Exploring the use of indigenous Western Cape plants as potential water and soil pollutant phytoremediators with a focus on green infrastructure

COPYRIGHT © The Author(s) Published under a Creative Commons Attribution 4.0 International Licence (CC BY 4.0) Urban water managers, engineers and conservation ecologists in the Western Cape (WC) Province of South Africa are faced with a major environmental and human health challenge, with urbanisation, industrialisation, population growth and agricultural development placing pressure on the limited water and soil resources. In addressing this resource degradation an effective, affordable and sustainable solution is required. The implementation of ‘green infrastructure’ (GI), such as phytoremediation, involves the use of plants to hinder pollutant transport and attenuate runoff flow, protecting the health of the human population and the environment. However, care must be taken when selecting plant species due to possible invasive behaviour, affecting ecosystem dynamics. As a result of the need for resource remediation in both urban and rural areas, the use of non-invasive indigenous species is vital to an efficient and sustainable technology, as urban areas are often the initial sites for introduction from which invasions spread. This paper proposes indigenous WC species for potential use in GI, identified from global bioremediation literature, as an aid to the practicing civil engineer and water manager responsible for the design and management of the phytotechnology. These indigenous species offer potential as phytoremediators in local GI, as well as suggest the types of plants that should be investigated further as alternatives to effective exotics. The investigation returned 56 non-invasive WC plant species likely to aid resource remediation without jeopardising the conservation and biodiversity of the administered area. The selected vegetation is potentially capable of increasing heterogeneity and adjusting to the dynamic biogeographic conditions of the recipient habitat. Thus, distinct species capable of remediating a wide range of environmental contaminants for GI, into the diverse habitats of the WC, at a fraction of the cost of conventional techniques, are promoted. Exploring the use of indigenous Western Cape plants as potential water and soil pollutant phytoremediators with a focus on green infrastructure


INTRODUCTION
Persistent and excessive release of heavy metal, organic and inorganic pollutants from anthropogenic activities is blamed for destroying entire natural ecosystems, posing a major human and environmental health problem (Cunningham et al., 1995;Salt et al., 1995;Dhir et al., 2009;Abhilash et al., 2012, Choudhary et al., 2015Tripathi et al., 2015). The Western Cape (WC) Province of South Africa has been subjected to similar deleterious impacts, with increased urbanisation, high population growth, agricultural development, dysfunctional water and wastewater treatment works, and industrialisation placing enormous pressure on the limited soil and water resources (Giliomee, 2006;Nomquphu et al., 2007).
The widespread contamination of soil, a non-renewable natural resource in the short term and very expensive to reclaim once physically or chemically degraded, has for decades been a global remediation and management challenge (Ali et al., 2015;Rada et al., 2019). This rampant degradation is evident with more than 25% of global soil resources highly degraded and roughly 44% moderately degraded, predominantly as a result of pollution (Peuke and Rennenberg, 2005). Alarmingly, South Africa is recognised as one of the regions experiencing the most rapid environmental degeneration (Abhilash, 2015). The WC receives various pollutants of anthropogenic origin to the soils, which include significant amounts of bioavailable heavy metals (Schloemann, 1995), fertilisers, pesticides, herbicides and fungicides (Malan et al., 2015), as well as stormwater, wastewater and sewage runoff (Chen et al., 2008;Müller et al., 2014).
Similarly, pollutant deposition deteriorates water-quality in urban and rural watercourses and groundwater aquifers, raising a plethora of public and environmental health concerns (Constantine et al., 2014;Sibanda et al., 2015;Pakdel and Peighambardoust, 2018). In South Africa, main rivers classified as having a poor ecological condition increased by 500% between 1999 and 2011, with some beyond the point of recovery, posing a threat to the country's potable water resources (Oberholster and Ashton, 2008;DWS, 2018). This is as a result of the nationwide failure to properly treat wastewater, where 56% of the sewage treatment works are failing and 11% are dysfunctional (DWS, 2018). De Villiers and Thiart (2007) found that 6 of the 20 largest river catchments in South Africa that are under eutrophication are located in the WC. In addition, the findings of Musingafi and Tom (2014) and Milandri et al. (2012) point to the fact that urban effluent discharge contributes a wide range of pollutants to freshwater ecosystems.
Green infrastructure (GI) is the interconnected set of natural and engineered ecological systems providing environmental services . In comparison with traditional grey infrastructure, these systems have been found to reduce financial costs by up to 42% (Vineyard et al., 2015). These systems offer a novel and sustainable approach to polluted water and soil systems, https://doi.org/10.17159/wsa/2021.v47.i3.11860 reducing pollutant transport and reducing resource toxicity, by combining resource remediation with sustainable civil engineering methods (Choudhary et al., 2015;Kondo et al., 2015;Malan et al., 2015;Malherbe et al., 2018). The GI engineered approach has the ability to complement or replace existing technological or grey infrastructure designs, presenting a more cost-effective, self-sustaining and versatile long-term alternative (Nel et al., 2014). In South Africa, GI is currently an under-realised asset (DWS, 2018).
Critical to GI efficiency, phytoremediation uses appropriately selected plants for the in-situ treatment of environmental contaminants from soils, sediments and water (Terry and Banuelos, 1999;Dietz and Schnoor, 2001;Visoottiviseth et al., 2002;Peuke and Rennenberg, 2005;Payne et al., 2014). A large number of plants possess potential to detoxify, degrade and/or remove pollutants from the environment, and have gained importance globally, receiving significant scientific and commercial attention (Salt et al., 1998;Gleba et al., 1999;Guerinot and Salt, 2001;Krämer and Chardonnens, 2001;Meagher, 2000;Peuke and Rennenberg, 2006). This green engineering solution is popular due to its cost-effectiveness, aesthetic advantages, longterm applicability, employment generation capacity and scope of pollutant remediation efficacy (Raskin et al., 1994;Abhilash, 2015;Marrugo-Negrete et al., 2015). Although effective, plants and their rhizosphere organisms phytoremediate in different ways, with various mechanisms suitable for different pollutants (Read et al., 2010). The use of field and vegetable crops cultivated for commercial use as phytoremediators has proven valuable in bioremediating municipal and industrial discharge from failing sewage treatments works, as well as rehabilitating old and abandoned mining sites (Poonam et al., 2014;Rizwan et al., 2016).
In designing a sustainable GI remediation technology, the responsible engineer must assess the potential adverse impacts induced by the introduction of chosen effective pollutant-specific remediators, based on the findings of previous investigations (Leguizamo et al., 2017). Prior to plant selection for use in remediation projects, the ecosystem-related and ecological functions of the plants need to be established (Budelsky and Galatowitsch, 2004). For instance, failure to consider invasion threat may lead to creating an artificial or altered environment in which alien species thrive (Castro-Diez et al., 2014). In particular, remediation measures must endeavour to utilise plants that are indigenous to the site, in an attempt to return the resource to a sustainable state (Peer et al., 2005). A practicing civil engineer who lacks expertise in plant behaviour or ecosystem dynamics which regulate plant's potential future invasiveness in phytogeographic environments may be ill-equipped to design an effective yet sustainable engineered remediation technology.
It is for this reason that this study focused on indigenous plant species naturally occurring in the WC for potential use in pollutant bioremediation technologies. The plants were selected on the basis of their presence in global phytoremediation literature, phytogeographic distribution, conservation status and invasiveness. This article discusses the phytoremediation process and WC vegetation whilst emphasising the impact their introduction may have on the recipient ecosystem. Finally, after investigating plant species from relevant literature, potential indigenous species and cultivated crops effective for GI technologies in the WC are recommended.

Phytoremediation processes within green infrastructure engineering
The term phytoremediation is applied to a technology that makes use of both wild and transgenic plant species for the treatment of contaminated soils, sediments, water and air, with the aim to effectively restore, protect or ameliorate environmental degradation by removing pollutants or rendering them harmless (Salt et al., 1998;Terry and Banuelos, 1999;Dietz and Schnoor, 2001). These pollutants are predominantly generated by agricultural and industrial products and practices, urban pollution, stormwater runoff and defective wastewater treatment facilities (Khan et al., 2000). All plants have the ability to accumulate heavy metals that are essential for growth and development from soil and water, including Cu, Fe, Mg, Mn, Mo, Ni and Zn, with some plants able to accumulate heavy metals such as Ag, Cd, Co, Cr, Hg, Pb and Se, that offer no biological function (Dhir et al., 2009). Various phytoremediation processes exist, making use of different species of plants for each contaminant (Cunningham et al., 1997;Dietz and Schnoor, 2001). There is, however, a limit to the extent to which plants can accumulate these metals, beyond which they become toxic, with plants varying in removal efficiency and tolerance (Read et al., 2010). The intricate planning and design of GI technologies must consider both the type of pollution and habitats of the specific site, to appropriately identify and select potential species for optimised phytoremediation. Therefore, the use of non-invasive species, adept at remediating the pollution at hand whilst limiting the risk of biodiversity loss, should always be promoted (Payne et al., 2015). Numerous laboratory and field assessments have developed a better understanding of heavy metal, petrochemical, ammunition waste, chlorinated solvent, landfill leachate, non-point source agricultural runoff (pesticides and fertilisers) and urban stormwater runoff phytoremediation (Salt et al., 1995;Chaney et al., 1997;Macek et al., 2000;Pilon-Smits, 2005;Abhilash et al., 2012;Milandri et al., 2012). GI capitalises on the innate abilities of photosynthetic plants to eliminate a variety of pollutants, by destruction, inactivation, extraction, volatilisation or immobilisation, with some extracted metals even recycled for value (Raskin et al., 1994;Peuke and Rennenberg, 2006;Leguizamo et al., 2017).

The effect of biological invasion
Biological invasions alter the structure and function of the natural ecosystem, causing the loss of an ecoregion's characteristic species, and may result in loss of indigenous biota . South Africa has a long history of plant introduction and invasion transforming ecosystems, posing a major threat to the country's biodiversity, impacting negatively on the ecosystem's capacity to deliver goods and services, and in some cases severely threatening human livelihoods . Alien plant invasion costs the country 2 billion ZAR annually for its control (Van Wilgen et al., 2020;Zengeya and Wilson, 2020). This widespread degradation is particularly evident in the WC Province with the most extensive as well as the greatest number of invasive species, notably reducing the value of fynbos ecosystems by over 195 billion ZAR due to a loss in ecosystem services (Van Wilgen et al., 2001;Pyšek and Richardson, 2010;Van Wilgen et al., 2020).
The WC is home to 70% of the Cape Floristic Region (CFR), recognised as one of the richest habitats in the world, regarding its floristic heterogeneity and endemism (Von Hase et al., 2003). With a prominent diversity of over 9 000 species of which 70% are endemic, the CFR is one of only 25 globally accepted biodiversity hotspots (Giliomee, 2006). For its tiny size, the area most likely has the richest flora worldwide (McDowell and Moll, 1992).

METHODOLOGY
This paper investigated 800 literature sources to guide the selection of proven effective phytoremediation species for potential use in GI technologies in the WC. In compiling a list of potential phytoremediators, peer-reviewed articles, books, reports, case https://doi.org/10.17159/wsa/2021.v47.i3.11860 studies, conference proceedings, theses and dissertations, as well as online databases, were investigated. The findings related to various studies of bioremediation initiatives for stormwater quality improvement, sustainable urban drainage systems, watersensitive urban designs, water and soil rehabilitation initiatives and plant affinity for heavy metal accumulation. From these findings, potential endemic and indigenous candidates for the WC were identified by comparing herbarium records from the South African National Biodiversity Institute's Red List of South African Plants (SANBI) and Plants of Southern Africa (POSA), Plants of the Greater Cape Floristic Region (PGCFR), and Stellenbosch University Botanical Garden (SUNBG) with the species from literature. Plants displaying invasive characteristics, whose introduction may threaten the sustainability of a phytoremediation project and increase ecosystem vulnerability, were excluded, based on local and international standards. By considering existing recommendations and reported presence in literature, spatial distribution, invasive threat and vulnerability to extinction, erroneous introduction of potential plants into engineered designs and natural ecosystems is minimised.

RESULTS AND DISCUSSION
The investigation delivered 1 410 plant species, from 582 genera with 136 subspecies and variations, for potential use in GI phytoremediation initiatives. Analysis returned 257 indigenous and naturalised South African plant species of which 174 were distributed phytogeographically throughout the WC Province. Of the species, 80 were found to be registered as either endemic or indigenous to the province, with 56 of these regarded as non-invasive and of least conservation concern (Table 1). These plants are likely to aid remediation without jeopardising the conservation and biodiversity of the recipient ecoregion. In reporting, emphasis is placed on effective phytoremediators, their distribution as endemics or indigenous to the WC, conservation status and invasiveness.
Of the 56 phytoremediators, 3 are endemic, and 53 are indigenous to the WC. None of the species are registered as invasive to the WC. The list includes both aquatic and terrestrial plants with varying seasonal activity, which benefit GI initiatives, by resisting dormancy during periods of drought. Although the species listed are registered as non-invasive, caution must be taken when introducing plants, as they may with new evidence be found to threaten the sustainability of the recipient ecosystem. In addition, the behaviour of non-endemic species should be assessed prior to introduction in recipient habitats, whereafter approval from relevant ecologists must be sought. The practising civil engineer is cautioned to enlist the help of plant experts when choosing a species for application in GI.

Phytoremediation and Western Cape vegetation
The results showed several WC plant species that are effective across a range of pollutants, which include urban and rural stormwater runoff, agricultural effluent in the form of pesticides, herbicides and fertilisers, heavy metals, explosives and ammunition wastes, radionuclides, organic pollutants, carcinogenic air, water and soil, chlorinated aliphatic hydrocarbons, petroleum contaminants, domestic and industrial wastewater effluent, landfill leachate, sewage discharge and tannery waste. The list in Table 1 includes 10 endemic South African species, which are: Agapanthus africanus, Agapanthus praecox subsp. minimus, Arctotis acaulis, Aristea capitata, Berkheya zeyheri subsp. rehmannii var. rogersiana, Carpobrotus edulis subsp. edulis, Carpobrotus edulis subsp. parvifolius, Cyperus textilis, Elegia tectorum and Prionium serratum. The number of endemics, in proportion to total indigenous species identified, suggests immense potential for South African phytoremediators yet to be investigated, specifically in the WC, which offers unparalleled biodiversity richness.

The risk of invasive plants in green infrastructure
The use of non-invasive indigenous plants in bioremediation technologies supports plant acclimatisation, alleviates specimen sourcing and contributes to ecosystem conservation initiatives, bolstering the biodiversity and heterogeneity of a habitat by limiting exposure to invasive species. A dynamic resident biota influenced by natural factors, as well as the recipient site's conditions, must be accounted for in order to determine potential invasiveness of introduced species. Thus, consultation with a number of disciplines within science and engineering is imperative to promote an amalgamation of knowledge for the creation of a sustainable design science for GI.
From the findings, some species can be selected based upon their agressive growth properties and hardiness, common traits shared among efficient remediators. The species demanding caution are Ceratophyllum demersum, Panicum repens, Phragmites australis and Pteris cretica, which have been identified as prospective problematic species. This, however, does not excuse caution for the remaining plants, which may be classified as invasive with new evidence. The traits linked to invasiveness in ecosystems, mirrored in traits supporting phytoremediation efficiency, e.g., rapid growth and spread, hardiness and disease and pest resistance, as well as interactions with resident biota in prevailing environmental conditions, were considered (Le Roux et al., 2020).
In investigating WC crops, 8 species were identified for potential phytoremediation from reported literature (not shown). These species are Beta vulgaris subsp. vulgaris, Brassica rapa subsp. rapa, Cannabis sativa, Daucus carota subsp. sativus, Linum usitatissimum, Nicotiana tabacum, Sorghum bicolor and Vigna unguiculata subsp. protracta. Although the WC crops offer prospects for phytoremediation, it is the authors' opinion that they should not be considered for animal feed or human consumption, pending continuation of research on environmental toxin accumulation. The removed toxins are stored in different vascular compartments of the plant until mortality (Vamerali et al., 2010). During cultivation, these toxins may still pose a threat to environmental and human health (Khan et al., 2015). For a site exposed to extreme contamination, the use of cultivated crops can, however, mitigate the immediate environmental threat, supporting their inclusion as potential remediators.
*Care to be taken when using in sustainable and sensitive systems, as they display aggressive properties. resource remediation, assisted by the recommended indigenous plant list, must consider ecosystem dynamism in fluctuating biotic conditions and a changing climate. The assessment of ecosystem characteristics and plant specific invasiveness, whilst incorporating precautionary principles of plant introduction, are crucial prerequisites for effective planning, demanding a multidisciplinary approach between engineering and science . It is for these reasons that the civil engineer assessing plant species inclusion in the proposed species list for phytoremediation initiatives must consult relevant specialists in environmental science, field botany, conservation ecology and soil science, prior to introducing plant species into GI. In considering the habitats and context of plant species located in the WC, this interdisciplinarity will equip the practising engineer with the necessary skills and knowledge backed by scientific literature to design an effective but sustainable phytoremediation system for runoff remediation and site rehabilitation initiatives.

CONCLUSIONS
Increased anthropogenic activities have exposed the urban and rural WC to sustained environmental degradation, posing a major environmental and human health problem (Malherbe et al., 2018).
In an effort to sustainably remediate soil and water polluted by deleterious contaminants, engineered strategies of an effective and affordable nature are required. GI offers a sustainable and cost-effective solution for various environmental contaminants in soil, sediment and water (Terry and Banuelos, 1999). In combating biodiversity loss, the selection of potential plants for bioremediation initiatives must consider species that are naturally acclimatised to the recipient ecosystem, and which do not threaten the natural biodiversity. This is particularly important for urban remediation technologies as urban areas are often the initial sites for introduction, and from which invasions spread (Zengeya and Wilson, 2020). Reduced ecosystem biodiversity poses one of the greatest ecological engineering challenges (Tilman and Lehman, 2001). The need to use effective plant species for remediation cannot overshadow the need to protect and conserve the ecosystem. Thus the use of indigenous plant species that ameliorate the degraded environment and contribute to preservation must be considered. Although a specific species may not be recorded as invasive to the WC phytogeographic area, with introduction to a new habitat, the non-invasive species may threaten habitat sustainability . For this reason, species sourced from similar habitat conditions must be preferred.
The listed potential phytoremediators may aid in regulating the natural ecosystem, maintain equilibrium, and increase heterogeneity. They are also capable of adjusting to dynamic biogeographic conditions of various recipient habitats. This heterogeneity between species and phytogeographic distributions creates opportunities for the introduction of distinct vegetation into the diverse habitats of the WC, capable of remediating a wide range of environmental contaminants, at a fraction of the cost of conventional techniques. The non-invasive plants offer an attractive alternative to known invasive alien plants, while supporting natural biodiversity and conservation initiatives (Pyšek and Richardson, 2010).
In selecting potential phytoremediators the dynamic factors that mediate plant species' interactions with resident biota and prevailing conditions, affecting potential invasiveness within a specific ecosystem, need to be considered. It is imperative that the practicing engineer receives input from relevant specialists, and that plant introduction is only undertaken with their satisfaction. It would be advantageous for the Water Research Commission (WRC) to work towards establishing a guideline, accounting for different pollution and ecosystem contexts, with input from the following specialists: hydrologists, town and regional planners, disaster management practitioners, ecologists and soil scientists, environmental and municipal managers, and landscape architects, which engineers could use with less oversight.
In decontaminating sites exposed to extreme and continuous pollution, field crops potentially contribute to phytoremediation in toxic conditions due to their short life cycle, large biomass production and adaptability to the changing environment (Ciura et al., 2005), in turn, mitigating severe environmental effects and greatly reducing environmental cost (Santos-Jallath et al., 2012). This can, however, only be done with greater monitoring of pollutant toxicity tolerance and rotation over a number of cropping cycles under strict circumstances.
Acknowledging the WC's global biodiversity richness status (Von Hase et al., 2003), the absence of research relating to the phytoremediation potential of the WC plant species for GI initiatives reinforces the need for further studies. In highlighting the remediation efficacy of endemic species, the vegetation may be granted an additional economic value that could aid the decisions which encourage protection and development (Barbier et al., 1997).