Research ArticleResearch Articles
Open Access

Field-deployed Extruded Pellets Containing Activated Carbon Improve Recruitment Potential in Native Banksia Woodland Species

Vanessa S. Brown, Todd E. Erickson and Alison L. Ritchie
Ecological Restoration, August 2026, 44 (3) 201-212; DOI: https://doi.org/10.3368/er.44.3.201

Abstract

Invasive species threaten the success of seed-based restoration by competing for resources during early-stage seedling recruitment. Herbicides are commonly used in restoration efforts to control invasive species. Seed enhancement technologies using activated carbon (AC) have been developed to protect native seeds from herbicide damage as a one-pass method to sow seeds directly and apply herbicides to control invasive species simultaneously. Our study assessed the effectiveness of field-deployed extruded pellets (FEPs) with AC compared to a bare seed control treatment in enhancing the emergence and survival of four Western Australian native species (Anigozanthos manglesii D.Don, Acacia pulchella R.Br, Gompholobium tomentosum Labill., and Jacksonia furcellata [Bonpl.] DC.) under Simazine, Fusilade®Forte, and Roundup® herbicide treatments. Field-deployed extruded pellets significantly increased emergence across all species and treatments, with a 13% higher final emergence percentage at 96 days compared to the bare seed control, particularly under herbicide application (p < 0.001). Emergence analysis using generalized linear mixed models showed that FEPs significantly improved emergence in A. pulchella, G. tomentosum, and J. furcellata. Mortality dynamics revealed species-specific differences, with FEPs consistently delaying mortality compared to the bare seed control with and without herbicide, particularly in A. pulchella and G. tomentosum under Simazine and Fusilade®Forte. Results indicate that FEPs can enhance seed survival across multiple herbicide treatments, offering a viable method for improving seedling establishment in restoration projects subjected to herbicide application.

Keywords

Restoration Recap

  • Invasive species threaten restoration by competing for resources during seedling recruitment. Herbicides are used to reduce invasion but can be detrimental to native seeds.

  • Carbon-based seed enhancement technologies can be used in a single-entry herbicide-seeding application approach.

  • Field-deployed extruded pellets (FEPs, a seed enhancement technology) containing four native species were tested against three herbicides (Simazine, Fusilade®Forte, and Roundup®) over 34 weeks under field conditions.

  • FEPs had significantly higher emergence (13%) compared to bare seed under all treatments: 11% greater in the presence of herbicide and 19% greater in the absence of herbicide. FEP seedlings survived longer, reaching 50% mortality up to five days later than those from bare seed under stressed conditions.

  • Our tested FEPs have multiple benefits: enhanced emergence under specific herbicide conditions, delayed mortality, and enhanced survival under environmental and herbicide stress.

  • Investment into scaling up FEP mechanisation should be considered for at-scale restoration and must test more species, environment, and herbicide combinations to see if these benefits are universally beneficial.

Invasive plant species impact ecosystems worldwide and hinder many attempts at ecosystem repair (Weidlich et al. 2020). Banksia woodland restoration efforts in Western Australia are a prime example of how invasive plant species seriously impede ecosystem recovery (Stanbury et al. 2018Ritchie et al. 2021, Rokich 2016). Banksia woodlands are species-rich, low-woodland communities characterized by a canopy dominated by Banksia species and a diverse, variable understory. As a nationally listed Endangered ecological community, they provide critical habitat for fauna within a highly fragmented, Mediterranean-climate landscape. This ecosystem is severely threatened by urban expansion, groundwater decline, and altered fire regimes (Ritchie et al. 2021).

Invasive species such as perennial Ehrharta calycina Sm. (veldt grass) negatively impact the Banksia woodland community through altering soil nutrients (Fisher et al. 2006) and shifting vegetation structure and composition (Fisher et al. 2009, Fowler et al. 2023) while also reducing the growth and establishment of native species (Stanbury et al. 2018). Invasive species are commonly removed in restoration programs during site preparation, prior to seeding. If this is done through the application of herbicides, there may be negative impacts to the topsoil seedbank and the seeded or planted native species (Rokich et al. 2009Munro et al. 2023).

Native plant restoration can be accomplished on a large scale with seeding. Seed is desired due to its scalability and the limitations associated with topsoil shortages and costly nursery stock (Berto and Brown 2023). For example, topsoil is not always available due to the site history, or in the volume required, and some key species are absent from the topsoil seed bank (Fowler et al. 2015, Rokich et al. 2000). Additionally, planted nursery stock is more labour intensive and costly, and plants can suffer high mortality rates, especially after the second summer (Rokich 2016). Therefore, seed-based restoration can be used as a complementary solution with topsoil and nursery tubestock, and in combination with the management of invasive species, improve the success of native plant reinstatement. Management of the Banksia woodland restoration often controls invasive species with post-emergent herbicide application using glyphosate (Roundup) and/or fluazifop-p (Fusilade) prior to seeding and/or planting (Rokich et al. 2009, Rokich 2016). Previous research has largely examined the use of activated carbon (AC) as a seed enhancement to enable simultaneous seeding and application of pre-emergent herbicides (Madsen et al. 2014, Madsen et al. 2016, Davies 2018, Brown et al. 2019, Svejcar et al. 2024a), with only a few studies investigating the technology with herbicides labelled solely for post-emergent application (Munro et al. 2023, Brown et al. 2023a).

Activated carbon (AC) has been used as a protective material from pre-emergent herbicides for native seeds due to its high porosity and surface area (Bansal and Goyal 2005, Marsh and Rodriguez-Reinonso 2006). The material can be incorporated into herbicide protection pods or pellets (Madsen et al. 2016, Brown et al. 2019, Baughman et al. 2021, Svejcar et al. 2024a, Svejcar et al. 2024b). Studies have shown similar benefits when using herbicide protection pods with post-emergent herbicides. Furthermore, the pods can provide an improved growth microsite for seedlings in the absence of herbicide when compared to non-pelleted seeds (Munro et al. 2023, Brown et al. 2023a). Previous trials have demonstrated that herbicide protection pods can mitigate the impact of herbicide damage. However, these herbicide protection pods can also inhibit seedling emergence due to their formulation and production process. The pod formulation typically contains a number of ingredients with fine particles such as clay and AC, and when produced through cylindrical extrusion using an industrial pasta machine, can result in a compact, hard pod or pellet (Clenet et al. 2019, Brown et al. 2023a, Munro et al. 2023, Brown et al. 2023b). In contrast, our study used a modified extrusion method based on Stock et al. (2020) in which we extruded a slurry in-situ to create field-deployed extruded pellet (FEPs). These pellets are wetter, non-compacted, and still offer the potential for herbicide protection without the requirement for laboratory production and storage prior to deployment.

The purpose of our study was to determine the extent of protection offered by field-deployed extruded pellets to Banksia woodland species (Anigozanthos manglesii D.Don, Acacia pulchella R.Br., Gompholobium tomentosum Labill. and Jacksonia furcellata [Bonpl.] DC.) from three different herbicides (Simazine, Fusilade Forte and Roundup). We hypothesized that following herbicide application, emergence and survival would be greater from FEPs compared to the control (bare seed sown in situ) due to the ability of AC to adsorb herbicide and minimise non-target deleterious effects to native seed. We also hypothesized that in the absence of herbicide, FEPs would provide additional benefits to seedling growing conditions (e.g., carbon availability during early establishment, reduced fungal attack and overall favourable growth microsite), thereby improving emergence and survival compared to bare seed used as a control.

The field site (31°23’29.37” S, 115°56’7.62” E) located near the township of Gingin, Western Australia, is part of Heidelberg Materials Australia’s Lennard Brook tenement, 63 km northeast of capital Perth CBD, Western Australia. This site was previously mined for sand and aggregates and subsequent rehabilitation efforts were unsuccessful.
Figure 1.

The field site (31°23’29.37” S, 115°56’7.62” E) located near the township of Gingin, Western Australia, is part of Heidelberg Materials Australia’s Lennard Brook tenement, 63 km northeast of capital Perth CBD, Western Australia. This site was previously mined for sand and aggregates and subsequent rehabilitation efforts were unsuccessful.

Methods

Study Location

The study location is the Lennards Brook Quarry owned by Heidelberg Materials Australia in Gingin, Western Australia. The site had previously undergone two rehabilitation efforts (approximately 18 and seven years ago), however most species failed to establish. The site (31°23’29.37” S, 115°56’7.62” E) (Figure 1) experiences a Mediterranean climate with dry, hot summers and cold, wet winters. The site was prepared for this study by scalping (i.e., mechanical removal of the top 100 mm of soil) followed by shallow ripping. This process is current practice to eliminate established weeds and reduce the weedy soil seedbank typical of unmanaged lands prior to seeding.

Species Selection and Preparation

We selected four Banksia woodland species commonly used in restoration seeding programs (Table 1) that had been previously studied with herbicide protection seed enhancement technologies (Brown et al. 2023a, Munro et al. 2023). Seeds were purchased from a commercial supplier (Tranen Pty. Ltd., Bayswater, WA, Australia).

Seed dormancy was broken based on the specific type of dormancy for each species (Table 1). We completed a germination trial to determine the germination capacity of each species, which included four replicates of 25 seeds plated on Petri dishes filled with agar (0.7% w/v). Germination was defined as radicle emergence to greater than one-third of the seed coat length. Germination was monitored once a week and final germination was recorded after four weeks (Table 2). Seeds were stored in a controlled environment (15°C and 15% relative humidity) until required for experimental use.

Experimental Design and Field Trial Installation

The experimental design had three blocks, and within each block we included paired treatments. These paired treatments included three herbicides (Simazine, Fusilade Forte and Roundup) each paired with a no herbicide application control. Within each herbicide treatment and its paired control, we tested the four species with two seed treatments: a bare seed control and field-deployed extruded pellets (FEPs) (see “Seed Enhancement Technology Production” section, below, for methods). Each species was seeded in one plot (60 cm × 40 cm) with paired rows (bare seed control and FEPs), and each plot was replicated four times at random (total of 12 plots per species per herbicide or paired control) (Figure 2). A different number of seeds were sown for each species to account for the final germination capacity (%) (Table 2). All bare seed controls and FEPs were sown on the soil surface. We used a paired design for seeding (i.e., five groupings of bare seed controls or five field-deployed extruded pellets in each row) to ensure similar seed density between seed treatments.

View this table:
Table 1.

Four Banksia woodland species used in this trial and information on their dormancy type (DT), either morphophysiological (MPD) or physical (PY), and method of dormancy alleviation.

View this table:
Table 2.

The number of seeds planted per row for the bare seeds and the field-deployed extruded pellets (FEP) adjusted for final germination capacity (FG) and the calculation of seeds per FEP to the nearest seed.

Schematic layout of the experimental landscape showing the block design, herbicide sub-plot treatments, seed treatment and species used. The field trial was arranged in a randomized block design with replicated blocks distributed across the site to account for spatial variability. Within each block, randomized sub-plots represented all herbicide treatments (e.g., paired control with Simazine, Fusilade and Glyphosate), each containing four replicates of each species. Two seed treatments (bare seed [control] and field-deployed extruded pellets [FEP]) were applied within each species subplot. The diagram illustrates the special arrangement of treatments used to assess the effects of herbicide exposure and seed enhancement on seedling emergence and survival under restoration conditions.
Figure 2.

Schematic layout of the experimental landscape showing the block design, herbicide sub-plot treatments, seed treatment and species used. The field trial was arranged in a randomized block design with replicated blocks distributed across the site to account for spatial variability. Within each block, randomized sub-plots represented all herbicide treatments (e.g., paired control with Simazine, Fusilade and Glyphosate), each containing four replicates of each species. Two seed treatments (bare seed [control] and field-deployed extruded pellets [FEP]) were applied within each species subplot. The diagram illustrates the special arrangement of treatments used to assess the effects of herbicide exposure and seed enhancement on seedling emergence and survival under restoration conditions.

Seed Enhancement Technology Production

Field-deployed extruded pellets were produced on site by incorporating the dry ingredients (AC, sand and vermiculite) with the wet ingredients (water with a fungicide) using a multi-purpose mixer (DMM-050, 1050W, Ozito Industries, Australia) to make a slurry (Table 4). Once all the ingredients were thoroughly mixed, the seeds were gently incorporated by hand until evenly distributed throughout the mixture. The FEPs were delivered via hand extrusion directly into the plots using a 60 mL Terumo syringe (Terumo Corporation, Philippines) with the end piece removed to create a 30 mm opening (as per Stock et al. 2020). Each FEP was approximately 15–20 mL in volume, producing a 6-cm diameter, 1-cm thick pellet. A subsample (n = 18) of FEPs were taken to the laboratory to calculate the average number of seeds per FEP (Table 2).

Herbicide Application

Three herbicides were applied at the recommended rates after seeding (Table 3). Simazine plots were lightly watered first as Simazine application requires moist soils for the herbicide to work effectively (as per label directions). Herbicides were applied with a calibrated 10 L battery powered knapsack (Solo 414 10 Litre Knapsack Sprayer, Newport News, VA, USA). During herbicide application, maximum temperature was 21°C, relative humidity was 35% and average wind speed was 4 km hr-1, with no spray drift observed.

Soil Sampling and Microclimate Monitoring

Soils were sampled from four replicates within each block. Each sample was taken from a 25 cm2 area, from the top 5 cm of the soil profile. Soils were air dried and then sieved to 2 mm before being tested for ammonium nitrogen (NH4-N), nitrate nitrogen (NO3-N), phosphorous (P) (Col-well), potassium (K) (Colwell), sulfur (S) (KCI 40), organic carbon (OC), electrical conductivity (EC), pH (water and CaC12), and phosphorus buffering index (PBI) (CSBP Soil and Plant Laboratory, Kwinana, WA, Australia) (Table S1). Microclimate data were logged with Temperature-Moisture-Sensors (TMS, TMS-4 logger, TOMST Prague, Czech Republic [Wild et al. 2019]) that measure air (12 cm above surface), surface, and soil temperature (–6 cm depth) as well as soil moisture every 15 min.

Data Collection and Statistical Analysis

Seedling counts (emergence) were collected weekly for eight weeks from day 23. We recorded survival data at three time points: one month following peak emergence, two months later on day 153, and finally on day 236. Peak seedling emergence was calculated as the number of seedlings counted in each plot at day 96 converted to a percentage. Survival was calculated as the number of seedlings counted in each plot at 153 days converted to a percentage of the cumulative seedlings emerged. On day 236 only a few individuals were alive.

All statistical analyses were conducted in R statistical software version 4.3.0 (R Core Team 2023). Mean emergence (%) over time was plotted for each seed and herbicide treatment for each species. Welch Two Sample t-tests were used to compare emergence between FEP and bare seed treatments with and without herbicide treatment across species. Cumulative final seedling emergence (%) for each species was analysed by fitting zero-inflated beta-binomial general linear mixed models (GLMM), where zero scores were common for emergence (glmmTMB package [Brooks et al. 2017]), to assess the effects of seed treatments (FEP or bare seed) and herbicide treatments (Simazine, Fusilade Forte, Roundup and paired controls). Block and plot were included as random effects and were removed if there was no significant improvement to the model fit (in which case a generalized linear model [GLM] was used in the absence of random effects).

View this table:
Table 3.

Herbicides used in the trial.

View this table:
Table 4.

Ingredients used to make a slurry which was extruded directly into the plot with a 60 mL Terumo syringe (Terumo Corporation, Philippines) to make a total of 380 field-deployed extruded pellets (each approximately 15–20 mL).

Mean emergence (%) of Anigozanthos manglesii (A), Acacia pulchella (B), Gompholobium tomentosum (C) and Jacksonia furcellata (D) over time (peak emergence period) for control and field-deployed extruded pellets (FEP) seed. Herbicide treatments (Simazine, Fusilade®Forte, Roundup®) on the left with paired control on the right. Y-axis scales differ, and error bars represent standard error (± SE).
Figure 3.

Mean emergence (%) of Anigozanthos manglesii (A), Acacia pulchella (B), Gompholobium tomentosum (C) and Jacksonia furcellata (D) over time (peak emergence period) for control and field-deployed extruded pellets (FEP) seed. Herbicide treatments (Simazine, Fusilade®Forte, Roundup®) on the left with paired control on the right. Y-axis scales differ, and error bars represent standard error (± SE).

All observed emergence and survival data were analysed using time-to-event curves by using the drcte (Onofri et al. 2022) and drc (Ritz et al. 2015) packages to compare emergence across herbicide treatments over time. Time-to-event curves were fitted to the emergence data over time using a three-parameter log-logistic model with the lower limit as 0 and two-parameter log-logistic models for survival data (Brown and Mayer 1988, Ritz 2010). Three-parameter Models provide estimates for parameters, d, e, and b, for three-parameter and e and b for two-parameter models for which d is maximum emergence, e is time to 50% emergence (T50e) or time to 50% mortality (T50m) and b is emergence rate or mortality rate (Ritz et al. 2015).

Survival (%) at day 153 for each species was analysed using GLMMs to assess the effects of seed treatment and herbicide treatment. For the GLMMs, assumptions of normality were assessed through visualising the fitted and residual values (Zuur et al. 2009). Model fit was verified using DHARMa residual diagnostics (Hartig 2022). ANOVAs of all models were performed to detect significance between seed treatments and herbicide treatment. Post hoc pairwise comparisons were completed with the emmeans package (Russel 2021) and Tukey-adjusted p-values were applied to control for multiple pairwise comparisons among herbicide treatments.

Results

Emergence

Mean emergence over time was higher in field-deployed extruded pellets (FEP) across all species and treatments, excluding low emerging A. manglesii under the Simazine treatment (Figure 3). Final cumulative mean emergence (at day 96) was on average 13% higher from FEPs in comparison to the control bare seed under all treatments (t = –7.10, df = 527.15, p < 0.001), higher in FEPs with herbicide (FEP = 28%; bare seed = 17%; t = –4.62, df = 273.77, p < 0.01) and without herbicide (FEP = 33%; bare seed = 19%; t = –5.43, df = 253.97, p < 0.01; Figure 3; Supplementary Material Table S2) for all species.

Climate profile from June 2021 to January 2022 measured by the TMS data logger device at three sensors measuring air (12 cm above surface), surface and soil temperature (6 cm belowground), and one sensor measuring soil moisture. Soil moisture was calibrated for sandy soils and converted to volumetric soil moisture (%).
Figure 4.

Climate profile from June 2021 to January 2022 measured by the TMS data logger device at three sensors measuring air (12 cm above surface), surface and soil temperature (6 cm belowground), and one sensor measuring soil moisture. Soil moisture was calibrated for sandy soils and converted to volumetric soil moisture (%).

Emergence analysis using GLMMs showed significant differences across the four species when comparing seed treatments and herbicide effects (Supplementary Material Tables S3 and S4). Notably FEPs significantly improved emergence in A. pulchella, G. tomentosum and J. furcel-lata (p < 0.001, Supplementary Material Table S3, S4 and S5). The interactions between seed treatment and herbicides further illustrate the complexity of these effects. For J. furcellata, significant interactions with Fusilade Forte, Roundup, and the control non-herbicide treatments suggest that the FEPs benefits are subject to specific herbicide applications (Supplementary Material Tables S3 and S4). Interestingly, while Fusilade Forte alone significantly improved emergence in A. manglesii and J. furcellata, its interaction with FEP did not yield additional significant improvements (Supplementary Material Tables S3 and S4). In contrast, G. tomentosum showed significant negative interactions between FEP and Fusilade Forte, indicating that FEPs may not have provided adequate protection with these herbicides (Supplementary Material Table S3). The type III ANOVA results supported these findings, showing significant seed treatment effects in all species except A. manglesii (Supplementary Material Tables S4). The inter-action between seed treatment and herbicide is significant across all species, highlighting the effect of combined treatments on emergence (Supplementary Material Table S4).

Modelling seedling emergence curves found each species responded differently to seed and herbicide treatments. Seedling emergence was not inhibited or delayed in FEPs in A. manglesii, A. pulchella and J. furcellata, with consistent positive responses to FEP in terms of both maximum emergence (d), less days to reach 50% emergence (T50e) and faster emergence rates without herbicide (Figure S1; Tables S6 and S7). Herbicide treatments interacted with seed treatments, influencing emergence. For some species, certain herbicides such as Simazine and Roundup in combination with FEP showed significant improvements in emergence factors (Supplementary Material Figure S1, Tables S6 and S7). In A. manglesii, FEPs generally did not significantly increase maximum emergence compared to bare seed; the highest emergence was observed with FEP under Fusilade Forte and the corresponding control (No-F) as well as No-S, however the difference was small (Supplementary Material Table S6). T50e values were consistent across treatments, with FEPs showing slightly lower T50e (i.e., faster emergence), specifically under Roundup application. Acacia pulchella FEPs had significantly higher maximum emergence compared to bare seed across all herbicide treatments, particularly under Simazine (d; FEP = 28%; bare seed = 11%) and its paired control treatment (No-S d; FEP = 40%; bare seed = 13%). In G. tomentosum FEPs did not improve overall emergence, as they did not increase maximum emergence, T50e values were similar, and emergence rates were slower in comparison to bare seed. Jacksonia furcellata, however, showed significant increases in maximum emergence in FEPs compared to bare seed across treatments, in particular under Simazine and Fusilade Forte (Supplementary Material Tables S6 and S7). T50e values for J. furcellata were lower, particularly under Simazine indicating faster emergence, and the emergence rate was consistently faster in FEPs compared to the control bare seed (Supplementary Material Tables S6 and S7).

Percentage survival (at day 153) by seed treatment (control bare seed and field-deployed extruded pellets [FEP]) for each species (Anigozanthos manglesii [A], Acacia pulchella [B], Gompholobium tomentosum [C)], Jacksonia furcellata [D]) and herbicide treatment (Simazine, Fusilade®Forte, Roundup® and the paired control [No -]). Panels include all margins that pool across species and experiments for control bare seed FEP with emergence adjusted for germinability. Each box shows the distribution of cumulative survival, with the centre line as the median. Wilcoxon rank-sum tests compare control bare seed and FEP within each panel. Significant differences shown with asterisks (*, p = 0.05; **, p = 0.001; ***, p < 0.001). Note y-axis scales across panels differ.
Figure 5.

Percentage survival (at day 153) by seed treatment (control bare seed and field-deployed extruded pellets [FEP]) for each species (Anigozanthos manglesii [A], Acacia pulchella [B], Gompholobium tomentosum [C)], Jacksonia furcellata [D]) and herbicide treatment (Simazine, Fusilade®Forte, Roundup® and the paired control [No -]). Panels include all margins that pool across species and experiments for control bare seed FEP with emergence adjusted for germinability. Each box shows the distribution of cumulative survival, with the centre line as the median. Wilcoxon rank-sum tests compare control bare seed and FEP within each panel. Significant differences shown with asterisks (*, p = 0.05; **, p = 0.001; ***, p < 0.001). Note y-axis scales across panels differ.

Survival

There were no significant differences in seedling survival between seed treatments at the end of the experiment (day 236). With extreme temperatures and low to nil moisture recorded from late November (see Figure 4) few seedlings remained. At 153 days, FEPs had on average 2.7% greater survival in comparison to bare seed under all treatments (t = –2.53, df = 558.96, p = 0.012) and 3.8% greater survival without herbicide treatment (t = –2.40, df = 280.02, p = 0.017) for all species (Figure 5 and Supplementary Material Table S8). For all species combined there was significantly greater survival in FEPs under Simazine (by 5 %; t = –2.77; df = 74.08, p = 0.007) and Roundup (by 3%; t = –2.25; df = 93.14, p = 0.027) (Figure 5 and Supplementary Material Table S8).

When considering the survival of individual species, FEPs under all treatments significantly improved survival of A. pulchella (by 8%, t = –4.59; df = 117.63; p < 0.001), but the impact was less pronounced or not significant in others. Herbicide treatments showed mixed effects, significantly improving survival of A. manglesii in FEPs with Roundup but not in other species (Supplementary Material Table S8). Interaction effects between seed treatments and herbicide were more complex and species-specific. Notably, G. tomentosum benefited significantly from the interaction between FEP and Fusilade Forte, but A. manglesii and J. furcellata did not exhibit similar benefits. For A. mangle-sii, survival did not differ between seed treatments but was significantly affected by herbicide treatments (p = 0.001). In contrast, neither the seed treatment nor the herbicide treatment had significant effects on survival of J. furcellata. Additionally, a significant interaction effect between seed treatment and herbicide was noted for A. pulchella (p = 0.001) and G. tomentosum (p < 0.001). These findings indicate that the effects of treatment combinations on seedling survival were not simply additive and varied considerably between species (Supplementary Material Tables S3 and S4).

FEPs largely extended the time to 50% mortality (T50m) compared to bare seed across most species and herbicide treatments (Supplementary Material Table S9 and Figure S2). The mortality rate (b) varied with seed treatment, often showing that FEP seedlings tended to have slightly lower mortality rate compared to seedlings planted as bare seed; however, this was not consistent across all species and herbicide treatments. Anigozanthos manglesii FEP seedlings had increased T50m (i.e., longer survival time) under Simazine (T50m; FEP = 101 ± 7 days, bare seed = 83 ± 5 d) and control no-herbicide treatments (T50m; FEP range = 105–131 d, bare seed range = 89–116 d), indicating delayed mortality; however, the mortality rates were inconsistent (Table S9), likely due to poor emergence numbers. The herbicide protection of FEPs was clearer in A. pulchella, with longer T50m and lower mortality rates under herbicide treatments (Supplementary Material Table S9 and S10), suggesting longer survival time and less death in FEPs in comparison to bare seeds. Gompholobium tomentosum and J. furcellata were similar to the other species, however G. tomentosum showed a slightly elevated mortality rate under Simazine (b; FEP = 3.79 ± 0.44, bare seed = 3.08 ± 0.34) (Supplementary Material Table S9).

Herbicide-specific responses varied between species. FEPs generally increased T50m across species exposed to Simazine suggesting that the effect of post-emergent herbicide on seedling emergence and survival was somewhat mitigated by the FEP treatment. Fusilade Forte showed varied effects on T50m, with FEPs often having a higher T50m and lower or similar mortality rate to bare seeds. The effect of Roundup showed FEPs had significantly higher T50m than bare seeds in all species except A. manglesii; however, the mortality rate was generally lower for FEPs under Roundup . Under control conditions, without herbicide, FEPs often resulted in higher T50m across species, indicating a protective effect of FEPs to environmental conditions in the absence of herbicides.

Discussion

This study demonstrated the potential benefits of seed enhancement technologies that contain activated carbon (AC) for herbicide protection of select Banksia woodland species during restoration seeding. Seedling emergence was positively affected by seed treatment, with emergence significantly higher in field-deployed extruded pellets (FEPs) in comparison to bare seed. However, seedling survival was negatively impacted by the herbicide treatment, and by the interaction between herbicide and seed treatment. Along with herbicide protection, this study has shown that modifying the AC pellet-making method to field-deployed extrusion can be used as an alternative to laboratory made pellets extruded through a caste die to help alleviate hardness that may impede seedling emergence (see Davies et al. 2017, Clenet et al. 2019, Brown et al. 2019, Baughman et al. 2021, Brown et al. 2023a, Munro et al. 2023, Davies et al. 2024). The hardness of laboratory made pellets is a considerable limitation, especially under field conditions, (i.e., in a water-limited environment) where a pre-made pellet may not break down efficiently, reducing pellet-seed-soil contact (Brown et al. 2023a). Improvements in seedling emergence from FEPs support the ongoing investment in this technology and the development needed to apply it at scale for the enhancement of restoration outcomes.

Seed enhancement technologies such as FEPs or herbicide protection pods (HPPs) were originally designed for protecting Poaceae species against pre-emergent herbicides (Madsen et al. 2016). Our study has investigated the modification of HPPs (as FEPs) and their application on two additional plant families (Haemodoraceae and Fabaceae), as well as expanding our understanding of the capacity of HPPs to protect seeds from both pre- and post-emergent labelled herbicides. The seed emergence curves of each species responded differently to seed treatments and herbicide treatments. Benefits of FEP were most apparent in A. manglesii, A. pulchella and J. furcellata, with consistent positive responses in terms of both maximum emergence and faster emergence rates. Herbicide treatments interacted with the seed treatment FEP, improving the emergence parameters. For some species, FEPs with the pre-emergent herbicide Simazine unsurprisingly led to significantly improved emergence, for which HPPs were originally designed. In G. tomentosum there were mixed results, indicating species-specific responses to the capacity of herbicide protection and/or the benefit of FEPs. Previous studies of Gompholobium have showed negligible increased emergence with enhancement or herbicide protection (Munro et al. 2023), whereas improvements have been seen with removal of the fungicide from the formulation (data unpublished), potentially indicating the importance of maintaining rhizobium symbionts (Chappill et al. 2008). As the review by Davies et al. (2024) indicated, many species tested have had no evidence of herbicide protection and a broader investigation of species that represents a wider range of seed sizes, morphology and family groups is warranted.

The effects of FEPs on seedling survival were less consistent than those observed for emergence. Generally, we observed delayed and lower mortality in FEP seedlings. But the magnitude of these effects varied significantly between species and herbicide treatments. For example, survival time to 50% mortality (T50m) in FEP seedlings of all species was significantly longer, apart from A. manglesii under Fusilade Forte and Roundup herbicide treatments, and J. furcellata under Fusilade Forte herbicide treatments, which were not significantly different from bare seed. However, the delay in mortality (i.e., a higher T50m) of FEPs compared to bare seeds across most species, suggests that FEPs containing AC provide a protective or mitigating effect from the negative impacts of herbicide application and environmental stress. Given the extreme heatwave at the end of the study year (Perth and surrounding areas in late December had five days of maximum temperatures persisting over 40°C and overnight temperatures remaining above 20°C [BOM 2022]), and the overall acidity of the site (see soil analysis results in Supplementary Material Table S1), our results warrant further investigation in multiple environments to confirm if these survival trends would occur under more favourable or average weather conditions. Furthermore, monitoring seedlings through the second summer would avoid drawing assumptions for HPP use from a single environment or trial.

Herbicide protection pods were designed for a single-entry restoration approach, allowing for simultaneous seeding and herbicide application by protecting the native seeds in AC (Madsen et al. 2016). The results from our FEPs highlight the importance for targeted approaches when applying treatments due to species-specific responses to herbicide. Perhaps selecting one herbicide treatment for a biodiverse seeding mix will not be appropriate and a two-herbicide approach may be beneficial. For example, a recent study of HPPs demonstrated the application of two herbicides applied in tandem to reflect the novel herbicide practices used by land managers in the sagebrush steppe of the western United States (Svejcar et al. 2024a). As invasive weed species become more dominant in Western Australian systems (O’Donnell et al. 2012, Funk et al. 2016), changes in restoration practices to meet these challenges are required along with research to understand which native seeds are susceptible to selected herbicides and would benefit from HPPs.

Our study showed that FEPs are beneficial for improving the restoration of certain species. FEPs showed multiple benefits: enhanced emergence under specific herbicide conditions, delayed mortality, and enhanced survival under environmental and herbicide stress, despite several nuanced interactions between seed treatments, herbicides, and species. Proof of concept results from this study favour the use of FEPs when controlling invasive species in restoration sites and highlight their importance as a consideration for large scale restoration using seed enhancement technologies. Research is currently being undertaken to mechanise the deployment of FEPs at scale for restoration field environments (Erickson et al. 2023). This apparatus extrudes the slurry mixture comprising seeds, growing medium and water through an industrial peristaltic pump. When dropped intermittently from a certain height the slurry forms a “pat” of predefined thickness (thereby controlling the seeding depth) on the soil surface. The growing medium may be tailored to include beneficial compounds for germination, inorganic and/or organic compounds for soil rehabilitation on disturbed lands (i.e., post-mining or post-agriculture) (Erickson et al. 2023), and/or herbicide or pathogen protection (Harshani et al. 2024) for seeds.

Acknowledgments

A.L.R., T.E.E. and V.S.B. acknowledge funding support from the Australian Research Council’s Linkage Program (LP170100075), Department of Industry, Innovation and Science’s Global Inno-vation Linkages program (GIL 53873) and The Mary and Elsie Stevens Scholarship awarded to V.S.B. by the Graduate Women WA. We thank Heidelberg Materials Australia, specifically the Gingin Operations team for supporting and facilitating the experimental trial. V.S.B. acknowledges D. St Jack, L. Merchant, Y. Kenna, K. Greenacre, P. Bathgate, C. Brown, F.M. Pourmo-barakeh, M. Masarei, J. Leighton, and B. Berto who helped to install and/or monitor the trial. ALR acknowledges Professor A. Onofri statistical assistance.

Illustration of Banksia ericifolia (heath-leaved banksia or lantern banksia)

Banksia ericifolia (heath-leaved banksia or lantern banksia), is a woody shrub of the family Proteaceae native to Aus-tralia. L.H. Bailey. 1917 Standard Cyclopedia of Horticulture (New York, New York, NY: The MacMillan Company. The Florida Center for Instructional Technology, fcit.usf.edu.

This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (https://creativecommons.org/licenses/by-nc-nd/4.0/) permitting copying and distributing the material in any medium or format in unadapted form only, for noncommercial purposes only, provided the original work is properly cited

References

  1. ↵
    1. Bansal, R.C.
    and M. Goyal. 2005. Activated carbon adsorption and environment: Adsorptive removal of organics from water. Pages 297–372 in Activated Carbon Adsorption. Boca Raton, FL: CRC Press, Taylor & Francis Group.
  2. ↵
    1. Baughman, O.W.
    , J. Griffen, J. Kerby, K.W. Davies, D. Clenet and C. Boyd. 2021. Herbicide protection pod technology for native plant restoration: One size may not fit all. Restoration Ecology 29(3)e13323.
    OpenUrl
  3. ↵
    1. Berto, B.
    and V.S. Brown. 2023. Ten years to restore the planet: A seedy situation. Restoration Ecology 31(3)e13755.
    OpenUrl
  4. ↵
    BOM. 2022. Greater Perth in December 2021: Very warm with an extreme heatwave in late December. Bureau of Meteorology, Last Modified Tuesday, 4 January 2022. https://www.bom.gov.au/climate/current/month/wa/archive/202112.perth.shtml.
  5. ↵
    1. Brooks, M.E.
    , K. Kristensen, K.J. Van Benthem, A. Magnusson, C.W. Berg, A. Nielsen et al. 2017. glmmTMB balances speed and flexibility among packages for zero-inflated generalized linear mixed modeling. The R Journal 9(2):378–400.
    OpenUrl
  6. ↵
    1. Brown, R.F.
    and D.G. Mayer. 1988. Representing cumulative germination 2: The use of the Weibull function and other empirically derived curves. Annals of Botany 61(2):127–138.
    OpenUrlCrossRef
  7. ↵
    1. Brown, V.S.
    , T.E. Erickson, R.J. Hobbs, S. Mastrantonis and A.L. Ritchie. 2023a. Carbon-based pelleting, soil ripping and herbicide application can be used to overcome plant recruitment barriers in Grey Stinkwood (Jacksonia furcellata). Ecological Management and Restoration 24(2–3)119–127.
  8. ↵
    1. Brown, V.S.
    , A.L. Ritchie, J.C. Stevens, T.D. Hanks, R.J. Hobbs and T.E. Erickson. 2023b. Seed positioning in extruded pellets: Does it matter? Restoration Ecology 31(1)e13784.
    OpenUrl
  9. ↵
    1. Brown, V.S.
    , A.L. Ritchie, J.C. Stevens, R.J. Harris, M.D. Madsen and T.E. Erickson. 2019. Protecting direct seeded grasses from herbicide application: Can new extruded pellet formulations be used in restoring natural plant communities? Restoration Ecology 27(3):488–494.
    OpenUrl
  10. ↵
    1. Chappill, J.A.
    , C.F. Wilkins and M.D. Crisp. 2008. Taxonomic revision of Gompholobium (Leguminosae: Mirbelieae). Australian Systematic Botany 21(2):67–151.
    OpenUrl
  11. ↵
    1. Clenet, D.R.
    , K.W. Davies, D.D. Johnson and J.D. Kerby. 2019. Native seeds incorporated into activated carbon pods applied concurrently with indaziflam: A new strategy for restoring annual-invaded communities? Restoration Ecology 27(4):738–744.
    OpenUrl
  12. ↵
    1. Davies, K.W.
    2018. Incorporating seeds in activated carbon pellets limits herbicide effects to seeded bunchgrasses when controlling exotic annuals. Rangeland Ecology and Management 71(3):323–326.
    OpenUrl
  13. ↵
    1. Davies, K.W.
    , M.D. Madsen and A. Hulet. 2017. Using activated carbon to limit herbicide effects to seeded bunchgrass when revegetating annual grass-invaded rangelands. Rangeland Ecology and Management 70(5):604–608.
    OpenUrl
  14. ↵
    1. Davies, K.W.
    , D.R. Clenet, M.D. Madsen, V.S. Brown, A.L. Ritchie and L.N. Svejcar. 2024. Activated carbon seed technologies: Innovative solutions to assist in the restoration and revegetation of invaded drylands. Journal of Environmental Management 371:123281.
  15. ↵
    1. Erickson, T.E.
    , A.L. Guzzomi, M.I. Masarei and D.J. Merritt. 2023. Method and apparatus for use in establishing or re-establishing plant life in a land area. United States Patent Application Publication. Kings Park, Australia: Botanic Gardens and Parks Authority. https://patentimages.storage.googleapis.com/d2/9f/fd/96a022765a66e0/US20230172097A1.pdf.
  16. ↵
    1. Fisher, J.L.
    , W.A. Loneragan, K. Dixon, J. Delaney and E.J. Veneklaas. 2009. Altered vegetation structure and composition linked to fire frequency and plant invasion in a biodiverse woodland. Biological Conservation 142(10):2270–2281.
    OpenUrl
  17. ↵
    1. Fisher, J.L.
    , E.J. Veneklaas, H. Lambers and W.A. Loneragan. 2006. Enhanced soil and leaf nutrient status of a Western Australian Banksia woodland community invaded by Ehrharta calycina and Pelargonium capitatum. Plant and Soil 284:253–264.
    OpenUrlCrossRef
  18. ↵
    1. Fowler, W.M.
    , J.B. Fontaine, N.J. Enright and W.P. Veber. 2015. Evaluating restoration potential of transferred topsoil. Applied Vegetation Science 18(3):379–390.
    OpenUrl
  19. ↵
    1. Fowler, W.M.
    , R.J. Standish, N.J. Enright and J.B. Fontaine. 2023. Extinction debt varies in two threatened Mediterranean-type woodland communities undergoing rapid urbanisation. Australian Journal of Botany 71(8):421–433.
    OpenUrlCrossRef
  20. ↵
    1. Funk, J.L
    , R.J. Standish, W.D. Stock and F. Valladares. 2016. Plant functional traits of dominant native and invasive species in Mediterranean-climate ecosystems. Ecology 97(1):75–83.
    OpenUrlPubMed
  21. ↵
    1. Harshani, H.S.
    , T. Burgess, G. Hardy and T.E. Erickson. 2024. Extruded seed pelleting offers an alternative direct seeding option to rehabilitate forested ecosystems impacted by a soil-borne plant pathogen. Ecological Management and Restoration 25(2):120–128.
    OpenUrl
  22. ↵
    1. Hartig, F.
    2022. DHARMa: residual diagnostics for hierarchical (multi-level/mixed) regression models (R package version 0.4. 6). Vienna, Austria: R Foundation for Statistical Computing. https://CRAN.Rproject.org/package=DHARMa.
  23. ↵
    1. Madsen, M.D.
    , K.W. Davies, C.S. Boyd, J.D. Kerby and T.J. Svejcar. 2016. Emerging seed enhancement technologies for overcoming barriers to restoration. Restoration Ecology 24:S77–S84.
    OpenUrl
  24. ↵
    1. Madsen, M.D.
    , K.W. Davies, D.L. Mummey and T.J. Svejcar. 2014. Improving restoration of exotic annual grass-invaded range-lands through activated carbon seed enhancement technologies. Rangeland Ecology and Management 67(1):61–67.
    OpenUrl
  25. ↵
    1. Marsh, H.
    and F. Rodriguez-Reinonso. 2006. Activated Carbon. Amsterdam, The Netherlands: Elsevier Science and Technology Books.
  26. ↵
    1. Munro, T.P.
    , A.L. Ritchie, T.E. Erickson, D.G. Nimmo and J.N. Price. 2023. Activated carbon seed technologies provide some protection to seedlings from the effects of post-emergent herbicides. Restoration Ecology 31(4)e13875.
    OpenUrl
  27. ↵
    1. O’Donnell, J.
    , R.V. Gallagher, P.D. Wilson, P.O. Downey, L. Hughes and M.R. Leishman. 2012. Invasion hotspots for non-native plants in Australia under current and future climates. Global Change Biology 18(2):617–629.
    OpenUrlCrossRef
  28. ↵
    1. Onofri, A.
    , M.B. Mesgaran and C. Ritz. 2022. A unified framework for the analysis of germination, emergence, and other time-to-event data in weed science. Weed Science 70(3):259–271.
    OpenUrlCrossRef
  29. ↵
    R Core Team. 2023. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing.
  30. ↵
    1. Ritchie, A.L.
    , L.N. Svejcar, B.M. Ayre, J. Bolleter, A.Brace, M.D. Craig et al. 2021. A threatened ecological community: Research advances and priorities for Banksia woodlands. Australian Journal of Botany 69(2):53–84.
    OpenUrl
  31. ↵
    1. Ritz, C.
    2010. Toward a unified approach to dose–response modeling in ecotoxicology. Environmental Toxicology and Chemistry 29(1):220–229.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Ritz, C.
    , F. Baty, J.C. Streibig and D. Gerhard. 2015. Dose-response analysis using R. PLoS One 10(12)e0146021.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Rokich, D.P.
    2016. Melding of research and practice to improve restoration of Banksia woodlands after sand extraction, Perth, West-ern Australia. Ecological Management and Restoration 17(2): 112–123.
    OpenUrl
  34. ↵
    1. Rokich, D.P.
    , K.W. Dixon, K. Sivasithamparam and K.A. Meney. 2000. Topsoil handling and storage effects on woodland restoration in Western Australia. Restoration Ecology 8(2):196–208.
    OpenUrl
  35. ↵
    1. Rokich, D.P.
    , J. Harma, S.R. Turner, R.J. Sadler and B.H. Tan. 2009. Fluazifop-p-butyl herbicide: Implications for germination, emergence and growth of Australian plant species. Biological Conservation 142(4):850–869.
    OpenUrl
  36. ↵
    1. Russel, V.L.
    2021. Emmeans: Estimated marginal means, aka least-squares means (R package version 1.7. 0). Vienna, Austria: R Foundation for Statistical Computing. https://CRAN.Rproject.org/package=emmeans.
  37. ↵
    1. Stanbury, K.E.
    , J.C. Stevens and A.L. Ritchie. 2018. Legacy issues in post-pine (Pinus pinaster) restoration environments: Weeds compromise seedling growth and function more than edaphic factors. Land Degradation and Development 29(6):1694–1704.
    OpenUrl
  38. ↵
    1. Stock, E.
    , R.J. Standish, M. Muñoz-Rojas, R.W. Bell and T.E. Erick-son. 2020. Field-deployed extruded seed pellets show promise for perennial grass establishment in arid zone mine rehabilitation. Frontiers in Ecology and Evolution 8:576125.
  39. ↵
    1. Svejcar, L.N.
    , D.R. Clenet, C.H. Guetling and K.W. Davies. 2024a. Activated carbon seed technology protects seedlings from two pre-emergent herbicides applied in tandem. Rangeland Ecology and Management 96:67–71.
    OpenUrl
  40. ↵
    1. Svejcar, L.N.
    , T.E. Martyn, H.R. Edlund and K.W. Davies. 2024b. A test of activated carbon and soil seed enhancements for improved sub-shrub and grass seedling survival with and without herbicide application. Plants 13(21):3074.
    OpenUrlPubMed
    1. Tieu, A.
    , K.W. Dixon, K.A. Meney and K. Sivasithamparam. 2001. The interaction of heat and smoke in the release of seed dormancy in seven species from southwestern Western Australia. Annals of Botany 88(2):259–265.
    OpenUrlCrossRef
  42. ↵
    1. Weidlich, E.W.A.
    , F.G. Flórido, T.B. Sorrini and P.H.S. Brancalion. 2020. Controlling invasive plant species in ecological restoration: A global review. Journal of Applied Ecology 57(9):1806–1817.
    OpenUrlCrossRef
  43. ↵
    1. Wild, J.
    , M. Kopecký, M. Macek, M. Šanda, J. Jankovec and T. Haase. 2019. Climate at ecologically relevant scales: A new temperature and soil moisture logger for long-term microclimate measurement. Agricultural and Forest Meteorology 268:40–47.
    OpenUrlCrossRef
  44. ↵
    1. Zuur, A.F.
    , E.N. Ieno, N.J. Walker, A.A. Saveliev and G.M. Smith. 2009. Mixed Effects Models and Extensions in Ecology with R, Vol. 574. Springer.
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Field-deployed Extruded Pellets Containing Activated Carbon Improve Recruitment Potential in Native Banksia Woodland Species
Vanessa S. Brown, Todd E. Erickson, Alison L. Ritchie
Ecological Restoration Aug 2026, 44 (3) 201-212; DOI: 10.3368/er.44.3.201
Twitter logo Facebook logo Mendeley logo
Bookmark this article

Jump to section

Keywords