New Zealand plant growth trials in rapid bio-filtration media

The purpose of this research was to:-

• (1)

  determine whether readily available New Zealand (NZ) plant species could establish and grow in a locally manufactured rapid bio-filtration medium (Filterra or NZF) in a range of NZ climates, and to assess their growth rates;

• (2)

  compare plants grown in rapid bio-filtration media with others grown in an Auckland Council approved raingarden mixture;

• (3)

  determine the capability of NZ plant species to sustain long-term growth in a rapid bio-filtration medium.

The upper limit of hydraulic conductivity (saturated, K_sat) of media in bio-filtration treatment devices is typically limited by two main factors.

• (1)

  Providing adequate contact time for pollutant removal mechanisms to act. The relationship between pollutant removal and K_sat is typically pollutant-dependent, being slowest for nitrogen and fastest for sediments and some metals (Fassman et al. 2013).

• (2)

  Plant survival (Payne et al. 2015) and the ability of plants to establish in the media (FAWB 2009). Fast draining media has a lower capacity to retain water and nutrients, and consequently a lower capacity to support plant growth.

A common maximum K_sat value used for a bio-filtration medium in design guidelines is 300 mm/hr (Woods Ballard et al. 2015). Design guidance in Australia was found to allow higher rates in tropical areas (up to 500 mm/hr) (Payne et al. 2015) Auckland, New Zealand guidance (2017) allows rates up to 1,000 mm/hr (Cunningham et al. 2017). Recently, rapid bio-filtration treatment devices using media with K_sat exceeding 2,500 mm/hr have been approved for use in Washington State (USA) by the Washington State Department of Ecology.

Stormwater360 (SW360) in partnership with Contech Engineered Solutions (USA) have produced Filterra media for the New Zealand market using locally sourced materials (Hannah et al. 2015). Five plant growth trials were set up to evaluate the ability of local plant species to establish and grow in the media. The plants were selected to cover a range of growth forms and rooting strategies.

The 26 plant species used in the trials can be grouped into four categories.

Dicotyledonous trees and shrubs – These are woody, flowering plants which grow with a pair of leaves from the embryo of the seed. Most are notable for fibrous branched root systems. The species of this type used in the trials were:-

• Coprosma propinqua

• Hebe speciosa

• Hoheria angustifolia

• Hoheria populnea

• Kunzea robusta

• Melicytus ramiflorus

• Metrosideros Crimson Glory

• Metrosideros excelsa

• Pennantia corymbosa

• Plagianthus regius

• Pomaderris apetala

• Sophora microphylla

Monocotyledonous Herbs – These are flowering plants with a single leaf growing from the embryo of the seed. The root systems are spread out with no single dominant root and stems usually have no secondary growths. The species of this type used in the trials were:-

• Arthropodium bifurcatum

• Dianella latissima

• Phormium cookianum

• Astelia banksii

• Xeronema callistemon

• Libertia grandiflora

Grasses – monocotyledonous flowering plants from the Poaceae or Gramineae family. They are versatile plants and are common species used in urban landscaping. The species of this type used in the trials were:-

• Chionochloa rigida

• Chionochloa rubra

• Festuca coxii

• Poa cita

Sedges – monocotyledonous flowering plants from the Cyperaceae family which resemble grasses and rushes, and are easy to propagate and fast growing. Their stems usually have a triangular cross-section. The species of this type used in the trials were:-

• Carex secta

• Carex virgata

• Carex flagellifera

• Carex testacea

Four native NZ, groundcover, plant species with rapid growth characteristics were evaluated over a 12-week period. Eight plants were grown in NZF media and eight in an Auckland Council approved raingarden medium (TP = Technical Publication 10, Auckland Regional Council 2003), used throughout NZ as a standardized rain garden media, as shown in Figure 1. The plant species were selected for the trial based on their resilience to a wide range of soil moistures, specifically including drought, ability to form a dense (weed-suppressing) groundcover size quickly, and ease of maintenance. The species selected had a mature height between 0.5 and 1 m and were thus generally suitable for places adjacent to roads.

Prior to planting, plants were saturated, then removed from their 1.7 liter planter bags. All the potting mix was washed from the roots before each plant was put into the selected medium. Plants were then immediately watered to establish root/media contact. The containers were 38 cm deep and 52 cm in diameter. A coarse, shredded-bark mulch was then placed over the surface to a depth of 7 cm. The plant trial was conducted on a grassed area at the SW360 premises that was fully exposed to the sun. Plant heights were measured weekly, and the individual change in plant biomass at the beginning and end of the 12-week period was measured.

The short-term plant growth trial lasted from 5 December 2015 to 22 February 2016. Over the first 6 weeks, until 18 January 2016, each container was watered with 7 L of tap water and 2–3 L of stormwater (taken from the local carpark catch pit) bi-weekly. This was the equivalent of 25 mm of rainfall a week and ensured plants were not drought stressed (visually). The plants were not watered after 18 January 2016.

Plant height was measured weekly by straightening (without pulling) the stems vertically and measuring the tallest vegetation. For Hebe speciosa, a relatively rigid, bush-like plant, the height and width were measured without manipulation. The container rim was the datum used for all measurements, which were taken by placing a straight rod across the container and extending a tape measure alongside the plant, perpendicular to the rod.

Soil moisture measurements were taken immediately before, and 24-hours after, watering (i.e. four times a week) during the bi-weekly watering period, and twice a week during the period when watering had ceased. Moisture levels were measured both 20 and 30 cm below the container rim, at a radial distance of 10–20 cm from the center of the container. Media moisture measurements were made in a new location each time, with each sequential measurement corresponding with the next hour position of a clock. A Frizzell Soil Moisture Probe (Model SMP3A) was used and operated by pulsing electricity into the medium and using the modified return signals to assess the percentage of water in the soil. The device compensates automatically for temperature variation.

Four long-term plant growth trials were set up around NZ – see Table 1.

The plants were watered occasionally to simulate the base level of water and nutrients that would enter a rapid bio-filtration treatment device in the field. Plants in PT2 and PT3 were watered fortnightly for the first 6 months with 2 L of stormwater taken from a local carpark catch pit sump, and once every 3 months after that. Plants in PT4 (Pegasus, Canterbury) and PT5 (Oratia, Auckland) were watered fortnightly with 2 L of diluted Yates Thrive All-Purpose Plant Food liquid plant fertilizer for the first 6 months, and monthly thereafter.

Plant heights were measured using the same method as for the short-term trial and qualitative assessments of plant health were made on a scale from 1 to 5, with 1 representing plant death, 3 stable growth, and 5 vibrant growth.

Changes in plant height were used to evaluate plant growth. Due to the different starting heights, the percentage difference (change) was used to evaluate and compare growth rates.

The growth rates are presented – Table 2 – over three date ranges:-

• (1)

  The full 12 week period 5 December 2015 to 23 February 2016,

• (2)

  the twice weekly watering period 5 December 2015 to 22 January 2016, and

• (3)

  the period during which the plants were not watered 18 January to 23 February 16.

Changes in biomass were also used to evaluate plant growth by extracting the plant from the growth medium, washing the roots to remove remaining soil or gravel, and drying them gently using paper towels. The plants were completely separated from growing media at both the beginning (Figure 2) and end of the experiment (Figure 2). Their weights were measured on a scale with an accuracy of ±0.1 g (Table 3).

At the end of the trial, plants were dried in a laboratory oven at 100°C for 24 hours. The root and shoots of each plant were separated and weighed to evaluate the root to shoot ratio. A high root to shoot ratio indicates a greater capacity to survive in dry and harsh conditions (Table 4).

One of the key properties, with respect to supporting plant growth, was the field capacity and the plant available water capacity of each medium. Field capacity is the total amount of water that can be held by the medium against gravity. It includes both plant available and hygroscopic water, the latter held below the nominal wilting point and not available to plants.

During the trial, the media received water from both rainfall and manual watering. Using a local rain-gauge at the nearby Rosedale wastewater treatment plant and converting the manual watering volumes into a rainfall depth equivalent, the amount of water the media received was recorded for each day. Media soil moisture levels, measured 20 and 30 cm below the container rim, were assessed alongside the rainfall and manual watering received.

The data showed that all the media stayed near full field capacity during most of the trial. The ability of both media to remain near to field capacity, even after weeks with no rainfall or watering arose from the presence of the mulch layer, which minimized moisture loss (Simcock & Dando 2013), and the low water uptake by plants due to their small size.

Table 5 shows a quantitative (plant height) and qualitative (visual inspection) performance summary for the long-term plant growth trials.

Detailed and regular measurements in the short-term plant growth trial showed vastly different plant species' growth performance in both NZF and TP media. The results show the importance of plant selection in implementing a rapid bio-filtration system and the susceptibility of plants species to transplant shock. Three native species established and grew in both the NZF and TP media, while Carex testacea struggled to establish in either.

Poa cita established best, as a species, in both the NZF and TP, after transplanting. Over the first 6 weeks, plant height increased by 30 and 36% respectively. Once watering stopped, the Poa cita in both media stopped growing and decreased slightly in height. It was noted that Poa cita in NZF developed many more reproductive stems than that grown in TP.

Hebe speciosa also established successfully in both media, although plants in TP grew larger and faster. The plants grew faster still in both media after watering ceased (than when being watered twice weekly). It is thought that this was because the plants had overcome transplant shock in the second half of the trial and not because of the drier media conditions.

The Dianella latissima plants also established and grew successfully in both media. New shoots could be seen on all four plants at the end of the trial. Those in NZF grew slowly and steadily during the bi-weekly watering period, and maintained their height when watering ceased. Despite the height decrease of the plants grown in TP during both 6-week periods, it could be seen clearly that they had established well and were growing vibrantly at the end of the 12-week trial.

Carex testacea plants did not establish in either media. All four plants halved in weight and had very small root systems. All four plants had died or were on the verge of dying by the end of the 12-week trial.

Over the 12 weeks, Hebe speciosa in TP tripled in weight while those grown in NZF doubled. The biomass of Dianella latissima was similar in both media. The biomass of the Poa cita plants grown in TP increased 59% while the plants in NZF stayed the same. Interestingly the root mass of the Poa cita plants showed root growth was the same regardless of medium and the 59% biomass gain was purely in shoot mass.

Plants grown in NZF had higher root to shoot mass ratios than those grown in TP. The lower water and nutrient content of NZF is thought to have resulted in the plants diverting a greater proportion of resources to root development.

Despite the lower root to shoot ratio in plants grown in TP (compared with those grown in NZF) the actual root biomass of plants was similar regardless of media. Hebe speciosa had the largest root mass, made up of fine, intertwined roots in a dense mat predominantly below the plant. The roots of Dianella latissima spread extensively, laterally and vertically, in both media, but especially in TP.

A higher root to shoot ratio usually equates to greater resilience to drought and nutrient stress but under ideal growing conditions plants with more shoots receive more sunlight and hence compete better.

The use of both quantitative (plant height) and qualitative (visual inspection) evaluation of long-term plant performance was informative, but also revealed care was needed in interpreting the results. 29 sets of plant species were grown in the 4 long-term trials. After plant trial periods of at least 11 months, 25 of the 29 species had successfully re-established in NZF; 18 of 29 both increased in height and had health scores of 3 or greater. In addition, 7 of 29 species showed vibrant growth but plant height decreased. On being transplanted from a rich potting mixture to a lean medium, some plants adapted to the lower nutrient environment by growing to a lower height and smaller size.

Out of the 29 sets of plants grown in the long-term plant trials, two grass species were not able to establish in NZF. This was unexpected, as Festuca coxii has been successfully used on living roofs in Auckland over at least 5 years, i.e. similar, highly permeable media and drought-stressed conditions.

The plant growth trial demonstrated successfully that 25 native NZ plant species can establish and grow in a rapid bio-filtration medium (NZF) with a hydraulic conductivity of 2,500 mm/hr. In the short-term trial, three of the four species trialed were able to establish and grow in the medium. In long-term ongoing trials, 27 of the 29 species established successfully in the rapid bio-filtration medium and were continuing to grow after 11 months or longer. The plants in most trials are expected to fully exploit the available root volume by about 24 months; plant growth and health assessment at that stage will be important to guide species selection.

The trials have shown conclusively that a wide variety of plants can become established and grow in NZF, although, as expected, some grew better than others. Research continues at SW360 to improve understanding and identify the key plant species characteristics that enable survival and growth in rapid bio-filtration media.