vertical farming systems with aquaponics with special emphasis on substrates. Aquaculture Research, 49(4), 1454–1468.

Kircher, W. (2004). Wetlands and water bodies. In N. Dunnett & J. Hitchmough (Eds.), The dynamic landscape. (pp. 294-347). London, England: Spon Press.

Kircher, W. (2007, April). Marginal wetland planting for oligotrophic swimming ponds. Paper presented at International scientific-practical conference – Formation of urban green areas 2007: Water and plants in landscape material, Klaipeda, Lithuania.

Kircher, W. (2008). Standortangepasste Bepflanzung nährstoffarmer Schwimmteiche [Plant-adapted planting of nutrient-poor swimming ponds]. 40. Veitshöchheimer Landespflegetage [Veitshöchheim State Care Day], Heft 115, p. 65-75.

Kircher, W., & Schönfeld, P. (2008). Sandfilter mit Moorpflanzen zur Wasseraufbereitung an Schwimmteichen [Sand filter with peat plants for water treatment at swimming ponds]. Versuche in der Landespflege [Attempts at Land Maintenance], 2008, 4.

Kircher, W., & Thon, A. (2015a). Einfluss der Bepflanzung durchströmter Filter von Schwimmteichen auf Verduns-tungsverluste und Fadenalgenbesatz [Influence of flow through filters of swimming ponds on evaporation losses and filamentous algae].

Kircher, W., & Thon, A. (2015b). Wachstum von Sphagnum Moosen an Schwimmteichen [Sphagnum mosses grow on swimming ponds].

Kircher, W., & Thon, A. (2016a). How to build a natural swimming pool: The complete guide to healthy swimming at home. London, UK: Filbert Press Limited.

Kircher, W., & Thon A. (2016b, April). Natural swimming pools (NSPs) – Principles and trials with site-conform vegetation. Paper presented at International Environmental Conference, Vilnius, Lithuania.

124

Koerselman W., & Verhoeven J. T. A. (1992). Nutrient dynamics in mires of various trophic status: Nutrient inputs and outputs and the internal nutrient cycle. In J.T.A. Verhoeven (Ed.). Fens and Bogs in the Netherlands (Geobotany, Vol. 18, pp. 397-432). Dordrecht, Netherlands, Springer.

Kontoleon, K. J., & Eumorfopoulou, E. A. (2010). The effect of the orientation and proportion of a plant-covered wall layer on the thermal performance of a building zone. Building & Environment, 45(5), 1287–1303.

Kyrö, K., Brenneisen, S., Kotze, D. J., Szallies, A., Gerner, M., & Lehvävirta, S. (2018). Local habitat characteristics have a stronger effect than the surrounding urban landscape on beetle communities on green roofs. Urban Forestry & Urban Greening, 29, 122–130.

Lee, J. G., & Heaney, J. P. (2008). Swimming pool water use analysis by observed data and long-term continuous simulation. Conserve Florida Water Clearinghouse, Department of Environmental Engineering Sciences, University of Florida. https://www.researchgate.net/profile/Joong_Lee3/publication/242783799_Measur e_4_Swimming_Pool_Water_Use_Analysis_by_Observed_Data_and_Long_term _Continuous_Simulation/links/53db8d8a0cf216e4210bf6be/Measure-4- Swimming-Pool-Water-Use-Analysis-by-Observed-Data-and-Long-term- Continuous-Simulation.pdf

Lee, K. E., Williams, K. J. H., Sargent, L. D., Williams, N. S. G., & Johnson, K. A. (2015). 40-second green roof views sustain attention: The role of micro-breaks in attention restoration. Journal of Environmental Psychology, 42, 182–189.

Liem, J. D. (2014). Biological water purification for indoor swimming pools: Towards chlorine-free swimming. Unpublished manuscript, Delft University of Technology, Delft, Netherlands.

Lin, M.-Y., Hagler, G., Baldauf, R., Isakov, V., Lin, H.-Y., & Khlystov, A. (2016). The effects of vegetation barriers on near-road ultrafine particle number and carbon monoxide concentrations. Science of the Total Environment, 553, 372–379.

Lin, Y.-F., Jing, S.-R., Wang, T.-W., & Lee, D.-Y. (2002). Effects of macrophytes and external carbon sources on nitrate removal from groundwater in constructed wetlands. Environmental Pollution, 119(3), 413–420.

Littlewood, M. (2005). Natural swimming pools: Inspiration for harmony with nature. Atglen, PA: Schiffer Publishing Co.

Loh, S. (2008). Living walls – A way to green the built environment. In Environment Design Guide (pp. 1-7). Australian Institute of Architects: Brisbane, Australia.

125

Loh, S. & Stav, Y. (2008) Green a city grow a wall. In R. Kennedy (Ed.), Proceedings of the Subtropical Cities 2008 Conference - From fault-lines to sight-lines - subtropical urbanism in 20-20 (pp. 1-9). Queensland, Australia: Centre for Subtropical Design.

Lott, R. B., & Hunt, R. J. (2001). Estimating evapotranspiration in natural and constructed wetlands. Wetlands, 21(4), 614-628.

Lucas, W. C., & Greenway, M. (2008). Nutrient retention in vegetated and nonvegetated bioretention mesocosms. Journal of Irrigation & Drainage Engineering, 134(5), 613–623.

Maas, J., Verjeij, Groenewegen, de Vries, & Spreeuweenberg. (2006). Green space, urbanity, and health: How strong is the relation? Journal of Epidemiology & Community Health, 60(7), 587–592.

Madre, F., Clergeau, P., Machon, N., & Vergnes, A. (2015). Building biodiversity: Vegetated façades as habitats for spider and beetle assemblages. Global Ecology & Conservation, 3, 222–233.

Magliocco, A. (2018). Vertical greening systems: Social and aesthetic aspects. In G. Pérez & K. Perini (Eds.), Nature based strategies for urban and Building Sustainability (pp. 263-271). Oxford: UK: Butterworth-Heinemann.

Malys, L., Musy, M., & Inard, C. (2014). A hydrothermal model to assess the impact of green walls on urban microclimate and building energy consumption. Building & Environment, 73, 187–197.

Manso, M., & Castro-Gomes, J. (2015). Green wall systems: A review of their characteristics. Renewable & Sustainable Energy Reviews, 41, 863–871.

Manuel, P. M. (2003). Occupied with ponds: Exploring the meaning, bewaring the loss for kids and communities of nature’s small spaces. Journal of Occupational Science, 10(1), 31-39.

Marchi, M., Pulselli, R. M., Marchettini, N., Pulselli, F. M., & Bastianoni, S. (2015). Carbon dioxide sequestration model of a vertical greenery system. Ecological Modeling, 306, 46–56.

Mårtensson, L.-M., Fransson, A.-M., & Emilsson, T. (2016). Exploring the use of edible and evergreen perennials in living wall systems in the Scandinavian climate. Urban Forestry & Urban Greening, 15, 84–88.

126

Masi, F., Bresciani, R., Rizzo, A., Edathoot, A., Patwardhan, N., Panse, D., & Langergraber, G. (2016). Green walls for greywater treatment and recycling in dense urban areas: A case-study in Pune. Journal of Water, Sanitation & Hygiene for Development, 6(2), 342–347.

Mayrand, F., & Clergeau, P. (2018). Green roofs and green walls for biodiversity conservation: A contribution to urban connectivity? Sustainability, 10(4), 985.

Mayrand, F., Clergeau, P., Vergnes, A., & Madre, F. (2018). Vertical greening systems as habitat for biodiversity. In G. Pérez & K. Perini (Eds.), Nature based strategies for urban and building sustainability (pp. 227-237). Oxford, UK: Butterworth- Heinemann.

Mazzali, U., Peron, F., Romagnoni, P., Pulselli, R. M., & Bastianoni, S. (2013). Experimental investigation on the energy performance of living walls in a temperate climate. Building & Environment, 64, 57–66.

Mazzali, U., Peron, F., & Scarpa, M. (2012). Thermo-physical performances of living walls via field measurements and numerical analysis. WIT Transactions on Ecology & the Environment, 165, 251-259.

McPherson, E. G., Scott, K. I., & Simpson, J. R. (1998). Estimating cost effectiveness of residential yard trees for improving air quality in Sacramento, California, using existing models. Atmospheric Environment, 32(1), 75–84.

Messer, U. J. (2008). Studies on the development and assessment of perennial planting mixtures (Doctoral dissertation). University of Sheffield, Sheffield, UK.

Mitsch, W. J., & Gosselink, J. G. (2007). Wetlands. (4th ed.). Hoboken, NJ: John Wiley & Sons, Inc.

Momas, I., Brette, F., Spinasse, A., Squinazi, F., Dab, W., & Festy, B. (1993). Health effects of attending a public swimming pool: Follow up of a cohort of pupils in Paris. Journal of Epidemiology & Community Health, 47(6), 464–468.

Meral, A., Başaran, N., Yalçınalp, E., Doğan, E., Ak, M., & Eroğlu, E. (2018). A comparative approach to artificial and natural green walls according to ecological sustainability. Sustainability, 10(6), 1995.

Morris, J. T. (1991). Effects of nitrogen loading on wetland ecosystems with particular reference to atmospheric deposition. Annual Review of Ecology & Systematics, 22(1), 257-279.

Mostert, E. S., & Grobbelaar, J. U. (1987). The influence of nitrogen and phosphorus on algal growth and quality in outdoor mass algal cultures. Biomass, 13(4), 219-233. 127

Müller, H. E. (2001). Sind Badeteiche eine echte Alternative zu Freibädern? [Are bathing ponds a real alternative to outdoor pools?]. Archiv des Badewesens [Archives of the Bathing Industry], 06, 311-318.

Nemery, B., Hoet, P. H. M., & Nowak, D. (2002). Indoor swimming pools, water chlorination and respiratory health. European Respiratory Journal, 19(5), 790– 793.

Nichols, D. S., & Brown, J. M. (1980). Evaporation from a sphagnum moss surface. Journal of Hydrology, 48(3-4), 289-302.

Olsen, K. (2007). Clear waters and a green gas: A history of chlorine as a swimming pool sanitizer in the United States. Bulletin for the History of Chemistry, 32(2), 129- 140.

ÖNORM L 1128. (2013). “Schwimmteiche und Naturpools – Anforderungen an Planung, Bau, Betrieb und Sanierung” [Swimming ponds and natural pools – requirements for planning, construction, operation, and renovation]. Vienna: Österreichisches Normungsinstitut (ON) – Austrian Standards Institute.

Ottelé, M. (2011). The green building envelope. Vertical greening (Unpublished doctoral dissertation). Delft University of Technology, Delft, Netherlands.

Ottelé, M. (2015). A green building envelope: A crucial contribution to biophilic cities. In F. Pacheco Torgal, J. A. Labrincha, M. V. Diamanti, C.-P. Yu, & H. K. Lee (Eds.), Biotechnologies and biomimetics for civil engineering (pp. 135–161). Springer International Publishing.

Ottelé, M., & Perini, K. (2017). Comparative experimental approach to investigate the thermal 127mmobili of vertical greened façades of buildings. Ecological Engineering, 108, 152–161.

Ottelé, M., van Bohemen, H. D., & Fraaij, A. L. A. (2010). Quantifying the deposition of particulate matter on climber vegetation on living walls. Ecological Engineering, 36(2), 154–162.

Ottová, V., Balcarová, J., & Vymazal, J. (1997). Microbial characteristics of constructed wetlands. Water Science & Technology, 35(5), 117-123.

Pettit, T., Irga, P. J., Abdo, P., & Torpy, F. R. (2017). Do the plants in functional green walls contribute to their ability to filter particulate matter? Building & Environment, 125, 299–307.

128

Pérez, G., & Coma, J. (2018). Vertical greening systems to improve water management. In G. Pérez & K. Perini (Eds.), Nature based strategies for urban and building sustainability (pp. 191-201). Oxford, UK: Butterworth-Heinemann.

Pérez, G., Coma, J., Barreneche, C., de Gracia, A., Urrestarazu, M., Burés, S., & Cabeza, L. F. (2016). Acoustic insulation capacity of vertical greenery systems for buildings. Applied Acoustics, 110, 218–226.

Pérez, G., Coma, J., & Cabeza, L. F. (2018). Vertical greening systems for acoustic insulation and noise reduction. In G. Pérez & K. Perini (Eds.), Nature based strategies for urban and building sustainability (pp. 157-165). Oxford: UK: Butterworth-Heinemann.

Pérez, G., Coma, J., Martorell, I., & Cabeza, L. F. (2014). Vertical greenery systems (VGS) for energy saving in buildings: A review. Renewable & Sustainable Energy Reviews, 39, 139–165.

Pérez, G., Rincón, L., Vila, A., González, J. M., & Cabeza, L. F. (2011). Green vertical systems for buildings as passive systems for energy savings. Applied Energy, 88(12), 4854–4859.

Pérez-Urrestarazu, L., Egea, G., Franco-Salas, A., & Fernández-Cañero, R. (2014). Irrigation systems evaluation for living walls. Journal of Irrigation & Drainage Engineering, 140(4), 04013024.

Pérez-Urrestarazu, L., Fernández-Cañero, R., Franco, A., & Egea, G. (2016). Influence of an active living wall on indoor temperature and humidity conditions. Ecological Engineering, 90, 120–124.

Pérez-Urrestarazu, L., Fernández-Cañero, R., Franco-Salas, A., & Egea, G. (2015). vertical greening systems and sustainable cities. Journal of Urban Technology, 22(4), 65–85.

Pérez-Urrestarazu, L., & Urrestarazu, M. (2018). Vertical greening systems: Irrigation and maintenance. In G. Pérez & K. Perini (Eds.), Nature based strategies for urban and building sustainability (pp. 55-63). Oxford, UK: Butterworth- Heinemann.

Perini, K., Magliocco, A., & Giulini, S. (2017). Vertical greening systems evaporation measurements: Does plant species influence cooling performances? International Journal of Ventilation, 16(2), 152–160.

Perini, K., & Ottelé, M. (2014). Designing green façades and living wall systems for sustainable constructions. International Journal of Design & Nature and Ecodynamics, 9(1), 31-46. 129

Perini, K., Ottelé, M., Giulini, S., Magliocco, A., & Roccotiello, E. (2017). Quantification of fine dust deposition on different plant species in a vertical greening system. Ecological Engineering, 100, 268–276.

Perini, K., Ottelé, M., Haas, E. M., & Raiteri, R. (2011). Greening the building envelope, façade greening, and living wall systems. Open Journal of Ecology, 01(01), 1–8.

Perini, K., Ottelé, M., Haas, E. M., & Raiteri, R. (2013). Vertical greening systems, a process tree for green façades and living walls. Urban Ecosystems, 16(2), 265– 277.

Perini, K., & Roccotiello, E. (2018). Vertical greening systems for pollutants reduction. In G. Pérez & K. Perini (Eds.), Nature based strategies for urban and building sustainability (pp. 131-140). Oxford, UK: Butterworth-Heinemann.

Perini, K., & Rosasco, P. (2013). Cost–benefit analysis for green façades and living wall systems. Building & Environment, 70, 110–121.

Picard, C. R., Fraser, L. H., & Steer, D. (2005). The interacting effects of temperature and plant community type on nutrient removal in wetland microcosms. Bioresource Technology, 96(9), 1039–1047.

Pradhan, S., Al-Ghamdi, S. G., & Mackey, H. R. (2019). Greywater recycling in buildings using living walls and green roofs: A review of the applicability and challenges. Science of the Total Environment, 652, 330–344.

Prodanovic, V., McCarthy, D., Hatt, B., & Deletic, A. (2019). Designing green walls for greywater treatment: The role of plants and operational factors on nutrient removal. Ecological Engineering, 130, 184–195.

Prodanovic, V., Wang, A., & Deletic, A. (2019). Assessing water retention and correlation to climate conditions of five plant species in greywater treating green walls. Water Research, 167, 115092.

Prodanovic, V., Zhang, K., Hatt, B., McCarthy, D., & Deletic, A. (2018). Optimisation of lightweight green wall media for greywater treatment and reuse. Building & Environment, 131, 99–107.

Pugh, T. A., MacKenzie, A. R., Whyatt, J. D., & Hewitt, C. N. (2012). Effectiveness of green infrastructure for improvement of air quality in urban street canyons. Environmental Science & Technology, 46(14), 7692-7699.

Pulselli, R. M., Pulselli, F. M., Mazzali, U., Peron, F., & Bastianoni, S. (2014). Energy based evaluation of environmental performances of Living Wall and Grass Wall systems. Energy & Buildings, 73, 200–211. 130

Radić, M., Arsić, J., Džaleta, M., & Stevović, S. (2012). Eco-architecture and sustainable design as a function of the quality of environment. In Sustainable buildings and urban oasis, proceedings of International Green Build Conference 2011 (pp. 109- 113). Belgrade, Serbia: Ecoist.

Radić, M., Brković Dodig, M., & Auer, T. (2019). Green facades and living walls—A review establishing the classification of construction types and mapping the benefits. Sustainability, 11(17), 4579.

Ramírez, Á., Ayuga-Téllez, E., Gallego, E., Fuentes, J. M., & García, A. I. (2011). A simplified model to assess landscape quality from rural roads in Spain. Agriculture, Ecosystems & Environment, 142(3-4), 205-212.

Richardson, C. J. (1985). Mechanisms controlling phosphorus retention capacity in freshwater wetlands. Science, 228(4706), 1424–1427.

Riedel, J., Kietsch, U., Heinrich, A., Messer, U., & Kircher, W. (2007). Perennemix – Attractive gardens for public and private spaces. Bernburg, Germany: Anhalt University of Applied Sciences.

Riley, B. (2017). The state of the art of living walls: Lessons learned. Building & Environment, 114, 219–232.

Riley, B., de Larrard, F., Malécot, V., Dubois-Brugger, I., Lequay, H., & Lecomte, G. (2019). Living concrete: Democratizing living walls. Science of the Total Environment, 673, 281–295.

Rosasco, P. (2018). Economic benefits and costs of vertical greening systems. In G. Pérez & K. Perini (Eds.), Nature based strategies for urban and building sustainability (pp. 291-306). Oxford, UK: Butterworth-Heinemann.

Roehr, D., & Laurenz, J. (2008). Living skins: Environmental benefits of green envelopes in the city context. Eco-Architecture II, I, 149–158.

Rysulova, M., Kaposztasova, D., & Vranayova, Z. (2017). Green walls as an approach in grey water treatment. IOP Conference Series: Materials Science & Engineering, 245, 072049.

Safikhani, T., Abdullah, A. M., Ossen, D. R., & Baharvand, M. (2014). A review of energy characteristic of vertical greenery systems. Renewable & Sustainable Energy Reviews, 40, 450–462.

Sakkas, P. (2013). Domestic greywater treatment through the integration of constructed wetlands in Living Wall Systems (LWS) (Unpublished master’s thesis). Delft University of Technology, Delft, Netherlands. 131

Scarpa, M., Mazzali, U., & Peron, F. (2014). Modeling the energy performance of living walls: Validation against field measurements in temperate climate. Energy & Buildings, 79, 155–163.

Schulz, L. (1981). Nährstoffeintrag in Seen durch Badegäste [Nutrient entry in lakes by bathers]. Zentralblatt fur Bakteriologie, Mikrobiologie und Hygiene 1. Abt. Originale [Central Journal for Bacteriology, Microbiology and Hygiene 1. Dept. Originals], 173, 528-548.

Schwoerbel, J. (1987). Handbook of limnology. Ellis Horwood.

Segovia-Cardozo, D. A., Rodríguez-Sinobas, L., & Zubelzu, S. (2019). Living green walls: Estimation of water requirements and assessment of irrigation management. Urban Forestry & Urban Greening, 46, 126458.

Shaikh, A. F., Gunjal, P. K., & Chaple, N. V. (2015). A review on green walls technology, benefits, & design. International Journal of Engineering Sciences & Research Technology, 4(4), 312-322.

Sheweka, S., & Magdy, Arch. N. (2011). The living walls as an approach for a healthy urban environment. Energy Procedia, 6, 592–599.

Sheweka, S. M., & Mohamed, N. M. (2012). Green facades as a new sustainable approach towards climate change. Energy Procedia, 18, 507–520.

Shimwell, D. W. (2009). Studies in the floristic diversity of Durham walls, 1959- 2008. Watsonia, 27(4), 323.

Singh, P. K., Jain, S., Mathur, A., & Kumar, Y. (2017). An analysis on the potentials of Vertical Greenery System (VGS) in context to the application viewpoint. 2017 4th International Conference on Signal Processing, Computing and Control (ISPCC), 632–636.

Solera Jimenez, M. (2018). Green walls: A sustainable approach to climate change, a case study of London. Architectural Science Review, 61(1–2), 48–57.

Spieles, D. J., & Mitsch, W. J. (2000). Macroinvertebrate community structure in high- and low-nutrient constructed wetlands. Wetlands, 20(4), 716–729.

Sternberg, T., Viles, H., Cathersides, A., & Edwards, M. (2010). Dust particulate absorption by ivy (Hedera helix L) on historic walls in urban environments. Science of The Total Environment, 409(1), 162–168.

132

Sudimac, B., Cukovic-Ignjatovic, N., & Ignjatovic, D. (2018). Experimental study on reducing temperature using modular system for vegetation walls made of perlite concrete. Thermal Science, 22 (Suppl. 4), 1059–1069.

Suparwoko, & Taufani, B. (2017). Urban farming construction model on the vertical building envelope to support the green buildings development in Sleman, Indonesia. Procedia Engineering, 171, 258–264.

Sutton, R. (2014). Aesthetics for Green Roofs and Green Walls. Journal of Living Architecture, 2, 1–20.

Svete, L. (2013). Vegetated greywater treatment walls: Design modifications for intermittent media filters (Unpublished master’s thesis). Norwegian University of Life Sciences, Ås, Norway.

Taha, H. (1997). Urban climates and heat islands: Albedo, evapotranspiration, and anthropogenic heat. Energy & Buildings, 25(2), 99–103.

Tanner, C. C. (1996). Plants for constructed wetland treatment systems—A comparison of the growth and nutrient uptake of eight emergent species. Ecological Engineering, 7(1), 59–83.

This Old House. (2020). What makes stone wool a smart insulation choice: How the natural power of stone can increase the comfort of your home. This Old House. https://www.thisoldhouse.com/insulation/21015697/what-makes-stone-wool-a- smart-insulation-choice

Thon, A. (2014). Untersuchung von Bepflanzungsvarianten und Durchströmung der Filterzone von Kleinbadeteichen [Investigation of planting variants and flow through the filter zone of small bathing ponds] (Unpublished doctoral dissertation). University of Vechta, Vechta, Germany.

Thon, A., Kircher, W., Pesch, R., Schmidt, G., & Thon, I. (2009). Functionality, appearance and water purification rates of perfused vegetation mats for water treatment at private swimming ponds. In R. Nekrosiene & J. Kucinskiene (Eds.), Proceedings of the international scientific-practical conference – Formation of urban green areas 2009: Environment of residential districts. Klaipeda, Lithuania: Klaipëda Business and Technology College.

Timur, Ö.B., & Karaca, E. (2013). Vertical gardens (Chapter 22). In M. Ozyavuz (Ed.), Advances in Landscape Architecture (1st ed., pp. 587–622). IntechOpen: London, UK.

Total Habitat. (2017). Bugarium photo gallery (ABQ Biopark in Albuquerque, NM). Total Habitat. https://www.totalhabitat.com/bugarium-photo-gallery.html 133

Total Habitat. (2019). Assessment of the aquatic macroinvertebrate fauna and algae in 2- year-old swim pond. Unpublished manuscript.

Ulrich, R. S. (1993). Biophilia, biophobia, and natural landscapes. In S.R. Kellert & E.O. Wilson (Eds.), The biophilia hypothesis (pp. 73–137). Washington, DC: Island Press.

Urban, N. R., & Eisenreich, S. J. (1988). Nitrogen cycling in a forested Minnesota bog. Canadian Journal of Botany, 66(3), 435–449.

Urban Land Institute. (2017). Harvesting the value of water: Stormwater, green infrastructure, and real estate. Washington, D.C.: Urban Land Insitutute.

Urrestarazu, M., & Burés, S. (2012). Sustainable green walls in architecture. Journal of Food Agriculture & Environment, 10(1), 792-4.

Van de Wouw, P. M. F., Ros, E. J. M., & Brouwers, H. J. H. (2017). Precipitation collection and evapo(transpi)ration of living wall systems: A comparative study between a panel system and a planter box system. Building & Environment, 126, 221–237.

Van Duren, I. C., & Pegtel, D. M. (2000). Nutrient limitations in wet, drained, and rewetted fen meadows: Evaluation of methods and results. Plant & Soil, 220(1-2), 35-47.

Van Renterghem, T., Hornikx, M., Forssen, J., & Botteldooren, D. (2013). The potential of building envelope greening to achieve quietness. Building & Environment, 61, 34–44.

Veisten, K., Smyrnova, Y., Klæboe, R., Hornikx, M., Mosslemi, M., & Kang, J. (2012). Valuation of green walls and green roofs as soundscape measures: Including 133mmobiliz amenity values together with noise-attenuation values in a cost- benefit analysis of a green wall affecting courtyards. International Journal of Environmental Research & Public Health, 9(11), 3770–3788.

Victorero, F., Vera, S., Bustamante, W., Tori, F., Bonilla, C., Gironás, J., & Rojas, V. (2015). Experimental study of the thermal performance of living walls under semiarid climatic conditions. Energy Procedia, 78, 3416–3421.

Viecco, M., Vera, S., Jorquera, H., Bustamante, W., Gironás, J., Dobbs, C., & Leiva, E. (2018). Potential of particle matter dry deposition on green roofs and living walls vegetation for mitigating urban atmospheric pollution in semiarid climates. Sustainability, 10(7), 2431. 134

Vohla, C., Alas, R., Nurk, K., Baatz, S., & Mander, Ü. (2007). Dynamics of phosphorus, nitrogen, and carbon removal in a horizontal subsurface flow constructed wetland. Science of the Total Environment, 380(1–3), 66–74.

Vohla, C., Kõiv, M., Bavor, H. J., Chazarenc, F., & Mander, Ü. (2011). Filter materials for phosphorus removal from wastewater in treatment wetlands—A review. Ecological Engineering, 37(1), 70–89.

Vohla, C., Põldvere, E., Noorvee, A., Kuusemets, V., & Mander, Ü. (2005). Alternative filter media for phosphorous removal in a horizontal subsurface flow constructed wetland. Journal of Environmental Science & Health, Part A, 40(6–7), 1251– 1264.

Von Berger, F. (2010). Swimming ponds: Natural pleasure in your garden. Atglen, PA: Schiffer Publishing Co.

Vymazal, J. (2007). Removal of nutrients in various types of constructed wetlands. Science of The Total Environment, 380(1–3), 48–65.

Wassen, M. J., Venterink, H. O., Lapshina, E. D., & Tanneberger, F. (2005). Endangered plants persist under phosphorus limitation. Nature, 437(7058), 547–550.

Weerakkody, U., Dover, J. W., Mitchell, P., & Reiling, K. (2017). Particulate matter pollution capture by leaves of seventeen living wall species with special reference to rail-traffic at a metropolitan station. Urban Forestry & Urban Greening, 27, 173–186.

Weerakkody, U., Dover, J. W., Mitchell, P., & Reiling, K. (2018). Quantification of the traffic-generated particulate matter capture by plant species in a living wall and evaluation of the important leaf characteristics. Science of the Total Environment, 635, 1012–1024.

Weerakkody, U., Dover, J. W., Mitchell, P., & Reiling, K. (2019). Topographical structures in planting design of living walls affect their ability to 134mmobilize traffic-based particulate matter. Science of the Total Environment, 660, 644–649.

Weinreich, C. (2005). Schwimmteiche als Modulsystem [Swimming ponds as a modular system]. Unpublished manuscript, Anhalt University of Applied Sciences, Bernburg, Germany.

Weinstein, N., Balmford, A., DeHaan, C. R., Gladwell, V., Bradbury, R. B., & Amano, T. (2015). Seeing community for the trees: The links among contact with natural environments, community cohesion, and crime. BioScience, 65(12), 1141–1153. 135

Wesner, W. (2013). Definition Schwimmteich: Hintergrundwissen zur Definition und Abgrenzung zu anderen Gewässern [Definition of swimming pond: Background knowledge on the definition and delimitation to other waters]. ASC-Formblatt T0001.

White, S. A., Taylor, M. D., Albano, J. P., Whitwell, T., & Klaine, S. J. (2011). Phosphorus retention in lab and field-scale subsurface-flow wetlands treating plant nursery runoff. Ecological Engineering, 37(12), 1968–1976.

Wolcott, S. (2015). Performance of green walls in treating brewery wastewater. Unpublished manuscript, Rochester Institute of Technology, Rochester, New York, USA.

Wong, N. H., Tan, A. Y. K., Chen, Y., Sekar, K., Tan, P. Y., Chan, D., Chiang, K., & Wong, N. C. (2010). Thermal evaluation of vertical greenery systems for building walls. Building & Environment, 45(3), 663-672.

Wong, N. H., Tan, A. Y. K., Tan, P. Y., Chiang, K., & Wong, N. C. (2010). Acoustics evaluation of vertical greenery systems for building walls. Building & Environment, 45(2), 411-420.

Wong, N. H., Tan, A. Y. K., Tan, P. Y., & Wong, N. C. (2009). Energy simulation of vertical greenery systems. Energy & Buildings, 41(12), 1401–1408.

Zhao, X., Zuo, J., Wu, G., & Huang, C. (2019). A bibliometric review of green building research 2000–2016. Architectural Science Review, 62(1), 74–88. 136

APPENDICES

137

Appendix A. Filling Water Data

Date 1 7 12 6 9 10 2 4 8 3 5 11 25-Jul 11 24 7 17 6 7 16 12 8 14 9 10 26-Jul 0 20 0 0 0 0 0 0 0 0 0 0 27-Jul 6 30 8 21 8 9 10 10 10 21 13 10 28-Jul 0 15 0 10 0 0 0 0 0 0 0 0 29-Jul 12 13 8 13 6 6 7 6 7 15 7 11 30-Jul 10 4 3 13 2 1 9 2 3 4 2 7 31-Jul 5 4 5 12 4 4 4 4 4 9 5 8 1-Aug 3 3 3 6 3 1 3 2 3 6 2 3 2-Aug 3 2 3 4 2 3 4 4 3 5 3 3 3-Aug 2 1 3 4 3 3 3 2 3 6 2 5 4-Aug 2 1 3 5 2 3 3 3 2 3 3 2 5-Aug 0 0 0 0 0 0 0 0 0 0 0 0 6-Aug 3 3 4 5 5 3 6 5 8 9 8 12 7-Aug 0 0 0 0 0 0 0 0 0 0 0 7 8-Aug 2 0 0 3 0 0 3 2 0 5 2 5 9-Aug 3 0 3 3 0 3 4 2 4 4 3 6 10-Aug 0 0 0 0 0 0 0 0 0 0 0 0 11-Aug 6 4 7 6 6 5 6 5 5 5 5 9 12-Aug 0 0 0 5 0 0 5 0 0 4 2 4 13-Aug 4 5 6 5 6 6 4 5 6 4 4 8 14-Aug 2 2 4 8 2 2 4 3 2 3 3 8 15-Aug 3 3 3 4 4 4 6 4 3 4 4 8 16-Aug 4 2 3 6 3 2 3 2 2 2 2 3 17-Aug 0 0 0 0 0 0 0 0 0 0 0 0 18-Aug 0 0 0 0 0 0 0 0 0 0 0 0 19-Aug 8 0 2 4 0 2 3 4 0 3 3 7 20-Aug 16 4 4 5 4 4 5 7 6 5 6 4 21-Aug 5 0 4 3 4 5 5 3 2 4 4 5 22-Aug 0 0 0 0 0 0 0 0 0 0 0 0 23-Aug 4 6 2 4 4 2 3 4 5 2 3 8 24-Aug 0 0 0 0 0 0 0 0 0 0 0 0 25-Aug 0 0 0 0 0 0 0 0 0 0 0 0 26-Aug 10 11 5 10 9 8 12 10 6 11 9 10 27-Aug 0 0 0 4 0 0 11 3 18 4 5 5 138

28-Aug 0 0 3 6 4 5 5 0 0 0 0 4 29-Aug 6 4 0 8 6 0 5 3 0 5 3 4 30-Aug 0 0 0 5 3 4 6 1 7 3 3 1 31-Aug 0 0 0 0 0 0 0 0 0 0 0 0 1-Sep 0 0 0 0 0 0 0 0 0 0 0 0 2-Sep 10 0 8 11 7 5 12 11 6 10 7 6 3-Sep 0 0 0 0 0 0 0 0 0 0 0 0 4-Sep 0 0 0 0 0 0 0 0 0 0 0 0 5-Sep 0 0 0 8 6 0 12 0 5 0 7 6 6-Sep 0 0 0 0 0 0 0 0 0 0 0 0 7-Sep 7 4 8 6 4 7 8 10 5 10 6 9 8-Sep 0 0 0 0 0 0 0 0 0 0 0 0 9-Sep 0 0 0 5 3 3 7 2 4 2 2 4 10-Sep 0 0 0 0 0 0 0 0 0 0 0 0 11-Sep 4 6 4 9 5 6 8 6 4 7 5 8 12-Sep 0 0 0 0 0 0 0 0 0 0 0 0 13-Sep 0 0 0 0 0 0 0 0 0 0 0 0 14-Sep 0 0 0 0 0 0 0 0 0 0 0 0 15-Sep 0 0 0 0 0 0 0 0 0 0 0 0 16-Sep 0 0 0 0 0 0 0 0 0 0 0 0 17-Sep 5 3 5 6 5 6 8 4 3 6 5 6 18-Sep 0 0 0 0 0 0 0 0 0 0 0 0 19-Sep 0 0 0 0 0 0 0 0 0 0 0 0 20-Sep 4 5 2 3 0 2 3 5 4 3 2 4 21-Sep 9 8 12 12 10 8 10 9 8 13 10 12 22-Sep 0 0 0 0 0 0 0 0 0 0 0 0 23-Sep 0 0 0 0 0 0 0 0 0 0 0 0 24-Sep 0 0 0 0 0 0 0 0 0 0 0 0 25-Sep 0 0 0 0 0 0 0 0 0 0 0 0 26-Sep 3 0 2 2 1 1 5 0 3 1 1 0 27-Sep 0 0 0 0 0 0 0 0 0 0 0 0 28-Sep 0 0 0 0 0 0 0 0 0 0 0 0 29-Sep 3 2 0 3 2 0 12 0 4 4 4 1 30-Sep 0 0 0 0 0 0 0 0 0 0 0 0 1-Oct 0 0 0 0 0 0 0 0 0 0 0 0 2-Oct 0 0 0 0 0 0 0 0 0 0 0 0 3-Oct 0 0 0 0 0 0 0 0 0 0 0 0 139

4-Oct 0 0 0 4 4 0 12 7 5 7 12 9 5-Oct 0 0 0 0 0 0 0 0 0 0 0 0 6-Oct 0 0 0 0 0 0 0 0 0 0 0 0 7-Oct 0 0 0 0 0 0 0 0 0 0 0 0 8-Oct 0 0 0 0 0 0 0 0 0 0 0 0 9-Oct 3 3 3 3 3 3 3 3 3 3 5 4 10-Oct 0 0 0 0 0 0 0 0 0 0 0 0 11-Oct 15 12 11 11 11 14 12 10 11 10 11 11 12-Oct 0 0 0 0 0 0 0 0 0 0 0 0 13-Oct 0 0 0 0 0 0 0 0 0 0 0 0 14-Oct 0 0 0 0 0 0 0 0 0 0 0 0 15-Oct 0 0 0 0 0 0 0 0 0 0 0 0 16-Oct 0 0 0 0 0 0 0 0 0 0 0 0 17-Oct 0 0 0 0 0 0 0 0 0 0 0 0 18-Oct 0 0 0 0 0 0 0 0 0 0 0 0 19-Oct 0 0 0 0 0 0 0 0 0 0 0 0 20-Oct 8 8 8 8 8 8 8 8 8 8 8 8 21-Oct 0 0 0 0 0 0 0 0 0 0 0 0 22-Oct 0 0 0 0 0 0 0 0 0 0 0 0 23-Oct 0 0 0 0 0 0 0 0 0 0 0 0 24-Oct 0 0 0 0 0 0 0 0 0 0 0 0 25-Oct 0 0 0 0 0 0 0 0 0 0 0 0 26-Oct 0 0 0 0 0 0 0 0 0 0 0 0 27-Oct 0 0 0 0 0 0 0 0 0 0 0 0 28-Oct 0 0 0 0 0 0 0 0 0 0 0 0 29-Oct 0 0 0 0 0 0 0 0 0 0 0 0 30-Oct 0 0 0 0 0 0 0 0 0 0 0 0 31-Oct 0 0 0 0 0 0 0 0 0 0 0 0 1-Nov 0 0 0 0 0 0 0 0 0 0 0 0 2-Nov 0 0 0 0 0 0 0 0 0 0 0 0 3-Nov 0 0 0 0 0 0 0 0 0 0 0 0 4-Nov 0 0 0 0 0 0 0 0 0 0 0 0 5-Nov 0 0 0 0 0 0 0 0 0 0 0 0 6-Nov 0 0 0 0 0 0 0 0 0 0 0 0 7-Nov 0 0 0 0 0 0 0 0 0 0 0 0 8-Nov 0 0 0 0 0 0 0 0 0 0 0 0 9-Nov 0 0 0 0 0 0 0 0 0 0 0 0 140

10-Nov 0 0 0 0 0 0 0 0 0 0 0 0 11-Nov 0 0 0 0 0 0 0 0 0 0 0 0 12-Nov 0 0 0 0 0 0 0 0 0 0 0 0 13-Nov 0 0 0 0 0 0 0 0 0 0 0 0 14-Nov 0 0 0 0 0 0 0 0 0 0 0 0 15-Nov 0 0 0 0 0 0 0 0 0 0 0 0 16-Nov 0 0 0 0 0 0 0 0 0 0 0 0 17-Nov 0 0 0 0 0 0 0 0 0 0 0 0 18-Nov 0 0 0 0 0 0 0 0 0 0 0 0 19-Nov 0 0 0 0 0 0 0 0 0 0 0 0 20-Nov 0 0 0 0 0 0 0 0 0 0 0 0 21-Nov 0 0 0 0 0 0 0 0 0 0 0 0 22-Nov 0 0 0 0 0 0 0 0 0 0 0 0 23-Nov 0 0 0 0 0 0 0 0 0 0 0 0 24-Nov 0 0 0 0 0 0 0 0 0 0 0 0 25-Nov 0 0 0 0 0 0 0 0 0 0 0 0 26-Nov 0 0 0 0 0 0 0 0 0 0 0 0 27-Nov 0 0 0 0 0 0 0 0 0 0 0 0 28-Nov 0 0 0 0 0 0 0 0 0 0 0 0 29-Nov 0 0 0 0 0 0 0 0 0 0 0 0 30-Nov 0 0 0 0 0 0 0 0 0 0 0 0 1-Dec 0 0 0 0 0 0 0 0 0 0 0 0 2-Dec 0 0 0 0 0 0 0 0 0 0 0 0 3-Dec 0 0 0 0 0 0 0 0 0 0 0 0 4-Dec 0 0 0 0 0 0 0 0 0 0 0 0 5-Dec 0 0 0 0 0 0 0 0 0 0 0 0 6-Dec 0 0 0 0 0 0 0 0 0 0 0 0 7-Dec 0 0 0 0 0 0 0 0 0 0 0 0 8-Dec 0 0 0 0 0 0 0 0 0 0 0 0 9-Dec 0 0 0 0 0 0 0 0 0 0 0 0 10-Dec 0 0 0 0 0 0 0 0 0 0 0 0 11-Dec 0 0 0 0 0 0 0 0 0 0 0 0 12-Dec 0 0 0 0 0 0 0 0 0 0 0 0 13-Dec 0 0 0 0 0 0 0 0 0 0 0 0 14-Dec 0 0 0 0 0 0 0 0 0 0 0 0 15-Dec 0 0 0 0 0 0 0 0 0 0 0 0 16-Dec 0 0 0 0 0 0 0 0 0 0 0 0 141

17-Dec 0 0 0 0 0 0 0 0 0 0 0 0 18-Dec 0 0 0 0 0 0 0 0 0 0 0 0 19-Dec 0 0 0 0 0 0 0 0 0 0 0 0 20-Dec 0 0 0 0 0 0 0 0 0 0 0 0 21-Dec 0 0 0 0 0 0 0 0 0 0 0 0 22-Dec 0 0 0 0 0 0 0 0 0 0 0 0 23-Dec 0 0 0 0 0 0 0 0 0 0 0 0 24-Dec 0 0 0 0 0 0 0 0 0 0 0 0 25-Dec 0 0 0 0 0 0 0 0 0 0 0 0 26-Dec 0 0 0 0 0 0 0 0 0 0 0 0 27-Dec 0 0 0 0 0 0 0 0 0 0 0 0 28-Dec 0 0 0 0 0 0 0 0 0 0 0 0 29-Dec 0 0 0 0 0 0 0 0 0 0 0 0 30-Dec 0 0 0 0 0 0 0 0 0 0 0 0 31-Dec 0 0 0 0 0 0 0 0 0 0 0 0 1-Jan 0 0 0 0 0 0 0 0 0 0 0 0 2-Jan 0 0 0 0 0 0 0 0 0 0 0 0 3-Jan 0 0 0 0 0 0 0 0 0 0 0 0 4-Jan 0 0 0 0 0 0 0 0 0 0 0 0 5-Jan 0 0 0 0 0 0 0 0 0 0 0 0 6-Jan 0 0 0 0 0 0 0 0 0 0 0 0 7-Jan 0 0 0 0 0 0 0 0 0 0 0 0 8-Jan 10 11 13 10 13 11 18 14 12 18 12 14 9-Jan 0 0 0 0 0 0 0 0 0 0 0 0 10-Jan 0 0 0 0 0 0 0 0 0 0 0 0 11-Jan 0 0 0 0 0 0 0 0 0 0 0 0 12-Jan 0 0 0 0 0 0 0 0 0 0 0 0 13-Jan 0 0 0 0 0 0 0 0 0 0 0 0 14-Jan 0 0 0 0 0 0 0 0 0 0 0 0 15-Jan 0 0 0 0 0 0 0 0 0 0 0 0 16-Jan 0 0 0 0 0 0 0 0 0 0 0 0 17-Jan 0 0 0 0 0 0 0 0 0 0 0 0 18-Jan 0 0 0 0 0 0 0 0 0 0 0 0 19-Jan 0 0 0 0 0 0 0 0 0 0 0 0 20-Jan 0 0 0 0 0 0 0 0 0 0 0 0 21-Jan 0 0 0 0 0 0 0 0 0 0 0 0 22-Jan 0 0 0 0 0 0 0 0 0 0 0 0 142

23-Jan 0 0 0 0 0 0 0 0 0 0 0 0 24-Jan 0 0 0 0 0 0 0 0 0 0 0 0 25-Jan 0 0 0 0 0 0 0 0 0 0 0 0 26-Jan 0 0 0 0 0 0 0 0 0 0 0 0 27-Jan 0 0 0 0 0 0 0 0 0 0 0 0 28-Jan 0 0 0 0 0 0 0 0 0 0 0 0 29-Jan 0 0 0 0 0 0 0 0 0 0 0 0 30-Jan 0 0 0 0 0 0 0 0 0 0 0 0 31-Jan 0 0 0 0 0 0 0 0 0 0 0 0 1-Feb 0 0 0 0 0 0 0 0 0 0 0 0 2-Feb 0 0 0 0 0 0 0 0 0 0 0 0 3-Feb 0 0 0 0 0 0 0 0 0 0 0 0 4-Feb 0 0 0 0 0 0 0 0 0 0 0 0 5-Feb 0 0 0 0 0 0 0 0 0 0 0 0 6-Feb 0 0 0 0 0 0 0 0 0 0 0 0 7-Feb 0 0 0 0 0 0 0 0 0 0 0 0 8-Feb 0 0 0 0 0 0 0 0 0 0 0 0 9-Feb 0 0 0 0 0 0 0 0 0 0 0 0 10-Feb 0 0 0 0 0 0 0 0 0 0 0 0 11-Feb 0 0 0 0 0 0 0 0 0 0 0 0 12-Feb 0 0 0 0 0 0 0 0 0 0 0 0 13-Feb 0 0 0 0 0 0 0 0 0 0 0 0 14-Feb 0 0 0 0 0 0 0 0 0 0 0 0 15-Feb 0 0 0 0 0 0 0 0 0 0 0 0 16-Feb 0 0 0 0 0 0 0 0 0 0 0 0 17-Feb 0 0 0 0 0 0 0 0 0 0 0 0 18-Feb 6 10 8 7 10 10 11 9 10 10 10 12 19-Feb 0 0 0 0 0 0 0 0 0 0 0 0 20-Feb 0 0 0 0 0 0 0 0 0 0 0 0 21-Feb 0 0 0 0 0 0 0 0 0 0 0 0 22-Feb 0 0 0 0 0 0 0 0 0 0 0 0 23-Feb 0 0 0 0 0 0 0 0 0 0 0 0 24-Feb 0 0 0 0 0 0 0 0 0 0 0 0 25-Feb 0 0 0 0 0 0 0 0 0 0 0 0 26-Feb 0 0 0 0 0 0 0 0 0 0 0 0 27-Feb 0 0 0 0 0 0 0 0 0 0 0 0 28-Feb 0 0 0 0 0 0 0 0 0 0 0 0 143

29-Feb 0 0 0 0 0 0 0 0 0 0 0 0 1-Mar 0 0 0 0 0 0 0 0 0 0 0 0 2-Mar 0 0 0 0 0 0 0 0 0 0 0 0 3-Mar 0 0 0 0 0 0 0 0 0 0 0 0 4-Mar 0 0 0 0 0 0 0 0 0 0 0 0 5-Mar 0 0 0 0 0 0 0 0 0 0 0 0 6-Mar 0 0 0 0 0 0 0 0 0 0 0 0 7-Mar 0 0 0 0 0 0 0 0 0 0 0 0 8-Mar 0 0 0 0 0 0 0 0 0 0 0 0 9-Mar 0 0 0 0 0 0 0 0 0 0 0 0 10-Mar 0 0 0 0 0 0 0 0 0 0 0 0 11-Mar 6 8 11 9 11 9 10 10 10 10 12 13 12-Mar 0 0 0 0 0 0 0 0 0 0 0 0 13-Mar 0 0 0 0 0 0 0 0 0 0 0 0 14-Mar 0 0 0 0 0 0 0 0 0 0 0 0 15-Mar 0 0 0 0 0 0 0 0 0 0 0 0 16-Mar 0 0 0 0 0 0 0 0 0 0 0 0 17-Mar 0 0 0 0 0 0 0 0 0 0 0 0 18-Mar 0 0 0 0 0 0 0 0 0 0 0 0 19-Mar 0 0 0 0 0 0 0 0 0 0 0 0 20-Mar 0 0 0 0 0 0 0 0 0 0 0 0 21-Mar 0 0 0 0 0 0 0 0 0 0 0 0 22-Mar 0 0 0 0 0 0 0 0 0 0 0 0 23-Mar 0 0 0 0 0 0 0 0 0 0 0 0 24-Mar 8 10 10 10 10 12 12 12 11 11 10 12 25-Mar 0 0 0 0 0 0 0 0 0 0 0 0 26-Mar 0 0 0 0 0 0 0 0 0 0 0 0 27-Mar 0 0 0 0 0 0 0 0 0 0 0 0 28-Mar 0 0 0 0 0 0 0 0 0 0 0 0 29-Mar 0 0 0 0 0 0 0 0 0 0 0 0 30-Mar 0 0 0 0 0 0 0 0 0 0 0 0 31-Mar 0 0 0 0 0 0 0 0 0 0 0 0 1-Apr 8 8 8 10 11 10 13 13 12 17 15 15 2-Apr 0 0 0 0 0 0 0 0 0 0 0 0 3-Apr 0 0 0 0 0 0 0 0 0 0 0 0 4-Apr 0 0 0 0 0 0 0 0 0 0 0 0 5-Apr 0 0 0 0 0 0 0 0 0 0 0 0 144

6-Apr 0 0 0 0 0 0 0 0 0 0 0 0 7-Apr 0 0 0 0 0 0 0 0 0 0 0 0 8-Apr 0 0 0 0 0 0 0 0 0 0 0 0 9-Apr 0 0 0 0 0 0 0 0 0 0 0 0 10-Apr 10 9 10 10 11 12 13 15 14 15 16 17 11-Apr 0 0 0 0 0 0 0 0 0 0 0 0 12-Apr 0 0 0 0 0 0 0 0 0 0 0 0 13-Apr 0 0 0 0 0 0 0 0 0 0 0 0 14-Apr 0 0 0 0 0 0 0 0 0 0 0 0 15-Apr 0 0 0 0 0 0 0 0 0 0 0 0 16-Apr 0 0 0 0 0 0 0 0 0 0 0 0 17-Apr 0 0 0 0 0 0 0 0 0 0 0 0 18-Apr 0 0 0 0 0 0 0 0 0 0 0 0 19-Apr 0 0 0 0 0 0 0 0 0 0 0 0 20-Apr 12 12 12 14 15 14 18 16 17 22 16 20 21-Apr 0 0 0 0 0 0 0 0 0 0 0 0 22-Apr 0 0 0 0 0 0 0 0 0 0 0 0 23-Apr 0 0 0 0 0 0 0 0 0 0 0 0 24-Apr 0 0 0 0 0 0 0 0 0 0 0 0 25-Apr 0 0 0 0 0 0 0 0 0 0 0 0 26-Apr 0 0 0 0 0 0 0 0 0 0 0 0 27-Apr 0 0 0 0 0 0 0 0 0 0 0 0 28-Apr 13 11 13 15 14 14 18 16 12 20 23 21 29-Apr 0 0 0 0 0 0 0 0 0 0 0 0 30-Apr 0 0 0 0 0 0 0 0 0 0 0 0 1-May 0 0 0 0 0 0 0 0 0 0 0 0 2-May 0 0 0 0 0 0 0 0 0 0 0 0 3-May 0 0 0 0 0 0 0 0 0 0 0 0 4-May 0 0 0 0 0 0 0 0 0 0 0 0 5-May 0 0 0 0 0 0 0 0 0 0 0 0 6-May 0 0 0 0 0 0 0 0 0 0 0 0 7-May 0 0 0 0 0 0 0 0 0 0 0 0 8-May 0 0 0 0 0 0 0 0 0 0 0 0 9-May 0 0 0 0 0 0 0 0 0 0 0 0 10-May 0 0 0 0 0 0 0 0 0 0 0 0 11-May 0 0 0 0 0 0 0 0 0 0 0 0 12-May 0 0 0 0 0 0 0 0 0 0 0 0 145

13-May 17 15 17 17 18 19 20 21 20 28 23 26 14-May 0 0 0 0 0 0 0 0 0 0 0 0 15-May 0 0 0 0 0 0 0 0 0 0 0 0 16-May 0 0 0 0 0 0 0 0 0 0 0 0 17-May 0 0 0 0 0 0 0 0 0 0 0 0 18-May 0 0 0 0 0 0 0 0 0 0 0 0 19-May 0 0 0 0 0 0 0 0 0 0 0 0 20-May 8 10 3 11 12 10 11 10 10 12 13 21 21-May 0 0 0 0 0 0 0 0 0 0 0 0 22-May 0 0 0 0 0 0 0 0 0 0 0 0 23-May 0 0 0 0 0 0 0 0 0 0 0 0 24-May 0 0 0 0 0 0 0 0 0 0 0 0 25-May 4 4 3 3 1 6 3 2 2 3 1 8 26-May 0 0 0 0 0 0 0 0 0 0 0 0 27-May 0 0 0 0 0 0 0 0 0 0 0 0 28-May 0 0 0 0 0 0 0 0 0 0 0 0 29-May 10 6 7 7 9 7 6 6 7 8 7 8 30-May 0 0 0 0 0 0 0 0 0 0 0 0 31-May 0 0 0 0 0 0 0 0 0 0 0 0 1-Jun 0 0 0 0 0 0 0 0 0 0 0 0 2-Jun 0 0 0 0 0 0 0 0 0 0 0 0 3-Jun 0 0 0 0 0 0 0 0 0 0 0 0 4-Jun 11 12 10 14 11 12 14 10 13 19 18 17 5-Jun 0 0 0 0 0 0 0 0 0 0 0 0 6-Jun 0 0 0 0 0 0 0 0 0 0 0 0 7-Jun 0 0 0 0 0 0 0 0 0 0 0 0 8-Jun 0 0 0 0 0 0 0 0 0 0 0 0 9-Jun 0 0 0 0 0 0 0 0 0 0 0 0 10-Jun 0 0 0 0 0 0 0 0 0 0 0 0 11-Jun 0 0 0 0 0 0 0 0 0 0 0 0 12-Jun 9 8 17 10 11 10 11 13 11 15 14 17 13-Jun 0 0 0 0 0 0 0 0 0 0 0 0 14-Jun 0 0 0 0 0 0 0 0 0 0 0 0 15-Jun 0 0 0 0 0 0 0 0 0 0 0 0 16-Jun 0 0 0 0 0 0 0 0 0 0 0 0 17-Jun 0 0 0 0 0 0 0 0 0 0 0 0 18-Jun 0 0 0 0 0 0 0 0 0 0 0 0 146

19-Jun 0 0 0 0 0 0 0 0 0 0 0 0 20-Jun 0 0 0 0 0 0 0 0 0 0 0 0 21-Jun 0 0 0 0 0 0 0 0 0 0 0 0 22-Jun 0 0 0 0 0 0 0 0 0 0 0 0 23-Jun 14 13 15 20 19 20 20 18 21 27 27 27 24-Jun 0 0 0 0 0 0 0 0 0 0 0 0 25-Jun 0 0 0 0 0 0 0 0 0 0 0 0 26-Jun 8 5 6 10 8 8 12 11 10 10 9 11 27-Jun 0 0 0 0 0 0 0 0 0 0 0 0 28-Jun 0 0 0 0 0 0 0 0 0 0 0 0 29-Jun 0 0 0 0 0 0 0 0 0 0 0 0 30-Jun 0 0 0 0 0 0 0 0 0 0 0 0 1-Jul 0 0 0 0 0 0 0 0 0 0 0 0 2-Jul 8 8 10 13 12 12 13 12 15 18 16 20 3-Jul 0 0 0 0 0 0 0 0 0 0 0 0 4-Jul 0 0 0 0 0 0 0 0 0 0 0 0 5-Jul 0 0 0 0 0 0 0 0 0 0 0 0 6-Jul 0 0 0 0 0 0 0 0 0 0 0 0 7-Jul 0 0 0 0 0 0 0 0 0 0 0 0 8-Jul 0 0 0 0 0 0 0 0 0 0 0 0 9-Jul 11 9 10 13 12 12 14 15 13 17 15 18 10-Jul 0 0 0 0 0 0 0 0 0 0 0 0 11-Jul 0 0 0 0 0 0 0 0 0 0 0 0 12-Jul 0 0 0 0 0 0 0 0 0 0 0 0 13-Jul 0 0 0 0 0 0 0 0 0 0 0 0 14-Jul 0 0 0 0 0 0 0 0 0 0 0 0 15-Jul 0 0 0 0 0 0 0 0 0 0 0 0 16-Jul 0 0 0 0 0 0 0 0 0 0 0 0 17-Jul 10 10 10 10 10 10 10 10 20 20 10 20 18-Jul 0 0 0 0 0 0 0 0 0 0 0 0 19-Jul 0 0 0 0 0 0 0 0 0 0 0 0 20-Jul 0 0 0 0 0 0 0 0 0 0 0 0 21-Jul 10 7 6 13 12 13 11 12 13 10 12 10 22-Jul 0 0 0 0 0 0 0 0 0 0 0 0 23-Jul 0 0 0 0 0 0 0 0 0 0 0 0 24-Jul 4 7 8 13 10 8 6 10 11 10 12 16 25-Jul 0 0 0 0 0 0 0 0 0 0 0 0 147

26-Jul 0 0 0 0 0 0 0 0 0 0 0 0 27-Jul 0 0 0 0 0 0 0 0 0 0 0 0 28-Jul 7 10 7 12 8 7 10 10 14 16 15 15 29-Jul 0 0 0 0 0 0 0 0 0 0 0 0 30-Jul 0 0 0 0 0 0 0 0 0 0 0 0 31-Jul 10 10 8 10 10 10 10 10 10 10 10 16 1-Aug 0 0 0 0 0 0 0 0 0 0 0 0 2-Aug 0 0 0 0 0 0 0 0 0 0 0 0 3-Aug 0 0 0 0 0 0 0 0 0 0 0 0 4-Aug 7 12 10 14 14 12 10 14 18 16 24 20 5-Aug 0 0 0 0 0 0 0 0 0 0 0 0 6-Aug 0 0 0 0 0 0 0 0 0 0 0 0 7-Aug 0 0 0 0 0 0 0 0 0 0 0 0 8-Aug 0 0 0 0 0 0 0 0 0 0 0 0 9-Aug 0 0 0 0 0 0 0 0 0 0 0 0 10-Aug 7 15 14 19 19 18 10 15 27 19 27 30 11-Aug 0 0 0 0 0 0 0 0 0 0 0 0 12-Aug 0 0 0 0 0 0 0 0 0 0 0 0 13-Aug 0 0 0 0 0 0 0 0 0 0 0 0 14-Aug 0 0 0 0 0 0 0 0 0 0 0 0 15-Aug 10 15 6 15 10 10 10 15 10 20 25 16 16-Aug 0 0 0 0 0 0 0 0 0 0 0 0 17-Aug 0 0 0 0 0 0 0 0 0 0 0 0 18-Aug 10 13 11 19 25 20 23 20 25 24 21 22 19-Aug 0 0 0 0 0 0 0 0 0 0 0 0 20-Aug 0 0 0 0 0 0 0 0 0 0 0 0 21-Aug 5 5 3 8 6 8 6 7 5 9 9 5 22-Aug 0 0 0 0 0 0 0 0 0 0 0 0 23-Aug 3 3 2 7 5 7 7 7 8 11 10 9 24-Aug 0 0 0 0 0 0 0 0 0 0 0 0 25-Aug 0 0 0 0 0 0 0 0 0 0 0 0 26-Aug 4 7 5 8 7 8 5 7 9 12 12 6 27-Aug 0 0 0 0 0 0 0 0 0 0 0 0 28-Aug 0 0 0 0 0 0 0 0 0 0 0 0 29-Aug 0 0 0 0 0 0 0 0 0 0 0 0 30-Aug 0 0 0 0 0 0 0 0 0 0 0 0 31-Aug 0 0 0 0 0 0 0 0 0 0 0 0 148

Appendix B. Water Chemistry Data

Raw data

Conduct- Total ivity NO3- Nitrite NH4+ Ptot hardness Date Sys # (μS/cm) (mg/) (mg/l) (mg/l) (mg/l) (mmol/l) 22-Jul 1 274 2.3 0.07 0.11 0.0078 1.6 22-Jul 7 219 0.4 0.01 0.08 0.0078 1.78 22-Jul 12 313 5.7 0.17 0.13 0.0078 1.78 22-Jul 6 157 1.2 0.05 0.1 0.1094 1.6 22-Jul 9 255 1.6 0.03 0.13 0.0313 1.78 22-Jul 10 212 1.2 0.01 0.13 0.0195 1.25 22-Jul 2 277 6.4 0.16 0.22 0.4475 0.53 22-Jul 4 338 7.2 0.15 0.09 1.1765 0.71 22-Jul 8 168 8.7 0.05 0.24 0.5355 0.53 22-Jul 3 215 6.4 0.05 0.11 0.0195 0.71 22-Jul 5 147 5.6 0.1 0.12 0.3303 0.71 22-Jul 11 235 8.2 0.05 0.1 0.0313 0.53 22-Jul Tap 49 3.9 0 0.07 0.0098 0.53 13-Nov 1 215 2.1 0.03 0.16 0.0039 1.78 13-Nov 7 465 0.3 too low 0.21 0.0156 1.78 13-Nov 12 316 3.7 0.01 0.24 0 1.78 13-Nov 6 168 too low too low 0.12 0.0078 1.6 13-Nov 9 170 0.1 0.01 0.12 0.0098 1.78 13-Nov 10 335 0.2 too low 0.14 0.0078 2.32 13-Nov 2 194 2.1 0.01 0.44 0 0.89 13-Nov 4 403 2.9 too low 0.24 0.0078 1.25 13-Nov 8 203 1.6 too low 0.52 0.002 1.07 13-Nov 3 305 2.8 too low 0.16 0 1.25 13-Nov 5 140 2 too low 0.3 0.0039 0.53 13-Nov 11 263 2.3 0.01 0.08 0.002 1.07 22-Jul 1 610 0.7 0.05 0.64 0.0017 2.14 22-Jul 7 540 4 0.02 0.1 0.0056 2.32 22-Jul 12 580 7.3 too low too low 0.0017 2.32 22-Jul 6 560 0.2 0.08 0.13 0.0083 2.5 22-Jul 9 560 0.3 0.1 0.3 0.0083 2.32 149

22-Jul 10 540 0.6 0.05 too low 0.0043 2.32 22-Jul 2 440 2.9 0.07 1.19 0.0056 1.07 22-Jul 4 610 6 0.09 0.35 0.0043 1.07 22-Jul 8 530 3 0.08 0.19 0.0017 1.78 22-Jul 3 600 3.9 0.08 too low 0.0096 1.96 22-Jul 5 610 3.1 too low 0.07 0.003 1.07 22-Jul 11 540 3.4 0.06 0.43 0.0096 1.96

Raw pH data

Date Sys # pH 31-Jul-19 1 8.2 31-Jul-19 7 8.3 31-Jul-19 12 8.1 31-Jul-19 6 8.3 31-Jul-19 9 8.3 31-Jul-19 10 8.2 31-Jul-19 2 4.4 31-Jul-19 4 4.2 31-Jul-19 8 4.3 31-Jul-19 3 4.3 31-Jul-19 5 4.4 31-Jul-19 11 4.4 31-Jul-19 Tap 7.7 2-Oct-19 1 8.2 2-Oct-19 7 8.4 2-Oct-19 12 8.4 2-Oct-19 6 8.4 2-Oct-19 9 8.5 2-Oct-19 10 8.5 2-Oct-19 2 4.8 2-Oct-19 4 4.4 2-Oct-19 8 4.7 2-Oct-19 3 4.6 2-Oct-19 5 4.5 2-Oct-19 11 4.6 14-Nov-19 1 7.8 14-Nov-19 7 7.9 150

14-Nov-19 12 7.9 14-Nov-19 6 7.9 14-Nov-19 9 8.0 14-Nov-19 10 8.0 14-Nov-19 2 4.7 14-Nov-19 4 4.3 14-Nov-19 8 4.6 14-Nov-19 3 4.4 14-Nov-19 5 4.5 14-Nov-19 11 4.6 3-Apr-20 1 7.9 3-Apr-20 7 8 3-Apr-20 12 8.1 3-Apr-20 6 8.1 3-Apr-20 9 8.1 3-Apr-20 10 8.2 3-Apr-20 2 4.7 3-Apr-20 4 4.2 3-Apr-20 8 4.5 3-Apr-20 3 4.3 5-Apr-20 5 4.4 3-Apr-20 11 4.5 29-Jul-20 1 8.1 29-Jul-20 7 8.1 29-Jul-20 12 8 29-Jul-20 6 8.1 29-Jul-20 9 8.1 29-Jul-20 10 8.2 29-Jul-20 2 4.9 29-Jul-20 4 4.4 29-Jul-20 8 4.8 29-Jul-20 3 4.4 29-Jul-20 5 4.5 29-Jul-20 11 4.6

151

Average water chemistry parameters.

Lime fen Lime fen Acidic bog Acidic bog Parameter Date Non-veg Veg Non-veg Veg 22-Jul-19 269 208 261 199 Conductivity 13-Nov-19 332 224 267 236 (μS/cm) 22-Jul-20 577 553 527 583

- 22-Jul-19 2.80 1.33 7.43 6.73 NO3 13-Nov-19 2.03 0.10 2.20 2.37 (mg/l) 22-Jul-20 4.00 0.37 3.97 3.47

+ 22-Jul-19 0.11 0.12 0.18 0.11 NH4 13-Nov-19 0.20 0.13 0.40 0.18 (mg/l) 22-Jul-20 0.25 0.14 0.58 0.17 22-Jul-19 0.0078 0.0534 0.7198 0.1270 Ptot 13-Nov-19 0.0065 0.0085 0.0033 0.0020 (mg/l) 22-Jul-20 0.0030 0.0070 0.0039 0.0074 22-Jul-19 1.72 1.54 0.59 0.65 TH 13-Nov-19 1.78 1.90 1.07 0.95 (mmol/l) 22-Jul-20 2.26 2.38 1.31 1.66

Two-way repeated ANOVA sources of variation for water chemistry values (excluding baseline).

Parameter Filter media Vegetation Conductivity 0.6103 0.4644 - NO3 0.1582 0.1268 + NH4 0.2180 0.0864 Ptot 0.2400 0.2564 TH 0.0001* 0.6954 *Significant

Two-way ANOVA sources of variation for water chemistry values.

Parameter Date Filter media Vegetation Interaction 22-Jul-19 0.8146 0.1123 0.985 Conductivity 13-Nov-19 0.6772 0.2979 0.5527 22-Jul-20 0.7384 0.5802 0.2039 22-Jul-19 0.0007* 0.2781 0.4397 - NO3 13-Nov-19 0.0542 0.1404 0.0876 22-Jul-20 0.1963 0.0940 0.1878 152

22-Jul-19 0.2237 0.2693 0.1248 + NH4 13-Nov-19 0.0493* 0.0251* 0.2209 22-Jul-20 0.4042 0.2366 0.4666 22-Jul-19 0.0145* 0.0624 0.0355* Ptot 13-Nov-19 0.1091 0.9048 0.5619 22-Jul-20 0.6862 0.0420* 0.8924 22-Jul-19 <0.0001* 0.5503 0.2415 TH 13-Nov-19 0.0009 >0.9999 0.4789 22-Jul-20 0.0026* 0.2551 0.5597 *Significant

153

Appendix C. Vegetation Data

Plant replacement/addition.

Date Species Element Count 12-Apr-19 Carex serotina 6 1 Mimulus jungermannioides 6 1 Mimulus jungermannioides 9 1 Mimulus jungermannioides 10 1 Leontopodium sp. 9 1 Leontopodium sp. 10 1 3-Apr-20 Schoenus ferrugineus 6 5 Schoenus ferrugineus 9 5 Schoenus ferrugineus 10 5 Carex davalliana 6 4 Carex davalliana 9 4 Carex davalliana 10 4 25-Apr-20 Aster flaccidus 6 1 Aster flaccidus 9 1 Aster flaccidus 10 1 Succisella inflexa 'Frosted Pearls' 6 3 Succisella inflexa 'Frosted Pearls' 9 3 Succisella inflexa 'Frosted Pearls' 10 3 Campanula cochlearifolia 6 3 Campanula cochlearifolia 9 3 Campanula cochlearifolia 10 3

Vegetation quality raw data Lime Lime Lime Acid Acid Acid Date Variable fen 1 fen 2 fen 3 bog 1 bog 2 bog 3 21-Jul-19 Coverage 2 2 2 3 2 2 Estimated richness ratio 66.67% 66.67% 66.67% 81.82% 81.82% 81.82% Vitality 3 3 3 5 4 4 14-Aug-19 Coverage 3 3 2 4 3 3 Estimated richness ratio 75.00% 58.33% 66.67% 81.82% 81.82% 72.73% Vitality 3 2 3 4 4 4 154

26-Sep-19 Coverage 3 3 2 5 3 4 Estimated richness ratio 66.67% 75.00% 58.33% 63.64% 54.55% 63.64% Vitality 4 3 3 4 4 4 30-Oct-19 Coverage 4 4 3 5 4 4 Estimated richness ratio 58.33% 50.00% 58.33% 54.55% 54.55% 63.64% Vitality 3 2 3 4 4 4 3-Apr-20 Coverage 4 4 2 5 4 5 Estimated richness ratio 37.50% 37.50% 37.50% 63.64% 54.55% 63.64% Vitality 3 3 2 4 4 4 8-May-20 Coverage 5 4 4 5 5 5 Estimated richness ratio 50.00% 50.00% 56.25% 54.55% 54.55% 54.55% Vitality 4 3 4 5 4 4 4-Jun-20 Coverage 5 5 4 5 5 5 Estimated richness ratio 50.00% 62.50% 50.00% 54.55% 54.55% 63.64% Vitality 4 3 3 4 4 5 19-Jul-20 Coverage 5 5 5 5 5 5 Estimated richness ratio 50.00% 56.25% 43.75% 45.45% 63.64% 54.55% Vitality 4 4 4 4 4 4 18-Aug-20 Coverage 4 4 4 5 5 5 Estimated richness ratio 37.50% 43.75% 50.00% 54.55% 54.55% 45.45% Vitality 3 3 3 4 4 5

155