Sagittaria Control Techniques

Sagittaria is a difficult plant to control due to its aquatic nature, tolerance to herbicide treatments, and prolific seed production. Most control programs aim to either reduce the impact of sagittaria or to contain and prevent further spread. Eradication is often only feasible in small outlier areas, therefore, detection of new infestations while they are small is critical. Control options are limited to herbicide application and physical removal. Currently, there are no biological control agents released in Australia, however, a research program
to identify and test potential agent is in progress by Agriculture Victoria Research (Australian Government
In catchments where sagittaria is already present the main goal is to prevent further downstream or inter‐
catchment basin spread. Preventing downstream spread is a difficult task because the main causes of spread,
floodwaters, ducks and possibly birds are difficult to control. However, strategic control programs to prevent annual seeding of plants can help reduce the rate of downstream spread (Australian Government 2012).
In Australia, there are no label recommendations specifically for Sagittaria platyphylla, but minor use permits
have been issued by the Australian Pesticide and Veterinary Medicines Authority (APVMA) (Australian
Government 2012). Several herbicides are registered for the control of Sagittaria montevidensis (arrowhead)
in rice crops (i.e. the active ingredients: Azimsulfuron, Bensulfuron, Bentazone, Benzofenap, Carfentrazone‐
ethyl and MCPA) (APVMA 2018).
Use of herbicides in or near water bodies is strictly regulated by label recommendations, and maximum
residue limits are recommended in the national water quality management strategy Australia and New
Zealand Guidelines for ‘Fresh and Marine Water Quality’ (ARMCANZ 1995; cited in Adair et al. 2012).
Herbicide control is difficult and will often only provide limited and temporary control of sagittaria. Foliar applications of herbicide helps remove the standing biomass of emergent plants but the herbicides, available under permit, don’t sufficiently relocate to the underground tubers or stolons, nor do they come into contact with the submerged rosettes. Regrowth from the rosettes, tubers and stolons can form new emergent plants in as little as six to 12 weeks after treatment. Herbicide control of emergent foliage before the plant flowers significantly reduces seed production, hence it can be a useful method to help prevent further spread of established infestations (Australian Government 2012).
Table 3 outlines the herbicides that have been used for control of sagittaria in Australia.
Use of herbicides for control of sagittaria in aquatic habitats should not exceed maximum residue limit guidelines or conditions specified under minor use permits. In billabongs where shallow water and lack of flow generally occur, herbicide application can be difficult without exceeding specified limits and may require progressive application in sections of the invaded water body. Herbicide applications for the control of
Sagittaria often result in variable levels of control that are not consistent between locations and time of application (Adair et al. 2012 and citations within).
Herbicide application by a single active ingredient remains a major component of current Sagittaria management and therefore, the risk of generating herbicide resistance is high and needs further investigation.
Research is required to improve our understanding of the physiological effects of herbicides, particularly the rates and pathways of herbicide translocation throughout single and multiple plants. Understanding the effects of herbicide application on demographic processes (i.e. mortality of different life stages and fecundity) will provide useful information with respect to optimal application timing for effective control.
Additional herbicide techniques that target submerged plant parts warrant further investigation (Adair et al.
Other potential herbicides: In New Zealand, the Environmental Protection Authority (EPA) in 2013 modified the approvals for herbicides containing the active ingredients; metsulfuron‐methyl, haloxyfop‐R‐methyl,
imazapyr isopropylamine or triclopyr triethylamine salt. These substances can now be applied onto or into the water in New Zealand as herbicides to control aquatic pest plants. The New Zealand EPA considered these
substances beneficial in the control of aquatic pest plants and more effective than other methods of control
(New Zealand Government 2013). Some of these herbicides are also used on other emergent aquatic weeds in the USA. Therefore, it would be beneficial to screen these formulations in Australia and determine their efficacy for control of sagittaria and potential for future use. The results of this screening could then be used to apply for a permit to use the best of these candidates in Australian waterways. It would be beneficial if such a permit could be obtained that allowed multiple agencies to utilise the permit.
As an example, imazapyr has been used recently in winter in irrigation channels in Australia to obtain long term control of sagittaria. There were problems with managing the residues in the water so that irrigated crops were not damaged. However, a study was done that indicated that it could be used effectively and safely under certain situations (Clements et al. 2017). Further, in non‐crop areas and when used in summer its
risk is substantially reduced and this is how it is used in NZ and USA – but imazapyr is not registered for use in this manner in Australia.
Further, research is underway in Queensland and Victoria to develop the herbicides flumioxazin and endothall, respectively, for the Australian aquatic weeds market (Clements et al. 2013; Clements et al.
accepted; DAF 2017). Both of this product have the potential for use against the submerged sagittaria growth
form. Finally, an application for registration of a new product (florpyrauxifen‐benzyl ester) has recently been
submitted for use in rice to control sagittaria, amongst other species, the potential application of this product
to control sagittaria outside of rice crops has not been investigated (Wells and Taylor 2016).
Physical removal
Mechanical control of sagittaria is utilised in channels and drains when the hydraulic capacity of water
delivery infrastructure needs to be restored quickly. The technique, which mostly uses excavation machinery,
is particularly useful where herbicide application is inappropriate, such as near sensitive crops or where
channels are in continual use and cannot be shut down for applications of herbicides. Mechanical control
methods can be costly due to high labour and transport costs. They may also fragment sagittaria plants, which
may then disperse through water delivery infrastructure. Viable propagules such as seed, tubers, stolons, or
crowns may also remain after treatment by mechanical methods, necessitating follow‐up suppression activity.
In irrigation systems, mechanical control can damage or re‐profile drains and channels, leading to leakage or
ponding, which can affect water delivery. Hygiene measures, including inspection and wash down, should be
implemented where machinery, such as excavators, have been working amongst sagittaria infestations. Such
measures can significantly reduce the likelihood of spread of sagittaria to new areas (Australian Government
2012). The design of irrigation channels can influence colonisation patterns of sagittaria. Where water levels
can be maintained at depths greater than the transition point from submerged to emergent forms of
sagittaria (c.a. 50 cm), water delivery benefits are obtained. Steeper slopes decrease infiltration of sagittaria
into deeper parts of a channel and reduce the impact of damaging emergent forms (Adair et al. 2012 and
citations within).
Various stubble management practices, such as retention, burning or incorporation have no immediate effect
on establishment and growth of Sagittaria calycina (closely related species), while nitrogen applications at
rates similar to those used for rice (between 20 and 200 kg N ha −1) enhanced growth. In rice crops,
allelopathic potential to selectively inhibit the development of sagittaria exists in rice germplasm, offering the
possibility of integrated weed management practices (Adair et al. 2012 and citations within).
Response to other manipulations
The roles of fire, grazing by domestic stock and slashing have not been evaluated in Australia for suppression
of sagittaria, but are unlikely to be of great benefit due to the asexual modes of reproduction in sagittaria and
non‐target impacts in rice‐growing areas. In natural ecosystems, competitive and desirable native aquatic
vegetation may offer potential for the suppression of sagittaria, particularly if combined with other forms of
control such as selective use of herbicides or classical biological control, but formal evaluation is required
(Adair et al. 2012 and citations within).
One water authority is currently investigating the use of microwave technology to heat and destroy plant
material and seeds that are in the sediment of irrigation channels and drains.
Biological control progress update
Classical biological control is a potential prospect for the long‐term and sustainable suppression of sagittaria
in Australia. Classical biological control may provide a method to supplement chemical control methods,
which are currently limited in effectiveness in Australia, with potential issues with persistent use of a single
chemistry and herbicide resistance.
Commencing in 2010, a collaborative project led by DEDJTR (Agriculture Victoria) and involving CSIRO, La
Trobe University and scientists from the USA completed the first developmental phase of research on the
biological control of sagittaria. They determined that sagittaria was a highly suitable target for biological
control and that the prospects for successful control were high (Kwong 2014). Surveys for prospective
biocontrol agents in the USA identified four insect species, all weevils, with considerable promise based on
their potential for reducing the growth, survival and/or reproductive output of sagittaria (Kwong et al. 2014).
A plant demography study identified the vulnerable stages of the weed that should be targeted for biological
control (Kwong et al. 2017a). A detailed genetic study comparing populations in the USA and Australia
determined that three main genotypes existed in Australia (Kwong et al. 2017b). Collections of biocontrol
agent colonies were targeted to these locations (Kwong 2016).
Prior to the introduction and release of biocontrol agents into Australia, it is a federal requirement that the
agents undergo stringent testing to ensure that they do not attack other plant species. Only natural enemies
that are proven to be ‘host‐specific’ are granted approval for release.
Colonies of the fruit feeding weevil (Listronotus appendiculatus) and the crown boring weevil (Listronotus
sordidus) (Figure 2) were collected from the USA and imported into Agriculture Victoria Research’s insect
quarantine facility at the Centre for AgriBiosciences Research in Bundoora, Victoria. The results of Listronotus
sordidus no‐choice trials demonstrated that plant species un‐related to S. platyphylla (i.e. rice, Eleocharis
dulcis, Eleocharis sphacelata or Triglochin procerum) were unsuitable hosts. The host‐range of L. sordidus
appears to be confined to plant species within the Alismatacae family, with closely‐related sagittaria species
being acceptable hosts. However, the ability of each species to support the development of L. sordidus larvae
may be limited by the availability of an abundant food supply, with tuber‐forming species making better
hosts. All of the sagittaria species tested are either declared weeds in many states or ornamental plants that
are not highly important species to the ornamental industry. Damage and some larval survival, although
negligible, was detected on some of the native species and further trials will be required to determine the risk
of attack of L. sordidus on these species (Kwong 2016).
The host specificity testing of the fruit‐feeding weevil, L. appendiculatus produced very promising results. No
oviposition or larval development was recorded on the native species, Alisma plantago‐aquatica, Caldesia
oligococca and Caldesia acanthocarpa. Eggs were laid, minor damage was done to the foliage and petioles,
and larvae developed on the fourth native species, Damasonium minus, but at rates much lower than S.
platyphylla and S. calycina. This is a common artefact of no‐choice laboratory testing and is not expected to
occur in the field. Importantly, oviposition of the adult progeny reared on D. minus was very low, indicating
that this plant is a poor host and unlikely to maintain viable successive generations. An application for release
of L. appendiculatus has been prepared and will be circulated to collaborating biocontrol scientists for peer
review prior to submission to the Australian Government.
The third agent, a tuber‐feeding weevil Listronotus frontalis, was imported into AgriBio in December 2016 and
basic biology and life history studies are underway prior to host testing in autumn 2018.

Leave a Reply

Your email address will not be published. Required fields are marked *