Report on invasive plant species on Tonga

Download, read and print this report in PDF format

Introduction

Dangerous species not known to be in Tonga

Species that are invasive elsewhere and are likewise invasive or have the potential to become so in Tonga

Species that are mentioned or listed as being weedy or invasive elsewhere and are common or weedy in Tonga

Native and naturalized species exhibiting aggressive behavior

Strategies for dealing with invasive species

Recommendations

Appendix 1.  Background material and references

Appendix 2.  Species by category

Appendix 3.  Other invasive plant species, mostly of agricultural concern, reported to be present in Tonga

Appendix 4. Invasive species present in American Samoa, the Cook Islands, Fiji, French Polynesia, Hawai‘i, Niue or Samoa but not present in Tonga

Appendix 5.   Scientific name synonyms

Appendix 6.   Invasive and potentially invasive species present on Tongatapu

Appendix 7.   Invasive and potentially invasive species present on ‘Eua

Appendix 8.   Invasive and potentially invasive species present on Vava‘u

Appendix 9.   Invasive and potentially invasive species present in the Ha‘apai island group

Appendix 10.   Invasive and potentially invasive species recorded as being present elsewhere in Tonga

Report to the Kingdom of Tonga on Invasive Plant Species of Environmental Concern

James C. Space and Tim Flynn ⁽¹⁾

The Kingdom of Tonga requested assistance from the US Department of Agriculture, Forest Service, Institute of Pacific Islands Forestry, to conduct a survey of invasive plant species of environmental concern, similar to surveys previously conducted in Micronesia, American Samoa and Niue. The survey was carried out from 5-20 July 2001. The islands of Tongatapu, ‘Eua, Vava‘u, Ha‘ano, Lifuka, Foa and ‘Uiha were surveyed. The latter four islands are in the Ha‘apai group of islands. The survey of Ha‘ano was not as thorough as we would have liked because stormy weather forced us to return to Lifuka. The objectives, as with previous surveys, were to: (1) identify plant species that are presently causing problems to natural and semi-natural ecosystems; (2) identify species that, even though they are not presently a major problem, could spread more widely or are known to be problem species elsewhere; (3) confirm the absence of species that are a problem elsewhere and, if introduced to Tonga, could be a threat there; and (4) make appropriate recommendations.

During our visit local experts (2) showed us sites of known infestations. We also had available copies of botanical surveys conducted in the past (see Appendix 1, References). The intent was only to conduct an overall survey and not an exhaustive survey of the weed flora of the islands. Additional surveys of individual species, sensitive areas or other islands in the archipelago can and should be conducted as needed. This report summarizes our findings and makes some suggestions for further action.

Invasive species occurring in or of threat to Tonga have been grouped into four categories:

1. Species that are invasive elsewhere in similar ecosystems but were not seen on our visit and are not reported in the literature as being present in Tonga (259 species).
2. Species that are invasive elsewhere and are also invasive or potentially invasive in Tonga (31 species).
3. Species that are invasive or weedy elsewhere and are common, weedy or cultivated in Tonga (135 species).
4. Native species that exhibit aggressive behavior (9 species).

These species are listed in Appendix 2. Additional information about each species is located on a World Wide Web site, http://www.hear.org/pier/, and on the PIER-CD, copies of which have been made available to the Ministry of Agriculture and Forestry.

There are a number of other species that are mostly invasive weeds in gardens, fields, and pastures and along roadsides but don’t seem to pose any particular threat to wildland ecosystems. While we did not specifically survey for them, a list of these species, compiled from the literature, is included in Appendix 3.

1. Dangerous species not known to be on Tonga

Tonga is fortunate that a number of troublesome species have yet to reach the country. These are listed in Appendix 2, Table 1. The following list summarizes the worst of these, which should be excluded from entry into the country or evaluated for eradication if found.

Two rubber trees, Castilla elastica (Panama rubber tree) and Funtumia elastica (African rubber tree), are species that have proven very invasive in Samoa. Castilla elastica is present in French Polynesia as well.

Cecropia obtusifolia and C. peltata are invasive tree species that are a problem in Hawai‘i and French Polynesia, respectively. Cecropia obtusifolia is also reported to be invasive on Rarotonga (Cook Islands).

Chromolaena odorata (Siam weed) is a highly invasive pan-tropical weed. It will likely show up in Tonga at some point in time and should be promptly eradicated if found. It has tiny, wind-dispersed seeds that can also travel on boots, clothing or used cars or equipment. Biological controls are available but are most effective in open areas, less so in shaded stands. This species would be a problem for agriculture, as well.

Cinnamomum verum (cinnamon) was found to be very invasive in our survey of American Samoa. It is also present in Samoa, Fiji, French Polynesia and the Cook Islands (Rarotonga). This is the true cinnamon tree, not the bay rum tree (Pimenta racemosa) called "sinamoni" in Tonga.

Clerodendrum chinense (Honolulu rose, known as losa Honolulu in American Samoa and Samoa, where it is a problem) is a shade-tolerant species. It reproduces from root suckers and can form dense thickets, crowding out other species.

Clerodendrum quadriloculare is suspicious because it appears to have the ability to invade intact or relatively intact native forests. This species is notorious for being a prolific producer of root suckers and, in fact, the plant is easily propagated by means of root cuttings. The species is an attractive yard plant and is commonly planted for that purpose. Cultivated specimens were observed in American Samoa and French Polynesia. In Hawai‘i it is becoming a problem ornamental, producing numerous root suckers that appear some distance from the parent plant.

Clidemia hirta (Koster’s curse) is a serious problem species in Hawai‘i and other locations. It was reported by Waterhouse (1997) to be present in Tonga. If so, it is probably on one of the islands that we didn’t survey, as it is a very distinctive plant and easily recognized. This is a very serious weed of the forest understory on many tropical islands and should be immediately eradicated if found. It is present in Samoa, American Samoa and Vanuatu. Other members of the family Melastomataceae (including Arthrostemma ciliatum, Dissotis rotundifolia, Heterocentron subtriplinervium, Medinilla magnifica, Medinilla venosa, Melastoma candidum, Melastoma sanguineum, Memecylon floribundum, Miconia calvescens, Ossaea marginata, Oxyspora paniculata and Tetrazygia bicolor) that are not native to Tonga should also be excluded.

Cryptostegia grandiflora (rubber vine, India rubber vine) is a climbing vine that has become a serious problem in northeastern Queensland, Australia. It is present in New Caledonia and Fiji.

Dodder (Cuscuta campestris) is a threat to agriculture as well as an environmental weed. It is parasitic on a wide range of host plants and is difficult to control. A similar-looking species (Cassythia filiformis) is present but is native throughout the Pacific. Dodder is present in the Cook Islands, Fiji, French Polynesia, Hawai‘i, New Caledonia, Niue and Samoa.

Macfadyena unguis-cati (cat’s claw climber) is an aggressive vine that climbs trees and also forms a dense mat on the ground. Control is difficult because it has tuberous roots and reproduces from pieces and cuttings. It is a problem species in Hawai‘i. It was observed on Niue and is reported to be moderately invasive in New Caledonia (Meyer, 2000). It is often planted as an ornamental.

Maesopsis eminii (musizi, umbrella tree) is a large African tree that has been introduced into other countries as a timber tree. Fruit-eating birds (and possibly fruit bats) spread its seed and it has become a problem in a number of countries. It was introduced as a timber tree to Fiji, where it is starting to naturalize.

Melaleuca quinquenervia (cajeput, paper bark tree) is a native of eastern Australia, New Guinea and New Caledonia. It produces large quantities of wind-dispersed seeds and reproduces profusely after fire or other disturbance. It is a major problem in the State of Florida (US) and is present in Fiji, French Polynesia (Tahiti) and Hawai‘i.

Merremia tuberosa (wood rose), a climbing, smothering vine, is notable for its aggressive behavior on Niue. It is also a problem species in Hawai‘i. It is most likely to be introduced as an ornamental species and spread through discarded cuttings and floral arrangements containing the seeds.

Miconia calvescens (the purple plague, velvetleaf) is undoubtedly the most destructive invasive plant in the Pacific. It has been a disaster to the forest ecosystem of Tahiti in French Polynesia and has subsequently spread to other islands in French Polynesia (Meyer and Florence, 1996). It has also escaped in Hawai‘i and is the subject of an intensive and costly eradication effort there. It recently was discovered in Queensland, Australia, were an eradication project is also under way. This species is an attractive garden plant and might be introduced this way or as tiny seeds on shoes or used equipment.

Mimosa diplotricha [=invisa] (giant sensitive plant, known as vao fefe palagi (American Samoa and Samoa), la'au fefe tele and la'au fefe palagi (Samoa)) is a particularly nasty plant covered with thorns, forming dense tangles that are difficult to walk through. It is present in a number of South Pacific locations (Cook Islands, Fiji, French Polynesia (Society Islands), New Caledonia, Niue, Papua New Guinea, Samoa, Solomon Islands and Vanuatu) but, fortunately, has not yet reached Tonga. It has only recently been introduced to Niue and eradication efforts there are ongoing. It will undoubtedly make its appearance in Tonga sooner or later. Mimosa pudica (sensitive plant, mateloi), a smaller plant with only small prickles, is present throughout Tonga and usually a component of the weedy vegetation of roadsides and other disturbed areas).

Passiflora tarminiana [mollissima] (banana poka, banana passionfruit), a smothering vine that is a problem in Hawai‘i and New Zealand, is also absent. It can smother the forest canopy when the sub-canopy vegetation is disturbed. Other Passaflora species not already present (Passiflora alata, Passiflora caerulea, Passiflora coccinea, Passiflora ligularis, Passiflora pulchella, Passiflora rubra, etc.) should also be excluded.

Pithecellobium dulce (Madras thorn) is a thorny tree that is a problem species in Hawai‘i and is present in New Caledonia, Fiji and French Polynesia (cultivated). The seeds are bird-dispersed.

Psidium cattleianum (strawberry guava) is a small tree that forms dense thickets. It is a major problem species in a number of island ecosystems including Hawai‘i, Fiji, Tahiti and the Cook Islands (Rarotonga and Mangaia) well as La Réunion and Mauritius in the Indian Ocean. Varieties with red and yellow fruits are known. Birds and pigs (and possibly fruit bats as well) disperse the seeds.

All Rubus species (raspberries, blackberries, thimbleberries, brambles) should be excluded. These include Rubus alceifolius, invasive in Australia (Queensland) and La Réunion; Rubus moluccanus, a serious pest of the Mascarine Islands and present in Australia, Fiji, New Caledonia and the Solomon Islands and Rubus rosifolius, very invasive in French Polynesia and Hawai‘i and present in New Caledonia, the Solomon Islands and Vanuatu as well. A number of other Rubus species are invasive. In Hawai‘i, R. argutus (prickly Florida blackberry) and R. niveus (hill or Mysore raspberry) are problems. Introduced Rubus species (in particular, R. niveus) are a major problem in the Galapagos Islands. Some other species that should be excluded are R. ellipticus, R. glaucus and R. sieboldii. In general, where Rubus species are not present on tropical islands, they should not be introduced. If already introduced, they should be evaluated as candidates for eradication. The vines form thorny thickets and the fruits are widely dispersed by birds.

Tibouchina herbacea (glorybush or cane ti) and T. urvilleana (glorybush, lasiandra, princess flower) are major problem species in Hawai‘i.

A number of potentially invasive grass species are not yet present in Tonga, including:

• Pennisetum setaceum (fountain grass), a grass that is a major problem in Hawai‘i. It is present in Fiji and French Polynesia. This species is often sold as an ornamental through seed catalogues and so could be imported into Tonga via the mail system.

• Imperata cylindrica (cogon grass) was reported to be present in Tonga by Yuncker (1959), but is not reported by other authors and was most likely I. conferta. I. cylindrica is a very invasive species elsewhere and a problem for agriculture as well as the environment.

• A number of other grass species of various degrees of invasiveness, including Andropogon gayanus, Andropogon glomeratus, Andropogon virginicus, Axonopus compressus, Cenchrus brownii, Cenchrus ciliaris, Cortaderia jubata, Cortaderia selloana, Digitaria insularis, Echinochloa polystachya, Hymenachne amplexicaulis, Hyparrhenia rufa, Ischaemum polystachyum, Ischaemum rugosum, Ischaemum timorense, Microlaena stipoides, Nassella cernua, Panicum repens, Paspalum urvillei, Pennisetum clandestinum, Pennisetum polystachyon, Pennisetum purpureum, Phyllostachys nigra, Saccharum spontaneum, Schizachyrium condensatum, Setaria palmifolia, Sporobolus elongates and Tripsacum latifolium. Grasses are easily introduced as contaminants in imported seed, imported sand and gravel or on used machinery, and by their nature tend to be invasive.

The best indicator that a species might be invasive is the fact that it is invasive elsewhere. However, each island ecosystem is unique and invasiveness cannot be predicted with certainty. A good strategy is to be extremely cautious and exclude these and other species known to be invasive or weedy elsewhere (although the best strategy is to exclude all species not shown by risk assessment to be of acceptable risk). The known problem species that have the potential to cause problems in tropical island ecosystems and are not yet present in Tonga are listed in Appendix 2, Table 1. These species should be excluded through plant quarantine and, if establishment is detected, promptly evaluated for eradication. In addition, species that are reported to be present in American Samoa, the Cook Islands, Fiji, French Polynesia, Hawai‘i, Niue and Samoa but are not present in Tonga are listed in Appendix 4. These species would be of high risk of introduction from any air and ship traffic between these points and Tonga.

2. Species that are invasive elsewhere and are likewise invasive or have the potential to become so on Tonga

Some known invasive plants that are causing trouble in similar ecosystems have been introduced into Tonga (Appendix 2, Table 2). Some of them are already causing problems while others are not. Some are cultivated plants that have not (yet) escaped and their potential for causing damage is so far unknown. However, one of the best predictors of invasiveness is the behavior of the species elsewhere, and these are known troublemakers.

Adenanthera pavonina (lopa, coral bean tree), invasive in secondary forests throughout the Pacific, is fairly widespread (and not native) in Tonga. It is quite invasive in American Samoa. Trees produce large quantities of seed and the tree will grow on a variety of soils. It may have not yet reached its full potential in Tonga. Although the seeds are eaten and many people consider it native, it was introduced from Southeast Asia and Malesia. Coral bean has the ability to tower above the native canopy (as on Mt. Talau) and eventually form monospecific stands. Its presence should be monitored and it may be appropriate to remove it from sensitive and protected areas such as the national parks.

Asparagus setaceus (taupo ‘ou, ornamental asparagus) was present on all the islands visited and, because birds spread its seeds, is probably prevalent throughout the archipelago. On Lifuka/Foa and ‘Uiha it has spread widely through the forest, both in the understory and climbing into the canopy. This species is a weed in Hawai‘i. Asparagus densiflorus was also noted in cultivation.

Coccinia grandis (ivy or scarlet gourd) is a smothering vine that is showing potential for serious damage to the forests of Saipan. The vines climb over trees and form such dense cover that the forest underneath is completely shaded out and destroyed. It is also invasive in Guam and Hawai‘i and is reportedly present in Fiji and Vanuatu. It is a vegetable commonly used in southeastern Asian cooking and the plant is often introduced for that reason. The variety seen on Tongatapu and ‘Eua may be a horticultural variety and less prone to spread, but nevertheless it is a plant of some concern. Since it is apparently only present on Tongatapu and ‘Eua, introduction to other islands should be discouraged.

Cordia alliodora (kotia, Ecuador laurel, salmwood) was introduced to Tonga as a forestry tree. It was similarly introduced into Vanuatu and has become a pest there (Tolfts, 1997). It is spreading where it is present in Tonga (Tongatapu, ‘Eua and Vava‘u). The value of Cordia as a tool for reforestation of badly degraded areas with little or no native vegetation is recognized. But this species should not be introduced into areas of intact native forest and its presence should be monitored in areas of special concern. If found in these areas (i.e. national parks) efforts should be made to remove it.

Dieffenbachia seguine (spotted dieffenbachia or dumb cane), a common house and yard plant, was noted on ‘Eua and may be under cultivation elsewhere. This species is a problem in American Samoa and is reportedly present in the Cook Islands, Fiji and French Polynesia. If found in natural areas, such as the National Park, this species should be promptly eradicated, as it reproduces vegetatively and can thrive in the dense shade of an intact native forest canopy.

Flemingia strobilifera (luck plant) was seen on Tongatapu, ‘Eua, Vava‘u and Lifuka/Foa. This species is a prolific seed producer and can form dense thickets. It is invasive in French Polynesia and Hawai‘i and is beginning to naturalize in Tonga. It has the potential to become a serious problem. A closely related species, F. macrophylla, was noted in cultivation on Vava‘u and at the MAF agricultural station on Tongatapu. Both species have naturalized in Jamaica and F. macrophylla is showing signs of naturalizing where it has been planted in American Samoa. These species should not be planted more widely or introduced to islands where they are not present.

Grevillea robusta (oke‘, silk oak) was noted on Tongatapu, ‘Eua and Vava‘u. This tree is commonly introduced as an ornamental and for forestry plantings. It has become a pest in Hawai‘i and is naturalized and starting to spread on the island of Rurutu in the Austral Archipelago, French Polynesia.

Hemigraphis alternata (metal leaf, red ivy) was seen at several locations. This species is shade tolerant and will spread in the forest understory. Two other species with similar behavior are Tradescantia spathacea (faina kula, oyster plant, boat plant, boat lily, moses in a boat) and T. zebrina (wandering jew). These species are widely planted in Tonga, both in gardens and as ornamentals in cemeteries. They are commonly found along roadsides where cuttings have apparently been dumped.

Indigofera suffruticosa (‘akau veli, indigo) is a major component of the weed vegetation of Tonga, the worst seen in the Pacific so far.

Several Ligustrum plants (L. sinense?) were noted in cultivation in a yard on Hala Fatafehi in Nuku‘alofa. Privets are notorious invaders and it would be desirable to eradicate any examples and exclude them in the future.

Lonicera japonica (Japanese honeysuckle) was seen in cultivation on Tongatapu, Vava‘u and Lifuka/Foa. This species is a serious pest in a number of countries and is on the New Zealand noxious weed list and banned from sale in that country. It can be spread both by birds and cuttings.

Melia azedarach (sita, Chinaberry) is a prolific producer of bird-dispersed seed. It is a problem species in Hawai‘i and some islands in French Polynesia (Mangareva, Akamaru, Taravai) (Meyer, 1998) and is invasive in South Africa. There is the potential that it could become much more plentiful in Tonga.

Melinis minutiflora (puakatau, molasses grass) is a species that is both invasive and can cause a serious fire hazard. It has fostered the establishment of fire regimes on many islands where it has been introduced. It is present but uncommon on Tongatapu and Vava‘u.

Mikania micrantha (mile-a-minute weed), a smothering vine, is a fairly recent introduction to Tonga (Tongatapu). It is spread both by seed (dispersed by wind or in clothing or hair of animals) and vegetatively from broken stem fragments. Each node of the stem can produce roots. This species is becoming widespread throughout the Pacific and is a pest wherever it occurs.

Panicum maximum (saafa, Guinea grass, buffalograss) is the most troublesome invasive grass in Tonga. It forms dense stands up to 2 m tall. The seeds are dispersed by wind, and it can survive long periods of drought. It spreads by seed and locally from underground rhizomes.

Paraserianthes falcataria [Falcataria moluccana] (Molucca albizia), known as tamaligi palagi in American Samoa and tamaligi uliuli in Samoa, is planted on ‘Eua and a number of specimens were seen on Vava‘u. This species has the potential, over time, of spreading more widely (as can be seen in Hawai‘i, Pohnpei and Tahiti). If this is not a desired species, further planting can be discouraged and emphasis given to cutting the existing trees.

Pimenta dioica (sipaisi, allspice, pimento) is widely planted and naturalized in Tonga. We were shown one area on ‘Eua where it had naturalized as a dense thicket of saplings. Given its demonstrated aggressive nature this species may well become a major problem in the future. The use of allspice as a desirable medicinal plant is recognized but its presence is areas of intact native forest should be monitored and efforts made to eradicate it from these areas. The seeds are bird-dispersed.

Piper auritum ("kava Hawai‘i" in Tonga, eared pepper, also called "false kava") has been introduced to Pacific islands as a fast-growing form of kava, but it is worthless in this regard. It is presently subject of an eradication campaign on the island of Pohnpei, Federated States of Micronesia. It has also been introduced into Hawai‘i but the local kava growers association is working with the authorities to eradicate it there as well. See also SPC Pest Alert No. 19, False Kava. This species suckers profusely and produces many small seeds that can be spread by birds, rodents and bats and can also be introduced into new areas on machinery. Locally, it spreads by suckers, forming large clumps. Piper auritum was noted on all the islands visited (Tongatapu, ‘Eua, Vava‘u, Lifuka/Foa, Ha‘ano and ‘Uiha) and is probably present in other locations as well. This species appears to be well on its way to becoming a major pest and should be evaluated for eradication. A special effort should be made to eradicate this serious invader from all park or reserve lands. It should be noted that it was seen in the National Park on ‘Eua.

Pluchea carolinensis (sour bush) was seen on Tongatapu between Alaivahamama and Taufa‘ahau roads and the lagoon as far south as the Tongan National Centre. The seeds are spread by wind and possibly by birds. It is a widespread pest species in Hawai‘i, commonly invading almost every habitat type. Since the infestation on Tongatapu appears to be limited to a relatively small area, eradication should be attempted as soon as possible and the area monitored for seedlings.

The dense stands of Psidium guajava (kuava, guava), especially on ‘Eua, were the worst seen so far in the Pacific. This is a major invasive species in the Galapagos Islands and a problem in French Polynesia (Marquesas Islands), New Caledonia, Hawai‘i and Fiji as well. Frugivorous birds, as well as rats and feral pigs, disperse the seeds.

Solanum capsicoides was seen on ‘Eua and Vava‘u. Although small, it is quite spiny and would not be a desirable addition to the vegetation of Tonga. The specimens we saw were producing large amounts of small, tomato-like fruit. Spread may be by birds or pigs or by humans who use the fruit in lei making. This species would be a good candidate for eradication.

Solanum mauritianum (pula, bugweed, wild tobacco, tree tobacco) is quite prevalent throughout Tonga. It is a noxious weed in South Africa (Henderson, 1995) and is reported to be moderately invasive on Rarotonga, Cook Islands (Meyer, 2000).

Solanum torvum (prickly solanum, devil's fig) is a large spiny species of disturbed areas and old fields that forms dense, impenetrable thickets. A fairly recent introduction according to local sources, it is becoming quite prevalent on Tongatapu and Vava‘u. Its seeds are bird-spread. A leaf-eating chrysomelid beetle, Leptinotarsa undecimlineata, is reported to be host-specific and might be a useful control agent. (Waterhouse and Norris, 1987).

Spathodea campanulata (African tulip tree) is occasionally seen, mostly planted as an ornamental. This tree has become a major problem in Fiji, the Hawaiian Islands and some other places. The seeds are wind-dispersed and it also propagates from root suckers and cuttings. Large trees do not stand up well to wind. Further planting of this species should be discouraged and existing trees monitored for spread. The possibility of biological control is being investigated in Fiji.

Syngonium angustatum (arrowhead plant, goosefoot plant), a climbing aroid, is mostly cultivated in Tonga, but a few escapes were noted. This species has the ability to spread in the deep shade of intact forests, forming a dense mat on the forest floor as well as climbing trees. It is difficult to eradicate as it is able to reproduce from a single node and bits and pieces of the stems or roots are easily overlooked. It spreads from dumped cuttings. It is a problem species in American Samoa, is widespread in Hawai‘i and is quite invasive in Niue.

Tecoma stans (piti, yellow bells, yellow-elder, yellow trumpetbush) is a serious invader of disturbed areas in Tonga, as in French Polynesia. It grows in dense stands, commonly with other weedy species. The seeds are wind-dispersed.

Wedelia [Sphagneticola] trilobata (Singapore daisy) has become a serious pest on many Pacific islands (Thaman, 1999) as well as in Australia. On Tonga, it was noted in cultivation at the entrance to the Paradise Shores Resort on the northwest side of Tongatapu Island and a few other locations. It forms dense mats along roadsides and in disturbed areas and is a problem in agriculture. Control by chemical means is difficult and mechanical removal often leaves numerous nodes that freely root and rapidly spread. Eradication will involve several visits to the site for follow-up action. Given its limited distribution and proven invasive nature, it is an obvious candidate for eradication.

Other common aggressive invaders include Lantana camara (talatala, talatala talmoa), Stachytarpheta cayennensis [urticifolia] (hiki ‘i kuma, ‘iku ‘i kuma, blue rat’s tail, blue mouse’s tail) and Mimosa pudica (mateloi, sensitive plant), often growing in a mixture with Indigofera suffruticosa, Psidium guajava, Solanum mauritianum, Tecoma stans and other weedy species. These plants commonly form dense thickets along roadsides and in disturbed areas.

3. Species that are mentioned or listed as being weedy or invasive elsewhere and are common or weedy in Tonga

A large number of other common or weedy introduced species were noted. Many of these species, which might best be termed aggressive weeds, are mostly prevalent along roadsides or on disturbed sites, although some species, particularly alien trees, can gradually spread into forested ecosystems. In the case of vines and plants that form dense ground cover, the regeneration of native species can be inhibited. Some of these species could become a problem in the future, since there is often a long lag time between introduction and when a species begins to cause serious impacts. These species (listed in Appendix 2, Table 3) should be monitored for spread and possible control measures, if necessary.

Antigonon leptopus (chain of hearts), a climbing vine often planted as an ornamental, is a widespread pest on Guam. Only a few examples, mostly in cultivation, were seen on Tonga.

Bryophyllum pinnatum [Kalanchoë pinnata] (pipi vao, life plant, air plant) was seen both in cultivation and naturalized. It reproduces vegetatively and can be invasive on the forest floor.

Cestrum nocturnum (lakau po'uli, night-flowering cestrum) is quite prevalent and weedy. Cestrum diurnum (vaitohi, day cestrum) is also present and was found naturalizing along coastal areas of Vava‘u.

Cyperus involucratus [=alternifolius] (umbrella sedge) is very prevalent in wet areas, especially on Tongatapu.

Hyptis pectinata (mint weed, comb hyptis) is a very common pan-tropical weed. It is listed as a noxious weed in Fiji and Hawai‘i.

Jatropha curcas (fiki, physic nut) is widely used as a living fence throughout Tonga, from whence it escapes and naturalizes.

Momordica charantia (meleni ‘ae kuma, vaine ‘initia, bitter-melon, balsam pear), a member of the cucumber family, is a climbing vine and a rather unimportant component of the weed vegetation. However, its fruit can be a host for fruit flies.

Murraya paniculata (orange jessamine, satin-wood, Chinese box) is widely planted as a hedge and ornamental plant throughout Tonga. While it has the potential to naturalize (its seeds are probably bird-dispersed), no spread was noted so far.

Passiflora foetida (love-in-a-mist) was fairly widespread. It is quite prevalent on most Pacific islands.

Stachytarpheta cayennensis [urticifolia] (blue rat’s tail) is ubiquitous, growing up to 2 m tall; Stachytarpheta jamaicensis is present in some locations as well. Both species are widespread in the Pacific.

Some exotic tree species that have been introduced to Tonga and might spread more widely include Acacia auriculiformis (earleaf acacia), Cedrela odorata (sita hina, cigar box cedar, Mexican cedar), Ceiba pentandra (kapok), Delonix regia (flame tree), Gliricidia sepium (mother of cacao, quickstick), Samanea saman (monkeypod), Thevetia peruviana (yellow oleander) and Toona ciliata (sita kula, Australian red cedar). Bauhinia monandra (orchid tree) is widespread and Leucaena leucocephala (siale mohemohe) is everywhere present, as is the case throughout the Pacific region.

A number of introduced grasses are established, including Arundo donax (kaho folalahi, giant reed), Axonopus fissifolius (narrow-leaved carpetgrass), Bothriochloa bladhii, (blue grass, Australian beardgrass),  Cenchrus echinatus (hefa, bur grass); Chloris barbata, (swollen fingergrass), Chloris radiata (plush-grass, radiate fingergrass), Chrysopogon aciculatus (matapekepeka, Mackie’s pest, lovegrass), Digitaria ciliaris (fingergrass, smooth crabgrass), Digitaria violascens (smooth crabgrass, violet crabgrass), Eleusine indica (takataka, takataka ‘a leala, mohuku siamane, goosegrass), Melinis repens (salapona, Natal grass), Paspalum conjugatum (vailima, T grass), Paspalum dilatatum (dallis grass), Paspalum fimbriatum (fimbriate paspalum), Paspalum paniculatum (Russell river grass, galmarra grass), Paspalum scrobiculatum (rice grass), Setaria pumila (foxtail), Sorghum halepense (kola, Johnson grass) Sporobolus indicus (fisihina, smutgrass, wiregrass, Indian dropseed), Urochloa mutica (puakatau, para grass) and Urochloa subquadripara (green summer grass).

Other weedy species include Annona muricata (‘apele ‘initia, soursop), Annona squamosa (‘apele papalangi, ‘apele Tonga, sweetsop, custard apple), Bidens pilosa (beggar’s tick), Canna indica (misimisi, canna lily), Centrosema pubescens (centro), Chamaecrista nictitans (partridge pea, Japanese tea senna), Clerodendrum buchananii (amo‘ula, red clerodendrum, pagoda-flower), Costus speciosus (crepe ginger), Crassocephalum crepidoides (fisi puna, thickhead, fireweed), Desmodium tortuosum (Florida beggerweed), Furcraea foetida (fau malila, Mauritius hemp), Ipomoea cairica (maile miniti) and I. hederifolia, Lablab purpureus (pini lae puaka, hyacinth bean, bonavist), Leonurus japonicus (Lion’s tail), Macroptilium atropurpureum (siratro, purple bushbean), Neonotonia wightii (glycine), Pueraria montana var. lobata (akataha, fue‘aepuaka, kudzu), Ricinus communis (lepo, lepohina, castor bean), Sambucus mexicana (elderberry, Mexican elder), Senna [Cassia] tora (te‘epulu, tengafefeka, peanut weed), Solenostemon [Plectranthus] scutellarioides (coleus, known as pate or patiale in Samoa), Thunbergia fragrans (fue hina, white lady, white thunbergia, sweet clock-vine), Triumfetta rhomboidea (mo‘osipo, Chinese burr, paroquet burr), Triumfetta semitriloba (Sacramento bur) and Urena lobata (mo‘osipo Tonga, hibiscus burr).

Species that are cultivated or of limited extent but have the potential to become more widespread include Allamanda cathartica (yellow trumpet vine), Breynia disticha (snowbush), Duranta erecta (‘olive, golden dewdrop), Eugenia uniflora (Surinam cherry, kafika palangi on Niue), Heliconia psittacorum, Odontonema tubaeforme (fire spike, cardinal flower), Schefflera actinophylla (octopus tree, umbrella tree), Schefflera arboricola (dwarf brassia, dwarf schefflera), Senna [Cassia] alata (la‘au fai lafa, candle bush), Sesbania grandiflora (hummingbird tree, scarlet wisteria tree, known as sepania in Samoa), Syzygium jambos (fekika papalangi, malabar plum, rose apple), Thunbergia alata (black-eyed susan vine, known as tagamimi in Samoa), Tillandsia usneoides (old man’s beard, Spanish moss) and Tithonia diversifolia (tree marigold, known as matala on Niue).

4. Native and naturalized species exhibiting aggressive behavior

Merremia peltata (fue mea), a native or early introduction, is quite invasive along forest edges where there has been disturbance, but its extent seems to be limited on Tonga, unlike many Pacific islands. The fact that it only occurs on some islands in the archipelago (we saw it only on ‘Eua and Vava‘u and it is reported to be present on Tafahi) is interesting and gives rise to the suspicion that the introduction might be more modern, but Whistler (1988) indicates that it was recorded from Tonga in 1890. In any case, given its aggressive nature, eradication of any introductions to non-infested islands would seem to be warranted.

Strategies for dealing with invasive species

It was not our purpose to perform a review of quarantine operations and other methodologies for excluding and managing invasive species. Rather, the following strategies are general operational principles that have proven effective in dealing with exotic pests.

The first line of defense against invasive species, and the most cost-effective, is to keep them out. Control at ports of entry is essential, and those concerned with the protection of natural ecosystems should work closely with plant protection and quarantine officials to combat known and potential invasive plant species. Plant quarantine officers should be familiar with both agricultural pests and those that threaten wildland ecosystems. At a minimum, a list of known noxious species to be excluded should be developed and exclusion of these species should be backed by the force of law and regulation. Better yet is to utilize the "precautionary principle" (now used by Australia and New Zealand and under serious consideration by a number of other counties) to exclude all alien species not shown to be of acceptable risk. Risk assessment and management techniques can be used to assess the likelihood and effects of possible introductions and to develop exclusion and eradication strategies. Tonga is fortunate to have effective quarantine measures in place and operating.

Tonga might consider setting up an invasive species committee consisting of concerned government agencies, organizations and individuals. Close and immediate coordination and cooperation between various government departments is essential when an invasive species problem is encountered, especially when there is a need to move quickly to eradicate an introduced species. Such a committee can be effective both for long-term strategic actions, such as review and strengthening of relevant laws and regulations, as well as short-term tactical and operational problems, such as action when a new species is found to have been introduced. The committee could also draw up a prioritized action plan. This would include critical areas to be protected and species that might lend themselves to control or eradication. Time, money and people are always in limited supply and must be directed to the places where they will do the most good. Some recommendations are made below as to possible management actions against some individual plant species, but these should be tested these against available resources and other priorities.

Education of the public about the danger of introductions and encouraging the use of native species needs to continue. People need to be encouraged to take responsible actions such as following quarantine regulations, not dumping garden cuttings in the woods and reporting suspicious plants. There are many instances where an invasive plant started out as a pretty flower planted in a yard or garden. Public service announcements on television or radio can be used and "wanted" posters can be prepared for critical species. Education of schoolchildren is especially important, as this is the most impressionable age. Children can also have a notable effect on the actions of their parents. Prompt follow-up to public reports and inquiries is essential to maintain the credibility of a public education program.

The public also needs to understand that the immediate eradication of a small area of a problem species, even if it involves the use of pesticides, may be better than living with a problem species forever. There are many instances where you hear people say, "I wish we'd taken action when this pest was first noticed". It may even be worthwhile to take people to a place where they can be shown the full extent of the problem if the infestation is allowed to spread so that they will understand, accept and support eradication. For example, anyone visiting Tahiti would very likely come away convinced that Miconia calvescens is an ecological disaster and that it should be prevented from becoming established on other Pacific islands. The public should be informed and involved in any proposed control or eradication actions.

Local nurseries, botanical gardens or plant importers can be sources of new introductions. A positive approach is to work together to develop a "white list" of both native and non-native species that the public can be encouraged to plant.

Foresters, conservation officers, extension agents and others that spend time in the field should be alert to new species that exhibit invasive behavior. Often, these species first show up in urban or farm areas because they are usually introduced by people and tend to first become established in gardens and disturbed areas. Suspicious plant species should be promptly reported. Periodically scheduled surveys can also be conducted for new or expanding infestations. An evaluation should be conducted for any new species that appears to be invasive or is known to be invasive elsewhere. Assistance by an expert who is familiar with the species and methods for its eradication or control should be requested if needed. Prompt action is essential, since once a species becomes widespread, control or eradication can be extremely costly or impossible. Assistance is also available on-line from experts through the Pacific Pestnet and Aliens list-servers.

Tonga appears to have adequate laws and regulations in place to deal with quarantine and new introductions, but it would be appropriate to periodically review them to see if they might need to be strengthened. In the case of Tonga, where most land is privately owned, the ability of government to require the control of noxious species on private lands or to take action on private lands if the landowner cannot be located or does not take prompt action is essential. Provision for emergency response procedures and funds to deal with immediate problems should also be in place. New Zealand and some of its town councils have strong laws and regulations that can be used as models.

Recommendations

In addition to the above general strategies, we offer the following specific recommendations:

• Make every effort to keep out all the species listed in Appendix 2, Table 1. All of these are known invasive species elsewhere, and there is no sense in running the risk that they will act the same in Tonga. As we are able to gather information on other species that might threaten tropical island ecosystems we will add them to the database. For those with Internet access, additional information and new listings can be found at http://www.hear.org/pier/. The information is also available on the PIER-CD for local use. Assistance from experts in identifying and managing invasive species can also be obtained by subscribing and posting inquiries to the Aliens and Pacific Pestnet list servers.

• Take special measures to keep Castilla elastica, Funtumia elastica, Cecropia obtusifolia and C. peltata, Chromolaena odorata, Cinnamomum verum, Clerodendrum chinense, Clerodendrum quadriloculare, Clidemia hirta and other members of the family Melastomataceae, Cryptostegia grandiflora, Cuscuta campestris, Macfadyena unguis-cati, Maesopsis eminii, Melaleuca quinquenervia, Merremia tuberosa, Mimosa diplotricha, Passiflora tarminiana [mollissima] and other Passaflora species not already present, Pithecellobium dulce, Psidium cattleianum, all Rubus species, Tibouchina herbacea and T. urvilleana, Pennisetum setaceum, Imperata cylindrica and other grass species out, to monitor for their occurrence, and to eradicate them immediately if found. These are all well-documented problem species that have had a major impact on natural ecosystems elsewhere. The potential impact of these species, if they are introduced and become established, is very severe.

• Extraordinary measures need to be employed against Miconia calvescens because its effect on Tonga’s ecosystems would be so devastating if introduced. At a minimum, quarantine officers should be alert to people who might have been in the woods or rural areas in French Polynesia (particularly the islands of Tahiti, Moorea, Raiatea and Tahaa) or Hawai‘i (especially the island of Hawai‘i), and inspect their shoes or boots for seeds. Any trucks or equipment coming from French Polynesia or Hawai‘i, particularly those that have been used in rural areas, must be power washed or steam cleaned. Any infestations picked up from public reporting or scouting should be promptly eradicated before the plants set seed.

• Coccinia grandis (ivy gourd, scarlet gourd) is apparently only present on Tongatapu and ‘Eua and may be a horticultural variety with weak invasive tendencies. However, given the species’ behavior on Saipan, it would be good policy to discourage its spread to other islands. Import of seed or cuttings should be prohibited to prevent introduction of more virulent varieties. Agricultural extension officers should discourage further planting of the species on Tongatapu and ‘Eua and existing plantings should be monitored for spread and evaluated for control or eradication if naturalization begins to take place.

• Cordia alliodora (kotia), while a desirable forestry tree, is readily naturalizing on Tongatapu, ‘Eua and Vava‘u. This species should not be introduced to locations where it is not wanted.

• Several shade-loving species, particularly Dieffenbachia seguine, Hemigraphis alternata, Syngonium angustatum, Tradescantia spathacea and Tradescantia zebrina have become established in forested areas. On other Pacific islands dense stands of these species have been noted, crowding out other species. At present, the infestations in Tonga are mostly along roadsides, apparently as a result of dumping of plants or cuttings. Serious consideration should be given to controlling these infestations in natural and sensitive areas, such as the National Parks. The public should be encouraged, through education, not to dump garden cuttings and waste.

• Flemingia strobilifera (luck plant) and F. macrophylla are potentially problem species. Further planting of F. strobilifera should be discouraged (especially introduction to islands where it is not present) and it should be controlled in natural and sensitive areas. F. macrophylla was only noted in cultivation (in the MAF agricultural station on Tongatapu and in the forestry nursery/arboretum on Vava‘u) and it might be best just to eliminate it.

• Grevillea robusta should be closely monitored for spread and not introduced to additional locations without due consideration of its invasive tendencies.

• The tentative identification of the Ligustrum plants seen in cultivation on Hala Fatafehi in Nuku‘alofa should be confirmed. If these are isolated examples, the owner should be requested to eliminate them.

• Lonicera japonica (Japanese honeysuckle) is a dangerous species that was seen only in cultivation. If the owners of these plants could be persuaded to give them up, it would eliminate the danger from this plant.

• Discourage further planting of Paraserianthes falcataria [Falcataria moluccana], especially near natural or sensitive areas. While it spreads only slowly, it has managed to become widespread on several Pacific islands.

• Given its invasive tendencies, Piper auritum ("kava Hawai‘i") is likely to become even more widespread in the future. The species should be targeted for control or eradication, perhaps by encouraging farmers and landowners to destroy it.

• Pluchea carolinensis (sour bush) is a known serious pest with a limited distribution in Tonga, as far as could be determined (between Alaivahamama road and Taufa‘ahau Road and the lagoon as far south as the Tongan National Centre). Its seeds are spread by the wind and possibly by birds. It should be evaluated for eradication.

• Likewise, the small, spiny Solanum capsicoides seen on ‘Eua and Vava‘u is an excellent candidate for eradication. It is likely to become a problem for agriculture if it becomes widespread.

• The effectiveness and feasibility of introduction of the biological control agent for Solanum torvum (prickly solanum, devil's fig) should be investigated.

• Planting of Spathodea campanulata (African tulip tree) as an ornamental should be discouraged and existing trees monitored for spread.

• Wedelia [Sphagneticola] trilobata is just getting a foothold on Tonga. If unchecked, it will spread rapidly throughout the island, becoming a problem along roadsides, in gardens and fields, or any other open, disturbed areas. Serious consideration should be given to an eradication campaign against this species.

• Be very cautious in introducing new grasses, as many of them are aggressive invaders. Also, grass seed is invariably contaminated with other, possibly invasive, grasses or weeds.

• Much of the land in Tonga is farmland or secondary forest where it is difficult to keep weedy species out. However, there are areas such as Mt. Talau National Park on Vava‘u and ‘Eua National Park where it may be desirable to keep them relatively free of invasive species for the protection of native biodiversity and tourism values. Where the forest canopy can be kept intact, this can generally be done at low cost. Keeping out shade-tolerant species that can invade closed forests is the biggest problem. Where it is necessary to remove invasive trees such as Adenanthera pavonina (lopa) that have invaded the forest, the public should be informed about the purpose of the cutting. Intact native forests are the most resistant to invasion. Any measures that limit the amount of disturbance will help keep invasive species out.

• Species listed in Appendix 2, Table 3 are pests and, although they may not be presently causing serious damage to natural ecosystems, they are certainly not desirable species. They should not be more widely distributed.

• A number of species used in ornamental plantings are, at least to some degree, invasive. While many of these species have desirable ornamental or physical characteristics, planting exotics as opposed to native species is a policy question that needs to be carefully considered. Most species that naturalize to any extent will gradually spread throughout available and suitable habitat, given enough time. Existing plantings should be monitored for spread and new introductions should be carefully assessed to minimize risk.

• Enforce steam cleaning or power washing of all used cars, trucks and equipment coming into Tonga. This is desirable not only from the standpoint of excluding invasive plants, but also insects and diseases. Especially suspect are roadbuilding machinery, military equipment and off-road vehicles. Inspectors should be especially vigilant to make sure used cars, trucks and equipment from Southeast Asia, Fiji, Hawai‘i and the Philippines are clean as many dangerous weeds are present in these locations. For instance, seeds of Chromolaena odorata can be transported in the radiators of used cars, trucks and other equipment.

• Closely inspect boots, camping equipment and other materials for soil and seeds, particularly when they have been used in countries where Chromolaena odorata, Miconia calvescens, Mimosa diplotricha and other small-seeded species are present.

• A risk-rating scheme would be useful in evaluating the various known and potential invasive species for their risk of introduction, spread and potential damage. The Institute of Pacific Islands Forestry is looking at the possibility of adapting the Australian risk assessment system to the needs of the Pacific. If so, our hope is to eventually rate all the species listed in the PIER database and provide this information to Pacific island countries. In the meantime, risk assessments can be performed on individual species to help evaluate their potential for invasion and spread.

• The purpose of this survey was to give an overall assessment of the situation. Individual species of concern should be more fully evaluated as needed as to extent, invasiveness, and the possibility for control or eradication. Technical assistance should be requested, if needed, to evaluate individual species.

Appendix 1.  Background material and references

Appendix 2.  Species by category

Appendix 3.  Other invasive plant species, mostly of agricultural concern, reported to be present in Tonga

Appendix 4. Invasive species present in American Samoa, the Cook Islands, Fiji, French Polynesia, Hawai‘i, Niue or Samoa but not present in Tonga

Appendix 5.   Scientific name synonyms

Appendix 6.   Invasive and potentially invasive species present on Tongatapu

Appendix 7. Invasive and potentially invasive species present on ‘Eua

Appendix 8.   Invasive and potentially invasive species present on Vava‘u

Appendix 9.   Invasive and potentially invasive species present in the Ha‘apai island group

Appendix 10.   Invasive and potentially invasive species recorded as being present elsewhere in Tonga

(2)  We would like to express our appreciation for the hospitality, assistance and support of the following members of the Ministry of Agriculture and Forestry without whose help this survey would not have been possible: Haniteli Fa‘anunu (Director), Sione Foliaki, Taniela Hoponoa, Sione Kaufusi, Leody Vainikolo, Taniela Foliaki and Vaea ‘Anitoni.

Need more info? Have questions? Comments? Information to contribute? Contact PIER!

[ Return to PIER homepage ]

This page new 20 October 2001.