CNMI Wildlife and Vegetatio,n Surveys and Feral Animal Control ANATAHAN 2002·2003 · 2008-11-24 · Anatahan Feral Animal Control Project 2002-2003 CNMI Wildlife and Vegetation Surveys

2003 Technical Report #10

CNMI Wildlife and Vegetatio,n Surveys and Feral Animal Control

ANATAHAN 2002·2003

Conducted by

CNMI Division of Fish and Wildlife,

CNMI Northern Island Mayor's Office,

and

US Fish and Wildlife Service

Justine de Cruz Laura Williams

Nathaniel Hawley Jacob Esselstyn

Curt Kessler Ortwin Bourquin

Anatahan Feral Animal Control Project 2002-2003

CNMI Wildlife and Vegetation Surveys and Feral Animal Control

ANATAHAN 2002-2003

This report is dedicated to the memories of Happy Gideon and Joseph Ogumoro (former Mayor of the Northern Islands) who contributed their sound judgement, good humor, versatility, and hard work to many of the Northern Island projects conducted by our combined agencies. They were good friends to us all and are greatly missed.

2003 Technical Report #10 2


CNMI Wildlife and Vegetation Surveys and Feral Animal Control ANATAHAN 2002·2003

EXECUTIVE SUMMARY

To attenuate the impact of feral animals on Anatahan's ecology, the CNMI and the US Fish and Wildlife Service (USFWS) are cooperating on a US Navy funded project to remove feral pigs and goats as an emergency measure to put a stop to the destruction of the island. This report covers the feral animal removal activities, faunal and vegetation surveys conducted by CNMI-Division of Fish and Wildlife (DFW), the Northern Island Mayor's Office (NIMO), and USFWS from April 2002 - April 2003.

Wildlife surveys to assess species diversity and abundance were conducted from 25 April - 2 May 2002 for birds, skinks, geckos, lizards, rodents, bats, invertebrates, goats, and pigs on the eastern side of Anatahan. Two vegetation surveys were conducted: a point-center quarter survey to establish forest species diversity and a baseline survey of long-term vegetation plots (with fenced and unfenced plots) to monitor habitat recovery with the removal of feral animals. The vegetation plots established in May 2002 were reassessed in November 2002 to evaluate the impact of removal on vegetative recovery.

Feral animal removal strategies, logistics, and activities were planned by the agencies involved during September 2001 - January 2002. Goats and pigs were removed systematically from 4 - 16 May 2002 on the south, east, and north sides of the island and in the volcanic crater, where most of the removal was conducted from a helicopter. Feral animals were removed opportunistically until 31 August 2002 on the western end of the island, near the village, mostly by capture or hunting on the ground. Following the major removal effort in 2002, trips were made monthly to Anatahan beginning in 2003 to control feral animals until a fence could be built across the island to separate subsistence hunting zones from no-ungulate zones. Eight goats were radio collared and released on the island to act as "judas" goats, leading the hunters using telemetry to herds of remaining animals. Shooting from an aerial platform was scheduled to occur monthly through calendar year 2003 while ground hunting was to continue from May through August 2003, rounding out the reduction phases of the project.

Survey Results In May 2002, forest birds appeared to be densely packed into the small amount of forest remaining on Anatahan. We crudely estimated that less than one-third of the island remained forested (about 1,050 ha out of the island's 3,230 ha) and birds were most numerous in the forests remaining on the eastern end. Only four species of forest birds occurred on Anatahan (the Micronesian Honeyeater, Micronesian Starling, Whitethroated Ground-dove, and Micronesian Megapode), or fewer species than on any of the other northern islands. Honeyeater and starling densities appeared to be quite high, but megapode and ground-dove densities were low. The diurnal hawks on Anatahan


/'


may be depressing these last two populations. However, as the forest recovers from the negative effects of feral animals and once again provides cover for ground-foraging species, we expect to observe an overall rise in bird numbers, especially of the endangered Micronesian megapode.

Herpetofauna surveys have just begun to establish a base-line inventory for Anatahan so it is not surprising that the April - May 2002 survey found the Rock Gecko (Nactus pe/agicus) for the first time on the island. Mariana Fruit Bat surveys in May 2002 and subsequent observations by goat hunters until April 2003 identified several major roosting sites for bats on the island. Aggregations of bats dispersed and reformed over this period, with colony sites located mostly in the southern, northern and eastern coastal areas. Nearly simultaneous counts from the helicopter of several of the colonies, combined with survey observations, indicate that fruit bats on Anatahan during this period numbered at least 1,200 individuals.

The invertebrate survey in April - May 2002 was one of the most extensive conducted in recent decades on the northern islands of the CNMI. The list of 144 previously recorded invertebrate species on Anatahan was increased by an additional 117 species during this collection trip, of which 10 were new records for the CNMI and one, a click beetle, is considered to be a new species and has been sent to a specialist for naming. A number of new distribution records were found including a dragonfly which had not previously been recorded from the CNMI, and a robber fly, a tick, a praying mantis and a scorpion not previously recorded from Anatahan. A fly index technique was developed during the April - May trip that was simple, fast, and required a minimum of equipment. The method appears to have merit as an easily applied index technique.

In May 2002, the number of tree species present in both the point center-quarter survey and the vegetation monitoring plots was very low and lacked diversity. The size and therefore age structure of the tree population was skewed heavily towards the larger and older tree classes, with virtually no juvenile recruitment. Additionally, several tree species that are known to occur on Anatahan and that are important to wildlife were not found during either survey. Without drastic and consistently maintained reductions in the feral animal population, the lack of recruitment will eventually lead to loss of the forest and the development of unpalatable grasslands.

The re-survey in November 2002 of the vegetation plots established to monitor habitat recovery showed that seedling recruitment had improved in both fenced and unfenced vegetation monitoring plots, as had ground cover, suggesting that even six months of feral animal suppression had a positive effect on forest recruitment. However, fenced plots that excluded ungulates entirely were acquiring more species and greater ground cover at a faster rate than unfenced plots, indicating that feral animal control had not reached levels sufficient to minimize the remaining ungulates' negative impact on the island's vegetation. If the goats and pigs are not eliminated as uncontrolled feral populations, there is a real danger that plant and animal species native to the island could be lost, that soil erosion will continue, and that overall natural productivity will be further reduced.



Feral Ungulate Removal Aerial and ground hunting effectively reduced the population of goats on Anatahan in May - August 2002. A total of 2,561 goats out of the estimated population of 4,604 were removed. Of the methods used (capture and live removal, aerial gunnery, and ground hunting), aerial gunnery was the most efficient method of removal, considering the cost per unit removed, and the number of man-hours involved for each type of removal effort. However, ground hunting is most effective in ferreting out goats that hide in caves or do not move into open areas where aerial targeting is possible. Live capture was the least efficient method employed and did not result in removal of large numbers of animals.

Feral ungulate removal efforts were renewed in January 2003 with monthly trips to Anatahan focused on removing animals using telemetry and aerial gunnery. Between January and April 2003 over 824 feral goats and 52 feral pigs were eliminated from the island. The number of goats removed has diminished with each of these monthly trips, indicating that the method has been effective in reducing the number of goats on the island, thus making them harder to find. A combination of both aerial gunnery and ground hunting efforts will be necessary to reduce goat numbers further toward eradication.

Postscript On May 10 2003, the eastern volcano on Anatahan erupted sending steam, ash and rocks over the majority of the island. The feral animal control project is currently postponed until it is safe enough to approach the island again.

2003 Technical Report #/0 5


CNMI Wildlife and Vegetation Surveys and Feral Animal Control

ANATAHAN 2002·2003

INTRODUCTION

Anatahan is the closest of the volcanic Northern Islands to Saipan (120 km north). It is large (approximately 3,230 ha) and long (9 km) with a maximum elevation of 788 m. Over the last half-century, large numbers of feral goats and pigs have inflicted incredible damage on Anatahan that will be hard to remedy. Overgrazing has left the upper ridges as largely swordgrass (Ohbra 1994) with the native forest common only in ravines, on the eastern slopes of the island, and in a few pockets in the large crater. By 1995, Marshal et al. (2001) reported that much of the native vegetation had been destroyed by ungulates, with little remaining understory and severe erosion on the steep slopes. Several permanent vegetation stations established in 1995 along the western slopes and re-surveyed in 2000 document the loss of whole tree groves to feral ungulate damage (Kessler 2000). Under constant pressure from feral ungulates there appears to have been a rapid destruction of forest cover with little or no recruitment of native trees.

Typical erosion slope on the eastern portion of the Anatahan, April 2002.



While the flora of Anatahan is unique, species diversity is low. Tree ferns, a species rare in the southern portion of the archipelago, have in the past been abundant and are still common within the main crater. Anatahan is the only island of the northern portion of the archipelago with the endemic tree Macaranga thompsonii. Several large individuals of this species remain in degraded, previously forested ravines (Cruz et al. 2000). However, Anatahan has the lowest level of tree species diversity of any of the northern islands surveyed by CNMI-DFW in 2000. With eradication of feral ungulates, tree species diversity is expected to improve on Anatahan.

Habitat degradation has also taken its toll on native wildlife. According to avian surveys conducted in the year 2000 (Cruz et al. 2000), the Micronesian Megapode population on Anatahan appears to be widely distributed over the island (including the craters), but the numbers encountered were low for an island of 3,230 ha. From variable circular plot counts in 2000, biologists estimated a population of a little over 1,150 megapodes on Anatahan (range 523-2,539). However, population estimates for Sarigan, an island one sixth the size of Anatahan and recently cleared of feral ungulates, suggests that the smaller island supports approximately 1,306 megapodes (CNMI-DFW 2002). Numbers of White-throated Ground-doves (Gallicolumba xanthronura) are the lowest recorded for any of the northern islands (CNMI-DFW 2002).

Mariana Fruit bats (federally listed as endangered on Guam, proposed for listing in the CNMI, and protected by its locally endangered status and a moratorium on hunting in the CNMI) were observed in fair numbers while on Anatahan in July 2000 and July 2001. However, the population estimate of 1,000 fruit bats (Cruz et al. 2000; Johnson 2001) was substantially lower than two surveys that estimated the presence of 3,000 bats in 1983 (Wiles et al. 1989) and 1,902-2,136 in 1995 (Marshall et al. 2001). Without a significant vegetative recovery, Anatahan's fruit bat population will continue to decline.

To attenuate the impact of feral animals on the island's ecology, USFWS and the CNMI have been cooperating on a US Navy funded project to begin reduction of feral pigs and goats immediately and to continue this over a three-year period (2001-2003). The goat and pig removal project was initiated as an emergency measure to put a temporary stop to the destruction. In this sense it has bought the time necessary for a final solution to be developed and agreed upon. The local residents, when present on the island, have a genuine need for the goat resource as food. Therefore, efforts are underway to obtain sufficient funds to construct a fence around the village area in the northwest of the island so that (1) with proper livestock managing techniques, goats remain as a resource to the local residents, and (2) feral ungulates can be eradicated from the remaining major land area. If the goats are totally eradicated before the fence is built, then other native species (such as fruit bats) likely would be more heavily utilized. With expectations that the fence will be in place by mid-2004, and the local residents' food source secure, the eradication of the rest of the island's feral populations can move forward with public support during 2004-2006.

CNMI-DFW, CNMI-NIMO, and USFWS jointly planned hunting strategies, logistics, and activities from September 2001-January 2002. A reconnaissance flight to Anatahan



was made in February 2002, to assess the feasibility of setting up vegetation plots and to select campsites and water drops. The U.S. Marines assisted in transporting water and some equipment for the project during Millennium Thrust activities in early March 2002. To effectively evaluate the impact of ungulate control on the habitat, base-line surveys of vegetation and wildlife were conducted from 25 April to 2 May 2002 prior to control efforts. NIMO then took responsibility for reducing feral animals on the western end of the island, near the village, conducting the majority of the hunting on the ground. CNMI-DFW and USFWS took responsibility for reducing feral goats and pigs on the south, east, and north sides of the island and in the volcanic crater, where most of the removal was conducted from a helicopter. Goats and pigs were removed systematically from 4 - 16 May 2002, and were removed opportunistically until 31 August 2002. The vegetation plots established in May 2002 were reassessed in November 2002 to evaluate the impact of removal on vegetative recovery.

Following the major removal effort in 2002, trips were made monthly to Anatahan beginning in 2003 to control feral animals until the fence could be built across the island. Eight goats were radio collared and released on the island to act as "judas" goats, leading hunters using telemetry to herds of remaining animals. Shooting from an aerial platform was scheduled to occur monthly through calendar year 2003 (CNMI-DFW and USFWS) and ground hunting was to continue from May through August 2003 by CNMINIMO, rounding out the reduction phases of the project. This report covers the feral animal removal activities, faunal and vegetation surveys conducted by CNMI-DFW and USFWS from April 2002-April 2003.

Editorial Note: On May 10, before the fence could be built, the small volcanic crater on Anatahan unexpectedly erupted. The eruption sent ash and steam roiling over 9,000 ft into the air and has deposited a layer of ash over the entire island. At this writing (midJune), the eruption continues and we have had to postpone the eradication project until it is safe to approach the island. Feral animals continue to persist on Anatahan, despite the eruption and the destruction of much of the island's vegetation. DFW and USFWS remain committed to removing the remaining ungulates so that remnant pockets of vegetation can persist as sources for reforestation of the island after the eruption ceases.

METHODS

I. Baseline Faunal and Floral Surveys

Wildlife surveys to assess either species diversity or population size were conducted from 25 April - 2 May 2002 for birds, skinks, geckos, lizards, rodents, bats, invertebrates, goats, and pigs on the eastern side of Anatahan. Survey transects (Fig. 1) were sited either across the terrain to hold elevation relatively constant, or with the terrain to gain the advantage of sampling habitats that change with elevation. Two vegetation surveys were conducted: a point-center quarter survey to establish forest



species diversity and a baseline survey of long-term vegetation plots (with fenced and unfenced plots) to monitor habitat recovery with the removal of feral animals.

AVIFAUNA ASSESSMENT METHODS

Forest birds were surveyed using variable circular plot (VCP) methodology on 25-29 April 2002. The surveys were conducted from 6:00 to 10:30 a.m. by teams of two with one person responsible for bird observations and the other recording. Birds were counted both visually and by song for 5 minutes at each station. Distances to each detection were estimated by experienced observers who had calibrated their estimates prior to the survey. Birds were identified primarily by their song or call notes, and in open areas by sight. Point counts were performed at stations spaced every 150 m along four transects. The transects were set to sample available forested habitat and were composed of between 8 and 12 count stations for a total of 37 stations (Fig. 1).

All bird sampling transects ran parallel to the coastline, cutting across severely eroded ravines (Fig. 1). Transects sampled mixed native and coconut forest. Point count stations were not permanently marked but can be approximately relocated using the GPS coordinates in Appendix 1.

T7

T9

T8

Figure 1. Location of campsite (indicated by red triangle), vegetation exclusion plots (circled in purple), bird survey transects (T7 - T1 0 indicated with black arrows), Mariana fruit bat colony (yellow oval) and reptile/rodent assessment areas (green circles) on the east side of Anatahan in April-May 2002.

Variable circular plot counts allowed the data to be analyzed using the computer program DISTANCE. The VCP methods and details of the analytical methods are attached as Appendix 2. Data for each species was examined using histograms and



truncated such that 90-95% of the observations were used to fit the mathematical models. Data were grouped into intervals that allowed a good fit to the half-normal model (intervals and number of groups were different for each species). The l goodness-of-fit test for all species indicated a very good fit between model and data (P > 0.9). Data from several surveys conducted during 2000 in the northern islands were run simultaneously to increase sample size and reduce variation, but densities were calculated for each island independently.

The fit of the models constructed during data analysis in 2000 was poor (in many cases due to low sample sizes) compared with the algorithms constructed using both 2000 and 2002 data. Thus, density estimates for 2000 reported in Cruz et at. 2000 should be replaced by those reported here (see Table 1). Avian data collected from counts in 2000 on transects in the west, south, and in the crater were combined with 2002 data from the east side of the island to obtain island-wide population abundance estimates. In calculating forest bird population estimates, we estimated that approximately 1,050 ha of Anatahan are forested at this time. From aerial photos it is known that much of Anatahan has become de-forested over the past 50 years, however we assumed that the amount of forested area had not decreased significantly over the past 2-year period for estimation purposes.

BAT ASSESSMENT METHODS We only encountered one Mariana Fruit Bat colony during the first half of the expedition. This colony was located on the eastern side of the island. Bats at this colony were counted directly on three occasions using 10 x 50 binoculars and once from digital photographs. All of the direct counts and the photographs were taken from an observation point with a clear view of the colony at a distance of ca. 50 m. Additional colonies were noted on the north and south sides of the island by hunters after the colony described above had dispersed.

An Ultra Sound Advice® Mini-3 Bat Detector was employed on one evening in an effort to locate the insectivorous Pacific Sheath-tailed Bat (Emballonura semicaudata). One experienced observer listened for calls from 1845 - 2045h on 30 April 2002 in a Barringtonia forest along bird transect #8.

REPTILE AND RODENT ASSESSMENT METHODS Three distinct habitats were sampled; coastal forest dominated by Barringtonia asiatica, a coastal scrub dominated by Cocos nucifera and a upland forest dominated by Hibiscus tiliaceus, to determine the presence, abundance, and habitat preference of diurnal and nocturnal lizards. Twelve trapping stations were established at 25 m intervals at each habitat type along a transect (Fig. 1). One Victor® sticky-glue trap was placed flush with the ground at each station to sample diurnal lizards (Bauer and Sadlier, 1992; Rodda et a/., 1993). Traps were placed in the morning (0700-0830) and run for four to six hours for three consecutive days. Nocturnal lizards were also sampled along the three established transects (Fig. 1) using one Victor® sticky-glue



trap stapled to the trunk of a tree at a height of 0.25-1.0 m at each station. Traps were placed one hour prior to sunset and checked the following morning (0700-0800 h).

Monitor lizards (Varanus indicus) were also sampled along three transects adjacent to those established to sample smaller reptiles. Twelve flat loop snare traps, similar to bird of prey traps (Berger and Mueller, 1959) were placed every 50 m and baited with approximately one pound of fresh goat meat. The loop snare traps were baited at 0800 h and run for three consecutive days. Each trap was monitored every ten to twelve hours and re-baited if necessary.

Victor® snap-traps were used along each of the main transects to document the species of Rattus present on the island. Two traps were placed at each station, one positioned in a tree at a height of approximately 0.5 - 1.0 m and the other flush to the ground. Each trap was baited with peanut butter, set one hour prior to sunset (1800 h), and checked the following morning (0730-0830 h).

INVERTEBRATE ASSESSMENT METHODS

A. Insect Diversity Invertebrate collecting methods included sweep-netting, netting targeted

individual specimens, capturing individuals by hand, by small-paint brush dipped in alcohol, by trapping with styrofoam cup pit-traps with a commercial engine coolant, and using malaise traps. Some earthworms were collected land itter samples were taken for later examination by digging for them using a hand trowel. Several of the expedition members, especially N. Hawley, assisted by bringing in specimens. All specimens were killed in 70% isopropyl alcohol (Aaron Brands commercial rubbing alcohol) and either preserved in this or subsequently mounted as dry specimens using standard pinning techniques. Some sight records were accepted, for example those for coconut crabs and several for butterflies, dragonflies and a few other invertebrates.

Invertebrate specimen identifications were done by Dr. O. Bourquin where possible, and the rest of the specimens were sent to the Bishop Museum, Honolulu, except for earthworms, spiders and crickets, which were sent to Drs. D. Pliska (Natal Museum, Pietermaritzburg, South Africa), M. Saaristo (Turku University, Finland) and Dan Otte (Philadelphia Academy of Sciences, USA), respectively. At the time of report preparation, many of the identifications have not been received.

B. Fly Index A repeatable method of estimating relative fly densities was devised following a

trial to record relative fly densities using baited commercial fly-traps. The commercial fly-traps were not found to be suitable, and were a somewhat bulky item to pack and transport. The following method was therefore implemented:

Equipment: A butterfly net (internal ring diameter, 30 em); a piece of cardboard or a paper plate, (preferably wider than 30cm in diameter or on any side); uncooked

2003 Technical Report #10 II


meat or offal 1-2 days old; an aerosol insecticide to kill flies; specimen containers with 70% alcohol; permanent marker pen.

Method: At a chosen sampling site the paper plate or cardboard base was laid on the ground and a piece of 1-2 day old offal or meat bait at least 5 x 5 x 5 cm was placed in the center of the base. The time taken for the first fly to arrive was noted, and after 60 seconds the butterfly net was placed as swiftly as possible over the bait and held there while insecticide was sprayed over the net. When all the flies inside were dead or incapacitated, the net was withdrawn and all the flies in the net, and on or around the bait were placed into a container with 70% alcohol. Site number and date were written on the container.

At each site, a fresh bait was used as the previous one was contaminated with insecticide. Trapping was carried out between 0930-1200 h, or during mid-afternoon, at six sites, and was repeated after 2 to 3 days. The site characteristics were as follows:

1. Slight slope at base of hill above camp; open; ground cover short grass or bare ground/rock.

2. First patch of Neisosperma oppositifolia on trail up hill, directly inland, from camp. Site was in slightly shaded area on flat ground.

3. In steep-sided gully immediately to south of site #2 - well shaded and protected from wind.

4. Next to WW2 Japanese cistern in next valley south of the camp, under shade of Hibiscus tiliaceus, on flat ground.

5. Next to south edge of Japanese cistern in camp site, no woody plant cover, a ground cover of Zoysia maffeI/a.

6. On rocky bed of drainage line leading to "Orty's Beach", just above "beach" area. Open, no vegetation cover.

VEGETATION ASSESSMENT METHODS Two separate vegetation studies were conducted. The first, a point center-quarter survey, coincided with the bird surveys, using the same transect lines, stations and occurring concurrently. The purpose of this survey was twofold: to sample habitat associated with each bird listening station; and to quantitatively describe the forest in this section of the island. The second survey, vegetation recovery monitoring, was based on comparing measures of species diversity and cover in fenced (exclusion) and unfenced (open) plots before and after the beginning of control. The objective of the study was to gauge the effectiveness of feral animal control activities.

A. Point-center quarter Vegetation surveys were conducted simultaneously with bird surveys. Four

transects were established (Fig.1, Transects 7-10) with sampling points 150 m apart, coincident with bird survey transects and stations. In order to obtain broader more accurate forest data, extra forest survey points were established along Transect 9 in addition to those used for the bird surveys. A total of 50 vegetation points were surveyed along the four transects.



The sampling stations were sUlVeyed using a modified point-center quarter method (Mueller-Dombois and Ellenberg 1974). All trees two meters and greater in height were identified. Distance from the center of the sampling station and diameter at breast height (DBH) were measured. Height of each tree was estimated in meters. Canopy cover was estimated using a densiometer following manufacturer's instructions. Percent ground cover was estimated using the line intersect method in a randomly placed, 1m2 area around the center of the station.

The data were analyzed for tree density, absolute frequency, importance value (MuellerDombois and Ellenberg 1974), and percent canopy and ground cover. The 22 stations along transect 9 were analyzed separately and then pooled with Transects 7, 8 and 10 (28 stations). Frequency was calculated by the number of plots in which a species occurred divided by the total number of plots. The absolute frequency was calculated as the frequency *100. The absolute density of all tree species in a specific area (usually 100 m2

) was calculated with the following formula, where D = the mean distance of all distances to all trees in all plots: 'absolute density' = area/D2

. In order to determine the density of specific tree species, the ratio of the number of quarters in which that species occurred to the total number of quarters was calculated. That ratio was then multiplied by the overall 'absolute density' of trees yielding the number of trees per 100 m2 of a specific species. Basal area (BA) was calculated by the formula BA = n(%DBHr Dominance is the mean basal area of each tree species multiplied by that species' density. The importance value was calculated as the sum of relative dominance, relative density and relative frequency. Relative values were derived by taking the ratio of individual species to the sum of all species (Mueller-Dombois and Ellenberg, 1974).

Fencing and rebar to be carried up the hill for construction of vegetation enclosures.



B. Vegetation Recovery Monitoring Two experimental groups of vegetation plots were established: permanently

marked unfenced vegetation plots and permanently marked fenced (exclusion) vegetation plots. Ten 5 x 5-m fenced plots were paired with ten unfenced plots. The fenced plots served as exclusion plots, and allowed for comparisons of vegetation growth in the complete absence of feral animals. The plots were established and surveyed prior to feral animal control activities. Assignment of fenced or unfenced condition was determined randomly. Plots were marked with rebar, their location determined using GPS, and photographed. The plots were surveyed for all trees, seedlings (7 cm to 0.75 m in height), and shrubs (plants lacking a main stem or multistemmed species), which were identified and quantified by count. Ground cover was determined using the point intercept method at 10 points spaced 10 cm apart in a randomly located quadrat. Canopy cover was determined from densiometer readings taken in each cardinal direction from the center of the plot. Herbaceous vegetation was identified by species within randomly placed 1-m2 nested quadrats inside the plot. Plots were surveyed initially when established in April 2002 and re-surveyed 6 months later in November 2002.

Setting up fencing for vegetation plots on Anatahan in April 2002.



The tree data were analyzed for basal area (calculated according to the formula BA = [I'2DBH]2n); frequency (absolute frequency is the number of plots a species occurs in out of the total number of plots x 100), species richness (# of separate species per plot), relative frequency (calculated as the frequency of a species divided by the total frequency of all species), density (number of trees/ha), dominance (calculated as the mean BA x density), relative density (the density of each species divided by the total density for all species), and importance (importance values combined all the above variables to determine the most influential tree species in a community and was calculated according to Mueller-Dombois and Ellenberg 1974, pp. 100-125).

FERAL UNGULATE ASSESSMENT METHODS Goat and pig numbers were assessed on the eastern side of the island prior to the beginning of control efforts using strip transects (Sutherland 1996), the majority of which were centered on bird transects. The length of five transects was measured with hip chain and traveled at a steady pace on 25 - 30 April 2002. Using a tape measure, we measured perpendicular distance to animals sighted on either side of the transect, up to 60 m. Density of goats and pigs was then calculated using the DISTANCE program.

Loading hunting gear onto marine conveyance from Anatahan in May 2002.



II. Feral ungulate control methods

Feral ungulate control was conducted using a combination of ground teams and an aerial platform. Twelve personnel arrived on island on May 3 2002, either aboard the Marine Vessel (MV) Carolinian (Saipan Crew Boats), or on a chartered jet ranger helicopter (Americopters). Supplies were moved directly from the boats' deck to the campsite during a helicopter sling operation and required six loads of approximately 750 Ibs each. In March 2002, 40 drums (50 gal ea) of aviation fuel along with 800 gallons of water (palates of one gallon containers), were pre-positioned on the island by US Marine helicopters. From 11 - 13 May, our crew was supplemented with an additional five personnel from the Navy SEAL team. The base camp and helicopter refueling area was set up on the plateau above Alatoong Saipan, which is commonly called the "Golf Course", due to its level ground and short grass.

Shooters were outfitted with guns in .223 cal., .220 swift cal. or 12 gauge. They were required to wear internationally recognized orange vests and carry handheld radios. Additionally they were supplied with MRE's, canteens, and other field equipment. Shooters worked in groups that were airlifted to areas around the island. Areas in which shooters were dropped were dependent on adequate landing sites and pickup points. These types of areas were not commonly found in the steep terrain of the island and limited what could be searched by ground. Typically, a ground operation would be scheduled for the morning with the shooters being airlifted to a high point along a ridge. Personnel would then move in a line down slope keeping in contact by radio or preferably by sight. These personnel would then finish at the shoreline and move to a collection point that could be reached by small boat. They would then be ferried back to camp in a number of trips. In the afternoon, shooters would either walk from camp up slope or be dropped off by helicopter to return to camp on foot.

Helicopter operations typically consisted of the pilot (Rufus Crowe) and two shooters. Shooters would be placed in the rear of the helicopter with a Ruger .223 cal. mini-14 and a Remington model 870 12 gao shotgun. One of the shooters was responsible to tally kills and record areas visited. Helicopter shooting usually took place in the last hours of the day, but shooting hours were varied to determine the best return on effort.

The island was broken into sections to record effort. Ground shooting sections were based on a kilometer grid system that is found on the map currently used for Anatahan (Fig. 2). Helicopter sections were based on helicopter search patterns and topography. The difference was due to the speed that the helicopter moved and the difficulty in trying to keep track of the kilometer squares on the map while also searching and shooting animals. Thus it was more efficient to base sections on major topographical features while shooting from the helicopter. In addition, a political boundary was established, due to the concerns of the local residents, to allow the collection of goats by personnel from the Northern Islands Mayor Office. This boundary went from the south side at Alatoong Saipan, north up the slope and along the contour of the west side of the crater rim running generally through the high point of the island and down to the north side.



NIMO staff conducted removal of live and dead goats to the west of this imaginary boundary. All animals to the east were to be shot.

f c J~ D

N

3

,.

~:o::"" ''/

r .x ] ··i~t .'.. , . - ~. '"

Figure 2. Map of Anatahan divided for convenience in recording amount of effort expended on control of feral ungulates. Numbers and letters along the edge denote kilometer square blocks to track ground shooting effort. Helicopter shooting areas are marked with lines and numbers on the interior of the island. Area to the west of aerial removal zone was patrolled by ground hunters for the Northern Island Mayor's Office (NIMO). Fruit bat colonies observed from the air during hunting are marked with yellow circles. One colony on the eastern side of the island was disturbed by hunters and relocated to the eastern crater (illustrated by dotted line).

NIMO personnel constructed corrals in the village area so that captured goats could be maintained and removed live from the island. Freezers and generators were installed for storage of feral ungulates that had been shot, thus providing both food for the village and removal of goats and pigs from the wild. NIMO ground hunters coordinated their efforts with the DFW and USFWS team in May, and then conducted further removals opportunistically until 31 August 2002.

During 2003, monthly trips to Anatahan were conducted to knock-down numbers of feral goats and pigs prior to the construction of an island-wide fence that would section off an area where animals would be maintained for subsistence of the villagers. Shooting efforts were maintained over the portion of Anatahan scheduled for eradication of feral animals through April 2003. Nine DFW and USFWS staff worked on Anatahan from 29 January-1 February 2003 attaching radio transmitting collars to seven goats, assessing photo vegetation plots established in 1994 and 1995, and pre-positioning supplies for


future eradication efforts. Follow-up trips by two to three staff, primarily to accomplish aerial gunning and telemetry, were conducted from February 18 - 19, March 25 - 26, and April 22 - 23 2003.

RESULTS

I. Bird surveys

Four forest birds were detected on Anatahan: Micronesian Honeyeater Myzomela rubratra, Micronesian Starling Aplonis opaca, White-throated Ground-dove Gallicolumba xanthonura, and Micronesian Megapode Megapodius laperouse. No Collared Kingfishers Halcyon chloris were found. Table 1 summarizes bird detections in the remnant forest on the eastern side of Anatahan. Compared with detections from surveys in 2000, birds appeared to be more numerous in the eastern remnant forest than in other areas of Anatahan.

Table 1. Total numbers offorest birds detected along 6 transects on western, southern and central Anatahan in 2000, and along 4 transects in eastern Anatahan in 2002. Note: bird numbers for 2000 have been revised slightly from Cruz et al. 2000.

Bird numbers Bird numbers Detections Detections Forest Bird Species detected on 6 detected on 4 per Station per Station

transects in transects in (SE) (SE) 2000 2002 2000 2002

Micronesian Honeyeater 130 173 3.17 0.306 4.65 0.313 Micronesian Megapode 14 46 0.34 0.096 0.92 0.152 Micronesian Starling 35 114 0.83 0.160 3.05 0.531 White-throated Ground-dove 0 7 0 0.2 0.10)

TOTAL 179 340 4.54 8.82

Hawks are also resident on Anatahan, a situation unique in the archipelago. Previous identification of eggs collected from a hawk nest in 2000 indicated that the Common Buzzard Buteo buteo was resident. However, an adult captured for positive identification during this expedition was identified by the Smithsonian Natural History Museum as a Black Kite Milvus migrans. Black Kites are regular migrants to the Marianas, so it is possible that two hawk species were present on the island simultaneously. Although several hawks were observed foraging in the survey area daily, small sample sizes precluded abundance estimates. It is supposed that 20 - 50 hawks reside on the island.



Table 2. Avian density estimates calculated from Variable Circular Plot data using DISTANCE calculations for Anatahan on data collected in 2000 on the western, central, and southern portions of the island and in 2002 on the eastern side. Data were combined omitting any inter-annual differences and density was re-calculated for islandwide estimates.

Bird Species Model Fit

Year Lower 95% Confidence

Interval (birds/ha)

Estimate of Bird

Density/ha (SE)

Upper 95% Confidence

Interval (birds/ha)

Micronesian Honeyeater

0.918

2000

--~---~-----------

2002

--- - - --- - - - -- -- -Island-wide

9.02

--------------------22.44

-. - - - - - - - - -- - -- - - - -15.98

13.62 ___ J?AE~~1.._.

28.51 ____ .<~:~!?1 ____

20.68 (2.724)

20.57

---- - - - - - - - - - - -36.22

- -- - --- - --- - -----26.77

Micronesian Megapode

0.890

2000

- - -- --~ - - _. - -- 2002

-- - - _. - ---_. ----Island-wide

0.50

- -- - -- -- --- - - -----2.86

--------------------1.74

1.10 ____ .<9:~~91 ____

4.66 _____O:_1~~1. ___

2.79 (1.699)

2.41

- -- -- --_. ------ -7.6

- --- - -- _. - -- --4.46

Micronesian Starling

0.933

2000

- - - -- -- - - -. - - - - -2002

-----------------Island-wide

2.39

-- - - - - - -- - - - -- - --13.50

- --- - - --- - - - - - - ---7.89

3.90 ____ '<9:~~~1 ____

18.37 _. ___(?:~~U ___ .

10.77 (1.699)

6.37

---- - - --- - - - - - -25.00

. _.--.-.---------14.69

White-throated Ground-dove

0.925

2000 -----------------

2002

-----------------Island-wide

0 --------------------

0.20

--------------------0.09

0 ------------------

0.56 ___ J9:~~?J ____

0.26

0 -----------------

1.52

-----------------0.71

(0.141)

For all surveys, the most numerous forest bird species was the Micronesian Honeyeater with an estimated island-wide density of approximately 20.7 birds/ha (Table 2). In order of declining abundance were the Micronesian Starling at 10.8 birds/ha, the Micronesian Megapode with 2.8 birds/ha, and the White-throated Ground-dove with 0.3 birds/ha. This means that if there were 1,050 ha of forest left on Anatahan in May 200, then island-wide populations would be about 21,717 honeyeaters (range 16,778 - 28,109), 11,304 starlings (range 8,285 - 15,242), 2,928 megapodes (range 1,830 - 4,685), and 273 ground-doves (range 100 - 749). Generally, the number of birds detected on Anatahan was low compared with other islands.


Anatahan Feral Animal ControL Project 2002-2003

II. Bat surveys

Nearly all of the Mariana Fruit Bats in the colony on the eastern side of the island were roosting in Ironwood (Casuarina equisitifolia) trees with only a few individuals in Breadfruit (Artocarpus altilis) and Pacific Almond (Terminalia cattapa). Results from the counts at this colony are shown in Table 3.

Table 3. The number of bats counted during three direct counts and one count taken from several photographs on eastern Anatahan in April-May 2002.

Direct Count #1

Direct Count #2

Direct Count #3

Count from Photographs

Date 04/23/2002 04/28/2002 05/02/2002 04/22/2002 # bats counted 678 673 694 530

Bats located in the tops of Casuarina trees on the eastern side of Anatahan in April 2002. This colony later moved to the interior of the eastern crater when disturbed by

vigorous feral ungulate control activity.

All of these counts are probably conservative to a small degree as some portion of the colony was undoubtedly hidden by vegetation. The view of the colony however, was excellent. The direct counts are probably more accurate than the count taken from photographs. During a direct count, the movement of the bats (i.e., fanning, fighting, grooming, etc.) facilitates quantifying those individuals that are partially hidden by



vegetation or other bats. In photographs, tight groupings of several bats are particularly difficult to quantify due to the lack of movement.

During the pre-feral animal control survey period, N. Hawley scanned the eastern crater from the rim for the presence of bat colonies on 1 May. Visibility was good, but no bats were observed. While undertaking feral animal control, two additional colonies were noted from the helicopter: one on the southern coast and one along the northern coast (Fig. 2). C. Kessler estimated that there were at least 100 individuals in each of these colonies. After these two colonies had been noted, the large, eastern coastal colony was disturbed by feral animal control activities and subsequently relocated to inside the crater.

Because two colonies of at least one hundred fruit bats each were observed along both the southern and northern coasts prior to the disturbance of the coastal colony on the eastern side of the island, a rough estimate of colonial bats on the island reaches a minimum of 900 individuals. This would make the 2000 and 2001 island-wide abundance estimates of 1,000 individuals (Cruz et al. 2000; Johnson 2001) seem conservative. Although, the data are insufficient to produce an accurate population estimate for the entire island, it is supposed that over 1,200 fruit bats resided on the island in 2002.

No Pacific Sheath-tailed Bats were encountered during the expedition, but effort to locate them was insufficient to suggest that they are not present.

Research and hunt camp on the northeastern side of Anatahan in 2002.



III. Lizard surveys

Two species from the family Scincidae were documented during this investigation (Table 4): Cryptoblepharis poecilopleurus (Snake-eyed Skink) and Emoia caeruleocauda (Blue-tailed Skink). Four species from the family Gekkonidae were collected: Gehyra oceanica (Island Gecko), Lepidodactylus lugubris (Mourning Gecko), Gehyra mutilata (Mutilating Gecko), and Nactus pelagicus (Rock Gecko).

Table 4. Presence, abundance, and habitat preference of skinks and geckos caught with adhesive trapping on Anatahan, 2002.

# of # of Lizards Habitat trap trap C. E. L. G. G. /100 Tr

sites hrs poecilopleurus caeruleocauda luaubris mutilata oceanica Hrs Barringtonia-Coastal Forest Diurnal traps 12 132 5 17 0 0 0 16.7 Nocturnal traps 12 432 0 0 4 2 0 1.4 Cocos-Coastal Forest Diurnal traps 10 120 7 22 0 0 0 24.2 Nocturnal traps 10 360 0 0 0 0 5 1.4 Hibiscus-Native Forest Diurnal traps 10 120 1 8 0 0 0 7.5 Nocturnal traps 10 360 2 1 3 0 3 2.5 Total 64 1404 15 48 7 2 8 5.7

Two specimens of N. pelagicus were hand collected in a decomposing Neisosperma oppositifolia tree and sent to the Smithsonian Institute (Steve Gotte) for confirmation. Both specimens were confirmed as new island records for Anatahan. The Blue-tailed Skink was the most abundant lizard collected (n = 48).

Table 5. Results of monitor lizard loop-snare trapping, Anatahan 2002.

Habitat # Trap Days # Lizards Lizards/100 trap days

Barringtonia-Coastal Forest 30 5 16.7

Cocos-Coastal Forest 30 1 3.3

Hibiscus-Native Forest 30 5 16.7

Monitor Lizards (Varanus indicus) were documented in each of the three habitat types sampled. Morphological measurements yielded the following averages for the eleven



monitors caught: weight =340g, snout-vent-Iength =32.5 em, and tail =51.0 em. Monitor Lizards were most abundant in the Barringfonia and Hibiscus forest, as compared to the Cocos dominated forest. Comparisons between adhesive trapping and loop-snare trapping indicate that where the trapping rate for monitors was low, trap rates of small diurnal lizards were high, as illustrated on the Hibiscus transect (Table 5).

IV. Rodent surveys

The small mammal survey resulted in two rodents captured during 2,304 trapping hours. The two specimens collected have been tentatively identified as the Pacific Rat Rattus exulans; both specimens have been sent to a taxonomist for confirmation.

Large size Coconut Crabs (Birgus lafro) were observed often on the eastern side of Anatahan in April - May 2002.



V. Invertebrate surveys

A. Biodiversity Identification of sorted collections is underway. Reports on material that was

sent off-island to experts for identification will augment the materials identified by Dr. Ortwin Bourquin (see Appendix 3 for detailed results).

There were numbers of taxa with representatives in the CNMI which were not found on Anatal1an in April-May 2002, but which may still be found there. The list of 144 previously recorded invertebrate species on Anatahan was increased by an additional 117 species during this collection trip, of which 10 were new records for the CNMI and one, a click beetle, is considered to be a new species and has been sent to a specialist for naming.

Table 6. Summary of terrestrial and freshwater invertebrate species known for Anatahan in 2002.

# Species Previously Recorded

# Species Recorded Previously

and in 2002

# Species that are new

Anatahan Records

# Species that are new

CNMI Records

# New Species Detected

Total Number of

Species

101 43 106 10 1 261

Table 6 indicates the numbers of species (including some unidentified species) that were previously known and that have now been recorded from Anatahan. This number will increase as more identifications are received and if more collecting is carried out. A number of new distribution records were found including a dragonfly that had not previously been recorded from the CNMI, and a robber fly, a tick, a praying mantis and a scorpion not previously recorded from Anatahan. The tick species has been tentatively identified as Boophilus microplus, the cattle tick, which can be hosted by several species including cattle, horses, deer and goats. The tick is a vector of Babesia, which can cause the diseases known as babesiosis and theileriosis. The movement of goats between islands should be controlled to prevent the possible spread of ticks and disease potential.

The detrimental effects of goats and pigs on the vegetation of Anatahan has led to a visible reduction in the larger flying insects (butterflies and dragonflies) by removal of ground cover and undergrowth and by the concurrent reduction of dragonfly prey and ground flowers. This was seen clearly following a visit to the village site, where the vegetation surrounding the village was not subject to goat depredations. There is no doubt that biodiversity has suffered greatly on Anatahan because of introduced ferals (goats and pigs).



B. Fly Index Flies arrived at the baits between 1 and 9 seconds at all sites. Average number

of flies per trap per 60 seconds was 28.7 (range 9-90), and muscid flies made up 81.4% of the total catch, followed by calliphorid flies at 17.4 % (Table 7). The differences in numbers of flies caught at individual trap sites appeared to be a reflection of weather conditions such as wind and temperature. The fly density index indicated that flies were exceedingly numerous on Anatahan in 2002. From the results of the applied technique, very high fly densities could therefore be defined as occurring when flies reached an uncooked, decomposing meat bait in less than ten seconds, and that no less than 15 flies would have reached the bait within 60 seconds, with an average of 30 flies per minute being captured at six trapping sites. The presence of many feral goats and pigs, resulting in high levels of dung and carcasses, is considered to have led to such high densities.

Table 7. Results of fly trapping to establish a relative fly density index in April-May 2002. Collections were made at 6 traps (see methods for a description of trap sites).

Trap# n

Date Muscidae (Musca)

Muscidae {AntheriQonaJ

Calliphoridae Chloropidae (Cadrema)

Acalyptrate

4/28 10 2 1 4/30 11

4/28 19 2 5/01 84 5 1

4/28 10 19 3 5/01 11 1 4

4/28 10 1 4 5/01 31 3 4 1

4/28 9 2 5 4/30 9

4/28 72 11 6 5/01 12 2

Totals 276 4 60 3 1 % 80.2 1.2 17.4 0.9 0.3

The fly index technique was simple and fast, and required a minimum of equipment. Between 10 to 30% of flies on the bait escaped while the net was being placed over the bait, or from under the rim of the net (because of uneven ground surface) once the net had been placed over the bait. All the numbers of flies given in Table 7 are undercounts, and it is estimated that none of the sites would have yielded less than 15 flies per 60 seconds if all flies on the bait had been captured. The trapping method


could be refined and tested further. However, even as it stands, it appears to have merit as a fast, simple, easily applied index technique.

VI. Vegetation surveys

A. Point center-quarter Sixteen tree species were identified on the four transects surveyed. The overall

tree density was 11.44 trees/100m2. The two species found in the greatest density

were Hibiscus tiliaceus and Cocos nucifera. Cocos nucifera had at least twice the density of any other species in this area (Fig. 3). Absolute frequency of four species (H. tiliaceus, C. nucifera, Artocarpus mariannensis, Neisosperma oppositifolia) showed that they were present in one half to two-thirds of the plots (Fig. 4). Erythrina variegata, Pouteria obovata and Casaurina equisetifolia had the greatest basal area. The four densest and most frequent species also had the greatest importance values, with the exception of C. equisetifolia (Table 8).

The tree population in the survey area was heavily skewed toward trees in the middle DBH size range, i.e., 10 - 30 cm (Fig. 5). Neisosperma oppositifolia and H. tiliaceous dominated the population of trees with a DBH of less than 10 em, however, Artocarpus mariannensis and Ficus tinctoria maintained populations over a range of size classes. The population of large trees was very small, composed primarily of A. mariannensis, F. tinctoria, and H. tiliaceous, with few other singular species (Fig. 5).

The tree survey showed that there were very few individuals in the seedling stage, i.e., 2 - 3 m in height. For trees less than 9 m tall, H. tiliaceous and N. oppositifolia dominated while Artocarpus mariannensis and C. nucifera had the largest population in the greater than 9 m height classes (Fig. 6).

The ground cover was extremely variable throughout the area surveyed. Canopy cover was more consistent and ranged from 77% - 91 % (Fig. 7). However, it wasn't expected that the canopy would be very open because the transects were established intentionally in forested areas where forest birds would be detected.

The forest we sampled on the eastern portion of Anatahan in 2002 lacked species richness (that is, the number of species present was very low). It was instead dominated by a few species: H. tiliaceus, C. equisetifolia and C. nucifera. These three species are not particularly desirable for wildlife and C. equisetifolia is an invasive species that could easily begin to dominate and develop into monotypic stands. This has been observed with C. equisetifolia on other CNMI islands. Additionally, there appeared to be virtually no recruitment of juvenile trees. An exception to the latter may be N. oppositifolia, found in low numbers, that may be less preferred as a food source than other tree species by the large population of feral animals. This lack of recruitment will eventually lead to the loss of forest and the development of grasslands that are unpalatable to feral goats and pigs.



Table 8. Tree species (n = 50 stations) surveyed along 4 transects in forested areas. Point center-quarter vegetation surveys were conduced in conjunction with forest bird surveys on Anatahan in April - May 2002.

Pago H. tiliaceus

Gago C. equisetifolia

Dukduk Ar. mariannensis

Niyok C. nucifera

Putting B. asiatica

Rodda F. tinctoria

Fagot N. oppositifo/ia

Lada M. citri/olia

Kafu P. tectorius

Atmahayan P. argenteus

Gago E. variegata

Talisai T catappa

Ahgao P. obtusifolia

Lala P.obovata

Mapunyao A. mariannensis

Alum M. multiglandulosa

549.90 14.54 18.97 18.97 52.48

8152.87 5.82 0.51 43.08 49.41

784.37 17.37 15.90 15.90 49.17

456.28 27.38 17.44 0.51 45.33

1019.73 13.11 9.74 9.23 32.09

550.57 8.65 11.28 11.28 31.22

101.81 1.96 13.33 13.85 29.14

177.98 1.53 5.64 6.15 13.32

151.18 0.43 2.05 2.05 4.53

1472.58 2.10 1.03 1.03 4.16

3959.41 2.83 0.51 0.51 3.85

191.53 0.41 1.54 1.54 3.49

3184.29 2.27 0.51 0.51 3.30

1912.82 1.37 0.51 0.51 2.39

254.34 0.18 0.51 0.51 1.21

49.74 0.04 0.51 0.51 1.06 *'Relative' compares all species against the group *Density refers to the number of trees per area *Frequency measures distribution; the higher the frequency the more widely distributed a species *Dominance is derived from Basal Area and density *Importance value attempts to combine all the above variables to determine the most influential tree species in a community.

Sampling vegetation monitoring plots on Anatahan April - May 2002.



5.0

4.5

4.0

3.5

~'f 3.0 III 0c:: 0 Q) ...

Cl ~ 2.5

2.0

1.5

1.0

0.5

0.0 .,?> (p .9) 9) .9)~e, ~e, ~e, fiy-'?>.~~'b 4' .~i:t'?> 0<:'- ~ .~~ i..~e, .~i:t'b ~oe,'b # .~~r}

,,0 (Q~ if '"q} .~" ",0 ,?>~<:::i. ~0 (Q<::- .(Q~ ~v 0':;' ;§>q} ~ ~~<:-v . 9)<::-'" o~ ~'" 'be, cJ -.$Jv ~ .9)<::-'" .:;.~'< f:::' &~~ ~<o .~q} q.«. <0' ~. q. ,,'<' &0' ~. ~~ o~~ q. ~'?>'" 00: ~. ~"J q.

~ ~' 0' ~v ~.~.

~.

Tree Species

Figure 3. The density (# trees per 100 m2) of trees in the forests of southern Anatahan in April 2002 (n =50).


80

70

60

ij'50 c Q) ;:, tr ~ ~ 40 ~ ;:, '0 w .0

30c(

20

10

0

~I I ~ ~ ~

.~ ll~~ • 0,,11 ~~ . !t1l ·u ~1I ,~~ ~1I ~~ ~~ ~1I ~ ~1I ~ ~1I 0 o~ c-!I.'Ii.~..ou u~ 0~ ,~ .~ o~ ~o~ ~ ~ # ~ ~ '0~ ~ ~o~1j <::-oS ~~ o~" ~<:' 11 ct" ~ # 0 I::' • <lJ I::' ~ 0 ~,Ili ~ ,~ ~ ~ #

-('-' v· i...,'Ii ~~ «. ~. ~. «. ~ ~'Ii 0 «' '" «.'Ii ~" 0(;); <V. ~C8

~. ~. ~ v· ~-s «' ~.""

Tree Species

Figure 4. Absolute frequency of tree species found in a forested area of southern Anatahan impacted by feral animals in April 2002 (n =50).



35 i I 0 cmlj11R1nftellID 0

30 ~

25

CIl I ii 20-1

:::J ~

:~ I ~

.5 '0 15 'II:

10

5

o P--' '-r-'

"-~~ <:I"\)'". ~<:I'

L I ' L T I I I I I I I I I oil.:!

• (Jyjn Iut:JJIl=liliI:m:i§ 0 a---.----. O[l,~ 0 [::lII~lh"f:Ulrll"~

..~ ",z:• .F._~ .l?,., ItTlTtUI

I I - - I I ,..........., .oo.~ • r---I I I I I I I I I []1~u:I1u.

~~ ~~ ~~ C?J~ ~~ ~~ ~~ C?J~ ~~ ~~ ~~ ~. bo· ~. ~. 'J,~. ?J<' ~~. p.bo· ~~. \:{~b.' ~~.

~c;f r::/' f::J''\ \:{ \:{ <:I' \:{ () <:I',,<:I' ,,~~ 'j..<:I' 'j..~' ~<:I' ~~. b.<:I' b.~. ~<:I' ~~.

DBH size class (em)

Figure 5. The number and species of trees in diameter at breast height (DBH in em) size classes found on southern Anatahan in April 2002 (n =50).



80

70

60

50

.. , --, ... • ,.-. T,........... I? ..... - ..............J _, "'.... III . • I -.:, ",," Gl • I?~ 1t1l1'U"u. ~ 40 Gl

C!la \'l:llll[;:r.'0 'It •

30

I I

0

0

0 IL r.l.I.I.20 1

10~ I I

o I I I I=~- -.1 , I I L I I.. I , I. ,

0- -

i ~

0-2.99 3.0-5.99 6.0-8.99 9.0-11.99 12.0-14.99 15.0-17.99

Height class (m)

Figure 6. Height size classes (in meters) and tree species found in southern Anatahan in April 2002 (n ;;; 50).



7 8 9 10

Transect

120 i •

• Ground cover

100 t-I---------------------------

80 1 I

~ Cl

Q) Cl 60 LI _ e Q)

>«

o Canopy cover

40-1-1---

20

o

Figure 7. The average ground and canopy cover of forest stations surveyed on southern Anatahan in April 2002 (n =50).



B. Vegetation Recovery Monitoring A total of 5 tree species were found during the initial baseline surveys conducted

before feral animal control began. The overall tree density was 4 trees /100m2. Fenced

plots had 1.9 trees/100m2 and non-fenced plots had 2.2 trees/100m2. Cocos nucifera

had the greatest overall importance and frequency within the survey area. The importance and frequency of the remaining tree species was much lower than that of C. nucifera (Table 9). The diameter at breast height (DBH) size class distribution indicates that the greatest number of trees in the plots were in the mid to large range. There were very few individuals in the smaller diameter ranges (Fig. 8).

Table 9. Characteristics of 20-25 m2 plots on Anatahan prior to feral animal control and prior to random fencing of 10 plots in April 2002.

Species Avg BA Density 1100 mOl Frequency Importance Value (cm2

)

C. nucifera 415.88 5.80 0.75 175.84

591.71 0.40 0.10 18.66ttinctorla . tiliaceus 583.73 0.10 18.570.40

. oppositifo/ia I 32.85 I 1.40 I 0.25 I 35.24

T. cataDDa I 9203.82 I 0.20 I 0.05 I 40.98

*SA =basal area and is the total area physically occupied by a species *Density refers to the number of trees per area *Frequency measure distribution; the higher the frequency the more widely distributed a species *Importance value factors density, frequency and dominance to determine the overall importance, or relative contribution of a species


natahan Feral Animal Control Project 2002-2003

16 i ,

Ul Cii ;:, ~ .:;: 8'6

-.S 6

4

2

o

14 +1----------------------,

12 +1----------------------1

10 +I----------------------l

Ir--:--~__:_:1--------------J

0-4.99 F 0-4.99NF 5.00-9.99 F 5.00-9.99 10.00- 10.00- 20.00- 20.00- 30.00- 30.00- 40.00 + F NF 19.99 F 19.99 NF 29.99 F 29.99 NF 39.99 F 39.99NF

DBH size class (em)

Figure 8. The number of individuals per DBH size class present in 20 plots (F =10 fenced and NF =10 unfenced) on Anatahan in April 2002.



The first follow-up vegetation monitoring survey was conducted in November 2002,6 months after the initiation of the feral animal control program. In November, seedlings of 9 species were found throughout the 20 monitoring plots (pooled irrespective of treatment) compared with 3 species found in the original April survey. Compared with the April survey, the number of seedlings for all species was greater in November (Fig. 9). Also, the frequency of seedlings was greater for all species detected in the November survey (Fig. 10). Morinda citrifolia and Neisosperma oppositifolia were the most widely distributed species occurring in approximately half of all plots in November.

Comparing the two treatments (fenced and unfenced), we found a significant difference in species richness between fenced and non-fenced plots surveyed in November (t-test, P = 0.00015). On average the fenced plots contained both a greater diversity of species and a greater number of seedlings per plot than did the non-fenced plots (Table 10).

Average canopy cover and average ground cover increased slightly, but not significantly, in both fenced and non-fenced plots from April (prior to control) to November (6 months after control began) (Fig. 11). Although there was no difference in the percent of ground cover between fenced and unfenced plots in April (P =0.10), a comparison of the two treatments indicated that the fenced plots had significantly greater (P = 0.04) average ground cover than the nonfenced plots by November (Fig. 12).

Table 10. The average species richness and number of seedlings (all species pooled) per plot from fenced and non-fenced plots (n = 10 per treatment) surveyed 6 months after initiation of feral animal control on Anatahan.

Species richness k# species/plot)

Individual seedlings/plot

November 2002 Fenced Non-fenced

Avg. sd Avg. sd 4.4 1.26 2.2 0.79

20.9 15.7 16.7 12.4

{-test P = 0.00015

P=0.26



60 i

>CJ s:::: Q) ~ C" ~

U.

S ~

o ~ 20<

10

o .~

....t-\O~~ ""''lJ.t-fl,~'O _~ t,\\\·\O\\'IJ. .<, ~\'IJ.~., 0'0\\\,0~.\W \\.W·

~.o~V

i

50 +1---------------,

40 -I-I-------------j

30 I I

I I

April-02 • Nov-02

'IJ.'l'IJ. \fl,~'O ~~''l,~\~\'O~.t-'lJ.\'IJ.~~'IJ. v>~v.~oC. ~'IJ.~ 'IJ.,,,.·l,,fl,~ ~.o~\ ~.

Seedling Species

Figure 9. The absolute frequency of seedlings 2 - 3 m in height in 20-25m2 vegetation monitoring plots prior to feral animal control (April 2002) and 6 months after control initiation (November 2002).



60

50. - F. tinctoria

D N. oppositifolia

.T. catappa 40 +

f/l0)

.S 't:l ~ 30 f/l

0 -~

20

f/l:::l

10 I ~ :0 i:

0-

III E... Q) Q. f/l0 f/l'Qj Z

III 't:l CI "i: o ~

• H. tiliaceus

.C. papaya

DT.orientalis

April 2002

OM. citifolia

Ell P. artgenteus

• Unknown

Date

III E---J~... ... Q)0 Q.

I~ ~ f/l

"Qj Z

I IJ

Nov. 2002

Figure 10. Both the number of seedlings and the diversity of seedlings increased in the vegetation monitoring plots (n ::: 20) between April 2002, prior to feral animal control, and November 2002, after 6 months of control.


60

50

40

30

20

Fenced


80

70

% canopy cover

Non Fenced

Figure 11. Average canopy cover of fenced and non-fenced plots in April (prior to feral animal control) and November 2002 (6 months after the initiation of feral animal control) on Anatahan.

2003 Technical Report #/0

15

10

5

o

Fenced


50

45

40

35

30

% Ground Cover 25

20

Non Fenced

Figure 12. Average ground cover of fenced and non-fenced plots in April (prior to feral animal control) and November 2002 (6 months after the initiation of feral animal control) on Anatahan. Average ground cover in fenced plots was significantly greater ("*") than in unfenced plots in November 2002 (P = 0.04).


In April 2002, the forest in which the 20 long-term monitoring plots were established was more degraded than the forested areas where point center-quarter surveys were conducted. The long-term monitoring plots lacked species richness more so than the other areas surveyed. The overall and individual tree species density was very low and there was minimal seedling recruitment.

By November 2002, six months after the initiation of feral animal control, seedling recruitment and seedling species richness increased markedly in the long-term monitoring plots. This was largely due to the increase in species richness and the density of seedlings within the fenced exclusion plots (Table 10). It is clear from the significant differences in ground cover between fenced and non-fenced plots in November that feral animals were still heavily impacting seedling recruitment in those areas from which they were not excluded. The fenced plots, which excluded all ungulates, demonstrated that feral animal control has not reached a level that minimizes ungulate impact on the vegetation.

VII. Feral animal surveys

A. Goat density In April-May 2002, over 27 groups of goats were observed while conducting

counts along five transects totaling 6.7 km in length on eastern Anatahan. Density of goats near the east-side campsite was calculated to be 1.4 goats/ha (range 0.6 to 3.4 per ha) using the computer program DISTANCE for transect methods and selecting a model with a fit of p =0.71. If this density was typical over the entire 3,230 ha of Anatahan, then an island-wide abundance estimate prior to control measures was 4,604 goats (range 1,925 to 11,014). Large herds of goats (numbering 17-23) were commonly observed from the campsite in the early morning and early evening hours.

B. Pig density Only five pigs were detected over the length of the five transects during daylight

hours. This sample size was not sufficient to estimate density or abundance. Despite this failure in sampling, small groups of pigs were commonly observed in the early morning and early evening hours from the campsite indicating the presence of a substantial population of pigs on the island. Pigs are often active nocturnally, so sampling methods employed during the day failed to produce accurate results.

VIII. Removal of Feral Animals

During 14 days in May 2002, DFW and USFWS personnel removed a total of 2,421 goats, 62 pigs, and 6 cats from the north, south, and eastern sides of Anatahan. Aerial shooting removed 1,740 goats, 32 pigs, and 5 cats during approximately 34.5 flight hours. Records of shooting results by area (Table 11) indicate that the largest numbers of feral goats and pigs were killed in area 1 along the southern coast (Fig. 2).


From June-August 2002, NIMO removed 140 goats (28 of these alive) and 4 pigs. The combined control efforts in 2002 resulted in the removal of 2,561 goats and 66 pigs, for a total of 2,627 feral ungulates at a cost of $38.07 apiece.

Table 11. Results of aerial hunting in May 2002 by DFW and USFWS in different areas of Anatahan.

Area Goats Pias 1 979 14 2 160 6 3 134 2 4 312 4 5 155 6

TOTAL 1,740 32

Table 12. Results of ground shooting in different areas of Anatahan, May - August 2002.

Area Goats Pigs Man hours DFW & USFWS

D-4 226 2 129.5 E-2 18 E-4

36 9 191 1 74

F-2 33 2 20 F-4 50 0 28 G-3 14 0 3 G-4 116 9 33 1-1 14 J-1

6 6 9 1 4.5

J-2 0 0 20 Sub 681 34430 total

NIMO Village 56 3 June Village July Village

40 0 August

Sub44 1 140 4

total TOTAL 821 34

Aerial hunting was more efficient than ground hunting in terms of number of ungulates removed from Anatahan. The DFW and USFWS ground crew removed, in



approximately 344 man-hours, 681 goats, 30 pigs and 1 cat. Out of 27 grids that could be hunted from the ground, only 10 (37%) grids were covered by foot (Table 12). We found that extremely rough terrain and accessibility issues limited the amount of area that could be searched by ground hunters. However, ground teams eliminated animals that were inaccessible from the air making an important contribution to the effort.

Aerial Hunting

~ 120

-J!!. 20

i: a >- >- >- >- >- >- >- >- >- >- >- >- >C1l C1l C1l C1l C1l C1l C1l C1l C1l C1l C1l C1l C1l ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

I I I I I I I I I I I I I

'V 1.0 <D /'- Q') en 0 ..- N C") 'V 1.0 <D ..- ..- ..- ..- ..- ..- ..

Date

Figure 13. Number of goats killed per unit effort by aerial shooters (afternoon flights only) over the removal period in May 2002. Black dashed line indicates a downward

trend in the number of goats over time.

Because most ungulates were eliminated by aerial shooters, and most flights were made in the afternoon, we examined the relationship between aerial hunting effort and the number of goats shot during afternoon flights (Fig. 13). The number of goats shot declined steadily during the second week of the project. This is the pattern expected if goat removal is decreasing the number of animals on the island forcing the hunters to expend greater effort to find and then eliminate the remaining animals. Control efforts were hampered toward the end of the removal period when mechanical problems with the helicopter and financial constraints limited the number of hours spent in the air.


0:s: 100 (1)

:: 80 c :::s 60 ~

(1) 40 1·V ~. • • •c. •.. ..

Anatallan Feral Animal Control Project 2002-2003

Similarly, goats shot by ground hunters (Fig. 14) declined in number over time indicating a decline in the number of goats in the ground hunting areas. The additional manpower supplied by the Navy SEALS during the last week of removal helped to bolster ground team efforts and kept moral high. The trends in Figures 13 and 14 suggest that both aerial and ground removal reduced the goat population.

Ground Hunting

t:: ~

7.00 6.00

Q) 5.00 .:t::: c 4.00 ::::J 'Q)

3.00 Co 2.00 ~ -. 1.00 ~

0.00 >. >. >. >. >. >. >. >. >. >. >. >. >. co co m m co co co m m m m co m ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

I I I I I I I I I I I I I ("i) ~ L() <0 to- co (J) a ~ N ("i) ~ L()

~ ~ ~ ~ ~. ~

Date

Figure 14. Number of goats removed by ground hunting efforts on Anatahan in May 2002. Black dashed line indicates a downward trend in the number of goats.

NIMO personnel completed the installation of chain-link and barbed-wire corrals in the village area for maintaining live captured goats, thus improving their capacity for future removal of feral ungulates. In addition, the installation of portable electric generators and chest freezers augmented the village's storage capacity and will contribute to the residents' continued removal of feral animals from the vicinity. NIMO ground hunters shot 112 goats, captured and removed 28 goats live, and killed 4 pigs between June and August.

Prior to reinitiating control activities in 2003, Jess Omar and Curt Kessler completed a helicopter gunnery training course to improve the safety of project personnel. Control efforts were renewed 29 January - 1 February 2003 when radio telemetry collars were attached to seven females goats that were then distributed around the island. Personnel were transported by boat to the island, and supplies (aviation fuel, water,


food, camping and hunting equipment) for the remainder of the year were prepositioned on Alatoong Saipan (the golf course camp).

Capture of the goats for radio collaring, producing "judas" goats, was assisted greatly by the helicopter on 29 January 2003. Night time capture of goats by hand did not prove effective. Seven female goats were collared and then released at different points around the island from the helicopter. A total of 190 feral goats and pigs were then shot or captured for removal by both aerial and ground teams (Table 13). In January, ground shooting was done in the area adjacent to the golf course camp on the south side of the island and adjacent to the research camp at Banku on the eastern side of the island. Helicopter shooting occurred only on the outer slopes of the south, east, and north sides, but not in the interior due to high winds.

During the January 2003 trip pictures were taken of the village area for helping to outline and plan the proposed route for the fence. Two out of three photo vegetation transects were also resurveyed. Wildlife observations during the trip included at least two cats, 10 Anatahan hawks, and numerous, widely dispersed Mariana Fruit Bats. There seemed to be many small colonies of under 60 bats at various locations around the island including colonies along the shore of the eastern side and in the bottom of the small crater. The larger bat colony on the eastern side appeared to support at least 75 individuals. Several females carrying young were observed in flight. Of the kinds of trees producing food for bats, Erythrina variegata was in bloom, as were Pandanus spp. No noddies or other tern species were observed on the small rocks near the golf course where nests had been observed on other occasions.

Table 13. Results of aerial and ground shooting by CNMI-DFW and USFWS in different areas of Anatahan, January - April 2003.

DATE/KIND OF REMOVAL

FERAL GOATS FERAL PIGS TOTAL UNGULATES

30-31 January 2003 190 Aerial team 144 1 145 Ground team 40 5 45

18-19 February 2003 333 Aerial team 318 15

25-26 March 2003 228 Aerial team 214 14

22-23 April 2003 125 Aerial team 108 17

TOTALS 824 52 876


During the February 18 - 19 2003 trip to Anatahan to remove feral goats and pigs, 333 animals were removed by aerial gunning (Table 13). Aerial shooting was conducted for approximately four hours each day and telemetered goats were successfully tracked from the helicopter. Unmarked goats found with the "judas" animals were dispatched. Also during the trip, the remaining photo vegetation transect was resurveyed. One of the photo plots has been overtaken by a very active erosion landslide that will soon cut off an additional plot along the cliff line that the transect follows. Wildlife observations during the trip included noting the recent construction of an Anatahan hawk nest on a cliff face (no eggs or young), Gray-backed Terns (Sterna lunata) recently arrived to the rocky islets off the southern coast, and numerous and widely dispersed fruit bats feeding on ripe Pandanus fruit. The major bat colony on the north east side of the island continues to be used and photos revealed a minimum of 235 bats roosting in a Casuarina tree.

From 25 - 26 March 2003, 228 feral goats and pigs were eliminated using aerial gunning and telemetry (Table 13). Aerial shooting was conducted for approximately four hours each day. Telemetered goats were tracked from the helicopter and wild goats found with the "judas" animals were dispatched. The telemetered animals proved difficult to sight and one of the collared animals was recaptured and moved to a different area. Wildlife observations during this trip included two cats (one of which was dispatched), and nesting Anatahan hawks with at least one nest containing two eggs. Fruit bats were numerous and widely dispersed on the island, with the large colony on the northeast side containing at least 255 bats as documented with photography. Graybacked Terns and noddies were observed mating on the rocky islets near the golf course camping area.

From 22 - 23 April 2003, 125 feral goats and pigs were eliminated using aerial gunning and telemetry (Table 13). Aerial shooting time remained consistent at approximately four hours per day. The decrease in number of goats removed from the island indicated that numbers of goats are decreasing on the island with a consequent increased difficulty in finding the remaining animals. An addition goat was captured, radio collared and released on the north side of the island. One cat was observed on the south slope, and two dogs were observed just west of the golf course camp. These are presumably the same two dogs observed near the eastern research camp and in the small crater in May 2002. Wildlife observations during this trip included two downy young in the Anatahan hawk nest observed with eggs during the March trip, and another hawk nest with a downy chick. Adult hawks were observed carrying what appeared to be in one case a White Tern (Gygus alba) and in another case a Mariana Fruit Bat. Approximately 30-50 Gray-backed Terns and noddies were again present on the rocky islets along the southern coast.

In April 2003, photographs of the fruit bat colony on the eastern side revealed a minimum of 308 bats roosting in a Casuarina tree. Nearly simultaneous observations from the helicopter of colonies containing 300-400 bats and 500-700 bats, respectively,


along with observations of smaller colonies, bring the estimated number of fruit bats on Anatahan that month to 920 - 1,220 individuals.

Artifacts made from a crashed 8-29 plane by Japanese holdouts after WWII who inhabited Anatahan from 1944-1951. These items were mistakenly collected by survey personnel in May 2002, removed from the island, and later housed for safekeeping at

the Saipan Museum.

SUMMARY AND DISCUSSION

Floral and faunal surveys In May 2002, forest birds appeared to be densely packed into the small amount

of forest remaining on Anatahan. We crudely estimated that less than one-third of the island remains in forest (about 1,050 ha out of the island's 3,230 ha) and birds were most numerous in the forests remaining on the eastern end. Only four species of forest birds occurred on Anatahan (the Micronesian Honeyeater, Micronesian Starling, Whitethroated Ground-dove, and Micronesian Megapode), or fewer species than on any of the other northern islands. Compared to 500 ha Sarigan, which is recovering from feral animal degradation, honeyeater and starling densities appeared to be quite high on Anatahan. Megapode and ground-dove densities, on the other hand, were lower on Anatahan than on Sarigan. The diurnal hawks on Anatahan may be depressing these



latter populations (more work remains in order to identify with surety the species of hawk resident on Anatahan). However, as forest recovers from the negative effects of excessive numbers of feral animals and once again provides cover for ground-foraging species, we expect to observe an overall rise in bird numbers, especially of the endangered Micronesian megapode.

Herpetofauna surveys have just begun to establish a base-line inventory for Anatahan so it is not surprising that the April- May 2002 survey found Nactus pelagicus, the Rock Gecko, for the first time on the island. It is recommended that transects established during this investigation be monitored during the course of the ungulate control in order to increase our awareness of these cryptic life forms. However, due to the difficult topographic relief of the island, such monitoring will be a cost and labor-intensive effort.

Mariana Fruit Bat surveys in May 2002 and subsequent observations by goat hunters until April 2003 have identified several major roosting sites for bats on the island. Aggregations of bats dispersed and reformed over this period, with colony sites located mostly on the southern, northern and eastern coasts. Temporary gatherings of bats were found in the volcanic craters during periods of prolonged gunfire on the island. Nearly simultaneous counts from the helicopter of several of the colonies, combined with survey observations, indicate that fruit bats on Anatahan during this period numbered at least 1,200 individuals.

The invertebrate survey in April-May 2002 was one of the most extensive conducted in recent decades on the northern islands of the CNMI. The list of 144 previously recorded invertebrate species on Anatahan was increased by an additional 117 species during this collection trip, of which 10 were new records for the CNMI and one, a click beetle, is considered to be a new species and has been sent to a specialist for naming. A number of new distribution records were found including a dragonfly which had not previously been recorded from the CNMI, and a robber fly, a tick, a praying mantis and a scorpion not previously recorded from Anatahan. The tick is a vector of Babesia, which can cause the diseases known as babesiosis and theileriosis and goats are a known host for the tick. The movement of goats between islands should be controlled to prevent the possible spread of ticks and disease potential.

The fly index technique developed during the April - May trip was simple and fast, and required a minimum of equipment. The method appears to have merit as a fast, simple, easily applied index technique. The index, if performed repeatedly over time, can be used by residents of the northern islands to support their requests for assistance in controlling this threat to human health.

In May 2002, the number of tree species present in both the point center-quarter survey and the vegetation monitoring plots was very low and lacked diversity. The size and therefore age of the tree population was skewed heavily towards the larger and older tree classes, with virtually no recruitment of juvenile trees. Additionally, several tree species that are known to occur on Anatahan and that are important to wildlife were not found during either survey. The point-center quarter survey was conducted specifically


to oover a broad area and to sample the diversity of arboreal vegetation. Without drastic and consistently maintained reductions in the feral animal population, the lack of recruitment will eventually lead to loss of the forest and the development of unpalatable grasslands.

Evidence of pig damage to coconut trees outside of fenced plots on Anatahan in May 2002.

During the survey in November 2002 of the vegetation plots established to monitor habitat recovery, it was observed qualitatively that the vegetation in the arid coastal zones was recovering. Aided by a strong wet season and a decrease in grazing pressure, grasses and beach strand vines had begun to reclaim some of the previously barren zone. However, it was also observed that pig numbers may have increased during the hiatus in hunting between May 2002 and January 2003. More pig sign was observed in November than during the April - May survey trip. Quantitatively, seedling numbers and species diversity had improved in both fenced and unfenced vegetation monitoring plots, as had ground cover, suggesting that even six months of feral animal suppression had a positive effect on forest recruitment. However, fenced plots that excluded ungulates entirely were acquiring more species and greater ground cover at a faster rate than unfenced plots, suggesting that feral animal control had not reached levels sufficient to minimize the remaining ungulates' negative impact on the island's vegetation.



A definition of 'minimal' feral impact has not been determined for this project. We suggest that such a definition be set, and that it be based on measures of habitat recovery (for instance, when there are no significant differences in species diversity and seedling recruitment between fenced and non-fenced plots, then feral animals are having a minimal impact on the habitat). We also suggest that the success of ungulate control efforts be determined by the same measure, e.g., ungulate control measures will be deemed sufficient when feral animals have a minimal impact on the habitat.

The increase in species diversity in the exclusion plots also demonstrates that there remains the potential, despite years of ruin, for a variety of tree species to be recruited into degraded areas. Over a six-month period, six new species took hold in the longterm monitoring plots. These species may have lain dormant in the seed bank, be the result of wind borne dispersal, or may have been deposited by bats or birds passing over the plots, among other possible methods of dispersal. Although the mechanisms remain unknown, it is important to note that the potential still exists for the recruitment of a variety of plants into this area of Anatahan at this point in time.

If the goats and pigs are not eliminated as uncontrolled feral populations, there is a real danger that plant and animal species native to the island could be lost, that soil erosion will continue, and that overall natural productivity will be further reduced. Removal of goats from an area for at least three months will rid that area of ticks, which will starve. Goats should be allowed on Anatahan (and on other islands) only as tethered animals, or as controlled, herded populations. This will enable disease control, and easy meat and milk harvesting.

Feral ungulate removal Our evaluation of the ungulate control effort showed that aerial and ground

hunting effectively reduced the population of goats on Anatahan in May 2002. A total of 2,561 goats out of the estimated population of 4,604 were removed in 2002. Aerial hunting was the most efficient method of removal, considering the cost per unit removed, and the number of man-hours involved for each type of removal effort. However, ground hunting is more effective in ferreting out goats that hide in caves or do not move into open areas where aerial targeting is possible. A combination of both types of efforts will be necessary to reduce goat numbers toward eradication.

During our evaluation of the control effort, taking into consideration hours of effort, shooting results, and field observations, we identified several trends that could be improved upon during the eradication phases of the project. First, we found that hunters shooting from the aerial platform performed better when shorter flights were made, probably because longer flights produced fatigue. Second, Anatahan's terrain is so treacherous that ground teams were far less effective in removing large numbers of goats than were hunters using the aerial platform, although they were able to dispatch animals that could not be accessed from the aerial platform. We surmised that if time and man-power were limited, the most effective use of grant funds would be to support



aerial gunning. And third, we found that capture and live removal of goats from Anatahan was the least cost effective method employed with very few animals removed from the island in this manner.

As a consequence of this evaluation, DFW and USFWS decided that the majority of the effort in 2003 should be focused on short, frequent trips to Anatahan by aerial gunnery crews. Ground work is necessary to radio telemeter the goats and will become important again later in the project. Because the residents of Anatahan are present on the island year-round, their hunting efforts will be continuous. It is also planned that teams of hunting dogs and handlers will be deployed in various zones on Anatahan for 2-3 month periods during the latter half of 2003 in order to better eliminate pigs. It is intended that this work will be coincident with the building of a large fence around the village portion of Anatahan so that the eastern sections of the island can then be brought to eradication.

We also note that toward the close of the two-week control period in 2002, mechanical difficulties developed with the helicopter because the rotor had been scored. This not only reduced the amount of time spent at the control effort and made necessary costly repairs to the helicopter, but also highlighted several safety issues. In order to prevent future similar incidents, we (1) spend fewer consecutive hours firing from the helicopter in order to reduce shooter fatigue because fatigue leads to accidents; (2) improved compliance with safety procedures and procured better safety equipment (e.g., bullet catchers and safety harnesses); and (3) shooters were given training in basic helicopter survival techniques and gunning from a helicopter before control work renewed.

Unfortunately, DFW was not able to immediately follow-up the large hunting effort in May 2002 as was intended. A delay in the processing of the grant supporting the project prevented funds from reaching the CNMI until October delaying feral animal reduction activities. Prior to reinitiating control activities in January 2003, Jess Omar and Curt Kessler completed a helicopter gunnery training course to improve the safety of project personnel. Then radio telemetry collars were attached to eight female goats ("judas goats") that were then distributed around the island. Aerial gunnery trips were made in February, March, and April for two days each month with a constant effort of eight hours of shooting per trip. Approximately 824 goats and 52 pigs were removed from Anatahan over this four-month period for a total of 876 ungulates. Although effort remained constant each month, the number of goats removed has declined over time, indicating that their numbers are decreasing on the island and that they are becoming harder to find. Aerial gunnery has been effective in reducing the numbers of feral ungulates, however, intensive ground hunting efforts will need to be combined with the aerial platform to bring the island to the desired eradication status.

Recommendations 1. Densities of flies in the village area can be reduced by: • Having garbage-disposal systems that are closed to exclude fly breeding and

feeding opportunities, such as 55-gallon metal drums fitted with lids. Non-metal or



non-glass contents can be burned. Metal and glass should be stockpiled for removal or burial in pre-selected places on the island.

• Having sanitary toilet systems which exclude fly breeding or feeding opportunities. There are many such commercial systems available, although with a little ingenuity, physical work could be employed to achieve requirements at low cost.

• Adopting household cleanliness ethics, especially with regard to food storage and animal slaughtering.

2. Reduce feral goats and pigs in order to: • reduce the risk to plant and animal species indigenous to the island, • reduce soil erosion, • increase natural productivity, and • reduce the risk of tick-borne disease.

3. Construct a large-area fence near the village on the western end of the island to ensure that pigs and goats can be eradicated on the eastern portion of the island while domestic animals are tethered close to the village.

4. Eradicate feral animals from the remaining portions of Anatahan. • Reduce length of each removal period and increase number of removal periods to

avoid hunter fatigue. • Increase investment in aerial shooting time to improve efficiency. • Focus on removing pigs from the eastern portion of Anatahan.

The small volcano on the eastern end of Anatahan erupted in May 2003.



POSTSCRIPT On May 10 2003, the eastern volcano on Anatahan erupted sending steam, ash and rocks over the majority of the island. The feral animal control project is currently postponed until it is safe enough to approach the island again.

From photographs and reports by the CNMI Emergency Management Office (John Camacho pers. comm.), a heavy volcanic ash lies several feet deep over the central, northern, western, and southern sections of the island. The ash deposit to the east appears to be only a few inches deep, but the heat and ash seem to have smothered most vegetation. To date, feral pigs and goats, Mariana Fruit Bats, and flies appear to persist on the island, based on eye-witness reports. Although it is unclear to what extent the eruption has affected the terrestrial vertebrate and invertebrate fauna, it is assumed that the heavy layer of ash will have catastrophic effects on their populations. We hope that the surveys reported here, particularly the work done by Dr. Bourquin on the invertebrate fauna, will serve as a baseline for comparisons with future documentation of the re-colonization of Anatahan in the decades and centuries to come.

Aviation fuel depot on Alatoong Saipan after the eruption shows depth of ash deposits and damage to the vegetation.

If by November 2003 the eruption has decreased in strength, the eradication project will be resumed. Because residents of the island will not be able to return to their homesteads for many years to come, the fence portion of the project has become



superfluous and so has been cancelled. However, CNMI-DFW and USFWS remain committed to working with returning residents to better manage populations of domestic animals when that time comes.

Alatoong Saipan or golf course camp site in July 2000 prior to eruption.

Alatoong Saipan or golf course camp site in May 2003 post eruption.



Research camp site at Banku on eastern Anatahan covered in ash, May 2003.

Extent of heat and ash damage to vegetation near research camp; note leafless Hibiscus ti/eaceus.


The urgency of removing feral animals from Anatahan has increased due to the eruption. To a large degree the remaining feral animals will prevent recovery of the remaining vegetation and limit colonization of new plants if allowed to remain on the island. Subsequently, erosion of the newly deposited ash would be greater than will occur naturally and have a larger effect on marine resources. Thus, it has become even more imperative to remove the remaining animals as soon as it is safe to approach the island.

ACKNOWLEDGMENTS

Our special thanks to Dr. Ortwin Bourquin for volunteering to inventory Anatahan's invertebrates and to Sid Castro for assistance with fence construction. Personnel from CNMI-DFW and NIMO assisting with surveys and ungulate removal included: Gus Aguon, Ignacio Aldan, Ben Camacho, Greg Camacho, Larry Eloi, Alvin Fitial, Tony Flores, Happy Gideon, Sylvan Igisomar, Joe Kaipat, Henry King, Jess Metao, Sylverio Metao, John Olepai, Jess Omar, Don Reyes, Paul Reyes, Juan Salas, Ed Santos, Lawrence Stahl, Combinario Taisacan, James Tanaka, Joe Tomokane, and Robert Ulloa. This project was supported by a grant to the CNMI from the US Navy (DAMD1702-2-0053), by a grant to the CNMI through Federal Aid to Wildlife Restoration (PittmanRobertson funds, grant W-2-R-4 and W-2-R-5), and by the USFWS through in-kind support to Curt Kessler and by supporting the gunnery training classes for CK and JO. Photographs were taken by various staff and were also obtained from the internet.

LITERATURE CITED

Bauer, AM., and R.A Sadlier. 1992. The use of mouse glue traps to capture lizards. Herpetological Review, 23:112-113.

Berger, D.O., and H.C. Mueller. 1959 The Bal-chatri: A trap for birds of prey. Bird Band. Assu.30:18-26.

Cruz, J., L. Arriola, N. Johnson, and G. Beauprez. 2000. Wildlife and vegetation surveys Anatahan 2000. Technical report #6, CNMI-DFW, unpubl.

Falanruw, M., Cole, T., and Alan, A 1989. Vegetation surveys of Rota, Tinian, and Saipan, Commonwealth of the Northern Mariana Islands. Resource Bull. PSW27. Berkley, CA: Pacific Southwest Forest and Range Experiment Station, Forest Service, U.S. Department of Agriculture.



Johnson, N.C. 2001. A Survey of Mariana Fruit Bats in the Mariana Islands, including recent minimum population estimates, July 2001. Marianas Conservation Unlimited, unpubl.

Kessler, C. 2000. Anatahan Island, Commonwealth of the Northern Mariana Islands: feral animals and forest destruction. Zoology Unlimited, unpubl.

Marshall, A.P., D.J. Worthington, G.J. Wiles, and C.C. Kessler. 2001. Abundance and management of Mariana Fruit Bats and feral ungulates on Anatahan, Mariana Islands. Pacific Conservation Biology 7:134-142.

Mueller-Dombois, D. and Ellenberg, H. 1974. Aims and methods of vegetation ecology. John Wiley and Sons Inc., New York. Pp 100-125.

Sutherland, W. 1996. Ecological Census Techniques, a handbook. Cambridge University Press, Cambridge, UK. 336 pp.

Ohba, T. 1994. Flora and vegetation of the Northern Mariana Islands, Micronesia. Natural History Research, Special Issue NO.1 :13-69. Chiba Natural History Museum, Japan.

Rodda, G.H., M.J. McCoid, and T.H. Fritts. 1993. Adhesive Trapping II. Herpetological Review, 24:99-100

Wiles, Gary J, Thomas O. Lemke, and Nicholas H. Payne. 1989. Population Estimates of Fruit Bats (Pteropus mariannus) in the Mariana Islands. Conservation Biology 3:66-76.



APPENDIX 1

GPS coordinates for bird survey transects:

TRANSECT POINT COUNT UTM COORDINATES STATION

7 1 362869 1809032 7 2 362798 1809148 7 3 362739 1809267 7 4 362599 1809244 7 5 362516 1809277 7 6 POOR-COVERAGE 7 7 362255 1809330 7 8 362214 1809289 8 1 POOR-COVERAGE 8 2 POOR-COVERAGE 8 3 363210 1808903 8 4 POOR-COVERAGE 8 5 363208 1808902 8 6 363437 1808683 8 7 363381 1808558 8 8 363354 1808411 8 9 363410 1808311 8 10 363528 1808241 8 11 363409 1808174 8 12 363290 1808207 9 1 362915 1808953 9 2 362558 1808794 9 3 362492 1808895 9 4 362445 1809038 9 5 362377 1809139 9 6 362314 1809174 9 7 362532 1809306 9 8 362439 180933 10 1 362777 1808807 10 2 362727 1808684 10 3 362719 1808557 10 4 362695 1808436 10 5 362789 1808398 10 6 362929 1808102 10 7 POOR-COVERAGE 10 8 362930 1808102 10 9 363144 1808010


APPENDIX 2

VARIABLE CIRCULAR PLOT SURVEY PROTOCOL FOR NORTHERN ISLAND BIRD SURVEYS 1999·2002

Count protocol is based on the Variable Circular Plot method as articulated by Reynolds et al. 1980. For the northern islands of the CNMI, the protocol was modified from the original to suit the types of vegetation encountered, the species under consideration, and the skill level of the observers.

Transects: Transects are established along pre-existing trails and old roads where possible

in forested habitats using Soil and Water Conservation topographic maps. Where possible, a single compass direction is followed, however, on pre-existing paths, no attempt should be made to cut additional forest for the sake of establishing a straight line. The knowledge of island residents is crucial in deciding where to establish the survey transects and the best way of reaching the desired locations. Transects are normally set to cut across all vegetation types and cross the major geological land formations in the forest. Transects range in length from a few hundred meters to over a kilometer and range in number from four to eight, reflecting the number of personnel available for the survey and length of stay on the island. Transects are measured using a hip-chain and flagged at convenient intervals, but are not permanently marked except on Sarigan.

Count Stations: Point count stations are laid out along the above transects at 150 m intervals.

The center of each station is identified with an additional piece of flagging which is marked with transect and station numbers for certain identification. Each station is identified with GPS co-ordinates taken using a hand-held unit for GIS mapping purposes. Transect locations and count stations are noted on a topographic map for future reference.

Observer Training: Because the accuracy of distance estimates increases with practice, all

participating observers calibrate distance estimations for one four hour period the day before the survey. Eight (8) trials each are performed by estimating the distance to a flag (a) without sound aids, and (b) using tape-recorded bird calls. Estimates are given orally, then the true distance to the flag is verified with a 50-m tape measure. A second round of distance estimates for flags with and without sound are performed using written estimates. This series of trials is repeated in several different forest types. Tape recordings play calls of birds expected to be present to ensure accurate identification.

Point Cou nts: Counts are conducted for 5 minutes at each station and do not include a

playback of pre-recorded birdcalls. The distance to each bird detected is registered as


well as whether it is heard (H), seen (S), or both heard and seen (H/S or S/H). Often, two observers perform the counts; one is responsible for counting and the other for recording the registrations. Counts are conducted between 6 and 10 a.m. on days without strong winds or inclement weather.

Other Data: Weather conditions are noted at each station. The protocol and coding of

weather conditions are as follows:

(a) Wind speed in the Beaufort Scale Beaufort Number Wind Speed in mph Visual Wind Speed Indicators

o Less than 1 Smoke rises vertically 1 1 to 3 Wind direction shown by smoke

drift 2 4 to 7 Wind felt on face; leaves

rustle 3 8 to 12 Leaves, small twigs in constant

motion 4 13 to 18 Raises dust and loose paper;

small branches on the move 5 19 to 24 Small leafy trees sway; crested

wavelets on inland waters(survey must be discontinued in this kind of wind).

(b) Rain conditions using Weather Bureau code numbers: oClear 1 Light rain/drizzle 2 Showers 3 Heavy rain (survey must be interrupted until heavy rain passes) 4 Fog or smoke 5 Drizzle 7 Snow 8 Showers

(c) Cloud conditions as percent of sky covered to the nearest 10%.

Data Analysis: Survey counts are statistically analyzed using the computer software known as

DISTANCE. A sample size of over 60 detections must be used in DISTANCE analysis in order to achieve reasonable results. Because each individual survey rarely detects this many of each species, several survey data sets are run at the same time in a stratification that allows density estimates for each individual survey but pools the surveys to estimate detection functions.

Prior to model fitting, 5-10% of the furthest distances are eliminated from the data to help fit the mathematical functions. Distribution of the data is examined by histograms


of frequency and density before statistical analysis. The remaining distance estimates from the survey are then grouped into intervals. This grouping (e.g., 5 intervals of 19 m each) is chosen for each survey based on the best fit between model and data. The program may be run several times before achieving the optimum grouping. This procedure assumes that there are no significant observer effects, but the assumption should be tested in the future.

Literature Cited:

Reynolds, R.T., J.M. Scott, and R. A. Nussbaum. 1980. A variable circular-plot method for estimating bird numbers. Condor 82:309-313.



Invertebrate survey - Anatahan, Northern Mariana Islands.

O. Bourquin CREES, Northern Marianas College

November 2002

Introd uction.

I was invited by Dr. Tina de Cruz, Division of Fish and Wildlife, Saipan, to join an expedition to Anatahan and to collect terrestrial invertebrates as well as establishing a method of recording relative fly densities. The former would contribute towards establishing the biodiversity of the Island, and towards identifying invertebrates which could affect man in one way or another, and the latter to have some way of measuring the effects of reduction of goats on fly populations. My transport was financed by DFW, and CREES funds paid for my food, water, collecting and camping 5uipment. As part of a working group of eleven people, J left Saipan at 0100 h on 19th April 2002 and arrived back at 1815 h on 3' May 2002.

Acknowledgements

I am grateful to Dr. Tina de Cruz and her team for their assistance and enthusiasm, to CREES for allowing me to take part in the expedition; to colleagues in CREES, especially Mr. Ed Benjamin, for their assistance; and to Dr Saaristo (Finland) and Dr Samuelson (Honolulu) for spider and beetle identifications.

Collecting methods

Collecting methods included sweep-netting, netting targetted individual specimens, capturing individuals by hand, by small-paint brush dipped in alcohol, by trapping with styrafoam cup pit-traps with a commercial engine coolant and using malaise traps. Some litter samples were taken for later examination, and earthworms were collected by digging for them using a hand trowel. Several of the expedition members, especially Nate Hawley, assisted by bringing in specimens. All specimens were killed in 70% isopropyl alcohol (Aaron Brands commercial rubbing alcohol) and either preserved in this or subsequently mounted as dry specimens using standard pinning techniques. Some sight records were accepted, for example those for coconut crabs and several for butterflies, dragonflies and a few other invertebrates.

Fly density index

See Appendix A.

Identification of specimens

Specimen identifications were done by the author where possible, and the rest of the specimens were sent to the Bishop Museum, Honolulu, except for earthworms, spiders and crickets which were sent to Drs D. Plisko (Natal Museum, Pietermaritzburg, South Africa), M. Saaristo (Turku University, Finland) and Dan Otte (Philadelphia Academy of Sciences, USA) respectively. At the time of report preparation, many of the identifications had not been received.

General conditions

The part of the island in which collecting was conducted was on the NE coast, generally from sea-level to about 380m asl, with most collecting done between 10-120 m asl. The vegetation complexes had been greatly affected by goats and pigs, causing destruction of both ground cover and vegetation diversity. Large areas of bare ground and bedrock were evident. Especially noticeable was the absence of the beach morning glory (Ipomoea pescaprae), usually a common lowland flowering plant of the Micronesian islands, and attractive to a variety of invertebrates. A few hours spent on the last day in the village area indicated that where goats and pigs were not disturbing the vegetation, many more invertebrates, and a greater diversity of species, were immediately evident.

Where tree or shrub canopy was absent, the areas were dry - however moist soils were found in shade areas w here litter was present and in the numerous deep gullies furrowing the steep slopes. Ground cover was dominated by the ferns Pteris qliadriaurita in the lower-lying and shaded areas and Nephrolepis hirsutula from about 200 m on open slopes to areas under partial cover. Grass was not common, but Zoysia matrella did a sterling job in keep in soils intact on the lower slopes, forming denses carpetlike swards - obviously, like the ferns, not being eaten by goats.

It was generally hot to very hot, and the constant attention by many flies during the daylight hours, in all habitats, was an unpleasant nuisance factor.


Results

Relative fly density index

See Appendix I.

Biodiversity

There were numbers of taxa with representatives in the CNMI which were not found on Anatahan, but which may still be found there. Only those taxa for which representative species have been recorded from Anatahan are given below.

"CNMI- 3 (4 - 5 ?) species" means that three species are definitely known from the CNMJ, with perhaps 4 to 5 being present.

I 17 species were added to the previously known list (of 144 species), of which 10 were new records for the CNM I and one, a click beetle, is considered to be a new species and has been sent to a specialist for naming.

Table I indicates the numbers of species (including some unidentified species) which are so far known from Anatahan. This number will increase as more identifications are received, and ifmore collecting is carried out.

Table I. Terrestrial and freshwater species numbers known for Anatahan

Recorded previously Recorded previously and in 2002 New Analahan record New CNM [ record New species record Total

101 43 106 10 261

Phylum: Platyhelminthes Class: Turbellaria Flatworms

CNMI - 3 (4 - 5 ?) species

Microptaninae sp. Collected from Anatahan in 1992. Austratopacifica sp. Collected from Anatahn in 1992.

References Kawakatsu, M. and R.E. Ogren. 1994. A preliminary report on land planarians from the Northern Mariana Islands

(Turbellaria, Tricladida, Terricola). Nat. Hist. Res., Special Issue, No. I: 107-112.

Phylum: Mollusca Class: Gastropoda Snails and slugs

CNMI - 73 species.

Surprisingly, no land snails were found in the survey area, although they were searched for in leaf litter, under logs and rocks and under bases of coconut leaves. Their absence is probably related to a general drying out of the area following the heavy depredations by goats and pigs. At least seven species have been recorded previously from Anatahan.

Class: Gastropoda Subclass: Pulmonata

Family: Achatinellidae Etasmias sp. Previously recorded on Anatahan, 1992

Family: Partulidae Partuta gibba Ferussac Humped tree snail Previously recorded on Anatahan, 1992.

Family: Sublinidae Subulina octona (Bruguiere) Introduced. Previously recorded on Anatahan, 1992. AI/opeas spp. Previously recorded on Anatahan, 1992.

Family: He1icarionidae Lamprocystis sp. A. Previously recorded on Anatahan, 1992. Lamprocystis sp B. Previously recorded on Anatahan, 1992.


Liardetia sp. Previously recorded on Anatahan, 1992.

References Best, Bruce R. and C. E. Davidson. 1981. Inventory and atlas of the inland aquatic ecosystems of the Mariana

Archipelago. University of Guam Marine Laboratory Technical Report # 75.226 pp. Kurozumi, T. 1994. Land molluscs from the Northern Mariana Islands, Micronesia. Natural History Research, Special

Issue I: I 13-1 I9. Smith, Barry D. 1995. Status of the endemic tree snail fauna (Gastropoda: Partulidae) of the Mariana Islands. Unpublished

report for U.S. Fish and Wildlife Service, Pacific Islands Ecoregion. 19 pp.

Phylum: Annelida Earthworms and leeches

CNMI - 77 Species

Class: Oligochaeta Earthwonns At least three species of earthworms were collected, inhabiting moist soils in well shaded, damp drainage lines to dryer soils under litter in semi-open areas. These were sent to Dr. D. Plisko (South Africa) for identification, but could not be identified owing to a poor state of preservation, although preserved in 70 % alcohol. CREES has been apparently unable to obtain fonnalin (which is used to properly preserve such specimens) for the last 18 months. No previous records for Anatahan (or for the CNMI) were found.

References Nil

Class: Arachnida Family: Scorpionidae Scorpions

CNMI - 2 species.

Family: Ischnuridae Hormurus australasia (Fabricius) Two adults collected from rotten logs in survey area. One female had 10 unpigmented young, 4.5-6.0 mm long. No previous records from Anatahan.

References Chapin, E.A. 1957. Scorpionida. In: Insects of Micronesia 3(2): 65 - 70.

Class: Arachnida Order: Acari Suborder: Ixodoidea Ticks

CNMI - 5 species.

Boophilus microplus Canestrini Cattle tick Two found on a goat at about 500m asl. First record for Anatahan. Vector for babesia, anaplasma.

References Kohls. Glen M. 1957. Acarina: Ixodoidea. In: Insects of Micronesia 3(3): 85 - 104.

Class: Arachnida Order: Acari Mites

CNMI - 72 species.

Collected mites were sent to the Bishop Museum for identification.

Family: Parasitidae Parasitus sp. Recorded from Anatahan during 1992.

Family: Ascidae Asca aphidioides (L.) Recorded from Anatahan during 1992. Asca quinquesetosa Wharton Recorded from Anatahan during 1992. Cheiroseius sp. Recorded from Anatahan during 1992. Lasioseius sp. Recorded from Anatahan during 1992.

Family: Ameroseidae



Ameroseius sp I Should be found on Anatahan, found on Sarigan during 1992.

Family: Rhodacaridae Rhodacarus sp. Recorded from Anatahan during 1992.

Family: Ologamasidae Gamasiphus sp I. Recorded from Anatahan during 1992. Gamasiphus sp 2 Recorded from Anatahan during 1992.

Family: Digamasellidae Dendrolaelaps sp Recorded from Anatahan during 1992.

Family: Podocinidae Podocinumjamaicensis Evans and Hyatt Recorded from Anatahan during 1992.

Family: Eviphididae Evimirus sp. Recorded from Anatahan during 1992. Family: Macrochelidae Macrocheles sp I Recorded from Anatahan during 1992. Macrocheles sp 2 Recorded from Anatahan during 1992.

family: Laelapidae Hypoaspis sp I Should be on Anatahan, found on Sarigan, 1992. Hypoaspis sp 2 Recorded from Anatahan during 1992. Hypoaspis sp 3 Recorded from Anatahan during 1992. Hypoastis sp 5 Recorded from Anatahan during 1992. Hypoastis sp 6 Recorded from Anatahan during 1992.

Family: Uropodidae Trigolluropoda sp A Recorded from Anatahan during 1992. Trigolluropoda sp C Recorded from Anatahan during 1992. Trigonuropoda sp D Recorded from Anatahan during 1992.

References Aoki, J. 1994. Orabatid mites of the northern Mariana Islands. Nat. Hist. Res., Special Issue, No. I: 181-194. Baker, E.W. and M.D.Delfinado. 1964. Spintumicidae of south east Asia and the Pacific region. Pacific Insects 6(4): 571

591. Ishikawa, K. 1994. Preliminary list of terrestrial Gamasid mites (Acarina) from the northern Mariana Islands, Micronesia.

Nat. Hist. Res., Special Issue, No. 1: 175-177.

Class: Arachnida Order: Araneida Spiders.

CNMI - 84+ species.

All records (33) are first records for Anatahan.

Cribellate sp. 1 Collected from Anatahan, 2002.

Family: Araneidae Argiope appensa (Walck.) Collected from Anatahan, 2002. Cyclosa bifida (Doleschall) Collected from Anatahan, 2002. Cyrtophora moluccensis (Doleschall) Collected from Anatahan, 2002.

Makes a large three dimensional labyrinth of threads supporting a central dome, a series of I to 4 egg pouches above the hub of the dome. Nests are also occupied by small C. moluccensis, Argyrodes spp., Leucage spp., Gea or Argiope spp. The adult females are about 2.5 cm long, the males are about 3 mm long. Female abdominal patterns are extremely variable with large areas of black and reddish brown to bright greenish orange mottling on white/whitish background. Abdominal patterns ofjuveniles always white green and silver.

Neoscona sp. Collected from Anatahan, 2002. Neoscona (heisi (Walck) Collected from Anatahan, 2002.



Family: Clubionidae Clubionidae sp. Collected from Anatahan, 2002.

Family: Gnaphosidae Gnaphosidae sp. A Collected from Anatahan, 2002.

Family: Linyphiidae Neollesiotes remiformis Millidge Collected from Anatahan, 2002. Micronetinae Collected from Anatahan, 2002.

Family: Lycosidae Lycosidae sp. A Collected from Anatahan, 2002. Lycosidae sp. B. Collected from Anatahan, 2002.

Family: Ochyroceratidae Speocera sp. I. Collected from Anatahan, 2002.

Family: Oonopidae Opopaea sp. I Collected from Anatahan, 2002.

Family: Pholcidae Pholcidae sp. I Collected from Anatahan, 2002. Pholcidae sp. 2 Collected from Anatahan, 2002.

Family: Salticidae Cosmophasis sp. Collected from Anatahan, 2002. Myrmaraclme sp. Collected from Anatahan, 2002. Salticid sp. 4 Collected from Anatahan, 2002. Salticidae Collected from Anatahan, 2002.

Family: Scytodidae "Scytodes"fusca (Walck.) Collected from Anatahan, 2002. "Scy/odes" lugubris (Thorell) Collected from Anatahan, 2002.

Family: Tetragnathidae "Leucauge" sp Collected from Anatahan, 2002. "Leucauge" sp 2 Collected from Anatahan, 2002. "Leucage" sp. B Collected from Anatahan, 2002. "Leucage" sp. C Collected from Anatahan, 2002. "Leucage " sp. 5 Collected from Anatahan, 2002.

Family: Therididae Argyrodes argenta/us (Pickard-Cambridge) Collected from Anatahan, 2002. Argyrodes sp. 3 Collected from Anatahan, 2002. Argyrodes sp. 4 Collected from Anatahan, 2002. ColeosomafloridallG Banks Collected from Anatahan, 2002. Theridiidae sp. 4 Collected from Anatahan, 2002.

Family: Thomisidae Thomisidae sp. I Collected from Anatahan, 2002.

References Roewer, von C. FR. 1963. Araneina: Orthognatha and Labidognatha. In: Insects of Micronesia 3(4): 105 - 132. Saaristo, Dr. M., Turku University, Finland 2002. Identified the spiders listed above.

Class: Crustacea Order: Isopoda Sowbugs, isopods

CNMI - 19 + species

All specimens collected were sent to the Bishop Museum for identification. At least two species were represented.

Family: Arrnadillidae



Sphaerillo sp. 2 Previously recorded in Anatahan during 1992.

Family: Trichoniscidae Trichoniscidae sp. Previously recorded in Anatahan during 1992.

Family: Philoscidae Setaphora sp. 1 Previously recorded in Anatahan during 1992.

Family: Trachelipodidae Nag14rus sp. 2 Previously recorded in Anatahan during 1992

References Anon 200 I. NMNH.si.edu/gopher-menusl World list of terrestrial Isopoda, 13 January 200 I Nunomura, N. 1994. Peliminary list of terrestrial Isopod crustaceans from the northern Mariana Islands, Micronesia. Nat.

Hist. Res., Special Issue No. I: 259-261

Class: Crustacea Order: Decapoda Crabs

CNM[ - 16 species.

Family: Coenobitidae (Land hennit crabs) Birgus lalro L. Coconut crab Many sight records from the study area, and photographs taken. No previous records known. Coenobita perlalus Milne-Edwards. Red hermit crab Several specimens found in low-lying areas - none collected. Previous records 1992. Coenobita breviman14S Dana Short-handed hermit crab. Previously recorded from Anatahan during [992. Seen during survey. Coellobita rugosus H. Milne-Edwards. Rough hermit crab. Previously recorded from Anatahan during 1992.

Family: Gecarcinidae Land crabs Cardisoma hirtipes Hairy-legged land crab Seen occasionally, two specimens collected, one from a well shaded stream forest area, and one from a gully near the camp. Not previously recorded.

References Asakura, A., T. Kurozumi and T. Komai. 1994. Anomura (Crustaces: Decapoda) collected from the northern Mariana

[slands, Micronesia. Nat. Hist. Res., Special Issue No.1: 275-283 Gressitt, J. Linsley. 1954. Insects of Micronesia Introduction. In: Insects of Micronesia I: I Takeda, M., Y. Fukui, K. Wada and A. Asakura. 1994. Brachyura (Crustaces: Decapoda) from the northern Mariana

Islands, Micronesia. Nat. Hist. Res., Special Issue No. I: 285-29

Phylum: Arthropoda Class: Diplopoda Millipedes

CNMI - 10 species

Specimens ofColobognathids, Polydesmids and Spirobolids were sent to the Bishop Museum for identification. No previous records for Anatahan were found. At least 5 species are expected.

References Lawrence, R.F. 1984. The centipedes and millipedes of southern Africa. A.A. Balkema, Cape Town. 148 pp.

Phylum: Arthropoda Class: Chilopoda Centipedes

CNMI - 8 species

A few specimens, representing at least two species were collected, and were sent to the Bishop Museum for identification. There appear to be no previous records for the island.

References Gressitt, J. Linsley. 1954. Insects of Micronesia Introduction. In: Insects of Micronesia 1: I - 257. Lawrence, R.F. 1984. The centipedes and millipedes of southern Africa. A.A. Balkema, Cape Town. 148 pp.

Pbylum: Arthropoda Class Sympbyla Symphylans



CNMI - 2 species.

Family: Scolpendrellidae Symphy/ella marianensis Scheller Previously found in Anatahan during 1992.

References Scheller, Ulf. 1994. A new scolopendrellidae species (Myriapoda: Symphyla) trom the northern Mariana Islands,

Micronesia, with a list of west Pacific Symphyla species. Nat. Hist. Res., Special Issue No. I: 251-256.

Class: Insecta Order: Collembola Springtails

CNMI - 4-8 species.

All specimens collected were sent to the Bishop Museum for identification.

Family: Neelidae Nee/us minimus (Willem) Previously recorded trom Anatahan during 1992.

Family; Sminthuridae Sphaeridia biniserrala (Salmon) Previously recorded trom Anatahan duringl992.

References Itoh, R. 1994. Symphypleona (Insecta: Collembola) trom the Northern Mariana Islands, Micronesia. Nat. Hist. Res., Special

Issue, No. I: 217-218.

Class: Insecta Order: Tbysanura Bristletails, silvertisb, fisb motbs

CNMI - 3 species

One specimen was found in a piece of luggage which had come from Saipan. It is not known if the bristletail entered the luggage on Saipan or on Anatahan.

References No specific references found.

Class: Insecta Order: Odonata Dragonflies

CNMI - 12 species

Suborder: Anisoptera Dragonflies Family: Aeschnidae A large green-bodied dragonfly was seen near the east rim of the crater, and was probably Anox piratieus Kennedy. This species has not been recorded from Anatahan before, and is known only from Saipan and Guam in the Marianas - both records being at least 50 years old.

Family: Cordulidae Hemieordulia mindana Needham and Gyger Anatahan ~ an adult male and female were collected in the survey area. This is the first record of a cordulid in the Northern Marianas. Previously known in the Marianas trom Guam only.

Family: Libellulidae Agrionoplera insignis guamensis Lieftienck Previously recorded trom Anatahan in 1992. Dip/aeodes hipunelala (Brauer) Previously recorded trom Anatahan in 1951 and 1992. This small species was seen flying at both the survey area and at the village site, where one specimen was collected. T11O/ymis tillagra (Fabricius) Previously recorded trom Anatahan in 195 I, seen flying at both the survey area and at the village site.

References Lieftinck, M. A. 1962. Odonata. In: Insects of Micronesia 5(1): I - 95. Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No. I: 199-215.



Class: Insecta Order: Phasmida Family: Pbasmatidae Stick insects, walking sticks

CNMI - I species

AcallthograefJea dellliculata (Redtenbacher) Denticulate stick-insect No specimens recorded during this survey, but previously recorded from Anatahan during 1992.

References Kevan, D. Keith McE and Vernon R. Vickery. 1997. An annotated provisional list ofnon-saltatorial

orthopteroid insects of Micronesia, compiled mainly from literature. Micronesica 30(2): 269-353. Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Class: Insecta Order: Orthoptera Suborder: Caelifera Short-horn grasshoppers

CNMI - 9 species

Family: Acrididae Subfamily: Cyrtacanthacrininae Valanga excavata (Stal) Previously collected in Anatahan during 1991. One specimen collected during this survey, several seen. Seemingly confined to semi-open, sheltered patches. Locusta migratoria mallilellsis (Meyen). Migratory locust Previously collected in Anatahan during 1992. This species was common in the survey area at all altitudes examined, but only in open areas. A number of specimens were collected.

Subfamily: Catantopinae Ai%pus tha/assinus tamu/us (Fabr.)/ Ai%pus thalassilllls dubius Willemse intergrades Previously collected in Anatahan during 1992. This small species was fairly common in open short-grass areas - two were collected. Stellocatalltops sp/endens (Thunberg) Previously collected in Anatahan during 1992. A common species found in both open areas and semi-shaded areas at the ecotones of forest and fern land/grassland.

References Gressitt, J. Linsley. 1954. Insects of Micronesia Introduction. In: Insects of Micronesia I: 1 Kevan, D. Keith McE, Vernon R. Vickery and Mary-Lynn English. 1997. Acridoidea and related Orthoptera

(Grasshoppers) of Micronesia. Micronesica 30( I): 127-168. Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No.1: 199-215.

Class: Insecta Order: Ortboptera Suborder: Caelifera Long-horn crickets

CNMI - II species.

Family: Phaneropteridae Phaneropterafurcifera Stal Previously collected in Anatahan during 1992. This species was common where good ground cover was available, and was collected.

Family: Meconematidae Phisis parva Kevan One specimen, the first record, collected from Anatahan during this survey.

References Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No.1: 199-215. Vickery, Vernon R., D. Keith McE. Kevan and Mary-Lynn English. 1999. Insects of Micronesia 5 (4), Gryllacridoidea,

Rhaphidophorioidea and Tettignoidea (Grylloptera). In: Micronesica 32( 1): II - 83

Class: Insecta Order: Orthoptera Suborder: Ensifera Crickets and mole crickets

CNMI - 5 (to 10?) species


All collected specimens sent to Dr. D. Olte (U.S.A.) for identification. Family: Gryllidae Subfamily: Trigonidiinae Sword-bearing crickets Anaxipha sp. Collected from Anatahan in 1992. Subfamily: Mogoplistinae Ornebius sp. Collected from Anatahan during 1992.

Family: Gryllotalpidae Gryflotalpa orientalis Burmeister. Several specimens of mole cricket were collected from pit traps in deep ravine forest, and one flying to lights at night in the campsite. First record for Anatahan.


Special Issue No. I: 199-215. Swezey,O.H. 1946. Orthoptera and related orders of Guam. In: Insects of Guam - II, pp 3-8. Bulletin 189, Bernice P.

Bishop Museum, Honolulu, Hawaii. 257 pp.

Class: Insecta Order: Mantodea. Mantids, praying mantises

CNMI - 4 species

Family: Mantidae Tenodora angustipennis Saussure A mantis nest was found at about 380m asl in an open grassy/weedy area in the survey site on 25 April 2002, and 112 juveniles hatched from it on 3 May 2002. This is a first record for Anatahan.

References Kevan, D. Keith McE and Vernon R. Vickery. 1997. An annotated provisional list of non-saltatorial orthopteroid insects of

Micronesia, compiled mainly from literature. Micronesica 30(2): 269-353.

Class: Insecta Order: Blattaria Cockroacbes

CNMI- 10 species

Family: Blattidae Periplaneta australasia (Fabricius) Australasian cockroach Recorded from Anatahan during 1992. Blattidae sp. A small species collected during this survey.

Family: Blaberidae (=Nauphoetidae) Pycnoscelus indicus (F.) Recorded from Anatahan during 1992, and collected during this survey.

Family: Blatellidae Balta (= Onychostylus) notulata (Stahl). Collected during this survey, not previously recorded. Blatellid sp. A small species, collected during this survey.

References Kevan, D. Keith McE and Vernon R. Vickery. 1997. An annotated provisional list of non- saltatorial orthopteroid insects of

Micronesia, compiled mainly from literature. Micronesica 30(2): 269-353. Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Class: Insecta Order: Isoptera Termites

CNMI - 4 species

Suborder: Afontanella Family: Kalotermitidae Cryptotermes domesticus (Haviland) Found in Anatahan during 1992. Neotermes connexus Collected from dead log, under closed canopy. Not previously recorded from Anatahan.

Suborder: Fontanella Family: Heterotermitidae (=Rhinotermitidae)


Prorhinotermes illopina/us Silvestri Found on Anatahan during 1992, and collected during this survey from dead tree trunks. Winged individuals were found in pit traps in woodland, and in malaise traps.

References

Kevan, O. Keith McE and Vernon R. Vickery. 1997. An annotated provisional list of non- saltatorial orthopteroid insects of Micronesia, compiled mainly from literature. Micronesica 30(2): 269-353.

Miles T. G. 1998. www.utoronto.ca Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Class: Jnsecta Order: Dermaptera Earwigs

Family: Chelisochidae Chelisoches mario (F.) Collected from camp area. First record for Anatahan.

Family: Anisolabididae (=Carcinophoridae) Euborelfia annulata (F.) Collected from litter under Barrangtonia asiatica, and from the dry skull ofa goat. First record for Anatahan.

Family: Spongiphoridae (=Labiidae) Chae/ospaniafuscata clavata Found under litter in Barangtonia asiatica woodland, and in the camp area. First record for Anatahan. Paralabel/ula (=Labia) curvicauda (Motshulsky) Found on Anatahan during 1992.

References Kevan, O. Keith McE and Vernon R. Vickery. 1997. An annotated provisional list of non- saltatorial orthopteroid insects of



Class: Insecta Order: Embiidina Webspinners

CNMI - 3 species.

Aposthonia microllesiae (Ross) Recorded from Anatahan during 1951, and collected during this survey. Olig%ma humbertiana (Saussere) Collected during this survey, first record unless the record below is of this species. Olig%midae sp. Recorded from Anatahan during 1992

References Kevan, O. Keith McE and Vernon R. Vickery. 1997. An annotated provisional list of non- saltatorial orthopteroid insects of


Special Issue No. I: 199-215. Ross, Edward S. 1955. Embioptera. In: Insects of Micronesia 8( I): 1 - 8.

Class: Insecta Order: Psocoptera Psocids

CNMI -41 species.

Collected specimens were sent to Bishop Museum for identification.

Family: Lepidopsocidae Echmepteryx madagascariensis (Kolbe) Collected from Anatahan during 195 I.

Family: Pseudocaeciliidae Lobocaecilius fellnecus Lee and Thornton Found on Anatahan during 195 t .

References Thornton, Ian W. B., Lee S. S. and W.O.Chin. 1972. Psocoptera. In: Insects of Micronesia 8(4): 45 - 144.



Hemipterans All collected hemipterans were sent to Bishop Museum for identification.

Class: Insecta Order: Hemiptera Family: Anthocoridae Minute pirate bugs

CNMI - 12 species.

Buchallalliella sodalis (White) Recorded from Anatahan during 1951.

References Herring, Jon L. 1967. Heteroptera: Anthocoridae. In: Insects of Micronesia 7(8): 392 - 414.

Class: Insecta Order: Hemiptera Family: Miridae Leaf bugs

CNMI - 25 species.

TrigOllotylus dohertyi (Distant) Collected on Anatahan during 1992. Creolltiades pallidifer (Walker) Sweetpotato yellow bug Collected on Anatahan during 1992.

References Carvalho, Jose C. M. Heteroptera: Miridae. In: Insects of Micronesia 7( t): 1-100. Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Class: Insecta Order: Hemiptera Family: Nabidae Damsel bugs

CNMI - 2 species

Nabis nigrolilleatus (Distant) Collected from Anatahan during 1992.

References Gross, Gordon F. 1963. Heteroptera: Coreidae, Neididae and Nabidae. In: Insects of Micronesia 7(7): 357-390 Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Class: Insecta Order: Hemiptera Family: Reduviidae Assassin bugs

CNMI - 2 species.

Scadra rufidens Stal Recorded from Anatahan during 1992.


Special Issue No.1: 199-215. Wygodzinsky, Pedro and Robert L. Usinger. 1960. Heteroptera: Reduviidae. In Insects of Micronesia 7(5): 231 - 283.

Class: Insecta Order: Hemiptera Family: Lygaeidae Seed bugs

CNMI - 16 species

Nysius pulchellus Stal Lygeid bug Collected from Anatahan, 1951-1992. Paromius gracilis (Ramber) Collected from Anatahan during 1992.

References Barber, Harry G. 1958. Heteroptera: Lygaeidae. In: Insects of Micronesia 7(4): 173 - 218. Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No. I: t99-215.

Class: Insecta Order: Hemiptera Family: Alydidae Broad-beaded bugs


Anatallan Feral Animal Control Project 2002-2003

CNMI - 1 species

Melanacanthus marginegullatus Distant Collected from Anatahan during 1992.



Class: Insecta Order: Hemiptera, Family:Pentatomidae Stink bugs

CNMI - 13 species

Bulbostethus transversalis Ruckes Collected from Anatahan during 1992

References Ruckes, Herbert. 1963. Heteroptera: Pentatomoidea. In: Insects of Micronesia 7(7): 307-356. Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Class: Insecta Order: Hemiptera Family: Plataspidae Plataspid stink bugs

CNMI - 3 species

Coptosoma xanthogramma (White) Collected from Anatahan during 1992, and during this survey in a malaise trap at about 380 m as!.


Special Issue No. I: 199-215. Ruckes, Herbert. 1963. Heteroptera: Pentatomoidea. In: Insects of Micronesia 7(7): 307-356.

Homopterans

All collected homopterans were sent to Bishop Museum for identification.

Class: Insecta Order: Homoptera Family: Cicadellidae Leafboppers

CNMI - 21 species

Orosius argentatus (Evans) Collected from Anatahan during 1992.

References Linnavuori, R. 1960. Homoptera: Cicadellidae. In: Insects of Micronesia 6(5): 231 - 344 Linnavuori, R. 1975. Homoptera: Cicadellidae supplement. In: Insects of Micronesia 6(9): 611-632 Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Class: Insecta Order: Homoptera Family: Delphacidae Delphacid planthoppers

CNMI - 17 species.

Ugyops anatahani Fennah Collected from Anatahan during 195[.

References Fennah, R. G. 1956. Homoptera: Fulgoroidea. In: Insects of Micronesia 6(3): 39 - 211. Fennah, R. G. 1971. Homoptera: Fulgoroidea.Supplement In: Insects of Micronesia 6(8): 563 - 609. Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,




Class: Insecta Order: Homoptera Family: Psyllidae Jumping plantlice

CNMI - 7 species

Mesohollloloma hibisci (Froggatt) Collected from Anatahan during 1951.

References Caldwell,l.S. 1942. Psyllidae from Guam. Pp. 20-22. In: Insects of Guam-I. Bernice P. Bishop Museum Bulletin 172.

Honolulu, Hawaii. Tuthill, Leonard D. 1964. Homoptera - Psyllidae. In: Insects of Micronesia 6(6): 353-376.

Class: Insecta Order: Homoptera Family: Margarodidae Cottony cushion scales

CNMI - 2 species.

Icerya aegypliaca (Douglas) Egyptian fluted scale Recorded from Anatahan during 1959.

References Beardsley, 1. W. 1m. 1966. Homoptera: Coccoidea. In: Insects of Micronesia 6(7): 377-562. Beardsley, J. W. Jnr. 1975. Homoptera: Coccoidea supplement. In: Insects of Micronesia 6(9): 657-662 Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the

expedition. Nat. Hist. Res., Special Issue No. I: 199-215.

Class: Insecta Order: Homoptera Family: Diaspididae Armoured scales

CNMI - 20 species

Lepidosaphes esakii Takahashi Recorded from Anatahan during 1951.

References Beardsley, J. W. Jm. 1966. Homoptera: Coccoidea. In: Insects of Micronesia 6(7): 377-562. Beardsley, J. W. Jnr. 1975. Homoptera: Coccoidea supplement. In: Insects of Micronesia 6(9): 657 - 662 Moore, A. and A. Tudela. 1999. http://WWW.Crees.orgiplantprotection.NorthernMarianasCollege.Saipan. Updated

February 16, 1999.

Class: Insecta Order: Thysanoptera Thrips

CNM I - 16 species

Family: Phaleothripidae Dexiolhrips madrasensis (Anauthakrishnan) Collected from Anatahan during 1992. Haplolhrips pal/escens (Hood) Collected from Anatahan during 1992.

References Bianchi, Fred A. 1965. New Thysanoptera records from the Caroline and Mariana Islands. Proc. Hawaiian Ento. Soc.

XIX( I): 69-72 Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No. I: 199-2/5. Moore, A. and A. Tudela. 1999. http://WWW.Crees.orgiplantprotection.NorthernMarianasCollege.Saipan. Updated

February 16, 1999. Sakimvra, K. 1971. A review of the Genus Rhaebot!lrips Karny (Thysanoptera : Phlaeothripidae). Pacific Insects 13(2): 291

-403.

Class: Insecta Order: Neuroptera Lacewings and antlions

CNMI - 8 species

Family: Myrmeleontidae



Myrmeleon acer Walker Two larvae were collected /Tom dry sand under an overhang on a steep slope, probably belonging to this species. The only other ant lion which might occur on Anatahan (Dis/oleo" bistrigatus) is a large species, while the larvae which were collected are relatively small. No anti ions have previously been recorded ITom Anatahan.

Family: Chrysopidae Chrysops( Mal/ada ?) basalis Walker One specimen collected, previously not known from Anatahan. Mal/ada alcestis (Banks) Previously collected from Anatahan during 1992 Chrysops(Plesiochrysa ?) oceanica Walker Previously collected from Anatahan 1992, and two specimens collected during this survey.

References Carpenter, F. M. 1961. Neuroptera: Hemerobiidae. In: Insects of Micronesia 8(3): 35 - 43. Adams,1. 1959. Neuroptera: Mynneleontidae and Chrysopidae. In: Insects of Micronesia 8(2): 13 - 33. Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Class: Insecta Order: Coleoptera Family: Carabidae including Cincidelinae Carabid beetles

CNMI - 16 species

EndYllomena pradieri (Fairmaire) Previously collected from Anatahan during 1941. Tachys brachys Andrewes Previously collected ITom Anatahan during 1992.

References Darlington, P.I. 1970. Coleoptera: Carabidae including Cincidelinae. In: Insects of Micronesia 15(1): I - 509 Kasahara, S. 1994. Carabidae (Insecta: Coleoptera) of the Mariana Islands, Micronesia. Nat. Hist. Res., Special Issue, No.

I: 221-222.

Class: Insecta Order: Coleoptera Family: Dysticidae Diving beetles

CNMI - 4 species

Rhalltus pulverosus Stephens Previously recorded from Anatahan during 1992.



Class: Insecta Order: Coleoptera Family: Histeridae Hister beetles

CNMI- 2 species

Platysoma anatahensis Ohara Previously recorded ITom Anatahan during 1992. Saprinus cf. cyaneus F. Collected, first record for CNMI.

References Ohara, M. 1994. A new histerid beetle (Insecta: Coleoptera: Histeridae) from the northern Mariana islands, Micronesia.

Nat. Hist. Res., Special Issue, No. I: 223-226.

Class: Insecta Order: Coleoptera Family: Staphylinidae Rove beetles

CNMI - 8+ species

Scopaeus sp. Previously recorded ITom Anatahan during 1992. Anofylus sp. Previously recorded ITom Anatahan during 1992, and during this survey. Carpelimus sp. Previously recorded from Anatahan during 1992, and during this survey. Edaphus sp. Previously recorded from Anatahan during 1992, and during this survey. Philonthus longicornis Stephens Collected, first record for CNMI Omaliinae sp. Collected, first record of subfamily for CNMI.





Class: Insecta Order: Coleoptera Family: Ptiliidae Feather-winged beetles

CN MI - I species

Acrotrichis sp. Previously recorded from Anatahan during 1992.


Special Issue No. I: 199-215

CI.ass: Insecta Order: Coleoptera Family: Lucanidae Stag beetles

Figulus integrico/lis Thomson Previously recorded from Anatahan during 1992. Figulus memo semperi Kriesche Collected, first record for CNMI.



Class: Insecta Order: Coleoptera Family: Passalidae Bess beetles

CNMI - 1 species.

Popilius sp ? No previous records. Collected during this survey, 2002.

References Nil

Class: Insecta Order: Coleoptera Family: Scarabeidae Scarab beetles

CNMI - 12 species

Adore/us siniCliS Burmeister Chinese rose beetle Not previously recorded from Anatahan, collected during this survey. Aphodius lividus (Olivier) Not previously recorded from Anatahan, collected during this survey A/aenius orbicularis Schmidt Not previously recorded from Anatahan, collected during this survey

References Cartwright O.L. and R.D.Gordon 1971. Coleoptera: Scarabeidae. In: Insects of Micronesia 17(4): 257-296. Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No. J: 199-215 Moore, A. and A. Tudela. 1999. http://WWW.Crees.orgiplantprotection.NorthernMarianasCollege.Saipan. Updated

February 16, 1999.

Class: Insecta Order: Coleoptera Family: Buprestidae Buprestid beetles

CNMI- 3 species

Cyphogastra auripennis Saunders Not previously recorded from Anatahan, one specimen found during this survey.




Class: Insecta Order: Coleoptera Family: Elateridae Click beetles

CNMI - 15 species.

Conoderus (=Monocrepidius?) pallipes Eschscholtz Collected during this survey, 2002. No previous records. Lacon modestus (Boisduval) Collected during this survey, 2002. No previous records. SimodactyLus paUidus Fleutiaux Collected during this survey, 2002. No previous records. MeLanoxfIlhus sp. nov. Collected during this survey. NEW SPECIES


Special Issue No. I: 199-215 Van Zwaluwenburg, R.H. 1957. Coleoptera: Elateridae. In: Insects of Micronesia 16(1): 1-66. Van Zwaluwenburg, R.H. 1942. Elaterid and Eucnemid beetles from Guam. In: Insects of Guam - I. Bernice P. Bishop

Museum Bulletin 172. Pp. 53-55.

Class: Insecta Order: Coleoptera Family: Eucnemidae False click beetles

CNMI - I species

Fomax coxalis Fleutiaux Collected, and first record for CNMI.

Class: Insecta Order: Coleoptera Family: Dermestidae Dermestid beetles

CNMI - 7 species.

Dermestes ater De Geir Not previously recorded, specimens collected from rotting goat carcases. Dermestes macula/LIs De Geir Not previously recorded, specimens collected from rotting goat

carcases

References Beal, R. S. 1961. Coleoptera: Dennestidae. In: Insects of Micronesia 16(3): 109 - 131.

Class: Insecta Order: Coleoptera Family: Bostrichidae Branch and twig borers

CNMI - 5 species

Collecled specimens sent to Bishop Museum for identification.

No previous records found, several specimens awaiting identification.

References Chujo, Michio. 1958. Coleoptera: Bostrychidae. In: Insects of Micronesia 16(2): 85 - 104. Swezey, H. O. 1942. Miscellaneous families of Guam Coleoptera. Pp. 150-171. In: Insects of Guam - I, Bernice P. Bishop

Museum - Bullelin 172.

Class: Insecta Order: Coleoptera Family: Cleridae Checkered beetles

CNMI - 1 species

Necrobia rufipes (De Geer) Collected and first record for island.

Class: Insecta Order: Coleoptera Family: Coccinellidae Ladybugs

CNMI - 15 species.

Chi/ocorus sp. nr nigritus (F.) Found in Anatahan during 1992. CoccineUa transversaLis F. Found in Anatahan during 1992 CoeLophora inaequaLis var il/aequa/is (Fabricius) Found in Anatahan during 1992



Mellochilus (=CJ,eilomelles?) sexmaculatus (F.) Found in Anatahan during 1992, and collected during this survey.

References Chapin, Edward A. 1965. Coleoptera: Coccinellidae. In: Insects of Micronesia 16(5): 189 - 254. Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No.1: 199-215

Class: Insecta Order: Coleoptera Family: Tenebrionidae Darkling beetles

CNMI - 20 species

GOllocephalum adpressiforme Kaszab. Collected from Anatahan during 1992, and during this survey.

References

Kulzer, Hans. 1957. Coleoptera: Tenebrionidae. In: Insects of Micronesia 17(3): 185 - 256. Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Class: Insecta Order: Coleoptera Family: Mordellidae Tumbling flower bettles

CNMI - I (2?) species

Dellamora castanea (Boheman) Collected, and first record for CNMI.

Class: Insecta Order: Coleoptera Family: Cerambycidae Longhorn beetles

CNMI - 23 species

Prosoplus marianarum Aurivillius Collected from Anatahan during 1951-1992. Pterolophia camura (Newman) Collected during this survey, 2002, no previous records. Ropica palauana (Matsushita) Collected from Anatahan during 1992. Sybra anatahana Gressitt Collected from Anatahan during 1951 and during present survey, 2002. Sciadella(=Phloeopsis ?) meridiana (Ohbayashi) Collected from Anatahan during 1992, and during present survey. Sciadella(Microllesiella) mariana Gressitt Collected from Anatahan during 1992, and during

present survey 2002.

References Gressitt, J. Linsley. 1956. Coleoptera: Cerambycidae. In: Insects of Micronesia 17(2): 61-183. Miyano, S. J994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No. I: 199-215 Saito, A. 1994. Records of some Cerambycidae (Insecta: Coleoptera) from the northern Mariana islands, Micronesia. Nat.

Hist. Res., Special Issue, No. J: 227-228.

Class: Insecta Order: Coleoptera Family: Chrysomelidae Leaf beetles

CNMI - 10 species.

Phytorus lineolalus Weise Phytorus leaf beetle Collected during this survey, no previous records.

References Gressitt, J. Linsley. 1955. Coleoptera: Chrysomelidae. In: Insects of Micronesia 17( 1): I - 60 Moore, A. and A. Tudela. 1999. http://WWW.Crees.orgiplantprotection.NorthernMarianasCollege.Saipan. Updated

February 16,1999.

Class: Insecta Order: Coleoptera Family: Anthribidae Fungus weevils

CNMI - 5 species.


Araecems vieillardi (Montrouzier) Collected from Anatahan during 1992. Mauia subnotafus (Boheman) Collected during this survey, no previous records.

References Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. His!. Res.,

Special Issue No. I: 199-215 Zimmerman, Elwood C. 1942. Anthribidae of Guam. In: Insects of Guam - I, Bernice P. Bishop Museum Bulletin 172. Pp.

65-73.

Class: Insecta Order: Coleoptera Family: Curculionidae Weevils

CNMI - 28 species.

Menectetoms sellilosus (Boheman) Collected from Anatahan during 1992. Microcryplorhync!lus sp. Collected from Anatahan duri ng 1992 Lop!lot!letes sp. Collected from Anatahan during 1992, and during this survey. Amorphoidea lata Motschulsky Collected from Anatahan during 1992. Camplorhinus dorsalis (Boisduval) Collected, and first record for Anatahan. Daealus sp. Collected, and first record for Anatahan. Euscepes postfascialus (Fairn1aire) Collected, and first record for Anatahan. Miocalles sp. Collected, and first record for Anatahan.


Special Issue No. I: 199-215 Moore, A. and A. Tudela. 1999. http://WWW.Crees.org/plantprotection.NorthernMarianasCollege.Saipan. Updated

February 16,1999.

Class: Insecta Order: Diptera Family: Tipulidae Crane flies

CNMI - I I species.

One species collected, as yet unidentified.

References Alexander, C. P. 1972. Diptera: Tipulidae. In: Insects of Micronesia 12(8): 733-863

Order: Diptera Family: Culicidae Mosquitoes

CNMI- 14 species

One species, feeding on man during daylight in well shaded and sheltered areas, was collected but not identified. Few instances of mosquitoes attacking man were found on Anatahan.

Aedes oakley; Stone Previously collected from Anatahan in 1951. Aedes vexans lloclurnus (Theobald) ,Previously collected from Anatahan in 1992. Culex annuliroslris marianae Bohart and Ingram Previously collected from Anatahan in 1951 Culex /itoralis Bohart Previously collected from Anatahan in 1951. Culex quinquejasciatlls Say Believed to occur on "most of the island groups ".

References Bohart, Richard M. 1956. Diptera: Culicidae. In: Insects of Micronesia 12(1): 1-85. Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Order: Diptera Family: Ceratopogonidae Biting midges

CNMI - 26 species



Dasyhelea per/Ida Johannsen Previously collected from Anatahan in 1951. Forcipomyia caro/il/el/sis Tokunaga Previously collected iTom Anatahan in 1951.

References Tokunaga, Masaaki and E. K. Murachi. 1959. Diptera: Ceratopogonidae. In: Insects of Micronesia 12(3): J03-434.

Order: Diptera Family: Asilidae Robber flies

CNMI - I species

Asilidae sp. (Asilinae) Several specimens of a robber fly were collected from open areas above 200 m as!. These are the first records for the area.

References Evenhuis, N.L. (Ed). 1989, and as updated. Catalog of the Diptera of the Australasian and Oceanian Region. www.

B ishopmuseum.orglbishop/entolaocat Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Order: Diptera Family: Dolichopodidae Long-legged flies

CNMI- 8 ( 9 7) species

A species of the subfamily Chrysosomatinae was collected from Anatahan during 1992


Bishopmuseum.orglbishop/ento/aocat Bickel D. J. 1994. Diptera: Dolichopodidae Part !. Sciapodinae, Medeterinae and Sympycninae (part). In: Micronesica

27( 1/2): 361-406 Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No. I: 199-2/5

Order: Diptera Family: Syrphidae Syrphid flies

CNMI - 6(7 7) species

Ischiodon sClIle/laris (Fabricius) No previous records, collected during this suurvey, 2002,


Bishopmuseum.orglbi shop/ento/aocat Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. His!. Res.,

Special Issue No.1: 199-215 Shiraki, T. 1963. Diptera: Syrphidae. In: Insects of Micronesia 13(5): 129-188.

Order: Diptera Family: Otitidae Otitid picture-wing flies

CNMI - 4(5-67) species

Two unidentified Olilidae sp. were collected on Anatahan during 1992


Bishopmuseum.orglbishop/ento/aocat Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. His!. Res.,


Order: Diptera Family: Platystomatidae. Picture-wing flies



CNM[ - 3 species

Scholastes bimaculatus Hendel Collected at Anatahan during this survey, 2002. There were no previous records. Scholastes hirtiventris Malloch Collected at Anatahan during this survey, first record for CNMI.

References Bohart, a.E. and J Linsley Gressitt. 1951. Filth inhabiting flies of Guam. Bernice P. Bishop Museum Bulletin 204. 152

pp, 17 plates. Miyano, S. )994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Order: Diptera Family: Tephritidae. Fruit flies

CNMI - 9(10?)

Bactrocera (=Dacus) ochrosiae Malloch Previously recorded from Anatahan in 1992 and collected during this survey.

An unidentified Tephritidae was recorded from Anatahan during 1992


Bishopmuseum.orglbishop/ento/aocat Hardy, D. Elmo and W. Adachi. 1956. Diptera: Tephritidae. In: Insects of Micronesia 14(1): 1-28. Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No.1: 199-215 Moore, A. 1992. Status of fruit fly populations on Aguigan. In: The Aguigan Expedition, R.J. Craig (Ed). Proceedings

Marianas Research Symposium, Vol. 1: 48-50.

Order: Diptera Family: Lonchaeidae. Lonchaeid flies

CNMI - 4(5?) species

A Lonchaeidae species was recorded from Anatahan during [992.

References Evenhuis, N.L. (Ed). 1989, and as updated. Catalog of the Diptera of the Australasian and Oceanian Region.

www.Bishopmuseum.orglbishop/ento/aocat

Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res., Special Issue No.1: 199-215

Order: Diptera Family: Chloropidae. Chloropid flies

CNMI - 2 species

Cadrema pal/ida bilineata (de Meijere) Collected during this survey, 2002. No previous record. Rhodesielfa boharti Sabrosky? Collected during this survey, 2002. No previous record.

References Bohart, G.E. and J Linsley Gressitt. 1951. Filth inhabiting flies of Guam. Bernice P. Bishop Museum Bulletin 204. 152

pp, 17 plates. Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No. ): 199-215

Order: Diptera Family: Xenasteiidae Xenasteiid flies

CNMI -I species

Xenasteia divergens Hardy Recorded in Anatahan during 1951.



Bishopmuseum.orglbishop/ento/aocat Hardy, D. Elmo. 1980. Xenasteiidae, a new family of Schizophora (Diptera} from the Pacific and Indian Oceans. Proc.

Hawaiian Ento. Soc for 1977, XXIII(2): 205-226.

Order: Diptera Family: Muscidae. House flies, muscid flies

CNMI - J8 species

Afltherigona excisa (Thomson) Recorded from Anatahan in 1952. Dichaetomyia saperoi Bohart and Gressit Collected during this survey, 2002. Limflophora p/llmiseta Stein Recorded from Anatahan in 1951. Musca sorbefls Wiedemann Recorded from Anatahan in 1992, and found to be very conunon during this survey.

A species ofCoenossinae was recorded from Anatahan in 1992.

References Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No. I: 199-215 Snyder, Fred M. 1965. Diptera - Muscidae. In: Insects of Micronesia 13(6): 191-327.

Order: Diptera FamiJy: Calliphoridae. _Blow flies

CNMI - 9 species

Chrysomya megacephala (Fabricius). Collected during this survey - a common species attracted to meat and offal. Not recorded previously. Chrysomya rllfufacies (Macquart) Collected during this survey. Not recorded previously Phaenia(=Lucilia) cuprina (Wiedemann) Collected during this survey. Not recorded previously.

References James, Maurice T. 1962. Diptera: Calliphoridae. In: Insects of Micronesia 13(4): 108 - 127. Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,


Order: Diptera Family: Sarcophagidae. Flesh flies

CNMI - 6 (7?8?) species

Sarcophaga dux Thomson ( previously called S. misera Walker, Parasarcophaga '(}Iabi (Parker) and Parasarcophaga orchidea (Boettcher» Recorded from Anatahan during 1951. A Sarcophagidae sp. was recorded from Anatahan during 1992.

References De Souza Lopes, H. 1958. Dipterd: Sarcophagidae In: Insects of Micronesia 13(2): 15 - 49. De Souza Lopes, H. 1965. Diptera: Sarcophagidae supplement. In: Insects of Micronesia 13(5): 189-190 Evenhuis, N.L. (Ed). 1989, and as updated. Catalog of the Diptera of the Australasian and Oceanian Region. www.

Bishopmuseum.org/bishop/ento/aocat Miyano, S. 1994. Insects of the Northern Mariana Islands, Micronesia, collected during the expedition. Nat. His!. Res.,


Order: Lepidoptera Family: Hyblaeaidae. Hyblaea moths..,

CNMI - 1 species

Hyblaea pllera Cramer Collected from Anatahan 2002, no previous records.

References



None.

Order: Lepidoptera Family: Noctuidae. Noctuid moths, underwings, cutworms

CNMI - 33 species

Chasmina tibialis (F.) Snowy noctuid No specimens collected, but one specimen seen in the Village area, 2002. Eudocima (Othreis) fullonia (Clerck) Fruit-piercing moth Collected from Anatahan, 2000.

References Denton G.R. W., R. Muniappen, L. Austin, and O.H.Diambra. 1999. Fruit piercing moths of Micronesia. Technical Report #

217, University of Guam, Guam, U.S.A. 26 pp. Moore, A. and A. Tudela. 1999. http://WWW.CREES.org/plantprotection.NorthemMarianasCollege.Saipan. Updated

February 16, 1999. Muniappen R., Denton G. W.R., Marutani M., Lali T. S. and C.A. Kimmons. 1993. Fruit piercing moths in Micronesia and

their natural enemies. Micronesica, Suppl. 4: 33-39

Order: Lepidoptera Butterflies

CNMI 17 species

Superfamily: Hesperoidea Family: Hesperiidae Badamia exclamationis (F.) Brown skipper Recorded previously 1992, two specimens collected this survey. Fairly commonly encountered near Terminalia catappa trees - the food-plant.

Superfamily: Papilionoidea Family: Papilionidae Papilio polytes L. Black citrus swallowtail Previously recorded as a specimen during 2000, several seen at the Village site, but none seen at the survey area.. Apparently a relatively recent arrival in the Marianas (first record during 1950's, in Guam).

Family: Pieridae Although no Pierids have been recorded from Anatahan, there is no reason why at least one species should not be there. Eurema blanda (Boisduval), the large grass yellow, occurs on neighbouring Sarigan.

Family: Lycaenidae There was one sighting ofa lycaenid in the survey area, and a number of Zizina otis (Lesser grass blue) or Zizula hylax (Tiny grass blue), or both were seen during a very brief visit in the Village area. Lampides boeticus (Bean butterfly) should also occur in the Village area as one of its food plants, the beach morning glory (Ipomoea pescaprae), is common there, and the species was recorded from Sarigan during 200 I.

Family: Danaidae Euploea eleutho (Latreille and Godart) Marianas brown crow This is an important Marianas endemic which was found on Anatahan during 1971. None were seen during this survey. Efforts should be made to confirm its continuing existence on the island. Euploea eunice Godart Blue-branded king crow Very common in the survey area, and collected there, also collected on Anatahan during 2000.

It is possible that Danaus plexippus (L.), (Monarch) is present although there are no records. Schreiner and Nafus (1997) say that the species occurs on Pacific islands provided its host plants (Asclepias curassavica and Caltropis gigalllea) are available.

Family: Nymphalidae Hypolimnas anomala (Wallace) Guardian eggfly Not commonly seen in the survey area, and one collected. Several were seen in the village area. This is the first known record from Anatahan. Hypolimnas bolina (L.) Blue moon eggfly Recorded from Anatahan during 1992, and numbers seen in the survey area, one being collected.

Family: Satyridae Melani/is leda (L.) Evening brown Collected from Anatahan during 2000, and encountered fairly frequently at low and high altitudes. Two collected.



References Anon. 1997. Federal Register Vol. 62 No. 182, September 19,1997. Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. Hist. Res.,

Special Issue No.1: 199-215 Schreiner, 1. H. and D. M. Nafus. 1997. Butterflies of Micronesia. College of Agriculture and Life Sciences, University of

Guam. 30 pp.

Order: Hymenoptera Family: Scoliidae Scoliid wasps

CNM1 - I species

Campsomeris margine/la modesla Smith Found to be fairly common at flowers of trees such as Premna obtusifolia during this survey. No previous records.

References Nil

Order: Hymenoptera Family: Formicidae Ants

All ants sent to Dr. J.K. Wetterer (USA) for identification and confirmation. Confirmations and/or corrections will be available probably February 2003.

Subfamily: Ponerinae Anochetus graejJei Mayr Previously recorded for Anatahan in 1992. Found in litter in well shaded forested or closed

woodland areas during this survey. Odontomaclllls simillimus Smith Very common in the study area in virtually all habitats except very dry, open and eroded

areas. Previously recorded for Anatahan in 1992. Plalythyrea paral/ela (Smith) Not previously recorded for Anatahan, collected during this survey.

Subfamily: Myrmicinae Cardioeondyla emeryi Forel Collected during this survey, previously recorded for Anatahan in 1992. Cardioeondyla obscurior Wheeler Not previously recorded for Mariana Islands, collected during this survey. Cardioeondyla tjibodana Karavajev Not previously recorded for Mariana Islands, collected during this survey. MOllomoriumjlorieola (Jerdon) Not previously recorded for Anatahan, collected during this survey. ?Pheidole megacephala(Fabricius) Not previously recorded for Anatahan, collected during this survey. Pheidole umbollata Mayr Previously recorded for Anatahan in 1992. Pheidole new sp. near P. vatu Previously recorded for Anatahan in 1992. Pheidole sp Unidentified species, possibly the species listed above (nr. P. vatu), collected during this survey. Sirumigellys goddefroyi Mayr Not previously recorded for Anatahan, collected during this survey. Stmmigellys rogeri Emery Collected during this survey, previously recorded for Anatahan in 1992. Telramorium bicarillatum (Nylander) Collected during this survey,p reviously recorded for Anatahan in 1992. Tetramorium lallugillosum Mayr Collected during this survey, previously recorded for Anatahan in 1992. ?Tetramorium simi/lium (Smith) Not previously recorded for Anatahan, collected during this survey. ?Tetramorium tOllgallum Mayr Not previously recorded for Anatahan, collected during this survey.

Subfamily: Dolichoderinae fridomyrmex anceps (Roger) Not previously recorded for Anatahan, collected during this survey. Tapilloma melanoeephalum (Fabricius) Not previously recorded for Anatahan, collected during this survey.

Subfamily: Formicinae Camponatus ehlorotieus Emery Collected during this survey, previously recorded for Anatahan in 1992. Camponotus sp. - reticulatus species group Previously recorded for Anatahan in 1992. Campanotus sp. A common yellow legged, dark bodied ant. Not previously recorded for Anatahan, collected during this

survey. Paratree!lilla bourboniea (Forel) Collected during this survey, previously recorded for Anatahan in 1992. Paratreclrina IOl/gicornis Latreille Not previously recorded for Anatahan, collected during this survey. Paratreehina vaga (Forel) Not previously recorded for Anatahan, collected during this survey.

References Terayama, M., S. Miyano and T. Kurozumi. 1994. Ant fauna (Insecta: Hymenoptera: Formicidae) of the Northern Mariana

Islands, Micronesia. Pp 231-236. In A. Asakura and T. Furuki (eds.): Biological expedition to the Northern Mariana Islands, Micronesia. Natural History Research, Special Issue, Number I. Chiba, Japan.

2003 Technical Reporl #/0 83


Wetterer J.K. and R. Snelling? 1999. Ants of the Northern Mariana Islands. Draft paper, 8pages +Table 2.

Order: Hymenoptera Family: Vespidae Paper and potter wasps

CNMI- 12 species

Polistes stigma (F.) Collected from Anatahan during 1992 and found to be common during this survey. around Ficus tinctoria in "fruit", Hibiscus tiliaceus and Prelllfla obttlsifilia in flower. Two collected.

Ropalidia marginata sundaica Vecht Collected from Anatahan during 1992, and during this survey. RYflchium haemorrihoidale Not collected previously from Anatahan , but found fairly frequently during this survey. Rynchium quinquecinctulII brunneum CF.) Not collected previously from Anatahan , but found fairly frequently during this

survey.



Order: Hymenoptera Family: Sphecidae Sphecid wasps

CNMI - 10 species.

Liris sp. 1 Previously collected from Anatahan during 1992.

Dicranorhilla ritsemae Ritsema Previously collected from Anatahan during 1992. Motes sp.? Not previously collected from Anatahan, one specimen found during this survey.



Order: Hymenoptera Family: Megachilidae Leafcutting bees

Megaclzile ju/lawayi Cockerell Collected previously from Anatahan during 1992 Megachile [aticeps Smith Collected from Anatahan during 1992.

One as yet unidentified MegacJzile sp. (possibly M.jullawayi) was collected during this survey.

Pachodynerus nasidens Latr. Not previously collected from Anatahan, found commonly during this survey.

References Miyano, S. 1994. Insects of the northern Mariana Islands, Micronesia, collected during the expedition. Nat. His!. Res.,

Special Issue No. I: 199-215 Swezey, O.H. 1942. Wasps of Guam. Pp.184-187. In: Insects of Guam - I, Bernice P. Bishop Museum - Bulletin 172.



Appendix A. Fly density index.

Following a trial of the method (devised prior to seeing the survey area) to record relative fly densities using baited commercial flytraps, it was found that the method was not suitable, and that the fly-traps were a somewhat bulky item to pack and transport. The following method was therefore devised:

Aim: To establish a repeatable method of estimating relative fly densities.

Equipment: A butterfly net (ring diameter, internal, 30 cm); a piece of cardboard or a paper plate, (preferably wider than 30cm in diameter or on any side); uncooked meat or offal 1-2 days old; an aerosol insecticide; specimen containers with 70% alcohol; permanent marker pen.

Method: At a chosen sampling site the paper plate or cardboard (the base) was laid on the floor and a piece of 1-2 day old offal or meat bait at least 5 x 5 x 5 cm was placed in the centre of the base. The time taken for the first fly to arrive was noted, and after 60 seconds the butterfly net was placed as swiftly as possible over the bait and held there while insecticide was sprayed over the net. When all the flies inside were dead or incapacitated, the net was withdrawn and all the flies in the net, and on or around the bait were placed into a container with 70% alcohol. Site number and date were written on the container. At each site a new bait was used, as the previous one was contaminated with insecticide. Trapping was carried out between 09301200 h, or during mid afternoon, at six sites (see Figure I), and was repeated after I to 2 days. The site characteristics were as follows:

#1 Slight slope at base of hill above camp; open; ground cover short grass or bare ground/rock. #2 First patch of Neisosperma opposilijolia on trail up hill, directly inland, from camp. Site was

in slightly shaded area on flat ground. #3 In steep-sided gully immediately to south of site #2 - well shaded and protected from wind. #4 Next to WW2 Japanese cistern in next valley south of the camp, under shade of Hibiscus

tiliaceus, on flat ground. #5 Next to south edge of Japanese cistern in Camp site, no woody plant cover, a ground cover of

Zoysia. #6 On rocky bed of drainage line leading to "Orty's Beach", just above "beach" area. Open, no

vegetation cover.

Results: Flies arrived at the baits in between I and 9 seconds at all sites. Overall average number of flies per trap per 60 seconds was 28.7 (range 9-90), and muscid flies made up 81.4% of the total catch, followed b y calliphorid flies at 17.4 % (See Table I). The differences in numbers offlies caught at individual trap sites is a reflection of weather conditions such as wind and temperature.

Table 1: Results of f1 .. - ... .- ----_ .._.. - . -- ...... -. - -- --_._--- ... _--_ . Trap # or 1/ Date Muscidae Muscidae Calliphoridae Chloropidae Acalyptrate

(Musca) (Alliherigona) (Cadrema)

I 28' 10 2 30th II

2 28t " 19

I" 84 5 I 3 28th 10 19

I" II I 4 4 28th 10 I

lSI 31 3 4 I 5 28'h 9 2

30th 9 6 281

" 72 II 1st 12 2

Totals 276 4 60 3 I % 80.2 1.2 17.4 0.9 0.3

Discussion: Thc technique was simple and fast, and required a minimum of equipment. Between 10 to 30% of flies on the bait escaped while the net was being placed over the bait, or from under the rim of the net (because of uneven ground surface) once the net had been



placed over the bait. All the numbers of flies given in Figure 1 are undercounts, and it is estimated that none of the sites would hav yielded less than 15 flies per 60 seconds if all flies on the bait had been captured. Because of the constant presence (and high nuisance factor) of numbers of flies around people during all daylight hours and in all habitats, it was considered that the fly densities could be arbitrarily classed as very high. From the results of the applied technique, very high fly densities could therefore be defined as occurring when flies reached an uncooked, decomposing meat bait in less than ten seconds, and that no less than 15 flies would have reached the bait within 60 seconds, with an average of30 flies per minute being captured at at least six trapping sites.

The trapping method could be refined and tested further. However, even as it stands, it appears to have merit as a fast, simple, easily applied index technique.


Documents

CNMI Wildlife and Vegetatio,n Surveys and Feral Animal Control ANATAHAN 2002·2003 · 2008-11-24 · Anatahan Feral Animal Control Project 2002-2003 CNMI Wildlife and Vegetation Surveys