Biology and Impacts of Pacific Island Invasive Species. 15. Psittacula krameri, the Rose-Ringed Parakeet (Psittaciformes: Psittacidae)1

Aaron B. Shiels,2,4 and Nicholas P. Kalodimos3

Abstract: The rose-ringed parakeet (RRP), Psittacula krameri, has become established in at least four Pacific Island countries (Hong Kong China, Japan, New Zealand, U.S.A.), including the Hawaiian islands of Kaua‘i, O‘ahu, and Hawai‘i. Most Pacific islands are at risk of RRP colonization. This species was first introduced to Hong Kong in 1903 and Hawai‘i in the 1930s–1960s, established since 1969 in Japan, and in New Zealand since 2005 where it has repeatedly established after organized removals. The founding birds were imported cage-birds from the pet trade. In native India, RRP are generally found associated with human habitation and are considered a severe agricultural pest. In the Hawaiian Islands, RRP are increasing and expanding their geographic ranges below 500 m elevation. Population estimates in 2018 on Kaua‘i were ∼6,800 birds, which was a three-fold increase and a 22.5% annual growth rate in the prior 6 years, whereas O‘ahu had ∼4,560 birds with a 21% annual growth rate the prior 9 years; these rates suggest a population doubling time of ∼3.5 years. Wild RRP can live 14+ years, can reproduce after 1.5 years, and have few effective predators. Breeding pairs produce 1–3 fledglings annually. RRP are seed predators and rarely seed dispersers; their flock-foraging behavior can result in severe damage to orchard and field agricultural crops including tropical fruit and corn (Zea mays), and such economic damages are especially pronounced on Kaua‘i. Island societies should prevent new introductions and consider RRP deterrents and population control methods to protect resources.

Keywords: agricultural crop and orchard loss, economic costs, human-wildlife interaction, parrot damage management, pest eradication, pet trade, ring-necked parakeet, urban bird ecology

PSITTACULA KRAMERI, THE ROSE-RINGED or ring- necked parakeet, has been introduced to over 1Manuscript accepted 13 August 2019. 2 40 countries, gaining its status as the most USDA, APHIS, National Wildlife Research Center, widely introduced parrot in the world. 4101 LaPorte Avenue, Fort Collins, CO 80521, USA. 3University of Hawai‘i at Manoa, 1910 East West Although regarded as a strikingly beautiful Road, Honolulu, HI 96822, USA. bird by many people, this species is a severe 4Corresponding author (e-mail: aaron.b.shiels@usda. agricultural pest that establishes and repro- gov). duces rapidly, aggregates nightly in large roosts in human-inhabited areas, and is Pacific Science (2019), vol. 73, no. 4:421–449 disruptive to humans via its noise and fecal doi:10.2984/73.4.1 pollution. Aside from a few locations in This article was created by a U.S. government northern Europe where some of the public employee and is on the Public Domain. Public domain cherish this species in the wild (Menchetti information may be freely distributed and copied, but it is requested that in any subsequent use the U.S. et al. 2016, Crowley et al. 2019), rose-ringed Department of Agriculture and the journal, Pacific parakeets are generally unwanted invaders Science, be given appropriate acknowledgment. where they have established.

421 422 PACIFIC SCIENCE • October 2019

Rose-ringed parakeets are native to the parakeet, and Senegal long-tailed parakeet. tropical and subtropical parts of Africa and The official American Ornithologists Union Asia, but they have established, expanded, and common name is rose-ringed parakeet. persisted in both tropical and temperate The rose-ringed parakeet (RRP) is classi- regions of the world (Butler 2005). On Pacific fied into four geographically distinctive sub- islands, four countries (Hong Kong China, species of which only the South Asian Japan, New Zealand, and U.S.A.) have subspecies, P. krameri manillensis (Bechstein, established rose-ringed parakeet populations; 1800) and P. k. borealis (Neumann, 1915), are however, most islands in the Pacific— thought to be naturalized in the State of specifically islands from southern Canada to Hawai‘i(Pyle and Pyle 2009, N.P.K., pers. central Chile in the eastern Pacific, to islands obs.), Hong Kong, Japan, and New Zealand; from northern Japan to central New Zealand genetically these two subspecies are the most in the western Pacific—are potentially at risk represented across worldwide RRP popula- of rose-ringed parakeet colonization. The tions (Strubbe and Matthysen 2009a), and climate and latitudinal ranges (ca. 56° Nto their predominance correlates with the higher 44° S) of these Pacific islands match the ranges numbers transported outside their native where they have established elsewhere in the range via the pet trade (Jackson et al. 2015). world. Islands with human habitation are most These South Asian subspecies are not easily at risk of rose-ringed parakeet establishment distinguishable from one another, but have (Strubbe et al. 2015), as humans often enhance subtle morphometric differences and geo- habitat (e.g., by planting trees used by graphic origins; P. k. manillensis originates parakeets for nesting and feeding) and provide south of latitude 20° N in India and in Sri bird feeders. This species’ prevalence in the Lanka, whereas P. k. borealis originates north pet trade and in zoos increases the chances of of 20° N in India, as well as in Pakistan, its wild establishment by escaping, or being Bangladesh, Nepal, and Myanmar (Juniper released, from cages (Reino et al. 2017, Avery and Parr 1998, Menchetti et al. 2016). The and Shiels 2018). two African subspecies are P. k. krameri While few data on non-native parrots exist (Scopoli, 1769) and P. k. parvirostris (Souance, for Pacific islands (Runde et al. 2007), our review 1856), and to our knowledge they have not draws on studies of the rose-ringed parakeet naturalized in the Pacific region. Psittacula k. from both its non-native range in the Pacific and krameri is from the equatorial regions in other parts of the world and its native Indian central and western Africa, and is found range. Due to the rapid increases in population in Guinea, Senegal, and southern Mauritania growth and damage resulting from this species in the east, to western Uganda and southern during the last 5–10 years in Hawai‘i, we feel this Sudan (Juniper and Parr 1998). Psittacula k. review is particularly timely to both prevent parvirostris is limited to the eastern equatorial more islands from being colonized by these region of the African continent—northwest invasive birds, and to motivate population Somalia west across northern Ethiopia to the management to help reduce and prevent future Sennar district of Sudan. These two African damage and economic losses. subspecies were exported in lesser numbers and are not naturalized in the Pacific; they were recorded as being wild-living in Puerto NAME AND DESCRIPTION Rico (Atlantic Ocean) but they failed to Species Description establish (M. Oberle, pers. comm.). Psittacula krameri is in the Class: Aves, Order: Psittaciformes, Family: Psittacidae, Subfamily: Intra-Specific Variation Psittacinae, Genus: Psittacula (Cuvier, 1800). The African subspecies differ physically from Some common English names for this species the two Indian subspecies in being smaller and are: green long-tailed parakeet, long-tailed lighter green with a yellow wash throughout. parakeet, ring-necked parakeet, rose-ringed The facial area is light yellow-green and Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 423 contrasts with the more intense grey-blue Distinguishing Features wash to the posterior head and nape. African subspecies’ bills are smaller and dark red or Psittacula k. manillensis is a 38–42 cm long maroon distally grading to black in contrast to (body + tail) parrot, with its tail accounting for the larger red bills of the Indian subspecies about half of its length (∼25 cm; Butler 2005). (Forshaw 1989, Juniper and Parr 1998). In Europe, adult RRP generally weigh about Pithon and Dytham (2001) examined 180 140 g (Butler and Gosler 2004). RRP on Kaua‘i museum specimens of RRP and determined averaged 133 g, but 12 of the 17 individuals that body length and bill length were the weighed were juveniles; the four adult males parameters that were most likely to differ averaged 139 g (Gaudioso et al. 2012). RRP are among the four subspecies of RRP, and toe generally uniform light emerald green and the and wing length were also helpful in upper and lower mandibles of the bill are red. distinguishing some of the subspecies. Thus, Likeallparrots, RRP have zygodatyl feet—they there are subtle morphometric differences have two forward-pointing and two backward- among the four subspecies, but due to P. k . pointing toes—are brightly colored, possess manillensis being the most common species shortand stoutstrongly hookedbeaks, and have introduced, including throughout the Pacific, relatively short legs (Figure 1). we have focused our review on this sub- RRP are sexually dichromatic, with the species. most obvious characteristic distinguishing

FIGURE 1. Rose-ringed parakeets (Psittacula krameri) on an ironwood tree (Casuarina equisetifolia) adjacent to Kapi‘olani Park, O‘ahu, Hawai‘i. The upper two individuals are males (notice distinctive dark ring around neck), and the lower individual is a presumed adult female (notice faint green ring around neck, and shorter tail). The uppermost bird is a rare blue variety, with no more than three individuals known on O‘ahu. Photograph by Phil Taylor. 424 PACIFIC SCIENCE • October 2019 adult males from females being that males Hong Kong and Japan have non-native possess a black ring around their neck. The Alexandrine parakeet (Psittacula eupatria) black ring extends from the chin to both sides established, which is a close relative of the of the head, and a thin rose-colored ring is RRP (diverging from a common ancestor present under the black ring around the nape. about 5 million years ago; Groombridge et al. Females lack a dark neck ring but instead have 2004). The Alexandrine parakeet could be a faint pale green neck ring that is indistinct confused with RRP because of the emerald (Figure 1). Males have a black line above their color, long tail feathers, red bill, and black bill that extends to their eyes, whereas this line neck ring on the males. However, the is absent or less pronounced in females. Tail Alexandrine parakeet is about one-third length of adult males is typically a few centi- longer (58 cm long), 80% heavier (250–260 meters longer than the tail of adult females g), and its bill is significantly larger than that and juveniles (Forshaw 1989; Figure 1). The of the RRP. Other common names for the central tail feathers, especially in older males, Alexandrine parakeet include the greater rose- are washed in cobalt blue near the terminus ringed parakeet, great-billed parakeet, and and central vein, whereas females usually Alexandrine ring-necked parakeet. The call of possess shorter central tail feathers with less the Alexandrine parakeet is also known to be blue coloration than males (Figure 1). Juve- louder and harsher than that of the RRP niles and individuals up to the age of 3 years of (https://www.parrots.org/encyclopedia/alex both sexes look similar to mature females. andrine-parakeet). Additional parrots estab- Adults of both sexes have a thin orange- lished on Pacific islands where RRP have colored skin ring surrounding the eye, and the invaded are easily distinguishable from RRP eye has a two-toned tan (outer) and grey (e.g., cockatoos (Cacatua spp.) in Hong (inner) colored iris, whereas immature birds Kong and Hawai‘i, budgerigar (Melopsittacus younger than six-months old have dark grey undulatus) in Japan and Hawai‘i, native irises and no eye ring coloration. Adult parrots of New Zealand). plumage starts to appear at 18 months and is completed by 32 months. Young males lacking distinguishing adult male plumage ECONOMIC IMPORTANCE AND may successfully pair and reproduce (Butler ENVIRONMENTAL IMPACT 2003). Feet are greenish-grey in both sexes Foraging and ages (Butler 2003). Theverylongtail,emeraldgreenmonotypic Negative economic and environmental coloration, and distinct call (listen at: https:// impacts by parrots are largely related to www.parrots.org/encyclopedia/ringneck-para foraging activities. In general, most parrot keet) make the RRP fairly easy to distinguish species that have naturalized in the Pacific, from other parrot species that have been including the RRP, are largely pre-dispersal established in Hawai‘i and other Pacific Islands. granivores (seed predators) (Galetti 1993, del In addition to the RRP, at least four other Hoyo et al. 1997, Corlett 1998, Mack and species of parrots have established localized Wright 2005, Kalodimos 2008, Shiels et al. reproducing populations in Hawai‘i, and these 2018; Figure 2). Parrots have specialized bill four species (blue-crowned parakeet Thectocer- and tongue morphologies to be able to extract, cus acuticaudatus, mitred parakeet Psittacara dismember, and process seeds for ingestion mitratus, red-masked parakeet Psittacara ery- (Collar 1997, Corlett 1998). Parrot foraging throgenys, and red-crowned parrot Amazona impacts in the wild are primarily related to viridigenalis) are detailed in VanderWerf and seed destruction and damage, serving as net Kalodimos (in review). Specifically, the blue- seed “sinks” rather than one of seed dispersing crowedparakeethasabluecrownandpalebeak; “sources.” Thabethe et al. (2015) assessed the mitred parakeet, red-masked parakeet, and whether RRP disperse seeds during captive red-crowed parrot each have green bodies but trials, and found that RRP were mainly seed have red on their heads. predators of the tested species; however, some Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 425

their ability, when foraging at high densities, to eliminate a tree’s ability to reproduce. Each season, RRP completely consume the seed crop of a rare 12 m tall Gigasiphon macrosiphon (Fabaceae) tree at the Foster Botanical Garden (located in downtown Honolulu; N.P.K. pers. obs.). RRP even extracted and killed the seeds enclosed in welded-wire mesh cages specially designed to protectthe large 2–3 seeded legumes from depredation (R. Silva, pers. comm.).

FIGURE 2. Generalized diet of the rose-ringed parakeet (Psittacula krameri), a strict herbivore, with arrow Degradation to Agricultural Crops thickness indicating average relative abundance of food items consumed. Dashed lines indicate food items that In its native Indian range, the RRP is have been recorded as being consumed but are extremely considered to be one of the most serious rare in the average diet. The thinnest solid arrows indicate avian pests of agricultural fruit and seed crops infrequent items by volume (in crop + gizzard contents) but commonly consumed seasonally on some islands (e.g., (Ali and Ripley 1969, Shivanarayan 1981, O‘ahu, Japan); the medium solid arrow (fruit) indicates Saini et al. 1994, Mukherjee et al. 2000). nearly ubiquitous presence but low/medium relative Bashir (1979) reported that the annual loss of volume; the thickest solid arrow (seed) indicates ubiqui- 97,000 tons of corn seed caused by RRP in tous presence and high relative volume. All categories and relationships are based on reviewed literature (see text). Pakistan was equivalent at the time to U.S. $15 million. RRP feed on a wide variety of grain products and fruit (Mukherjee et al. seeds were passed intact and germinated, and 2000, Reddy 1998a, Reddy 1998b). For others were regurgitated like seeds of cam- example, RRP reduced yields of corn and phor (Cinnamomum camphora), which has sorghum in parts of India by 74–81% (Reddy 0.75 cm diameter seeds. Shiels et al. (2018) 1998a, Reddy 1999). Through RRP crop and analyzed crop and gizzards of 64 RRP wild- gizzard analysis in India, adults and nestlings harvested on Kaua‘i and found an average of were shown to have fed primarily on sorghum, 11.6 ± 3.3 (min: 0, max: 150) intact seeds per corn, and sunflower (Shivanarayan et al. 1981, individual, but were unable to determine if the Saini et al. 1994). Also in India, RRP find bean grinding action of the gizzard would destroy and oily seed crops highly palatable, such as all intact seeds. Therefore, although RRP may red gram Cajanus cajan, green gram Phaseolus occasionally disperse seeds, their main aureus, black gram P. mungo, Bengal gram behavior when handling seeds is predation Cicer arietinum, sunflower Helianthus anuus, (Figure 2). safflower Carthamus tinctorius, rapeseed Bras- Gregarious flock foraging behavior of RRP sica napus, rice Oryza sativa, and maize Zea magnifies their impacts to individual trees, mays (Bashir 1979, Rao and Shivanarayan orchards, and crop fields. On Kaua‘i, hundreds 1981). In 1975, the California Department of of birds may flock to corn fields to feed on the Agriculture estimated that the potential crop corn (Gaudioso et al. 2012), and when they feed losses due to a well-established RRP popula- in orchards on Kaua‘i farmers report fruit loss tion could reach $735,000 annually, a value and spoiling from RRP defecation while that resulted from an estimate of RRP perched on their trees. Such fruit spoiling damaging 0.1% of crops that they are known may account for more crop and economic loss to eat and that are grown in the area (cited in than the direct take of fruit and seed by RRP Paton et al. 1982). Vineyard damage by RRP (Novak 2018). Additionally, in Japan, the main in Surrey, U.K., was estimated at £5,000 per complaint of RRP as pests is their scattering of year (Fletcher and Askew 2007), a 0.8 ha apple flowers, especially cherry blossoms, when orchard in Germany lost 10–15% of its apples foraging for nectar (Matsunaga and Fujii to RRP (Van Kleunen et al. 2010 cited in 2018; Figure 2). RRP on O‘ahu have proven Menchetti et al. 2016), and a 0.8 ha almond 426 PACIFIC SCIENCE • October 2019

(Prunus dulcis) orchard suffered about 30% RRP damage these same fruit crop species, as fruit damage from RRP in rural Italy (Mentil well as papaya (Carica papaya), bilimbi (Aver- et al. 2018). rhoa bilimbi), and strawberry guava (Psidium Although RRP are often mistaken to be cattleianum) (N.P.K., pers. obs., Runde et al. indiscriminate eaters of fleshy fruit, they are 2007; Figure 3). In general, RRP have a most often targeting the seed within the fruit flexible plant diet, and food choice and usage for consumption, and therefore commonly appear to be partly related to temporal and dismember fleshy fruit (ripe and unripe) to spatial food abundance (Shiels et al. 2018) and crack into the seeds (Figure 2). In India (Saini nutritional quality. Threats to agricultural et al. 1994) and on Kaua‘i(Shiels et al. 2018), plantings from this and other parrot species is yellow guava (Psidium guajava) seeds are eaten related to (1) how close the population’s from January to March and July and August. evening roost or active nest is to the Mulberry, Morus sp., seeds are eaten during agricultural area, and (2) the desirability of April and May; and pearl millet Pennisetum the agricultural crop as a food source relative glaucum, sorghum Sorghum bicolour, and maize to other food sources (Gaudioso et al. 2012, from August to December (Saini et al. 1994). Shiels et al. 2018). On Kaua‘i, most radio- Although poorly investigated, it does not collared RRP left the roost and went into appear that RRP damage to agriculture nearby corn fields to feed, and daily movements currently extends to Pacific islands besides appeared to reflect the spatial ripening of corn Kaua‘i and possibly O‘ahu. On Kaua‘i, RRP infields(Gaudiosoetal.2012).OnotherPacific are increasing their agricultural damage to islands where RRP have established, they roost seed and orchard fruit (Koopman and Pitt inurbanareasthatarenotneartoagriculture(S. 2007, Gaudioso et al. 2012, Shiels et al. 2018; Matsunaga, pers. comm.; L. Maria, pers. Figure 2). One of the biggest agricultural comm.; Leven and Corlett 2004). In Belgium, economies on Kaua‘i is seed corn (Z. mays), RRP do not nest or forage near field agricul- and RRP feed on the kernels of corn cobs just tural lands (Strubbe and Matthysen 2009a), but prior to harvest, costing farming companies instead prefer lands classified as orchards hundreds of thousands of dollars in losses each (Strubbe and Matthysen 2011). Field observa- year (Gaudioso et al. 2012, Avery and Shiels tions from valley ridges and tall buildings on 2018). From a recent diet study on Kaua‘i and O‘ahu has revealed that individual RRP make analysis of 64 birds harvested from five daily movements of up to about 7 km from their locations, RRP ate a range of plant food items evening refuge to nesting locations (N.P.K. with the primary species consumed being corn unpubl. data). In Japan, RRP distribution (67% of individuals, 31% average diet by mass; appears to shift seasonally depending on food mostly seed) and yellow guava seeds (97% of resource availability, and in some cases RRP individuals, 30% average diet). Additionally, mayhave widenedtheirdietaryspectrumdue to there were differences in food consumption occasional foraging on the ground (Matsunaga among sites and between sexes (with males and Fujii 2018). generally consuming more than females). As in India, RRP on O‘ahu forage on many There was no evidence of RRP consumption leguminous tree species (Cassia spp. Caesalpi- of insect or other animals in the 64 crop and nia pulcherrima, Prosopis spp., Pithecellobium gizzard samples analyzed (Shiels et al. 2018). dulce; Figure 3) but ignore other, heavily In addition to seed crops of corn and fruiting and abundant food species (e.g., sunflower on Kaua‘i(Shiels et al. 2018), RRP Leucaena leucocephala, Albizia saman, Ficus are significant pests of fruit trees, as they spp.), indicating a possible hierarchy of food consume and damage yellow guava (P. gua- preference among plant species in their java), lytchi (Litchi chinensis), rambutan activity area on O‘ahu Island. Undoubtedly, (Nephelium lappaceum), mangoes (Mangifera agricultural plantings are more desirable food indica) and starfruit (Averroha carambola) (J. sources than most, if not all, naturally Denny, pers. comm., Koopman and Pitt 2007, available non-agricultural foods due to their Novak 2018, Shiels et al. 2018). On O‘ahu hyper-abundance in a concentrated area; Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 427

FIGURE 3. Rose-ringed parakeets consuming fruit of mango (Mangifera indica; upper picture) and ‘kiawe (Prosopis pallida; lower picture) on O‘ahu. Upper picture is an adult male, and lower picture is an adult female or juvenile of either sex (notice lack of dark ring around neck distinct to adult males). Photographs by N.P.K. however, RRP appear to have preferences species, RRP roost colonially, often number- even among agricultural crop types. ing tens to hundreds of individuals in single trees and thousands of individuals in groups of trees. The loudness of RRP calls is magnified Human Interactions and Perceptions from their simultaneous communications as RRP are often perceived as a nuisance to flocks both arrive and exit roosts. On Kaua‘i, human residents. Like many other parrot RRP roost only in palm trees, and when the 428 PACIFIC SCIENCE • October 2019 island’s RRP population was <3,000 they only at Līhu‘e Airport, Kaua‘i, to prevent potential roosted in the royal palms (Roystonea spp.); airstrikes (W. Bukoski, pers. comm.). From currently RRP on Kaua‘i roost in royal palms, 2005 to 2006 at the Heathrow Airport in coconut palms (Cocos nucifera), and Manila U.K., three of the 98 airplane bird strikes palms (Adonidia merrillii) (A.B.S., pers. obs.). involved RRP, and the cost per bird strike was Palms that line parking lots and shopping reported to be £20,000 (Fletcher and Askew centers are regular roost sites for RRP in 2007). As RRP expand in numbers and Kaua‘i and southern California (Avery and proximity to airports, airplane strikes with Shiels 2018). In Japan, RRP have been RRP may become more important to manage. observed roosting within university campuses and residential areas; however, in recent years Degradation to Natural Systems: Plants the roosting locations have not stabilized (S. Matsunaga, pers. comm.). Tree species RRP impacts to natural areas on Pacific used for roosts in Japan include Himalayan islands primarily involve seed predation cedar (Cedrus deodara), ginkgo (Ginkgo biloba), (Figure 2), although their destruction to trees and bamboo (Poaceae) groves (S. Matsunaga, in Australia by stripping bark has resulted in pers. comm.). Similarly, on Mahé (Seychelles, some trees being killed and thereby possibly Indian Ocean), the main roost was a large shifting the local tree community composition clump of non-native bamboo that was about (Fletcher and Askew 2007). Additionally, 30 m high and would contain about 300 RRP; there is evidence that non-native seeds may an additional smaller roost was in albizia be dispersed by RRP in Kaua‘i(Shiels et al. (Falcataria moluccana) trees (N. Bunbury, pers. 2018) and South Africa (Thabethe et al. 2015), comm.). highlighting RRP as a possible vector for On Kaua‘i, property owners of apartments, invasive species spread. Plants in the family condominiums, and hotels complain about the Melastomataceae contain some of the most noise from RRP calls when assembling in unwanted invasive plant species in the Hawai- roosts each night at sunset, and dispersing ian Islands (Medeiros et al. 1997), and these from the roost in the early morning (Avery species are commonly consumed by birds and and Shiels 2018, Novak 2018). Similar mammals, which aids in their spread (Shiels complaints have been voiced on O‘ahu, 2011). It has not been documented that RRP particularly from apartment residents adja- feed on Melastomataceae, but a captive mitred cent to the largest RRP evening roost on parakeet pet fed Miconia calvescens berries O‘ahu that is a large Ficus sp. tree on Beretania defecated viable seed (Gassman-Duvall and Punahou Streets (A.B.S. and N.P.K., pers. 2002), leaving this as a strong possibility for obs.). Hotel owners and local residents in RRP to do the same in the wild. Kaua‘i also complain about the RRP defeca- In their introduced range, RRP feed on and tion below roosting trees and the associated destroy seeds of both native and non-native pests (ants, cockroaches, and mice) that they plants (Cramp 1985, Strubbe and Matthysen believe are elevated because of the RRP 2007, Clergeau and Vergnes 2011). In Japan, communal and repeated defecation at perches RRP damage flowers of native cherry (Prunus and roosts (M. Martin, pers. comm.). Addi- spp.) blossoms when feeding on nectar tionally, RRP on Kaua‘i damage non-native (Matsunaga and Fujii 2018; Figure 4). In palms and Cook pines (Araucaria columnaris) France and the U.K., RRP have been by stripping leaves (M. Martin, pers. comm.). documented feeding on native species such In addition to the noise complaints from as berries of holly (Ilex spp.) and elder RRP calls in flight and at the communal (Sambucus spp.), and fruits and seeds of ash roosts, RRP are hazards to airplanes and (Fraxinus excelsior), hornbeam (Carpinus betu- accompanying passengers. Although we know lus), maple (Acer campestre), yew (Taxus of no airplane strikes involving the RRP on baccata), birch (Betula pubescens), alder (Alnus Pacific islands, lethal control by shooting RRP glutinosa), and willow (Salix alba)(Cramp is practiced near the runways and flight paths 1985, Clergeau and Vergnes 2011). Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 429

FIGURE 4. An adult female, or juvenile of either sex (notice lack of dark ring around neck distinct to adult males), rose- ringed parakeet destroying cherry blossoms (Prunus serrulata) to feed on nectar in Japan. Photograph by Satomi Matsunaga. There has been discussion about parrots as trees in Kaua‘i (W. Bukoski, pers. comm.). a threat to palm species (Runde et al. 2007, Despite their documented consumption of Pennybacker 2016), yet RRP rarely feed on some native plants in Europe, flower destruc- the fruit of native Hawaiian palms (Pritchardia tion in Japan, and anecdotal evidence from spp.). RRP feeding on Pritchardia spp. on observations in Hawai‘i, no reports yet exist Kaua‘i has been observed on only a few indicating that RRP are altering the natural or occasions, and only in planted areas (W. semi-natural environment through their feed- Bukoski, pers. comm.). Part of the very low ing ecology (Tayleur 2010). frequency of RRP feeding on native Hawaiian RRP are known to consume nectar and Pritchardia spp. may be due to the abundant flower buds (Ali and Ripley 1969, Clergeau fiber and lack of fleshy pulp on their fruit. and Vergnes 2011, Matsunaga and Fujii 2018; Many native Hawaiian palm species are Figure 2). In Japan, flower damage by RRP is threatened and endangered and therefore the main complaint from residents regarding infrequently exist in the disturbed habitats these birds, and particularly so for the where RRP exist. RRP do feed on low-fiber scattering of the blossoms of cherry trees fruit pulp of some non-native, ornamental, while RRP are foraging on nectar (Matsunaga palm species, such as Livistona chinensis, and Fujii 2018). On O‘ahu, RRP regularly Archontophoenix alexandrae, and Livistona visit flowers of Erythrina species and flowers rotundifolia; however, the palm seed, cleaned of many of the Myrtaceae, Fabaceae, and of its pulp, is discarded unharmed (N.P.K., Bignoniaceae; RRP are destructive to some pers. obs.). RRP have also been observed tree flowers especially of Cassia spp. and feeding on a few occasions on the fruits and Pithecellobium saman (N.P.K., pers. obs.). seeds of the leguminous native koa (Acacia koa) However, general flower foraging by RRP 430 PACIFIC SCIENCE • October 2019 lacks the thoroughness or persistence reserved avian bornavirus, chlamydiosis, Newcastle for tree seed or fruit foraging. Any use of disease virus, avian pox virus, avian influenza, flowers by parrots in native Hawaiian forest avian psittacosis, and pulmonary disease would probably threaten restricted-range (England 1974, Tozer 1974). Two wild RRP endemic Hawaiian honeycreeper food sup- in England were diagnosed via PCR and skin plies and would be unacceptable. The poten- inspections with PBFD (Sa et al. 2014), and a tial for RRP and other parrot use of native recent genetic study showed that this disease is Hawaiian forests should be cause for concern, now common among various RRP worldwide, though we have no evidence that RRP which has probably transmitted via the pet are foraging in native forest on any Pacific trade and includes wild RRP in the tropical Island. islands of Mauritius and Seychelles (Fogell et al. 2018). Field studies show that wild parrots in native ranges are free from several Degradation to Natural Systems: Animals diseases such as Psittacid herpes virus and for Most RRP interactions with other wildlife are antibodies to paramyxovirus serotypes 1 and 3 avian, yet on Pacific islands the influence of and internal parasites (Stone et al. 2005). RRP on native birds has not been clarified. Parrots are not competent hosts for West Nile Potential competitors of RRP in Japan may be virus and thus probably could not serve as cavity nesters such as white-cheeked starling functional carriers or sources of this virus (Spodiospar cineraceus) and the brown hawk owl (Komar et al. 2003). That said, RRP accept (Ninox scutulata) (S. Matsunaga, pers. comm.). food from residential bird feeders (Clergeau In Europe, RRP are known to compete with and Vergnes 2011, S. Matsunaga, pers. native cavity-nesting birds, and outcompete comm., N.P.K., pers. obs.), which enables native nuthatches (Sitta europaea)(Strubbe disease exchange from direct physical contact and Matthysen 2007, Strubbe and Matthysen with dozens of other bird species and their 2009b). Thus, a major concern for native birds fecal material. An additional behavior of RRP on islands or regions where RRP have been that may aid in disease transmission is their introduced is the risk of being outcompeted defecation on agricultural crops while fora- for nesting cavities. Nest cavity competition ging on them, and beneath communal roost from larger and more aggressive RRP could sites. drive one or more species of endemic parrots The extent to which RRP are disease of the Pacific Islands (e.g., threatened and carriers on Pacific islands is unknown. The endangered lories and lorikeets, such as only wild RRP in the Pacific that were Trichoglossus johnstoniae in the Philippines, screened for disease that we are aware of T. rubiginosus in Micronesia, and Vini ultra- are those by Gaudioso et al. (2012) on Kaua‘i marina in the Marquesas) to extinction. RRP where 15–18 individuals were tested and have even proven deadly to cavity roosting bat found negative for avian Psittacosis and avian species by cornering and killing them when influenza virus. Through physical inspection driving them out of tree cavities (Menchetti et of recently harvested RRP on Kaua‘i(Shiels et al. 2014, Hernandez-Brito et al. 2018). RRP al. 2018), and observations of live-birds at bird also have been observed attacking and killing feeders, evening roosts, and nest cavities in squirrels (e.g., S. vulgaris) in France (Mench- Honolulu (N.P.K. unpubl. data), there was no etti and Mori 2014), and non-native black rats physical evidence of disease-like symptoms. (Rattus rattus) in Spain (Menchetti and Mori However, a wild RRP in Honolulu, Hawai‘i, 2014) as the RRP pairs were apparently was recently observed with symptoms of avian protecting their nests. pox or PBFD (Figure 5). Several red-vented bulbuls (Pycnonotes cafer) that shared the same perching trees as the parakeet also had DISEASE mucosal lesions along the gape and nares Pathogens that parrots may contract include (N.P.K., pers. obs.). Naturalized, wild-living Psittacine beak and feather disease (PBFD), non-native parrots elsewhere exhibited Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 431

FIGURE 5. A wild rose-ringed parakeet with signs of either avian pox disease or Psittacine beak and feather disease (specific virus confirmation requires PCR genetic screening) in Honolulu, Hawai‘i. Notice wart-like growths at the base of the beak (yellow arrows inserted). Both avian pox and Psittacine beak and feather disease can be spread between wild and captive birds, among bird species, and can be lethal to birds. Photographs by N.P.K. in January 2019.

similar symptoms in a location where they PERCEIVED BENEFITS interacted with caged birds suffering from the same affliction (Kalodimos 2013). In contrast RRP are colorful birds of tropical origin and to the findings in Hawai‘i, Mase et al. (2001) this can bring an exotic or otherwise favorable isolated avian influenza (H9N2 influenza A feeling towards their local establishment in virus), which has the potential to be directly urban environments by some human residents, transferred to humans, from respiratory as has been the case with some European organs of RRP that had been imported from locations (Menchetti et al. 2016, Crowley et al. Pakistan to Japan, and were very similar 2019) and appears to be so in Hong Kong (>97% match) to the influenza strain found in (M. Leven, pers. comm.). Parrots are generally the 1997 human influenza outbreak in Hong perceived as intelligent birds with lively Kong. Mase et al. (2001) further concluded characters. Additionally, RRP visitation to that avian influenza may be circulating in RRP backyard birdfeeders in Europe (Butler 2003, distributed through the pet trade. Pet RRP Clergeau and Vergnes 2011), O‘ahu (N.P.K., held in captivity also present the possibility of pers.obs.),and Japan (Matsunaga, pers.comm.) introducing other diseases into populations can be welcomed by some human residents established in the wild, and this appears to be even if they are aware that the birds are non- the case with PBFD in many countries native. Such welcoming of these birds outside worldwide where wild RRP have established their native range can make their removal (Fogell et al. 2018). Although confirmed cases and management extremely difficult (Mench- of disease in RRP in Hawai‘i or other Pacific etti et al. 2016, Novak 2018, A.B.S., pers. obs.). islands have been absent (but see Figure 5), Although some may suggest that RRP have the potential interaction of wild RRP with the beneficial effect of reducing non-native captive or wild birds should be viewed as a plant spread by their high levels of seed serious disease transmission concern (e.g., predation, such perceived benefits of the RRP Fogell et al. 2018). are unlikely to be true or at least fall short of 432 PACIFIC SCIENCE • October 2019 eliminating weedy plant reproduction about US$1 each. Although this has appar- entirely. For example, RRP flocks on O‘ahu ently helped reduce RRP abundance and exhibited daily foraging on strawberry guava therefore damage in parts of their native range (P. cattleianum), with each bird in a flock (Scalera 2001), it is facilitating the global killing 100 or more seeds per feeding (N.P.K. dispersal of this potential pest species. It unpubl. data). However, there is no evidence should be emphasized that the reason RRP suggesting that RRP kill all of the seeds in a have invaded islands, countries, and ecosys- stand or otherwise limit the spread of such a tems outside their native range is due to highly invasive plant. Additional species of escaped individuals from the pet trade; there- non-native weedy plants that have been fore, this industry should not be supported for observed having their seed crops significantly these birds. reduced by RRP in Hawai‘i, yet continue to thrive and spread, include yellow guava (P. guajava; Shiels et al. 2018), and high yielding REGULATORY ASPECTS Fabaceae such as Prosopis pallida, Acacia Regulations on the importation and posses- formosa, and F. moluccana (N.P.K., pers. sion of RRP differ depending on state and obs.). Because RRP are not known to colonize country laws. In Hong Kong, Japan, and New or forage in continuous native forest, which is Zealand, there appear to be no restrictions on where non-native plants are most unwanted, it the breeding or sale of RRP, but technically it is unlikely that RRP will help reduce the non- is prohibited to release them into the wild in native plant problems that are nearly ubiqui- New Zealand (N. Mihi, pers. comm.) and they tous in some island settings in the Pacific, such are unwanted organisms under the New as in the Hawaiian Islands. Zealand Biosecurity Act (C. Miskelly, pers. Field studies in native habitats have comm.). All birds in Hong Kong are protected determined that many parrot species may by law, and therefore RRP cannot be harmed, provide pollination services (Skutch 1983, trapped, killed, or harassed (M. Leven, pers. Vicentini and Fischer 1999, Cotton 2001). comm.). RRP in the State of Hawai‘i are However, RRP interaction with flowers of officially classified as injurious wildlife species present on Pacific islands generally (DLNR 2014), pestiferous (Koopman and leads to destruction or partial dismemberment Pitt 2007), invasive (Lohr 2012), and their of the flower. import into the State (i.e., all Hawaiian islands) There is a possibility for human economic has the designation of ‘restricted – for private gain with RRP in the pet trade industry, but and commercial use’ http://hdoa.hawaii.gov/ markets appear quite saturated and the import pi/pq/import-program/). Interestingly, RRP and sale of RRP is prohibited in some areas are not on the ‘prohibited list’ for animal such as the Hawaiian Islands (Hawaii Depart- importation into the State of Hawai‘i, perhaps ment of Agriculture statement in 2016 news- because they are allowed in some cases such as paper article: https://www.13abc.com/home/ for research animals or for zoos (http://hdoa. headlines/Parakeet-problems-at-Hawaii-park- hawaii.gov/pi/pq/import-program/), or because 365537961.html). Avery and Shiels (2018) the online importation lists have not been presented the RRP import data into the updated since 2006. Furthermore, a 2016 U.S. from 1980 to 2009, which showed Hawai‘i news article quoted a Hawai‘iDepart- imports had peaked at >40,000 birds during ment of Agriculture employee saying RRP are 1985–1990. Despite widely practiced captive banned for importation as pets (https:// breeding of RRP, wild capturing of RRP in www.13abc.com/home/headlines/Parakeet-pro their native range for subsequent sale to the blems-at-Hawaii-park-365537961.html), and pet trade industry still occurs (Menchetti et al. pet store owners in Hawai‘i have confirmed that 2016). In fact, Scalera (2001, cited in they are not allowed to sell RRP in Hawai‘i Menchetti et al. 2016) reported up to 5,000 (various pet store employees in 2019, pers. RRP were caught each day in Northern India comm. with A.B.S.). Although RRP are classi- and were sold wholesale to the pet industry for fied as injurious wildlife in the State of Hawai‘i, Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 433 manipulation of this species in the wild is Nagano, Shizuoka, Gifu, Osaka, and Hyogo. In controlled by State law and a special permit is the past, RRP have been recorded in Ehime on necessary to take, harm, capture, or harass RRP Shikoku Island, and further south in Saga and (http://dlnr.hawaii.gov/dofaw/permits/). Land- Miyazaki prefecture on Kyushu Island. Appar- owners on Kaua‘i have applied and subsequently ently, there have been few records recently received temporary State permits to shoot except for within the Kanto District, which RRP that are damaging their crops or property includes Tokyo and its surroundings (Gorden- (A.B.S., pers. obs.). ker 2009). Currently (in 2018), there are three surviving groups of RRP in Japan, and they are restricted to: (1) Tokyo, Kanagawa prefecture, GEOGRAPHIC DISTRIBUTION AND HISTORY and Saitama prefecture (>1500 individuals), (2) The RRP has formed reproducing popula- Gunma prefecture (<50 individuals), and (3) tions in at least 40 countries (Strubbe and Chiba prefecture (<20 individuals; S. Matsu- Matthysen 2009a), which has earned it naga, pers. comm.). classification as the most successful natura- In New Zealand, RRP have repeatedly lized parrot in the world. Some of the island established and in most cases have been nations outside of the Pacific for which RRP subsequently removed since 2005 in the have naturalized include those in Spain, North Island (Takapuna to Havelock North), Portugal, Greece, United Kingdom, Maur- and one population in Christchurch (South itius, Seychelles, Singapore, and Tanzania Island) was released in 2016 and is still being (Strubbe and Matthysen 2009a, Menchetti managed today (Miskelly 2018; L. Maria, et al. 2016). On Pacific islands, RRP were first pers. comm.). Miskelly (2018) has summar- introduced in Hong Kong in 1903 (Leven and ized the incidences of RRP in the wild from Corlett 2004) and Hawai‘i in the 1930s–1960s the North Island as: breeding and later (Avery and Shiels 2018); RRP have established eradication of RRP occurred in Beachland in the southwestern part of Tokyo, Japan, (South Auckland) in 2005–2006; Okere Falls, since 1969 (Gordenker 2009), and in New Rotorua, had five birds including a fledgling Zealand first in 2005 where they have been from 2007 (eradication attempted in 2012 but repeatedly established and removed (Miskelly birds still present in 2015); Havenlock North 2018; L. Maria, pers. comm.). In all cases that from 2010 (eradication attempted in 2016); we are aware of, RRP were introduced into Tamahere, Hamilton had at least two birds in the wild through intentional or accidental 2013; Hikutia, north of Paeroa, had at least 15 release of captive or pet birds. Wild parrot birds from 2011 to 2014, and 12 of these were species were imported to many countries in removed in 2014–2015; and Takapuna had large numbers to be sold as pets in the 1960s four birds in 2015. On the South Island in through 1980s (Traffic 1987, Collar 1997), yet Christchurch, there was a fire at a zoo in 2016 the pet trade continues to supply RRP to most causing 30 birds to be released and not all countries of the world (Menchetti et al. 2016, were recaptured (Miskelly 2018; L. Maria, Avery and Shiels 2018). RRP have continued pers. comm.). An update in 2018 has con- to be a relatively inexpensive small parrot firmed that the Christchurch birds had been species that have been readily available in the successfully breeding in the wild since 2016 global pet trade. and a trapping and shooting program had not In Japan, established RRP are currently successfully removed all of them yet (L. only known on the largest island, Honshu. Maria, pers. comm.). Similarly on the North RRP originated from escapees (Eguchi and Island, escaped RRP have been removed from Amano 2004), and have been known to the wild except for a few birds surviving an successfully reproduce in Tokyo, Chiba, eradication attempt in Havenlock North, Kanagawa, Aichi, Kyoto, and Hiroshima Hawkes Bay (L. Maria, pers. comm.). prefecture (region). Additional areas of In Hong Kong, RRP were recorded Honshu Island where RRP have been recorded throughout the territory during the 1970s include Niigata, Tochigi, Gunma, Saitama, and 1980s in flocks of up to 87 birds. 434 PACIFIC SCIENCE • October 2019

However, the RRP population substantially In the Hawaiian Islands, RRP are common declined and has become restricted to two on Kaua‘i and O‘ahu (Avery and Shiels 2018), urban areas on Hong Kong Island, and a third and there is a small (perhaps incipient) subpopulation is in Kowloon, which is on the population on Hawai‘i Island in the Puna mainland-China portion of Hong Kong District (I. Leinbach, pers. comm.). All RRP (Leven and Corlett 2004). Currently there in the Hawaiian Islands occupy lowland are an about 15 RRP on Hong Kong Island habitats disturbed by humans (Figure 6). (M. Leven, pers. comm.), separated into two Kaua‘i Island has the largest population of subpopulations (centered on Hong Kong RRP, with a minimum number identified in Park, and centered on Ap Lei Chau/Ocean early 2018 of ∼6,800 individuals (A.B.S., Park). The Kowloon population has about 10 unpubl. data); they occupy the southern birds, and the population appears to be portion of the island, from Kekaha to Līhu‘e, increasing at the Kowloon roost (M. Leven, but have been sighted in almost all lowland pers. comm.). The cause for the decline from habitats along the perimeter coast (Figure 6). the 1980s to present is unknown, and all birds There are currently two known RRP roosts on are protected by law in Hong Kong. Accord- the island that each span approximately 1 km; ing to the Hong Kong Bird Report, the the highest density roost is in the Lāwai- significant entries since 1999 stated that the Po‘ipū-Kōloa area, and second highest density RRP abundance “appears to be continuing to roost is in Līhu‘e(Shiels et al. 2018). The decline” in 1999–2000, and there were multi- origin of the Kaua‘i population is believed to ple entries during 2009–2015 stating their be the individuals released by workers at a presence on Hong Kong Island (M. Leven, bed-and-breakfast inn near Lāwai in the 1960s pers. comm.). (Pyle and Pyle 2017). By 1981, RRP were

FIGURE 6. Rose-ringed parakeet (RRP) distribution maps for the Hawaiian islands of Kaua‘i (left island) and O‘ahu (right island), as of June 2019. The pink shading is where RRP have been recorded in the wild, according to observations posted on eBird (https://ebird.org/home) or by authors A.B.S. and N.P.K. The black lines on the maps are major roads, and the two red arrows point to the main airports on each island (Līhu‘e Airport on Kaua‘i, and Honolulu International Airport on O‘ahu). The distance from Kaua‘itoO‘ahu is about 115 km, and RRP have never been recorded flying between these islands. Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 435 regularly reported foraging in Hanapēpē reported in 1988–1990 in Kona (western Valley and roosting in Kalāheo, just west of Hawai‘i Island), and that population had Lāwai. Following crop damage in Hanapēpē apparently decreased substantially by 2016 Valley and permits issued to some residents for andarenotconsideredestablishedbyPyle RRP lethal control, the birds abandoned their and Pyle (2017). RRP are currently uncommon Kalāheo roost and were frequently observed and rarely seen on Hawai‘i Island; it is speculated south of Lāwai at the National Tropical that fruit and nut farmers have been lethally Botanical Gardens and east at the Waita shooting them there, which may explain the Reservoir near Kōloa (Pyle and Pyle 2017). By sparse or lack of sightings in recent times (Pyle 1997, the RRP on Kaua‘iwerewellestablished and Pyle 2009). On Maui Island, a few RRP have from Hanapēpēto Līhu‘e, with roosts in ‘Ōma‘o, been noted in the wild since the 1980s, but they Kalāheo, and Līhu‘e(Pyle and Pyle 2017). By have not established on this island, including 2012,thereweretwolargeroosts(Lāwai-Po‘ipū- the two escaped RRP observed in Makawao in Kōloa and Līhu‘e), with a combined total of 1988, one in Wailua in 1990, and one in Nāpili about 2,000 birds counted (Gaudioso et al. Bay in 2014 (Pyle and Pyle 2017). 2012), and by 2015 the Kaua‘i RRP distribution Because parrots are generally non-territor- had extended further west, spanning from ial, social, flocking birds, increasing population Kekaha to Līhu‘e(Shiels et al. 2018). sizesdoesnotinitiallypromoterangeexpansion On O‘ahu, RRP absolute numbers (∼4,650 (Butler 2003). In England where RRP now individuals) are less than on Kaua‘i and there number in the many thousands (e.g., ca. 10,000 are at least three population centers (evening in 2004, Butler 2005; ca. 30,000 in 2011, Butler roosts: Olomana, Waikele, and Honolulu) et al. 2013), their geographic expansion was widely distributed across the southern and minor (0.4 km per year) through year 2000 central portion of the island (N.P.K. unpubl. (Butler 2003). Searching for nest cavities is a data; Figure 6). RRP were first released into main driver of parakeet geographic expansion the wild on O‘ahu as early as the 1930s, but and once a cavity is occupied by a pair, foraging were not consistently and repeatedly observed activities are conducted around it (Strubbe and until the 1970s (Pyle and Pyle 2009). The Matthysen 2009a). Food availability can also early reported centers of abundance on O‘ahu influence range dynamics, and Matsunaga and in the 1970s were in the Makiki-Punahou Fujii (2018) have observed RRP near Tokyo areas and Waimānalo areas (Pyle and Pyle alter their distribution seasonally to capitalize 2009); local numbers continued to grow in on preferred food resources. these areas through the 2000s. In 2000, the Central Union Church in downtown Honolulu HABITAT AND CLIMATIC REQUIREMENTS had 75 RRP roosting at night (Pyle and Pyle 2017), and that roost now has expanded to RRP exist over a very wide climatic and about 2,600 individuals and spans approxi- latitudinal range (ca. 56° Nto44° S), and are mately 0.5 km (N.P.K. unpubl. data). Besides thus considered to be generalist species. Their this Honolulu roost site, the other active roosts native geographic region includes cool habi- in 2018 on O‘ahu were at Olomana Golf Links tats in Afghanistan and northern India in the and Waikele (N.P.K. unpubl. data). Himalayas to hot and dry savannas and RRP on Kaua‘iandO‘ahu have strong parkland south to Sri Lanka (Juniper and population growth (see text below, and popula- Parr 1998). In their native Indian range, RRP tion growth figures) though there is less known inhabit mesic to semi-arid deciduous and about population trends and distributions on evergreen open forest and savannah. They Hawai‘i Island. Paton et al. (1982) observed a typically shun dense forest and elevations flock of five RRP in eastern Hawai‘i Island greater than approximately 1,000 m, yet (10 km south of Hilo), and a pair of those Menchetti et al. (2016) stated that they can subsequently nested in a cavity of a native ‘ōhi‘a be found up to 2,000 m elevation. Their tree (Metrosideros polymorpha) in February 1981. presence is often associated with human Additionally, there was a flock of 25–30 RRP modified environments such as residential areas, 436 PACIFIC SCIENCE • October 2019 parkland, and agricultural land. Commensalism exclusively rely on pre-existing hollows for with humans is believed to enhance the nesting, though they commonly widen the probability of an introduced species becoming existing cavity by picking with their strong established (Lockwood 1999, Cassey 2002), and bill. RRP prefer large diameter trees for RRP thrive in human-altered landscapes nesting rather than randomly sampled trees (Strubbe et al. 2015, Shiels et al. 2018). (Tayleur 2010). RRP often re-use the same On each of the Pacific islands where RRP cavities in subsequent years. Bluekens (2002) have established, they are active in habitats determined cavities were re-used at a rate of modified by humans. RRP populations center 61–70% during a 3-year study in Europe. On around suburban and human created park- O‘ahu, the same tree cavities are used each land, which typically has large ornamental year, but it is not known if it is the same trees used as food, perching, and nesting breeding birds using the same cavities from sources. Like in India, RRP on islands are not year to year. RRP in their native range nest in highly attracted to closed canopy, dense tree cavities in many different forest habitats forest, or exceptionally dry habitats that lack as well as occasionally nest in cavities in roofs many trees for nesting or foraging (N.P.K., or rock walls when suitable tree cavities are pers. obs.). On both Kaua‘i and O‘ahu, RRP not available (Forshaw 1989, Juniper and Parr are most commonly seen in lowland mesic 1998). In the Hawaiian Islands, RRP are only areas of open suburban forest, parkland, known to nest in tree hollows (Figure 7); farmland, and residential areas up to around 500 m elevation (Gaudioso et al. 2012, N.P.K. unpubl. data; Figure 6). Knowledge of the spatial organization and proximity of RRP to agricultural plantings is paramount in evalu- ating RRP risk to agriculture, and predictive niche- and spatial-modeling can help identify areas most at risk of RRP population spread and habitat usage over time. The range of climates that this species can survive in is broad (temperate to tropical), but it does have limitations. RRP cannot survive in geographic locations where there are pro- longed periods of freezing temperatures. RRP introduced into New York City (U.S.A.) suffered from frostbite during the winter (Roscoe et al. 1976),inhibitingtheirestablishment.Similarly, RRP introduced into Belgium have been knowntosuffermortalityduetowinterweather (Temara and Arnhem 1996). Following a European survey, Strubbe and Matthysen (2009c) determined that RRP establishment success declines with decreasing temperature.

FIGURE 7. Rose-ringed parakeet at nest cavity. Upper RESOURCE REQUIREMENTS AND picture is an adult female or juvenile of either sex (notice LIMITATIONS lack of dark ring around neck distinct to adult males) emerging from a nest cavity in an ironwood tree Nesting Cavity Resource (Casuarina equisetifolia) adjacent to Kapi‘olani Park, ‘ ‘ The availability of suitable nesting cavities O ahu, Hawai i (Photograph by Phil Taylor). Lower picture is an adult female returning to her nest to feed her strongly influences the presence of RRP in an young (two of the three chicks are pictured) on O‘ahu, area. RRP are secondary cavity nesters and Hawai‘i (Photograph by N.P.K.). Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 437 however, these birds have been observed partially rely upon food from bird feeders, investigating building ventilation, roof drai- especially during the winter months (Garrett nage holes, and highway viaduct drainage et al. 1997, Bittner 2008, Clergeau and pipes (N.P.K.,pers. obs.). Treecavities used by Vergnes 2011, S. Matsunaga, pers. comm.). this species for nesting in the Hawaiian Islands During a radio-tracking study in suburban are most commonly ironwood (Casuarina France, Clergeau and Vergnes (2011) deter- equisetifolia; Figure 7), albizia (Gaudioso et mined that approximately half of the foraging al. 2012), African tulip (Spathodea campanulata) time of RRP was at residential bird feeders, monkeypod (Samanea saman), and silk oak and the availability of this constant anthro- (Grevillea robusta); on one occasion on Hawai‘i pogenic food source could be a key to its Island they were found nesting in a cavity of success in the region. On O‘ahu, RRP also use the native ‘ōhi‘a tree (Paton et al. 1982). bird feeders, but they do not rely upon them for Successful nesting appears to occur in cavities survival (N.P.K., pers. obs). Water sources are 5 m or higher (N.P.K. unpubl. data). Multiple undoubtedly important and may influence this pairs often nest in the same tree if it has species’ strong affinity for mesic environments multiple tree hollows. Cavities are focal points whereit canaccess water resources from pooled for flock socialization, courtship, seasonal pair water in the canopy, or from flowers or fruit in formation and pairs often compete for nest trees. RRP have not been seen to descend to the cavities (Figure 1); fierce fights between RRP ground in the Hawaiian Islands despite this when choosing their nest cavities can be beinga regular occurrenceinJapan (Matsunaga observed (Bluekens 2002). The lack of avail- and Fujii 2018) and France (Clergeau and able (unoccupied) cavities in an area likely Vergnes 2011). Artificial feeding of RRP motivates pairs to explore new geographic should be discouraged especially in areas that areas for usable cavities, as nesting cavities are do not yet have RRP but are near a RRP a limited resource for RRP (Bluekens 2002, population; it only takes a few birds to find a Strubbe and Matthysen 2007). food source and more will quickly follow.

Food Resource DEMOGRAPHY The presence of not only nesting resource but Reproduction and Aging also food species is important for RRP presence (Matsunaga and Fujii 2018). Trees The breeding biology of the RRP on the with legume seed crops such as Prosopis spp., Indian sub-continent is well known, though it Ceasalpinia spp., Pithecellobium dulce, and is poorly known on Pacific islands. In south- Cassia spp., as well as other seed crops ern India, RRP begin nesting as early as including a few with fleshy pulp such as December while in northern India they begin Terminalia spp., Livistona chinensis, and Man- nesting in February (Lamba 1966, Shivanara- gifera indica appear to be indicative of areas yan et al. 1981, Forshaw 1989, Juniper and this species frequents on a daily or weekly Parr 1998). Clutch size ranges from two to six basis on O‘ahu (N.P.K., pers. obs.). Guava and eggs (Lamba 1966, Shivanarayan et al. 1981), corn comprised the majority of the RRP diet with the median being four eggs, two of which on Kaua‘i, and corn is planted year-round so are generally fertile (Lambert et al. 2009, this food resource is readily available (Shiels et Tayleur2010, Butler et al. 2013). RRP spend a al. 2018). Matsunaga and Fujii (2018) have total of 9–10 weeks in the nest: 3 weeks suggested that the RRP distribution shifts incubating eggs, and 6–7 weeks feeding the around Tokyoaccording to availability of food young (Lamba 1966, Shivanarayan et al. 1981, resources, including nectar sought from Forshaw 1989, Juniper and Parr 1998). In several species of flowering trees. In tempe- their Indian range, RRP enjoy a relatively rate areas where this species has naturalized high rate of breeding success. Lamba (1966) (e.g., Japan, California, England, Nether- examined 33 nests and found that an average lands, Belgium, Germany), RRP appear to of 3.0 young fledged per nest. Shivanarayan 438 PACIFIC SCIENCE • October 2019 et al. (1981) examined 66 nests and found that Population Dynamics an average of 1.7 young fledged per nest. This A repeated pattern for RRP introductions is lower reproductive success was attributed to that there appears to be a pronounced lag time predation by crows and snakes (Shivanarayan between initial introduction and before a et al. 1981). In England, Butler (2003) studied rapid and exponential population increase. 108 RRP nests during 2001–2003 and con- For example, RRP temporarily established in firmed that fledging rates averaged 1.9 (±0.1 the U.K. in 1855, and it took ∼140 years to SE) young per nest. Twelve RRP nests establish a self-sustaining population (Lever inspected in the Greater London area during 1987, Tayleur 2010). The lag time from initial 1997–1998 revealed an average of 0.8 young introduction to rapid and exponential increase fledged per nest (Pithon and Dytham 1999). was ∼40 years on Kaua‘i and ∼70 years on On O‘ahu, nesting begins uniformly about the O‘ahu (Figure 8). Many factors can influence middle of January and the chicks fledge during lag times and examples of the Allee effect (e.g., the first half of April (N.P.K, unpubl. data). number of individuals introduced and main- Only a single clutch is raised though re- tained, number of introduction events, male clutching is possible if eggs are lost early in to female ratio, available resources; Lockwood the incubation period (Lambert et al. 2009). et al. 2007). In the lag time years, RRP During the 2012 and 2013 breeding seasons, out populations appear to fluctuate and may even of seven nests on O‘ahu there was an average temporarily decrease in some years. Estab- fledgling success rate of 3.0 chicks per nest with lished RRP populations on both Hong Kong each pair producing two to four fledglings (N.P. Island in the 1980s and 1990s (Leven and K, unpubl. data). RRP Reproduction rates on Corlett 2004), and Honshu Island recently (S. other Pacific islands are unknown aside from Matsunaga, pers. comm.), showed significant evidence from one nest cavity on Hawai‘iIsland reductions over several years despite an containing four eggs that resulted in just one absence of any population control of the birds fledgling (Paton et al. 1982). by humans. Aside from the well-established RRP are known to reach maturity at about RRP populations on the islands of Kaua‘i, 1.5 years old, which is prior to acquiring their O‘ahu, and Honshu that each have >1,000 mature plumage pattern (Butler 2003). Adult plumage is usually complete at about 2–2.5 years of age (Butler 2003). Senar et al. (2019) followed 156 wild RRP for 13 years in Barcelona, Spain, and estimated annual sur- vival probabilities for adults (0.83, 95% confidence interval [CI]: 0.77–0.87) and juveniles during their first (0.57, 95% CI: 0.37–0.79), second (0.79, 95% CI: 0.58–0.87), and third winter (0.83, 95% CI: 0.65–0.88); adult females had slightly higher survival (0.87, 95% CI: 0.78–0.93) than adult males (0.80, 95% CI: 0.73–0.86). Average life-span of these wild RRP was estimated as 4.8 years (95% CI: 3.6–6.4 years), although one bird was 12 years old and one was 14 years old at FIGURE 8. Rose-ringed parakeet (Psittacula krameri) population estimates of established birds in the wild on the end of the study. In captivity, RRP O‘ahu (1930–2018), and Kaua‘i (1960–2018). Estimates generally live for 20 years (Pithon 1998) are minimum number known alive based on visual and may live for as long as 34 years (Brouwer surveys. The pre-2012 (O‘ahu) and pre-2010 (Kaua‘i) et al. 2000). RRP do not maintain strong pair data points were reported by Pyle and Pyle (2017) using Bishop Museum records, and the 2012–2018 counts were bonds outside of the breeding season and it is by author N.P.K. (O‘ahu) and the 2011 and 2018 counts not clear if pairs consist of the same partners were by author A.B.S. (Kaua‘i); see text for methods from year to year. description. Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 439

RRP, the remaining islands in the Pacific with produce a single count for this roost. At the established RRP have highly fluctuating Līhu‘e roost, there was a single count on 23 populations as a result of management January 2018, and the observer was stationed (New Zealand) or presumably natural causes on the ground, about 40 m northeast of the (Hawai‘i, Hong Kong; Table1). Furthermore, backside of the Historic County Building Kaua‘i and O‘ahu are the only two islands in (Rice Street), which was the same location as the Pacific with recent and regular island-wide in 2011. population estimates during the last 6–12 RRP on O‘ahu have experienced the same years. population growth patterns as on Kaua‘i, yet Kaua‘i has the largest population of RRP in the O‘ahu population is just behind Kaua‘iin the Pacific. The first birds introduced into the timing and total numbers of birds (Figure 8). wild were in the 1960s, and by 1982 the The current estimate (in 2018) for RRP on population had grown to 90 birds (Pyle and O‘ahu is 4,557 individuals, of which 2,600 are Pyle 2017). By 1994, late 2000s, and 2011, the at the Honolulu roost, 1,723 are at the population was estimated at 150–200 birds, Waikele roost, and 234 are at the Olomana 500–1,000 birds, and over 2,000 birds, roost (N.P.K. unpubl. data). The author respectively (Gaudioso et al. 2012, Pyle and N.P.K. sampled the Honolulu roost since Pyle 2017). The most recent count of RRP in 2005 using the same sampling techniques January 2018 on Kaua‘i was 6,801 birds (A.B. during the post-fledging, non-nesting season. S., unpubl. data; Figure 8); therefore, the RRP Briefly, the sampling method used by N.P.K. population on Kaua‘i more than tripled within was similar to that described above for Kaua‘i, 6 years and experienced an average annual but flocks at the Honolulu roost were growth rate of 22.5%. Whereas the count at photographed as they arrived in the early each of the two roosts on Kaua‘i(Lāwai- evening, typically 18:30–19:30, depending Po‘ipū-Kōloa and Līhu‘e) was nearly evenly upon the season, and tallying the birds was split in December 2011 (Gaudioso et al. subsequently completed using a computer 2012), the Lāwai-Po‘ipū-Kōloa roost monitor. This sampling of the Honolulu roost accounted for two-thirds of the birds on the yielded an average annual growth rate of 21% island in 2018 (A.B.S., unpubl. data). A brief based on roost counts of 181 ± 56 (mean ± description of the methods that A.B.S. used in SE; n = 3), 213 ± 63, (n = 4), 448 ± 20 (n = 2), 2011 and 2018 to count the RRP population 1,210 (n = 5) in non-nesting portions of years on Kaua‘i are summarized here. RRP at the 2005, 2006, 2010, and 2014, respectively. The two roosts (Lāwai-Po‘ipū-Kōloa and Līhu‘e) O‘ahu and Kaua‘i estimated annual growth were counted during the same week, and by rates are similar to that in the U.K. (27%) the same observer (A.B.S) in December 2011 estimated by Butler et al. (2013) where RRP (see Gaudioso et al. 2012) and January 2018. totaled 10,000 individuals by 2004 and may Each RRP count occurred at dawn (0600– now be as high as 30,000 birds (Butler et al. 0715 hours), as individuals began to leave the 2013). Upon author N.P.K. counting the roost area, until the last individual left the Honolulu roost during the incubation nesting roost. As in Gaudioso et al. (2012), and from a phase of 2015, only 789 (n = 3) RRP returned vantage point nearby the roost (generally in to the nightly colonial roost, meaning about the direction the birds flew, which was north 35% (or 421 individuals) of the RRP were or inland towards the mountains), RRP were missing at the nighttime roost compared with counted individually, but if flock size was large roost counts before the nesting season in (>50 individuals) RRP were counted in 2014. The large number of missing birds at increments of five birds. During the 2018 the colonial evening roost probably can be count at the Lāwai-Po‘ipū-Kōloa roost, the attributed to these birds being females that observer was stationed on the rooftop of a were incubating eggs at their nest sites. When 12 m tall building east of Lāwai Beach Resort incubating and feeding their young, females (Lāwai Road) to complete the counts on 22 remain at the nest day and night despite male and 24 January; these counts were averaged to mates returning to the colonial roost each TABLE 1 Summary of Characteristics Associated with Rose-Ringed Parakeet (RRP) Currently Established on the Seven Pacific Islands (Four Countries). Information Based on 2018 Evidence (i.e., Physical Surveys, Literature, Local Communications). RRP Establishment was defined as Confirmed Successful Breeding in the Wild

Pacific Island Original Local Human Main Type of Minimum # Minimum # Current Recent Population Reference Date Perception Damage of Current of Individuals Management Growth Introduced Roosts Kaua‘i, USA 1960s Mostly disliked by Agriculture, 2 6,801 Some local control Tripled in last 6 Shiels, unpubl. data; farmers & land primarily corn & in corn fields, and years; 22.5% annual Shiels et al. 2018; owners; some tropical fruit scant deterrence at growth rate Avery & Shiels residents like roost 2018; Gaudioso et al. 2012 O‘ahu, USA 1930s Unknown but Ornamental trees 3 4,557 None 21% annual growth Kalodimos, unpubl. probably mixed (flower, fruit, seed rate data; Pyle & Pyle destruction) 2017 Hawai‘i, USA 1980s Unknown but Unknown 1 6 None Unknown I. Leinbach, pers. probably mixed comm.; Pyle & Pyle 2017 Honshu, Japan 1969 Mixed; enjoyed at Flowers, primarily 3 1,570 None Unknown S. Matsunaga, pers. backyard feeders, cherry blossoms comm.; Matsunaga but dislike scattering & Fijii 2018 of flowers Hong Kong, China 1903 Mostly liked Unknown 2 15 None Unknown M. Leven, pers. comm.; Leven & Corlett 2004 North Island, New 2005 Dislike Unknown 1 6 Aggressive Unknown L. Maria, pers. Zealand eradication efforts comm. South Island, New 2016 Dislike Unknown 1 30 Aggressive Unknown L. Maria, pers. Zealand eradication efforts comm. Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 441 night. This suggests approximately 70% of MANAGEMENT the RRP population (male-female pairs) were The two main management techniques prac- reproductively active during the 2014–2015 ticed on RRP include population reduction survey; such a high proportion of the and deterrence. Common deterrents that population being reproductively active would largely remain untested for their efficacy greatly contribute to the exponential increase against RRP include: gas- or audio-cannons, seen in RRP populations (Figure 8). RRP are pyrotechnic exploders, hand-held lasers, auto- the most numerous, most widely dispersed, mated water sprays at canopy roosts, reflec- and have the most rapid population growth of tors, mirrors, ribbons, bird of prey or dead any parrot species in the Hawaiian Islands. RRP effigies, and covering resources with nets (Gaudioso et al. 2012, Khan et al. 2013, Klug PREDATORS et al. 2019). In Pakistan, where RRP com- monly eat and destroy fruit trees, Khan et al. Few effective predators of RRP have been (2013) had success deterring RRP using an documented on Pacific islands where RRP ultrasonic sound player in the center of fruit have established. Birds of prey, such as hawks orchard fields. The sound player acted as a and owls, would be potential predators of bird repellent and had audio playback of RRP, and observations in Honshu have sounds of “alarming noises of some fearsome revealed that northern goshawk (Accipiter animals”. Using this technology, RRP visits gentilis), peregrine falcon (Falco peregrinus), were reduced by 50% in yellow guava and large-billed crow (Corvus macrorhyncos) orchards and 90% in mango orchards, and have caught and eaten RRP (S. Matsunaga, apparently the sound player also helped pers. comm.). In Europe, the main predator of reduce damage to sunflower fields (Khan et RRP are squirrels (Sciurus spp.), which prey al. 2013). Residents near evening roosts on upon the eggs and chicks in the nesting Kaua‘i have used hand-held lasers to annoy cavities (Mori et al. 2013). Introduced grey and scatter the RRP just after they have squirrels (Sciurus carolinensis) depredated five entered their roost tree (M. Martin, pers. RRP tree-cavity nests in the U.K. (Butler et al. comm.). A common challenge with using any 2013), and owls (Strix aluco) and raptors deterrent is that animals can become habi- (peregrine falcon, Eurasian sparrowhawk tuated and therefore the efficacy of the [Accipiter nisus], and hobby falcon [Falco deterrent often decreases over time (Avery subbuteo]) have been documented taking and Shiels 2018). RRP in central London (Hancock and Martin Population reduction or eradication of 2015). Common mynas (Acridotheres tristis) introduced RRP populations has been chal- are the dominant cavity nester on several lenging, largely due to the difficulty in lethally Hawaiian islands and these non-native birds controlling their populations when they live overpower RRP for nest cavities as well as near humans, the lack of restrictions for their potentially depredate RRP eggs and young breeding or sale, and the substantial human nestlings in tree cavities (N.P.K., pers. obs.). regard for them in many settings (Crowley Feral honeybees (Apis mellifera) have been et al. 2019). Even when RRP populations are observed occupying tree hollows and there- small and all individuals can be removed from fore preventing RRP from using cavities for the wild on an island, new introductions can nesting on O‘ahu (N.P.K. unpubl. data) and in easily occur due to RRP maintained as pets Europe (Menchitti and Mori 2014). RRP and caged birds. New Zealand has lethally appear threatened by some seabirds, as managed several incipient RRP populations, observed at a RRP evening roost on O‘ahu removing them quickly after their establish- where RRP were taking off in masses and ment. In 2015 in Hikutaia (North Island), 12 circling in a tight flock in response to birds were successfully removed by tracking, overhead-soaring great frigate birds, Fregata trapping, and shooting; in 2016 in Havelock minor, and white terns, Gygis alba (N.P.K., North (North Island), six birds were successfully pers. obs.). 442 PACIFIC SCIENCE • October 2019 shot and there is a follow-up plan to eliminate difficult, RRP maintain a high altitude and the few that were missed; and in 2016 in resting height that is often out of range for Christchurch (South Island), about 30 birds shotgun control (particularly on Kaua‘i), and were purposefully released from an aviary at a relative to the 545 RRP on Mahé Island there zoo when a fire occurred, and these birds are many more RRP on O‘ahu (4,560) and successfully bred over a few years and they are Kaua‘i (6,800) and these islands are much now under a trapping, shooting, and monitor- larger than Mahé. ing program (L. Maria, pers. comm.). Of the Because invasive species eradications from Pacific islands with RRP currently established islands are not always possible or successful, (Ta b l e 1 ), it appears that North and South population suppression is often the next best Island of New Zealand are the most likely to be management practice. Because RRP have few RRP-free in the near future, and this is due to effective predators and long lifespans of 14+ their early, aggressive, and persistent RRP years (Pithon 1998, Senar et al. 2019), the removal efforts. most effective way to reduce population The only eradication of RRP on islands growth is by removing breeding aged birds that we are aware of has recently occurred in from the population (Butler 2003). Whereas the Seychelle Islands, Indian Ocean, summar- RRP are protected by law in Hong Kong, ized by Bunbury et al. (2019). There, 545 RRP there was a period of RRP lethal control in were removed from Mahé Island (157 km2), Japan in the 1980s to protect agricultural largely by shooting after poor mist-netting crops from RRP damage. Approximately 100 success. Of particular importance to this RRP were shot in Chiba prefecture to help island-wide eradication of RRP was to reduce damage to peanuts being grown in the advertise these efforts and their importance region (S. Matsunaga, pers. comm.). In the to the local community early in the planning Hawaiian Islands, parrot population reduc- stages. Bunbury et al. (2019) determined that tion has occurred for RRP on Kaua‘i and 80% of the respondents found out about the mitred parakeets on Maui. On Kaua‘i where local RRP problems and eradication efforts RRP are particularly destructive to the seed through the informational advertisements corn industry that is managed by multi- that were broadcast on television, particularly national companies, hundreds of thousands just before the nightly news. The RRP of dollars are spent each year in efforts to eradication strategy that worked best was to reduce crop damage by RRP (Koopman and not disturb the birds at their vulnerable roost Pitt 2007). Despite contractors such as USDA sites because that would scatter the population APHIS Wildlife Services harvesting a few from the very few reliable roost sites, but hundred parakeets each year from corn fields instead they recommended identifying the in Kaua‘i since 2002, it has had no obvious flight patterns of the birds returning to roosts effect on the island’s RRP population (Gau- each evening and then shooting the birds in dioso et al. 2012, Avery and Shiels 2018). strategic locations along the flight paths as Additional lethal control methods attempted they stopped briefly to congregate. Mahé to reduce the RRP population on Kaua‘i have Island is approximately nine times smaller included shooting at their nighttime roost, than Kaua‘i, and about 10 times smaller than and live-trapping. Aside from live-trapping O‘ahu, and both of these Hawaiian islands attempts, which yielded no catch (Gaudioso have been informally discussed for RRP et al. 2012), shooting can result in harvests of eradication but no formal evaluation for several birds a day (corn fields) to ∼120 in a significant control or eradications needs has night (roosts while RRP are sleeping). On occurred. Significant barriers to replicating O‘ahu, quadcopter drones have been shown to the Mahé Island RRP eradication techniques attract perching and undetected RRP in a on O‘ahu and Kaua‘i include the typical RRP geographic area to synchronously become flight patterns to roost (particularly on O‘ahu) airborne, then circle the drone, and eventually are through residential and urban areas that mob the drone (A.B.S. and N.P.K., pers. obs.); would make safe and persistent shooting this may be a method to move large flocks of Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 443

RRP to low altitude locations that are safe for safeguard against RRP depredation of coffee shooting or netting. However, RRP quickly fruit and seed. It is unknown if coffee orchard become aware of these flock reduction cullings were proactive measures to eliminate techniques after just a few days of control the individuals or reactionary measures in efforts and then change their visitation or response to crop depredations (Pyle and Pyle roosting patterns (W. Bukoski, pers. comm.). 2009). Unlike the RRP population, the mitred As a result of low efficacy of previously parakeet population on Maui was reduced attempted RRP damage control methods, primarily via shooting by approximately 85% corn companies on Kaua‘i have recently (from about 150 to 20 individuals) in 8 years begun to cover their corn fields with large (Radford and Penniman 2014). amounts of netting (Figure 9), which has The use of toxicants to control RRP has helped to reduce damage and seed loss. not been tested to our knowledge, despite We know of no attempts to control RRP avian toxicants commonly field tested to populations on O‘ahu, and the Hawai‘i Island control agricultural pests (Linz 2013). Tox- RRP population is so small that land managers icants mixed into an attractive bait-matrix are appear complacent if they even know the typically unpopular with the public. Prior to population exists. RRP in the coffee growing any use of a toxicant, a thorough risk region of Kona, Hawai‘i Island, historically assessment would be needed to ensure effec- may have been culled by orchard farmers to tiveness on RRP and prevent non-target species exposure. There are no known species-specific biological control agents for the RRP, and therefore introducing RRP diseases is not currently a management option. Unlike toxicants, fertility control is a non-lethal method that could be trialed for RRP population reduction. The contraceptive Diazacon has shown success at limiting reproduction of RRP in a laboratory setting in the U.K. when RRP were dosed with 18 mg/kg of Diazacon for 10 d (Lambert et al. 2010). Although promising from these labora- tory results, Diazacon still requires testing to ensure an attractive formulation and delivery system for RRP, an appropriate dosing frequency, and a risk assessment that accounts for non-target species effects (Avery and Shiels 2018). RRP on Kaua‘i could not be drawn to feeding stations or live-decoy traps that were placed in corn fields frequented by RRP (Gaudioso et al. 2012). Therefore, a major barrier to effective use of toxicant or contraceptive bait is successfully drawing RRP into a control device or bait dispensary. Management actions for RRP are recom- mended to be focused on specific agricultural fields, locations where they are causing damage, potential nesting areas, and on their flight path approaches to evening roosts. FIGURE 9. Netting is currently used to cover whole fields of corn (Zea mays) on Kaua‘i to prevent seed damage and Toxicants or fertility control baits could be consumption by rose-ringed parakeet (Psittacula krameri). deployed in bait stations to reduce or Photographs by Sarah Thompson. eliminate non-target exposure. Because nest 444 PACIFIC SCIENCE • October 2019 cavities are often a limiting resource for RRP, tropical feelings that they may bring to some artificial cavities with self-resetting lethality people is expected to diminish as local RRP devices inside could be crafted and deployed populations exponentially grow. RRP estab- in attempt to target adult breeders. If RRP are lishment and population growth generates lethally controlled at the roost (e.g., via human complaints and suffering from their shooting or large mist nets deployed using loud calls in large flocks, defecation beneath long poles or a bucket truck; Avery and Shiels mass-roosting, extreme crop damage, orna- 2018), it is likely that the population will mental tree damage, potential predation of scatter and become more difficult to predict native seed plants and spread of non-native roosting sites and flight routes into roosts seeds, and potential to transfer pathogens to (N. Bunbury, pers. comm.). Thus, it is other wildlife and to humans. Therefore, the recommended that final evening roosting generally complacent attitude towards RRP locations not be used as management locations establishment outside their native range is, in because birds will desert evening roost sites for our view, inappropriate. Instead, aggressive months if they feel threatened (Kalodimos management strategies where action is 2008) and a level of predictability of population immediate and science-based, like practiced movements will have been lost (Bunbury et al. in New Zealand, is the most responsible way 2019). Instead of lethal control activities at the to address new introductions of RRP, and roost, it is recommended that the RRP flight additional science-based management strate- patterns on the way to their evening roosts are gies should be employed to reduce and exploited and RRP are removed along their hopefully eliminate large populations like regular approach to roost. those established on Kaua‘i and O‘ahu. In particular, local government and biosecurity authorities should take measures to prevent PROGNOSIS AND RECOMMENDATIONS this species from establishing on new islands Parrots have loud and persistent vocalizations, and in new areas with valued crops, rare plant gregarious flocking and evening roosting species that reproduce by seed, and endemic behaviors, and their activity is routine and parrot and/or cavity-nesting fauna. predictable making detection, monitoring, and management possible and worthwhile. ACKNOWLEDGMENTS Approximately 16% of the world’s parrot population (60 of 335 species) has established We are grateful for the local status updates on breeding populations outside of their native RRP in 2018 provided by Lisa Maria (New ranges (Menchetti and Mori 2014); by far the Zealand), Satomi Matsunaga (Japan), Michael most widespread and damaging of all of these Leven (Hong Kong), and Israel Leinbach parrot colonist species is the RRP. Given their (Hawai‘i Island). We thank Nancy Bunbury, proven ability to survive in tropical and Diederik Strubbe, and two anonymous temperate climates, and from Japan to New reviewers for helpful suggestions that Zealand in the Pacific, most human-colonized improved the manuscript. Findings and con- islands in the Pacific could support RRP populations if they were to become acciden- clusions in this publication have not been tally or purposefully introduced. The like- formally disseminated by the U.S. Depart- lihood of such introductions is high given the ment of Agriculture and should not be popularity of this species in the pet trade and construed to represent any agency determina- in zoos and aviaries. Urban areas and tion or policy. agricultural lands at low elevation (500 m) are preferred RRP habitats within an island Literature Cited for foraging, nesting, and roosting. Thus, human interactions with these birds are Ali, S., and S. D. Ripley. 1969. Handbook of certain once RRP establish, and their aesthetic the birds of India and Pakistan. Vol. 3. beauty of bright green plumage and exotic/ Oxford University Press, Oxford. Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 445

Avery, M. L., and A. B. Shiels. 2018. Monk the rose-ringed parakeet Psittacula krameri in and rose-ringed parakeets. Pages 333–357 England during a period of rapid population in W. C. Pitt, J. C. Beasley, and G. W. expansion. Bird Study 60:527–532. Witmer, eds. Ecology and management of Cassey, P. 2002. Life history and ecology terrestrial vertebrate invasive species in the influences establishment success of intro- United States. Taylor and Francis, CRC duced land birds. Biol. J. Linn. Soc. Press, New York. 76:465–480. Bashir, E. A. 1979. A new “parotrap” adapted Clergeau, P., and A. Vergnes. 2011. Bird from the MAC trap for capturing live feeders may sustain feral rose-ringed parakeets in the field. Bird Control parakeets Psittacula krameri in temperate Seminars Proceedings, Paper 23. Europe. Wild. Biol. 17:248–252. Bittner, M. 2008. The wild parrots of Collar, N. 1997. Family Psittacidae (Parrots). Telegraph Hill: A love story . . . with Pages 296–339 in J. del Hoyo, A. Elliott, wings. Harmony. and J. Sargatal, eds. Handbook of birds of Bluekens, K. 2002. Inleidende studie naar het the world. Vol. 4. Lynx Edicions, nestgebruik van halsbandparkieten en de Barcelona. concurrentie met inheemse volgelsoorten. Corlett, R. T. 1998. Frugivory and seed M.S. Thesis. Department of Biology, dispersal by vertebrates in the oriental University of Antwerp. (Indomalayan) region. Biol. Rev. Brouwer, K., M. L. Jones, C. E. King, and H. Cambridge Phil. Soc. 73:413–448. Schifter. 2000. Longevity records for Cotton, P. A. 2001. The behavior and Psittaciformes in captivity. Int. Zoo Yearb. interactions of birds visiting Erythrina fusca 37:299–316. flowers in the Colombian Amazon. Bunbury, N., P.Haverson, N. Page, J. Agricole, Biotropica 33:662–669. G. Angell, P. Banville, A. Constance, Cramp, S. 1985. Handbook of the birds of J. Friedlander, L. Leite, T. Mahoune, Europe, the Middle East and North Africa. E. Melton-Durup, J. Moumou, K. Raines, The birds of the Western Palearctic. Vol. J. van de Crommenacker, and F. Fleischer- IV. Oxford University Press, Oxford. Dogley. 2019. Five eradications, three Crowley, S. L., S. Hinchliffe, and R. A. species, three islands: overview, insights McDonald. 2019. The parakeet paradox: and recommendations from invasive bird understanding opposition to introduced eradications in the Seychelles. Pages 282– species management. J. Environ. Manag. 288 in C. R. Veitch, M. N. Clout, A. R. 229:120–132. Martin, J. C. Russell, and C. J. West, eds. del Hoyo, J., A. Elliot, and J. Sardatal. 1997. Island invasives: Scaling up to meet the Handbook of the birds of the world. Vol. 4: challenge. Occasional Paper SSC no. 62. Sandgrouse to cuckoos. Lynx, Barcelona. IUCN, Gland, Switzerland. DLNR. 2014. Hawai‘i Department of Nat- Butler, C. J. 2003. Population biology of the ural Resources 2013. Accessed April 2016. introduced rose-ringed parakeet Psittacula http://dlnr.hawaii.gov/dofaw/files/2013/ krameri in the U.K. Ph.D. diss., University 09/HAR-13-124-Draft.pdf. of Oxford, Oxford. Eguchi, K., and H. E. Amano. 2004. Spread of ———. 2005. Feral parrots in the continental exotic birds in Japan. Ornithol. Sci. 3:3–11. United States and United Kingdom: past, England, M. D. 1974. Feral populations of present, and future. J. Avian Med. Surg. parakeets. British Birds 67:393–394. 19:142–149. Fletcher, M., and N. Askew. 2007. Review of Butler, C. J., and A. Gosler. 2004. Sexing and the status, ecology, and likely future spread aging rose-ringed parakeets Psittacula of parakeets in England. Central Science krameri in Britain. Ringing Migr. 22:7–12. Laboratory, York, U.K. Butler, C. J., W. Cresswell, A. Gosler, and Fogell, D. J., R. O. Martin, N. Bunbury, C. Perrins. 2013. The breeding biology of B. Lawson, J. Sells, A. M. McKeand, 446 PACIFIC SCIENCE • October 2019

V. Tatayah, C. T. Trung, and J. J. Groom- Jackson, H., D. Strubbe, S. Tollington, bridge. 2018. Trade and conservation R. Prys-Jones, E. Matthysen, and J. J. implications of new beak and feather Groombridge. 2015. Ancestral origins and disease virus detection in native and invasion pathways in a globally invasive introduced parrots. Conserv. Biol. bird correlate with climate and influences 32:1325–1335. from bird trade. Mol. Ecol. 24:4269–4285. Forshaw, J. M. 1989. Parrots of the world. Juniper, T., and M. Parr. 1998. Parrots: A Blandford Press, London. guide to parrots of the world. Yale Uni- Galetti, M. 1993. Diet of the scaly-headed versity Press, London. parrot, Pionus maximiliani, in a semidecid- Kalodimos, N. 2008. Determination of move- uous forest in southeastern Brazil. Biotro- ment and foraging patterns of Aratinga pica 25:419–425. erythrogenys (Aves: Psittacidae) on O‘ahu, Garrett, K. L., K. T. Mabb, C. T. Collins, and Hawai‘i, using mist-net live capture and L. M. Kares. 1997. Food items of natur- radio telemetry. Report to the Division of alized parrots in southern California. West. Forestry and Wildlife, Department of Birds 28:196–201. Land and Natural Resources, State of Gassman-Duvall, R. 2002. Naturalized Hawai‘i, Honolulu, Hawai‘i. mitred conures on Maui and other par- ———. 2013. First account of a nesting rots—pest evaluation and management population of Monk parakeets (Myiopsitta proposal for non-native, invasive bird monachus) with nodule-shaped bill lesions in species. Unpublished draft report, Maui Katehaki, Athens, Greece. Bird Pop. 12:1–6. Invasive Species Committee, Makawao, Khan, H. A., M. Javed, A. Tahir, and M. Maui, Hawai‘i. Kanwal. 2013. Limiting the ring-necked Gaudioso, J. M., A. B. Shiels, W. C. Pitt, and parakeet (Psittacula krameri) damage on W. P. Bukoski. 2012. Rose-ringed parakeet guava (Psidium guajava) and mango (Man- impacts on Hawaii’s seed crops on the gifera indica) with an ultrasonic sound island of Kauai: Population estimate and player in a farmland in Faisalabad, Paki- monitoring of movements using radio stan. Afr. J. Agric. Res. 8:6608–6614. telemetry. Final report QA 1874, USDA, Klug, P. E., W. P. Bukoski, A. B. Shiels, B. M. National Wildlife Research Center, Hilo, Kluever, and S. R. Siers. 2019. Rose-ringed Hawai‘i. parakeets. Wildlife Damage Management Gordenker, A. 2009. Feral parakeets. The Technical Series. USDA, APHIS, WS Japan Times, March 19, 2009: https:// National Wildlife Research Center. Fort www.japantimes.co.jp/news/2009/03/19/ Collins, Colorado. 16 pp. https://www.aphis. reference/feral-parakeets/#.WszUyf4U usda.gov/wildlife_damage/reports/Wildlife mot. %20Damage%20Management%20Techni- Groombridge, J. J., C. J. Jones, R. A. Nichols, cal%20Series/Rose%20Ringed%20Para- M. Carlton, and M. W. Bruford. 2004. keet_WDM%20Technical%20Series.pdf Molecular phylogeny and morphological Komar, N., S. Langevin, S. Hinten, change in the Psittacula parakeets. Mol. N. Nemeth, E. Edwards, D. Hettler, B. Phylogenet. Evol. 31:96–108. Davis, R. Bowen, and M. Bunning. 2003. Hancock, R., and J. R. Martin. 2015. Preda- Experimental infection of North American tion of rose-ringed parakeets by raptors birds with the New York 1999 strain of west and owls in Inner London. British Birds Nile virus. Emerg. Infect. Dis. 9:311–322. 108:349–353. Koopman, M. E., and W. C. Pitt. 2007. Crop Hernandez-Brito, D., M. Carrete, C. Ibanez, J. diversification leads to diverse bird pro- Juste,andJ.L.Tella.2018.Nest-site blems in Hawaiian agriculture. Human- competition and killing by invasive Wild. Conflicts 1:235–243. parakeets cause the decline of a threatened Lamba, B. S. 1966. Nidification of some bat population. R. Soc. Opn. Sci. 5(5):172477. common Indian birds: 10. The rose-ringed Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 447

parakeet, Psittacula krameri scopoli. Proc. Medeiros, A. C., L. L. Loope, P. Conant, and Zool. Soc. Calcutta 19:77–85. S. McElvaney. 1997. Status, ecology, and Lambert, M. S., G. Massei, J. Bell, C. Haigh, management of the invasive plant, Miconia and D. P. Cowan. 2009. Reproductive calvescens DC (Melastomataceae) in the success of rose-ringed parakeets Psittacula Hawaiian Islands. Bishop Museum Occa- krameri in a captive UK population. Pest sional Papers 48:23–36. Manag. Sci. 65: 1215–1218. Menchetti, M., and E. Mori. 2014. World- Lambert, M. S., G. Massei, C. A. Yoder, and wide impact of alien parrots (Aves Psitta- D. P. Cowan. 2010. An evaluation of ciformes) on native biodiversity and Diazacon as a potential contraceptive in environment: a review. Ethol. Ecol. Evol. non-native rose-ringed parakeets. J. Wild. 26:172–194. Manag. 74: 573–581. Menchetti, M., R. Scalera, and E. Mori. 2014. Leven, M. R., and R. T.Corlett. 2004. Invasive First record of a possibly overlooked birds in Hong Kong, China. Ornithol. Sci. impact by alien parrots on a bat (Nyctalus 3:43–55. leisleri). Hystrix, Ital. J. Mammal. 25:61–62. Lever, C. 1987. Naturalised birds of the Menchetti, M., E. Mori, and F. M. Angelici. world. Longman Publishing, Harlow, U.K. 2016. Chapter 12: Effects of the recent Linz, G. M. 2013. Blackbird population world invasion by ring-necked parakeets management to protect sunflower: A Psittacula krameri. Pages 253–266 in history. Proc. Wild. Damage Manag. Conf. F. M. Angelici, ed. Problematic wildlife. 15:42–53. Springer International Publishing, Lockwood, J. L. 1999. Using taxonomy to Switzerland. predict success among introduced avifauna: Mentil, L., C. Battisti, and G. M. Carpaneto. relative importance of transport and estab- 2018. The impact of Psittacula krameri lishment. Conserv. Biol. 13:560–567. (Scopoli, 1769) on orchards: first quanti- Lockwood, J. L., M. F. Hoopes, and M. P. tative evidence for Southern Europe. Belg. Marchetti. 2007. Invasion ecology. Black- J. Zool. 148:129–134. well Publishing, Malden. Miskelly, C. M. 2015 [updated 2018]. Rose- Lohr, C. A. 2012. Human dimensions of ringed parakeet. In C. M. Miskelly, ed. introduced terrestrial vertebrates in the New Zealand Birds Online. www.nzbird Hawaiian Islands. Ph.D. diss., University sonline.org.nz of Hawai‘iatMānoa, Honolulu. Mori, E., L. Ancillotto, M. Menchetti, Mack, A. L., and D. D. Wright. 2005. The C. Romeo, and N. Ferrari. 2013. Italian frugivore community and the fruiting red squirrels and introduced parakeets: plant flora in a New Guinea rainforest: victims or perpetrators? Hystrix, Ital. J. identifying keystone frugivores. Pages 184- Mammal. 24:195–196. –203inJ.L.Dew,andJ.P.Boubli,eds.Tropical Mukherjee, A., C. K. Borad, and B. M. fruits and frugivores: The search for strong Parasharya. 2000. Damage of rose-ringed interactors. Springer, The Netherlands. parakeet, Psittacula krameri Bordeat, to Mase, M., T. Imada, Y. Sanada, M. Etoh, safflower, Carthamus tinctorius L. Pavo N. Sanada, K. Tsukamoto,Y. Kawaoka, and 38:15–18. S. Yamaguchi. 2001. Imported parakeets Novak, S. 2018. Invasive parakeets are harbor H9N2 influenza A viruses that are threatening every crop on the Hawaiian genetically closely related to those trans- Island of Kauai—and multiplying fast. mitted to humans in Hong Kong. J. Virol. Modern Farmer, January 22, 2018. 75:3490–3494. https://modernfarmer.com/2018/01/inva Matsunaga, S., and T. Fujii. 2018. Dietary sive-rose-ringed-parakeets-threatening- preference of invasive rose-ringed parakeet every-crop-on-hawaiian-island-kauai/ Psittacula krameri manillensis around the Paton, P. W. C., C. R. Griffin, and L. H. Tokyo area. FBN 5:13–16. MacIvor. 1982. Rose-ringed parakeet 448 PACIFIC SCIENCE • October 2019

nesting in Hawai‘i: a potential agricultural in a maize cultivated crop in Rajenderna- threat. ‘Elepaio 43:37–39. ger, Hyderabad, Andhrapradesh. Indian Pennybacker, M. 2016. The rising parakeet J. Forestry 3:16–18. population imperils farms and native Reino, L., R. Figueira, P. Beja, M. B. Araujo, plants, officials say. Star Advertiser, January C. Capinha, and D. Strubbe. 2017. Net- 10, 2016. http://www.staradvertiser.com/ work of global bird invasion altered by features/the-rising-parakeet-population- regional trade ban. Sci. Adv. 3:e1700783. imperils-farms-and-native-plants-officials- Roscoe, D. E., W. B. Stone, L. Petrie, and say/ (Accessed May 1, 2016). J. L. Renkavinsky. 1976. Exotic Psittacines Pithon, J. A. 1998. The status and ecology of in New York State. N.Y. Fish Game the ring-necked parakeet Psittacula krameri J. 23:99–100. in Great Britain. Ph.D. diss., University of Runde, D. E., W. C. Pitt, and J. T. Foster. York, York. 2007. Population ecology and potential Pithon, J. A., and C. Dytham. 1999. Breeding impacts of emerging populations of exotic performance of ring-necked parakeets parrots. Pages 338–360 in G. W. Witmer, Psittacula krameri in small introduced W. C. Pitt, K. A. Fagerstone, eds. Mana- populations in southeast England. Bird ging vertebrate invasive species: Proceed- Study 46:342–347. ings of an international symposium. USDA ———. 2001. Determination of the origin of National Wildlife Research Center, Fort British feral rose-ringed parakeets. British Collins, CO. Birds 94:74–79. Sa, R. C., A. A. Cunningham, M. P. Dagleish, Pyle, R. L., and P.Pyle. 2009. The birds of the N. Wheelhouse, A. Pocknell, N. Borel, H. Hawaiian Islands: occurrence, history, dis- L. Peck, and B. Lawson. 2014. Psittacine tribution, and status. B.P. Bishop Museum, beak and feather disease in a free-living Honolulu, HI, U.S.A. Version 1 (31 ring-necked parakeet (Psittacula krameri)in December 2009). http://hbs.bishopmu Great Britain. Eur. J. Wildl. Res. 60:395- seum.org/birds/rlp-monograph/ –398. ———. 2017. The birds of the Hawaiian Saini, H. K., M. S. Dhindsa, and H. S. Toor. Islands: occurrence, history, distribution, 1994. Food of the rose-ringed parakeet and status. B.P. Bishop Museum, Hono- Psittacula krameri: a quantitative study. lulu, HI, U.S.A. Version 2 (1 January J. Bombay Nat. Hist. Soc. 91:96–103. 2017). http://hbs.bishopmuseum.org/ Scalera, R. 2001. Il parrocchetto dal collare birds/rlp-monograph/ Psittacula krameri. Pages 195–199 in R. Radford, A., and T. Penniman. 2014. Mitred Scalera, ed. Invasioni biologiche. Le intro- conure control on Maui. Pages 61–66 in R. duzioni di vertebrati in Italia: un problema M. Timm and J.M. O’Brien, eds. Proc. Vert. tra conservazione e globalizzazione. Corpo Pest Conf. Univ. Calif. Press, Davis, CA. Forestale dello Stato e Ministero delle Rao, G. S., and N. Shivanarayan. 1981. Note Politiche Agricole e Forestali, Rome. on the food of nestlings of rose-ringed Senar, J. C., L. Arroyo, A. Ortega-Segalerva, parakeets in Hyderabad. Pavo 19:97–99. J. G. Carrillo, X. Tomás, T. Montalvo, and Reddy, V. R. 1998a. Bird damage to maize A. Sanz-Aguillar. 2019. Estimating age- crop on the student’s research farm dependent survival when juveniles resem- at Rajendranagar, Hyderabad, Andhra ble females: invasive ring-necked parakeets Pradesh. Pavo 36:77–78. as an example. Ecol. Evol. 9:891–898. ———. 1998b. Studies on damage to sorghum Shiels, A. B. 2011. Frugivory by introduced by the rose-ringed parakeet, Psittacula black rats (Rattus rattus) promotes dispersal krameri, at Rajendranagar, Hyderabad, of invasive plant seeds. Biol. Invasions Andhra Pradesh. Pavo 36:79–80. 13:781–792. ———. 1999. Intensive damage to maize crop Shiels, A.B., W. P.Bukoski, and S. R.Siers.2018. by rose-ringed parakeet at a student’s farm Diets of Kauai’s invasive rose-ringed parakeet Psittacula krameri, Pacific Island Invasive Species • Shiels and Kalodimos 449

(Psittacula krameri): evidence of seed preda- within-taxon niche structure and associa- tion and dispersal in a human-altered land- tion with humans in the native range. scape. Biol. Invasions 20:1449–1457. Diversity Distrib. 21:675–685. Shivanarayan, N. 1981. Note on the food of Tayleur, J. R. 2010. A comparison of the rose-ringed parakeet in Hyderabad. Pavo establishment, expansion and potential 19:97–99. impacts of two introduced parakeets in the Shivanarayan, N., K. S. Babu, and M. H. Ali. United Kingdom. Proceedings of the Con- 1981. Breeding biology of rose ringed ference: The Impacts of Non-Native species. parakeet Psittacula krameri at Maruteru. British Ornithologists’ Union, Peterbor- Pavo 19:92–96. ough, UK. https://www.bou.org.uk/bouproc Skutch, A. F. 1983. Birds of tropical America. net/impacts-of-non-native-species/. University of Texas Press, Austin. Temara, K., and R. Arnhem. 1996. Perruches Stone, E. G., G. Montiel-Parra, and T. M. a collier (Psittacula krameri) victimes des Pérez. 2005. A survey of selected parasitic conditions climatiques en region Bruxel- and viral pathogens in four species of loise. Aves 33:128–129. Mexican parrots, Amazona autumnalis, Thabethe, V., A.-L. Wilson, L. A. Hart, and Amazona oratrix, Amazona viridigenalis, C. T.Downs. 2015. Ingestion by an invasive and Rhynchopsitta pachyrhyncha. J. Zoo parakeet reduces germination success of Wild. Med. 36:245–249. invasive alien plants relative to ingestion by Strubbe, D., and E. Matthysen. 2007. Invasive indigenous turaco species in South Africa. ring-necked parakeets Psittacula krameri in Biol. Invasions 17:3029–3039. Belgium: habitat selection and impact on Tozer, R. B. 1974. Feral parakeets and control native birds. Ecography 30:578–588. of introductions. British Birds 67:484–485. ———. 2009a. Predicting the potential dis- Traffic U.S.A. 1987. U.S. imports of parrots tribution of invasive ring-necked parakeets from Latin America. Traffic Bull. 7:5–8. Psittacula krameri in northern Belgium VanderWerf, E. A., and N. P. Kalodimos. In using an ecological niche modeling review. Status of naturalized parrot popu- approach. Biol. Invasions 11:497–513. lations in the Hawaiian Islands. Pages xx–xx ———. 2009b. Experimental evidence for in S. Pruett-Jones, ed. Naturalized parrots nest-site competition between invasive of the world. ring-necked parakeets (Psittacula krameri) Van Kleunen, A., L. Van den Bremer, and native nuthatches (Sitta europaea). Biol. R. Lensink, and P. Wiersma. 2010. De Conserv. 142:1588–1594. Halsbandparkiet, Monniksparkiet en ———. 2009c. Establishment success of Grote Alexanderparkiet in Nederland: invasive ring-necked and monk parakeets risicoanalyse en beheer. SOVON- onder- in Europe. J. Biogeography 36:2264–2278. zoeksrapport 2010/10 Dit rapport is ———. 2011. A radiotelemetry study of samengesteld in opdracht van Team Inva- habitat use by the exotic ringed-necked sieve Exoten van het Ministerie van Land- parakeet Psittacula krameri in Belgium. Ibis bouw, Natuur en Voedselkwaliteit. 153:180–184. Vicentini, A., and E. A. Fischer. 1999. Strubbe, D., H. Jackson, J. Groombridge, Pollination of Moronobea coccinea (Clusia- and E. Matthysen. 2015. Invasion success ceae) by the golden-winged parakeet in the of a global avian invader is explained by central Amazon. Biotropica 31:692–696.