Final Report Template

Home , Brasenia

Biological Control and Ecology of Cabomba and Alligator Weed

Cabomba. Source: Wikipedia Alligator weed. Source: Wikipedia

www.rirdc.gov.au

Biological Control and Ecology of Cabomba and Alligator Weed

by Shon Schooler and Richard Chan, CSIRO Ecosystem Sciences

May 2011

RIRDC Publication No 11/029 RIRDC Project No 08-53

ISBN 978-1-74254-215-7 ISSN 1440-6845

Biological Control and Ecology of Cabomba and Alligator Weed Publication No. 11/029 Project No. AWRC08-53

The information contained in this publication is intended for general use to assist public knowledge and discussion and to help improve the development of sustainable regions. You must not rely on any information contained in this publication without taking specialist advice relevant to your particular circumstances.

While reasonable care has been taken in preparing this publication to ensure that information is true and correct, the Commonwealth of Australia gives no assurance as to the accuracy of any information in this publication.

The Commonwealth of Australia, the Rural Industries Research and Development Corporation and the authors or contributors expressly disclaim, to the maximum extent permitted by law, all responsibility and liability to any person arising directly or indirectly from any act or omission or for any consequences of any such act or omission made in reliance on the contents of this publication, whether or not caused by any negligence on the part of the Commonwealth of Australia, RIRDC, the authors or contributors.

The Commonwealth of Australia does not necessarily endorse the views in this publication.

This publication is copyright. Apart from any use as permitted under the Copyright Act 1968, all other rights are reserved. However, wide dissemination is encouraged. Requests and inquiries concerning reproduction and rights should be addressed to the RIRDC Publications Manager on phone 02 6271 4165.

Researcher contact details

Name: Shon Schooler Address: CSIRO Ecosystem Sciences, PO Box 2583, Brisbane QLD 4001

Phone: 07-3833-5662 Fax: 07-3833-5504 Email: [email protected]

In submitting this report, the researchers have agreed to RIRDC publishing this material in its edited form.

RIRDC contact details

Rural Industries Research and Development Corporation Level 2, 15 National Circuit BARTON ACT 2600

PO Box 4776 KINGSTON ACT 2604

Phone: 02 6271 4100 Fax: 02 6271 4199 Email: [email protected]. Web: http://www.rirdc.gov.au

Electronically published by RIRDC in May 2011 Print-on-demand by Union Offset Printing, Canberra at www.rirdc.gov.au or phone 1300 634 313

Foreword

Alligator Weed (Alternanthera philoxeroides) and Cabomba (or Caroline fanwort, Cabomba caroliniana) are amongst the worst aquatic threats in Australia. Alligator weed was introduced to Newcastle (NSW) as a culinary herb in the 1940s, but it has now has spread to all States. Inundation allows Alligator Weed to outcompete other species and desirable forage grasses.

Cabomba was most likely introduced to Australia as an aquarium plant, but it is now a serious aquatic weed throughout the country. Its dense infestations interfere with recreational and agricultural use of water bodies and crowd out native aquatic plants. Both Alligator Weed and Cabomba are very difficult to control after they invade water bodies.

In view of the economic, social and environmental impacts and the difficulty of controlling both weeds by using conventional methods, biological control is recognised as an important research component in the national strategies for dealing with the weeds.

This project extends the work of two previous projects, begun in 2003, aimed at finding safe and effective biological agents to control these weeds. In the earlier work, the researchers had identified, prioritised and started the host specificity testing of several potential invertebrate agents.

The goal of this project was to provide safe and sustainable control of Alligator Weed and Cabomba through biological control and improvements in best-practice methodology. The researchers found the beetle Systena nientula is not a suitable biological control agent for Alligator Weed as it can complete its life cycle on several native species. The host specificity testing of the weevil Hydrotimetes natans was not completed due to insufficient numbers of the weevil being able to be reared in quarantine. However, several additional potential agents were identified for subsequent testing.

In terms of non-biological controls, shading of Cabomba reduced the plant’s biomass, but it is not an effective eradication method because vegetative propagules remain in the sediment.

This project was funded in Phase 1 of the National Weeds and Productivity Research Program, which was managed by the Australian Government Department of Agriculture, Fisheries and Forestry (DAFF) from 2008 to 2010. The Rural Industries Research and Development Corporation (RIRDC) is now publishing the final reports of these projects.

Phase 2 of the Program, which is funded to 30 June 2012 by the Australian Government, is being managed by RIRDC. Further research will be funded on the ecology and dispersal of Cabomba and on the ecology and biological control of Alligator Weed.

RIRDC is commissioning some 50 projects in Phase 2 that both extends on the research undertaken in Phase 1 and moves into new areas. Reports on these projects will be published in the second half of 2012.

This report is an addition to RIRDC’s diverse range of over 2000 research publications which can be viewed and freely downloaded from our website www.rirdc.gov.au. Information on the Weeds Program is available online at www.rirdc.gov.au/weeds

Most of RIRDC’s publications are available for viewing, free downloading or purchasing online at www.rirdc.gov.au. Purchases can also be made by phoning 1300 634 313.

Craig Burns Managing Director Rural Industries Research and Development Corporation

iii

Acknowledgments

In connection with the cabomba shading trial, thanks to Ms Henrika Sonne for the use of her property; Phil Moran (Noosa & District Landcare) managed the project and assisted with the construction, installation and removal of the shade fabric; Vanessa Moscato (Noosa & District Landcare and Noosa Water Watch) monitored water quality; Terry Stokes (Noosa Landcare volunteer) assisted with preparing and installing the shade fabric and propagating the native species for revegetation; and Andrew Petroeschevesky (National Aquatic Weeds Management Group) and Geoff Black (Sunshine Coast Regional Council) provided useful advice.

Contents

Foreword ...... iii

Acknowledgments ...... iv

Executive Summary ...... vii

1 Introduction ...... 1

2 Host specificity testing of Systena nitentula ...... 2

Methods...... 2 Origin of the colony ...... 2 Biology ...... 2 Host testing ...... 3 Results ...... 4 Biology ...... 4 Host testing ...... 4 Conclusions ...... 5

3 Host specificity testing of Hydrotimetes natans ...... 6

Methods...... 7 Origin of the colony ...... 7 Biology ...... 7 Host testing ...... 7 Discussion ...... 9

4 Shade as an eradication strategy for Cabomba caroliniana ...... 10

Methods...... 10 Results and discussion ...... 10

5 Effect of inundation, selective herbicide use and mowing on alligator weed ...... 12

References ...... 13

Tables

Table 1 Life stages of S. nitentula ...... 3

Table 3 Number of larvae completing each larval instar and developing into adults, by replication . 5

Table 4 Results of plant preference trial: H. natans ...... 8

Table 2 Results of survival tests: H. natans and C. caroliniana and B. schreberi ...... 9

Table 6 Water quality ...... 11

Figures

Figure 1 S. nitentula adult ...... 2

Figure 2 S. nitentula larva: third instar ...... 2

Figure 3 S. nitentula pupa ...... 4

Figure 4 H. natans adult on cabomba stem ...... 6

Figure 5 H. natans larva Figure 6 H. natans pupa ...... 6

Figure 7 C. caroliniana and Brasenia schreberi test plants ...... 8

Figure 8 Cabomba dry weight before and after shade fabric was placed over the dam ...... 11

Executive Summary

What the report is about?

Alligator weed (Alternanthera philoxeroides) and Cabomba (or Caroline fanwort, Cabomba caroliniana) are two aquatic weeds of national significance. Due to the weeds’ economic, social and environmental impacts and the difficulty of controlling them by using conventional methods, biological control is recognised as an important research component in the national strategies for dealing with the weeds.

This project represents the culmination of two previous projects, begun in 2003, which identified, prioritised and started the host specificity testing of several potential invertebrate agents. The goal of this project was to provide safe and sustainable control of the two introduced aquatic weeds, focusing on two priorities: biological control and improvements in best-practice methodology.

What are the relevant regions in Australia?

Cabomba (Cabomba caroliniana) is a serious aquatic weed, introduced to Australia as an aquarium plant, that has become widely naturalised in Australia where it invades lakes, potable water supply dams, farm dams, creeks, rivers and drainage canals, often forming dense submerged mats. Alligator Weed was introduced to Newcastle (NSW) as a culinary herb, but is now a serious aquatic weed throughout Australia.

Aims/objectives

This project had four primary components:

• testing the host specificity of the beetle Systena nitentula, a potential biological control agent for alligator weed

• testing the host specificity of the weevil Hydrotimetes natans, a potential biological control agent for cabomba

• examining the effect of shading as a management tool for cabomba

• examining the interaction between inundation, selective herbicide use and mowing in the management of alligator weed.

Methods used

In testing S. nitentula, 10 newly hatched larvae were placed on the washed roots of test plants in a 20- centimetre Petri dish lined with filter paper. Each Petri dish was sealed with paraffin film to retain moisture. More roots were added to each dish as required, to ensure that plenty of root material was available for the developing larvae.

Larval development to each instar and the number of adults that emerged were recorded. The adults were allowed to continue to develop in the boxes for observation, mating and oviposition. Eggs were collected and placed in a Petri dish on moist filter paper. The number of eggs hatched was noted in order to determine fecundity. Newly hatched larvae and the adults from each test plant were then placed in a cage with the same species of test plant for observation of continuing development.

vii

Eight plant species were tested:

• Alternanthera angustifolia—native (narrow leaved joyweed)

• denticulata (green)—native (Lesser joyweed)

• denticulata (New Zealand, large-leafed)—exotic

• denticulata (red-leafed)—native

• nana—native (Hairy joyweed)

• nodiflora—native (Common joyweed)

• sessilis (Australia)—native (Sessile joyweed or dwarf copperleaf)

• sessilis (New Zealand)—exotic.

It should be noted that multiple plant species have been included that have the same epithet. These “species” have differing morphology or colour, which may indicate that they may actually be different species or different varieties. Therefore, they have been included as separate entities in the host testing for added safety. The authors are currently collaborating with New Zealand Landcare on a project examining the genetic differences among these plants.

Six H. natans adults and one stem of a test plant were placed in 2-litre jars. The jars were sealed with a piece of gauze and rubber bands. The adults were removed from the test plants after four days and were placed in a holding container with fresh cabomba plants. The test plants were checked for feeding damage and any larval development.

Three replications of the following plants were used:

• Cabomba caroliniana (Cabomba)

• C. aquatica (Fanwort or Giant cabomba) • Brasenia schreberi (Watershield) • Hydrilla verticillata (Hydrilla) • Nymphoides indica (Water snowflake) • Ludwigia peploides (Water primrose) • Myriophyllum papillosum (Water milfoil) • Monochoria virginalis (Monochoria) • Ottelia ovalifolia (Swamp lily) • Utricularia gibba (Humped or floating bladderwort) • Ceratophyllum demersum.(Rigid hornwort or coontail)

The test was terminated after eight weeks.

A choice test was conducted to determine whether H. natans adults would choose C. caroliniana over the most closely related Australian species, B. schreberi. Twenty Hydrotimetes adults were placed in the centre of an aquarium tank with the C. caroliniana and B. schreberi plants at each end separated by

viii

a mesh screen with openings large enough for the weevils to move through it. The location of each adult was recorded after 24 hours.

Researchers also conducted a no-choice test to determine how long adults could survive on C. caroliniana compared with B. schreberi. Four adult weevils were placed in an aquarium with an abundant amount of either C. caroliniana or B. schreberi. Adults were left on the material for two months, fresh plant material being added as needed.

For the shade trial for cabomba, researchers covered a small farm dam with black builders’ plastic. Before covering the dam, researchers examined biomass by collecting plant material in underwater quadrats (0.25 x 0.25 metres). The material was taken to the CSIRO laboratories in Brisbane, sorted into species and dried to constant weight. Researchers also took water quality data four times during the project in order to determine if the plastic was affecting dissolved oxygen, turbidity, conductivity, nitrogen and phosphorus.

The plastic stayed in position for more than four months, after which researchers surveyed the dam internally by snorkelling in a grid pattern and externally by examining the perimeter. Any material found was collected and dried at CSIRO, and its dry weight was recorded.

Researchers revegetated the dam with native aquatic and edge species two weeks after removal of the plastic. The species used were Philydrum lanuginosum (frog’s mouth), Nymphoides indica (water snowflake), Lepironia articulata (lepironia) and Baumea rubiginosa (soft twigrush). The vegetation was then monitored at monthly intervals.

For the inundation trial, researchers investigated the effect of anthropogenic disturbance (herbicidal and mechanical) along a natural inundation gradient (20 to 282 days) on the biomass and resource allocation of the invasive wetland plant alligator weed (Alternanthera philoxeroides) and two co- occurring competitor plants, the introduced grass kikuyu (Pennisetum clandestinum) and the native grass couch (Cynodon dactylon), over two years.

Results/key findings

S. nitentula was imported from Argentina for testing on native and economically important non-target plants for the host specificity testing process. The beetle was tested and it was found it could complete its life cycle on several native Alternanthera species. It is therefore not suitable for release in Australia and no application has been submitted.

H. natans was imported from Argentina for testing on native and economically important non-target plants for the host specificity testing process. Preliminary testing was completed on the weevil and it was found it does not complete its life cycle on the most closely related species in Australia, Brasenia schreberi. However, testing was unable to be completed on the other species because sufficient numbers of the weevil could not be reared in our quarantine.

Several additional potential control agents were identified for subsequent testing.

Data was collected on the shade experiment with cabomba. It was found that, although shade will reduce the biomass of cabomba, it is not an effective eradication technique because vegetative propagules remain in the sediment.

Researchers also found that inundation duration is an important mechanism that allows alligator weed to outcompete desirable forage grasses: about 30 days’ inundation allows the weed to outcompete kikuyu.

1 Introduction

Alligator weed (Alternanthera philoxeroides) and cabomba (or Caroline fanwort, Cabomba caroliniana) are two aquatic weeds of national significance. Because of the weeds’ economic, social and environmental impacts and the difficulty of controlling them by using conventional methods, biological control is recognised as an important research component in the national strategies for dealing with the weeds. This project represents the culmination of two previous projects, begun in 2003, aimed at finding safe and effective biological agents. We had previously identified, prioritised and started the host specificity testing of several potential invertebrate agents. The goal of the current project was to provide safe and sustainable control of the two introduced aquatic weeds, focusing on two priorities—biological control and improvements in best-practice methodology.

The four primary objectives of the project were as follows:

• testing the host specificity of the beetle Systena nitentula, a potential biological control agent for alligator weed

• testing the host specificity of the weevil Hydrotimetes natans, a potential biological control agent for cabomba

• examining the effect of shading as a management tool for cabomba

• examining the interaction between inundation, selective herbicide use and mowing in the management of alligator weed.

2 Host specificity testing of Systena nitentula

Systena nitentula was found on alligator weed in surveys carried out by researchers at the US Department of Agriculture’s South American Biological Control Laboratory in Buenos Aires, Argentina. It was identified as a potential biological control agent. A re-description, biology and distribution study was carried out by Cabrera et al. (2005), who found that S. nitentula could be a monophagous flea beetle feeding on alligator weed under terrestrial conditions. Figure 1 shows the S. nitentula adult; Figure 2 shows the larva.

Figure 1 S. nitentula adult Figure 2 S. nitentula larva: third instar

Methods

Origin of the colony

The colony of S. nitentula (45 adults) was collected from San Ramón de la Nueva Orán in Argentina on 14 January 2009. The adults from this shipment did not live long, and we managed to obtain only six new adults. A second shipment, of 22 adults, collected from San Ramón de la Nueva Orán, was received on 7 April 2009. We established a colony of S. nitentula in our Brisbane quarantine.

Identification of the material was carried out by Dr Nora Cabrera from the División de Entomología, Museo de Ciencias Naturales de La Plata, in Buenos Aires.

Biology

The S. nitentula life cycle consists of egg, three larval stages, pre-pupa, pupa and adult. New adults emerging from the soil were pale in colour and darkened over time. The adults fed on the leaf margins of alligator weed and made shallow lesions on the stems. Mating began about seven days after emergence, and eggs were laid randomly on alligator weed leaf axils, in the fine root hairs and on paper towels (substrate). The larvae fed on and burrowed into the roots and crown of alligator weed. Pupation occurred in the roots.

The colonies of S. nitentula were reared in boxes (25 x 19 x 10 centimetres). The floor of each box was layered with paper towelling, and 20-centimetre stems of alligator weed were added. Twenty pairs of newly emerged adults were collected from different boxes to set up a new box for mating and

oviposition. The eggs were collected weekly and placed in Petri dishes lined with moist filter paper. Table 1 shows the life stages of the beetle.

Table 1 Life stages of S. nitentula

Stage Length (mm) Width (mm) Mean duration (days) Egg 0.78 0.39 1st instar larva – 0.22 (head capsule) 6.7 2nd instar larva – 0.29 (head capsule) 9.7 3rd instar larva – 0.40 (head capsule) 12.7 Adult (female) 3.98 2.31 –

Host testing

Ten newly hatched larvae were placed on the washed roots of test plants in a 20-centimetre Petri dish lined with filter paper. Each Petri dish was sealed with paraffin film to retain moisture. More roots were added to each dish as required, to ensure that plenty of root material was available for the developing larvae.

• Alternanthera angustifolia—native • A. denticulata (green)—native • A. denticulata (New Zealand, large-leafed)—exotic • A. denticulata (red-leafed)—native • A. nana—native • A. nodiflora—native • A. sessilis (Australia)—native • A. sessilis (New Zealand)—exotic.

Results

Biology

Systena adults fed on the leaves and tips of alligator weed. The larvae fed on the roots, and some burrowed into the thicker roots during larval development. They also moved from root to root in search of fresher material. The late third instar larvae, or pre-pupae, burrowed into the crown or thicker roots to pupate, although some completed pupation on the paper towelling. The pupation period was about nine days. Adults emerged from the roots and began feeding on the leaves and stems. Figure 3 shows the pupal stage.

Figure 3 S. nitentula pupa

Host testing

Table 3 shows the results of larval and pupal development and the number of adults that emerged. Tests on each test plant species were replicated three times, and S. nitentula larvae fed and developed into fertile adults on five of the main Alternanthera species—namely A. angustifolia, A. denticulata (three forms), A. nana, A. nodiflora and A. sessilis (two forms).

The newly emerged larvae fed on the tender young roots and progressively moved on to thicker, fleshier roots as they developed. Survival to adulthood was low (less than 27 per cent). Larvae also fed and developed on the stems and on the undersides of leaves but with less than 10 per cent survival. The egg hatch from 1192 eggs collected between 15 June and 20 July 2009 was about 61.4 per cent.

Adults and larvae from each plant species were set up in individual cages to determine whether colonies would develop on A. angustifolia, A. denticulata (red, green and New Zealand), A. nana, A. nodiflora and A. sessilis (Australian and New Zealand). S. nitentula populations were able to increase and develop colonies in the absence of alligator weed on each of the Alternanthera species tested. Observation of the plants showed no differences in the vigour and fecundity of the beetle. The colonies were terminated after 50 days.

Table 3 Number of larvae completing each larval instar and developing into adults, by replication

Plant species Replication 1 Replication 2 Replication 3 1st 2nd 3rd Adult 1st 2nd 3rd Adult 1st 2nd 3rd Adult A. philoxerioides 10 4 4 2 10 6 6 5 10 5 5 4 A. angustifolia 10 3 3 3 10 8 8 5 10 3 2 2 A. denticulata (red) 10 6 5 1 10 7 5 4 10 5 4 2 A. denticulata (green) 10 5 4 4 10 5 5 4 10 5 3 3 A. denticulata (NZ) 10 2 2 2 10 7 4 4 10 4 4 3 A. nana 10 4 4 2 10 7 7 4 10 4 4 3 A. nodiflora 10 8 6 3 10 3 2 2 10 6 6 2 A. sessilis (Aus) 10 8 6 4 10 8 3 2 10 5 2 2 A. sessilis (NZ) 10 5 3 3 10 4 4 4 10 3 3 3

Conclusions

We found that S. nitentula is a difficult beetle to culture compared with the other two chrysomelids tested previously, Agasicles hygrophila and Disonycha argentinensis. S. nitentula adult feeding is similar to that of the other two beetles, but the larvae were delicate and less vigorous. Mortality was high during larval development, and larvae were hard to locate among the roots. Development from first instar to adults in the culture was less than 30 per cent on the host plant Alternanthera philoxeroides.

The tests revealed that five other Alternanthera species were able to support colonies of S. nitentula in the absence of alligator weed. This suggests that S. nitentula is not specific to alligator weed and might pose a risk to non-target species if released as a biological control agent.

3 Host specificity testing of Hydrotimetes natans

Cabomba (Cabomba caroliniana) has fans of purple-backed lacy foliage that are divided into narrow segments, with green tops crowned with white flowers in summer. It is a serious weed of eastern Australia and was rated number 11 in the weeds of national significance priorities exercise (Thorp & Lynch 2000). An approved target for biological control in Australia, cabomba is native to the South American subtropics but has become widely naturalised in North America, southern Asia and Australia. In Australia it invades lakes, potable water supply dams, farm dams, creeks, rivers and drainage canals, often forming dense submerged mats.

The cabomba weevil, Hydrotimetes natans (see Figures 4, 5 and 6), is known from Argentina to Brazil. Observations in the field in Argentina show that the larvae tunnel through the stems of cabomba and the adults feed on the leaves, reducing growth. The weevil has never been used anywhere as a biological control agent. It was described by H Kolbe in 1911; little else about it was recorded in the literature until 2004. Dr Willie Cabrera-Walsh collected the weevil in Buenos Aires in a cabomba survey; Dr C O’Brien identified it as Hydrotimetes natans.

Figure 4 H. natans adult on cabomba s tem

Figure 5 H. natans larva Figure 6 H. natans pupa

Methods

Origin of the colony

Material for the Australian colony was collected from Corrientes and Entre Ríos Provinces in Argentina. We received a shipment of 101 adults on 7 April 2009.

Biology

The biology and life cycle of H. natans have not been studied in detail. Field survey reports by our collaborator Dr Cabrera-Walsh at the US Department of Agriculture’s South American Biological Control Laboratory described the H. natans adult as feeding on the plant tips while the larvae mine the plant stems. The life cycle from egg through three larval instars, pre-pupa and pupa to adult is about 40 days. At high population densities, this weevil causes severe tip and stem damage.

In Argentina the adults are present year round and survive at least a year in the laboratory. The mating season (from December to February) is easy to detect because the adults are found at dusk mating or alone on the plant flowers. During the rest of the year they are found only underwater. The larvae have been found in the cabomba stems from October to April–May, peaking from the beginning of summer; this suggests that gravid females that have laid eggs in summer overwinter and start laying eggs again toward the end of the following spring.

Host testing

The no-choice test: oviposition and life cycle completion

Six H. natans adults and one stem of a test plant were placed in 2-litre jars (see Figure 7). The jars were sealed with a piece of gauze and rubber bands. The adults were removed from the test plants after four days and were placed in a holding container with fresh cabomba plants. The test plants were checked for feeding damage and any larval development. Three replications of the following plants were used:

• Cabomba caroliniana • C. aquatica • Brasenia schreberi • Hydrilla verticillata • Nymphoides indica • Ludwigia peploides • Myriophyllum papillosum • Monochoria virginalis • Ottelia ovalifolia • Utricularia gibba • Ceratophyllum demersum.

The test was terminated after eight weeks.

Figure 7 C. caroliniana and Brasenia schreberi tes t plants Light feeding scars were recorded on all test plants other than L. peploides. No H. natans larvae development was observed on any of the test plants. All plants were dissected at the end of the experiment: no larval activity was observed. We concluded that H. natans might not have had sufficient time to settle on the plants for oviposition or that insufficient numbers of adults were used for each treatment.

The tests were repeated using 12 adults per treatment over 10 days. In this case feeding scars were recorded for C. caroliniana, C. aquatica, B. schreberi, O. ovalifolia and Ceratophyllum demersum. Despite this, no larval development was observed.

The choice test: plant preference

The choice test was conducted to determine whether H. natans adults would choose C. caroliniana over the most closely related Australian species, B. schreberi. Twenty Hydrotimetes adults were placed in the centre of an aquarium tank (60 x 30 x 30 centimetres), the C. caroliniana and B. schreberi plants being placed at each end of the tank, separated from each other by a mesh screen with openings large enough for the weevils to move through it. The location of each adult was recorded after 24 hours.

The result was that, although H. natans adults were found on both plants, they showed a clear preference for cabomba (see Table 4).

Table 4 R esults of plant preference trial

Position of H. natans Aquarium Brasenia Mesha Centre Mesha Cabomba 1 2 3 0 7 8 2 1 5 0 7 6 3 0 3 0 8 9 4 0 5 0 6 9 5 3 0 0 12 5 6 1 2 0 7 9 a. Gutter guard was placed in the tank to create the three chambers and keep the plants separate.

The no-choice test: adult survival test

We conducted a test to determine how long adults could survive on C. caroliniana compared with B. schreberi. Four adult weevils were placed in an aquarium with an abundant amount of either C. caroliniana or B. schreberi, replicated three times for each plant species (six aquaria in all). Adults were left on the material for two months, fresh plant material being added as needed.

The result was that adults survived equally well on the B. schreberi material and the C. caroliniana material (see Table 5). We observed, however, larvae and adults produced on C. caroliniana but not on B. schreberi.

Table 2 Results of survival tests: H. natans and C. caroliniana and B. schreberi

Aquarium Plant No. of adults survived No. of adults produced 1 B. schreberi 3 0 2 B. schreberi 3 0 3 B. schreberi 2 0 4 C. caroliniana 4 1 (2 larvae) 5 C. caroliniana 4 1 (2 larvae) 6 C. caroliniana 3 0

Discussion

The results of the experiments are inconclusive, although they do suggest that H. natans is sufficiently host specific for release in Australia. Host specificity testing of potential classical biological control agents requires progressing through a number of sequential steps. For H. natans this involves collecting the weevil from lakes in northern Argentina, obtaining export permits from the Argentinean provincial and federal governing bodies, obtaining import permits from the Australian government, rearing the weevil in a building approved by the Australian Quarantine and Inspection Service, collecting and propagating native species for testing, and carrying out the host specificity tests. The key to this testing is to show that, under the same conditions, the weevil can complete its life cycle on cabomba, while it cannot complete its life cycle on Australia’s native aquatic plants.

In our primary no-choice experiments we were unable to produce second-generation H. natans adults on cabomba or any of the non-target plants. This invalidates the idea that H. natans cannot complete its life cycle on non-target plants. We were, however, able to test the host specificity of the species most closely related to cabomba in Australia—watershield (B. schreberi)—in our adult survival experiments. These results were more conclusive: we found that H. natans did not complete its life cycle on B. schreberi while it did produce a second generation of weevils on cabomba. This suggests that H. natans is probably sufficiently host specific to allow its release in Australia.

4 Shade as an eradication strategy for Cabomba caroliniana

Cabomba, a submerged aquatic macrophyte native to South America, is spreading throughout the world and is considered a serious weed in many countries. In Australia it is one of the 20 weeds of national significance and is found from Darwin in the Northern Territory to Benalla in Victoria. At present there are no effective herbicides registered for use on cabomba, and the only control method available is physical removal. This takes the form of specifically designed aquatic weed harvesters or underwater divers using a vacuum device. New infestations of cabomba are regularly being found, and alternative methods of control are being investigated. A previous experimental study found that use of floating shade covers (5 x 5 metres) resulted in a significant reduction in cabomba biomass. This current project examined whether such an approach could be used to eradicate cabomba from small dams, thus providing a method of eradicating small outlying infestations of the weed.

Methods

For this trial we covered a small farm dam (35 x 40 metres) at Kin Kin in south-east Queensland with black builders’ plastic. The edges of the plastic were stitched with a portable bag-sewing machine (from Ezypack, in Perth) so that rope could be threaded through and anchored to the shore. Before covering the dam, we examined biomass by collecting plant material in underwater quadrats (0.25 x 0.25 metres). The material was taken to the CSIRO laboratories in Brisbane, sorted into species and dried to constant weight. We also took water quality data four times during the project in order to determine if the plastic was affecting dissolved oxygen, turbidity, conductivity, nitrogen and phosphorus.

We covered the dam on 18 December 2007. It took four people and the landowner most of the day to complete the operation. The procedure went well, despite problems with site access and the ground being wet because of previous rainfall. The slope leading down to the dam was sufficiently steep to make it difficult even for a four-wheel-drive vehicle. The site was suitable, though, because there was little wind and the dam was inaccessible to the public. Further, the landowner was committed to the project.

The plastic stayed in position for more than four months, being removed on 23 April 2008. The removal was difficult, necessitating the use of four-wheel-drive vehicles to tow each section of plastic off the dam in muddy and slippery conditions. We then surveyed the dam internally by snorkelling in a grid pattern and externally by examining the perimeter. Any material found was collected and dried at CSIRO, and its dry weight was recorded.

We revegetated the dam with native aquatic and edge species two weeks after removal of the plastic. The species used were Philydrum lanuginosum (frog’s mouth), Nymphoides indica (water snowflake), Lepironia articulata (lepironia) and Baumea rubiginosa (soft twigrush). We then monitored the vegetation at monthly intervals until May 2009.

Results and discussion

Before shading, the cabomba dry weight averaged 462 grams per square metre, although it varied according to depth (see Figure 8). After the shade was removed water clarity was high and no cabomba was immediately evident. In our surveys, however, we found two rooted plants along the edge where the plastic went around a tree and one floating piece of stem in the centre of the dam. These materials were removed by hand; they weighed 0.44 grams dry weight. Combing the sediment did not result in the discovery of any intact root or stem material. Dissolved oxygen and pH had

increased and turbidity had decreased compared with pre-shade conditions (see Table 6), meaning the reduction in cabomba resulted in improved water quality.

700 before shade

) after shade 2 600

500

400

300

200

Cabomba biomass (g/m biomass Cabomba 100

0 0.5 1.0 1.5 2.0 Water depth (m)

Figure 8 Cabomba dry weight before and after shade fabric was placed over the dam

Table 6 Water quality

Parameter 17 October 2007 24 January 2008 23 April 2008 6 August 2008 Dissolved oxygen (mg/L) 0.17 0.77 3.14 6.05 pH 5.76 5.69 6.29 6.65 Electrical conductivity (mS/cm) 0.169 0.139 0.114 0.106 Turbidity (NTU) 133 240 10 12 Nitrate–nitrogen (mg/L) 0.1 0.2 0.2 0.3 Filterable reactive phosphorus (mg/L) 0.02 0.05 0.02 0.07

Monthly sampling until December 2008 resulted in no cabomba regrowth being detected, but new patches of cabomba were seen in January 2009. Underwater surveys found that this was not an isolated patch: cabomba stems were evident across the bottom of the dam. Samples were taken, and we found no evidence that these were seedlings (no remaining seed capsules were attached, as we had found in seedling samples in the Northern Territory). The new growth was probably from remaining stem nodes that had become buried in the sediment.

The results showed that cabomba can be reduced in abundance through shading but that this is not a useful eradication tool. Among the difficulties associated with adopting this method are the cost of materials, the effort involved in constructing, installing and removing the plastic, and the limited duration of control. The technique could, however, be used effectively to reduce cabomba abundance in public access areas and over irrigation pumping areas.

5 Effect of inundation, selective herbicide use and mowing on alligator weed

Disturbance is an important component of many successful plant invasions. Interactions between natural and anthropogenic disturbances and the effects of these disturbances on invasive plants and desired vegetation are, however, rarely examined.

We investigated the effect of anthropogenic disturbance (herbicidal and mechanical) along a natural inundation gradient (20 to 282 days) on the biomass and resource allocation of the invasive wetland plant alligator weed (Alternanthera philoxeroides) and two co-occurring competitor plants, the introduced grass kikuyu (Pennisetum clandestinum) and the native grass couch (Cynodon dactylon), over two years. In the absence of additional disturbance, kikuyu biomass was negatively affected by inundation disturbance, alligator weed biomass was positively affected, and couch biomass was not affected. In addition, kikuyu was not affected by the selective removal of alligator weed, while couch increased in wetter habitats where kikuyu was absent as a result of inundation stress. This suggests that kikuyu is a superior competitor in drier habitats and inundation facilitates the invasion of alligator weed, while couch is an inferior competitor relative to both kikuyu and alligator weed and is therefore suppressed across its entire niche by these two introduced species.

Mowing alone had no effect on the biomass of the species, suggesting the plants are equally tolerant of shoot removal. Selective herbicide use reduced alligator weed biomass by 97.5 per cent, and the combination of selective herbicide use and mowing reduced the biomass of alligator weed more than herbicide alone—by 98.6 per cent compared with unmanipulated controls. To predict community change and prevent sequential exotic plant invasions after weed removal, it is necessary to consider the interacting effects of disturbance and the niche space of invasive species in the local propagule pool.

The complete paper dealing with this research has been published—Schooler, SS, Cook, T, Prichard, G & Yeates, A 2010, ‘Disturbance-mediated competition: the interacting roles of inundation regime and physical and herbicidal control in determining native and invasive plant abundance’, Biological Invasions, vol. 12, pp. 3289–98.

References

Cabrera, N, Sosa, AJ, Dorado, J & Julien, M 2005, ‘Sistena nitentula (Coleoptera: Chrysomelidae), a flea beetle injurious to Alternanthera philoxeroides (Amaranthaceae): redescription, biology and distribution’, Annals of the Entomological Society of America, vol. 98, no. 5, pp. 643–52.

Thorp, JR & Lynch, R 200, The Determination of Weeds of National Significance, National Weeds Strategy Executive Committee, Launceston, Tasmania.

Biological Control and Ecology of Cabomba and Alligator Weed by Shon Schooler and Richard Chan, CSIRO Ecosystem Sciences.

Pub. No. 11/029

Alligator weed (Alternanthera philoxeroides) and cabomba (or Caroline Government Department of Agriculture, Fisheries and Forestry (DAFF) fanwort, Cabomba caroliniana) are two aquatic weeds of national from 2008 to 2010. The Rural Industries Research and Development significance. Because of the weeds’ economic, social and environmental Corporation (RIRDC) is now publishing the final reports of these impacts and the difficulty of controlling them by using conventional projects. methods, biological control is recognised as an important research component in the national strategies for dealing with the weeds. This report is an addition to RIRDC’s diverse range of over 2000 research publications which can be viewed and freely downloaded from our This project represents the culmination of two previous projects, begun website www.rirdc.gov.au. Information on the Weeds Program is available in 2003, aimed at finding safe and effective biological agents to control online at www.rirdc.gov.au/weeds these weeds. In earlier research we had previously identified, prioritised and started the host specificity testing of several potential invertebrate Most of RIRDC’s publications are available for viewing, free agents. The goal of this project was to provide safe and sustainable downloading or purchasing online at www.rirdc.gov.au. Purchases can control of the two introduced aquatic weeds, focusing on two priorities— also be made by phoning 1300 634 313. biological control and improvements in best-practice methodology.

This project was funded in Phase 1 of the National Weeds and Productivity Research Program, which was managed by the Australian

This publication can be viewed at our website—www.rirdc.gov.au. Contact RIRDC: Ph: 02 6271 4100 All RIRDC books can be purchased from: Level 2, 15 National Circuit Fax: 02 6271 4199 Barton ACT 2600 Email: [email protected] web: www.rirdc.gov.au www.rirdc.gov.au PO Box 4776 Kingston ACT 2604

Cabomba. Source: Wikipedia Alligator weed. Source: Wikipedia