POPULATION ESTIMATES AND MONITORING GUIDELINES FOR ... · POPULATION ESTIMATES AND MONITORING GUIDELINES FOR ENDANGERED LAYSAN TEAL, ANAS LAYSANENSIS, AT MIDWAY ATOLL: PILOT STUDY

Technical Report HCSU-021

POPULATION ESTIMATES AND MONITORING GUIDELINES FOR ENDANGERED LAYSAN TEAL,

ANAS LAYSANENSIS, AT MIDWAY ATOLL: PILOT STUDY RESULTS 2008–2010

Michelle H. Reynolds¹, Kevin W. Brinck², and Leona Laniawe¹

¹U.S. Geological Survey, Pacific Island Ecosystems Research Center, Kilauea Field Station, Hawaii National Park, HI 96718

²Hawai‘i Cooperative Studies Unit, University of Hawai‘i at Hilo, Pacific Aquaculture and Coastal Resources Center, P.O. Box 44, Hawai‘i National Park, HI 96718

Hawai‘i Cooperative Studies UnitUniversity of Hawai‘i at Hilo

Pacific Aquaculture and Coastal Resources Center (PACRC)200 W. Kawili St.

Hilo, HI 96720(808) 933-0706

January 2011

The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the opinions or policies of the U.S. Government. Mention of trade names or commercial products does not constitute their endorsement by the U.S. Government.

Technical Report HCSU-021

POPULATION ESTIMATES AND MONITORING GUIDELINES FOR

ENDANGERED LAYSAN TEAL, ANAS LAYSANENSIS, AT MIDWAY ATOLL:

PILOT STUDY RESULTS 2008–2010

Michelle H. Reynolds¹, Kevin W. Brinck², and Leona Laniawe¹

¹U.S. Geological Survey, Pacific Island Ecosystems Research Center, Kīlauea Field Station,

Hawai'i National Park, HI 96718

² Hawai'i Cooperative Studies Unit, University of Hawai'i at Hilo, Pacific Aquaculture and

Coastal Resources Center, P.O. Box 44, Hawai‗i National Park, HI 96718

KEY WORDS

mark-resight population estimators, Laysan Duck, Midway Atoll, closed populations, Lincoln-

Petersen, detection probability, monitoring protocols

CITATION

Reynolds, M.H., Brinck, K.W., and Laniawe, L. 2011. Population estimates and monitoring

guidelines for endangered Laysan Teal, Anas Laysanensis, at Midway Atoll: Pilot study results

2008-2010. Hawai'i Cooperative Studies Unit Technical Report HCSU-021. University of

Hawai'i at Hilo. 67 pp., incl. 4 figures, 5 tables, & 3 appendices.

Hawaiì Cooperative Studies Unit

University of Hawaiì at Hilo

Pacific Aquaculture and Coastal Resources Center (PACRC)

200 W. Kawili St.

Hilo, Hawaiì 96720

808-933-0706

January 2011

ii

This product was prepared under Cooperative Agreement CA03WRAG0036 for the Pacific Island Ecosystems Research Center of the U.S. Geological Survey

Reynolds, Brinck, and Laniawe

Population monitoring of Laysan Teal at Midway Atoll

2

EXECUTIVE SUMMARY ......................................................................................................................................... 4

ACKNOWLEDGEMENTS ........................................................................................................................................ 5

INTRODUCTION ....................................................................................................................................................... 6

METHODS ................................................................................................................................................................... 7

STUDY SITE ............................................................................................................................................................... 7 FIELD METHODS ........................................................................................................................................................ 7

Population monitoring ......................................................................................................................................... 7 Capture and marking ........................................................................................................................................... 9

STATISTICAL ANALYSES ............................................................................................................................................ 9 Lincoln-Petersen style abundance estimates ........................................................................................................ 9 Simple count index of population ....................................................................................................................... 10 Power to detect a decline ................................................................................................................................... 11 Intrinsic growth rate .......................................................................................................................................... 11

RESULTS ................................................................................................................................................................... 11

Lincoln-Petersen style abundance estimates ...................................................................................................... 11 Summary of resight statistics ............................................................................................................................. 14 Simple count index of population ....................................................................................................................... 16 Power to detect a decline ................................................................................................................................... 17 Intrinsic growth rate and carrying capacity ...................................................................................................... 17

DISCUSSION ............................................................................................................................................................. 17

MANAGEMENT GOALS ............................................................................................................................................. 17 BEST ESTIMATES ...................................................................................................................................................... 18

Power analysis ................................................................................................................................................... 18 Mark-recapture likelihood models ..................................................................................................................... 19 Simple index tracking ......................................................................................................................................... 19 Hybrid methods .................................................................................................................................................. 19 Mark-recapture model data collection ............................................................................................................... 20 Lincoln-Petersen style estimates ........................................................................................................................ 20 Indices ................................................................................................................................................................ 21

MONITORING STANDARDS ....................................................................................................................................... 21 Power ................................................................................................................................................................. 21 Continuous resights ............................................................................................................................................ 21 All-wetland counts and incidental resights ........................................................................................................ 22

SUGGESTIONS FOR MONITORING .............................................................................................................................. 22

REFERENCES .......................................................................................................................................................... 24



3

APPENDIX 1. ASSUMPTIONS OF LINCOLN-PETERSEN MODELS AND THEIR VIOLATIONS IN

SURVEYS OF LAYSAN TEAL ON MIDWAY ATOLL ...................................................................................... 26

APPENDIX 2. ESTIMATING THE NUMBER OF MARKED LAYSAN TEAL AVAILABLE TO BE

COUNTED: MARKS ESTIMATED ALIVE (MEA) ............................................................................................. 28

APPENDIX 3. MIDWAY ATOLL LAYSAN TEAL RESIGHT SURVEY PROTOCOL ................................. 30

INCIDENTAL RESIGHTS ............................................................................................................................................ 30 Possible Band Numbers, Letters, and Symbols .................................................................................................. 31 Permanent Bands ............................................................................................................................................... 31 Temporary Bands ............................................................................................................................................... 32 Aluminum Only .................................................................................................................................................. 32 Band Discrepancy .............................................................................................................................................. 32

PROCEDURE FOR STANDARDIZED RESIGHT SURVEY OF ALL WETLANDS ................................................................ 34 Materials Needed: .............................................................................................................................................. 34 Procedure ......................................................................................................................................................... .34 Double Counts.................................................................................................................................................... 37 Temporary Bands ............................................................................................................................................... 38 Resight Survey Data Entry ................................................................................................................................. 38

PROCEDURE FOR PILOT ATOLL-WIDE ALL-WETLANDS COUNT .............................................................................. 43 Atoll-wide All-Wetlands Count Data Entry ........................................................................................................ 44

ATTACHMENT 1. REVISED DATA SHEETS: ............................................................................................................... 47 Resight data sheet .............................................................................................................................................. 48

ATTACHMENT 2. MIDWAY ATOLL LAYSAN TEAL SURVEY START LOCATIONS AND DIRECTION OF TRAVEL TABLES

................................................................................................................................................................................ 52

ATTACHMENT 3. MAPS ILLUSTRATING SAND ISLAND AND EASTERN ISLAND SURVEYS ............................... 53

ATTACHMENT 4. MIDWAY ATOLL LAYSAN TEAL RESIGHT REFERENCE,

MIDWAY_RESIGHT_REFERENCE_20AUG2010.XLS .............................................................................................. 56



4

EXECUTIVE SUMMARY

Accurate estimates of population size are often crucial to determining status and planning

recovery of endangered species. The ability to detect trends in survival and population size over

time enables conservation managers to make effective decisions for species and refuge

management. During 2004–2007, the translocated population of endangered Laysan Teal (Anas laysanensis; also Laysan Duck) was fitted with radio transmitters providing known (―gold

standard‖) measures of survival and reproduction. However, as the population grew, statistically

rigorous monitoring protocols were needed that were less labor intensive than radio telemetry. A

population die-off and alarmingly high number of carcasses (181) were recorded during a

botulism epizootic in August–October 2008, which further reinforced the need for effective

monitoring protocols since this endangered species is vulnerable to catastrophic population

declines. In fall 2008, we initiated a pilot study using standardized surveys with uniquely

marked birds to monitor abundance and estimate the population growth rate of the reintroduced

Laysan Teal. Since few birds carried marks (leg bands) after the 2008 botulism die-off (only

about 15% of the population), and standardized surveys were not yet implemented, the

magnitude of the die-off on the population size was unknown.

To learn more about this endangered species' status and develop monitoring protocols

useful to refuge managers and recovery planners in the U.S. Fish and Wildlife Service (USFWS),

we marked (banded) 252 new Laysan Teal for this pilot project. With skilled refuge staff and

trained volunteers, we conducted counts of marked, unmarked, and unknown birds during bi-

monthly surveys from Oct 2008 to Jan 2010. We recorded the identities of marked birds

observed, recovered carcasses, and then used the last date a bird was detected alive and the

median resight frequency to conclude if a bird was likely to be alive on a given survey date.

Using mark-resight data and individual resight frequencies, we produced a series of abundance

estimates from surveys that met accuracy criteria and approached ―closed population‖

assumptions. Since only one year of standardized, atoll-wide surveys were conducted, we

analyzed data selected from multiple surveys using Lincoln-Petersen (LP) estimates instead of

multi-year likelihood estimators. We adjusted surveys to account for unknown birds (e.g.,

swimming birds), temporary band loss, and described the frequency of double counting. Double

counting is an important consideration in the population estimate because we found a maximum

of 13% of marked birds were counted multiple times during a survey.

These survey protocols allowed us to estimate the species' post-fledging population

(combined adults and juveniles), and the methods are comparable to those used on Laysan

Island. The Laysan Teal population increased 91% from 247 (95% CI, 233–260) in 2007 to

439–508 in early 2010. There was no change from 2009 to 2010 indicating that there was no

population growth, however, our 2010 estimate should be considered preliminary since only one

month of 2010 resight data was used. We compared a series of direct counts to their

corresponding population estimates during 2008–2009 to evaluate if counts could serve as an

unbiased ―index‖ of population abundance. There was a moderate correlation between

abundance estimates and total birds counted (r2

= 0.51) during resight surveys but a low

correlation with all-wetland counts (r2

= 0.02). This indicated that using direct all-wetland

counts to predict abundance would result in confidence intervals on the order of ± 200 birds,

which is equal to 50% of the estimate. With such large confidence intervals, it would be unlikely

to detect annual changes in abundance or determine the magnitude of a catastrophic decline.



5

To improve the Laysan Teal population estimates, we recommend changes to the

monitoring protocol. Additional years of data are needed to quantify inter-annual seasonal

detection probabilities, which may allow the use of standardized direct counts as an unbiased

index of population size. Survey protocols should be enhanced through frequent resights, regular

survey intervals, and determining reliable standards to detect catastrophic declines and annual

changes in adult abundance. In late 2009 to early 2010, 68% of the population was marked with

unique color band combinations. This allowed for potentially accurate adult population

estimates and survival estimates without the need to mark new birds in 2010, 2011, and possibly

2012. However, efforts should be made to replace worn or illegible bands so birds can be

identified in future surveys. It would be valuable to develop more sophisticated population size

and survival models using Program MARK, a state-of-the-art software package which uses

likelihood models to analyze mark-recapture data. This would allow for more reliable adult

population and survival estimates to compare with the ―source‖ Laysan Teal population on

Laysan Island. These models will require additional years of resight data (> 1 year) and, in some

cases, an intensive annual effort of marking and recapture. Because data indicate standardized

all-wetland counts are a poor index of abundance, monitoring efforts could be improved by

expanding resight surveys to include all wetlands, discontinuing the all-wetland counts, and re-

allocating some of the wetland count effort to collect additional opportunistic resights.

Approximately two years of additional bimonthly surveys are needed to validate the direct count

as an appropriate index of population abundance. Additional years of individual resight data will

allow estimates of adult population size, as specified in recovery criteria, and to track species

population dynamics at Midway Atoll.

Key words: mark-resight population estimators, Laysan Duck, Midway Atoll, closed populations,

Lincoln-Petersen, detection probability, monitoring protocols

ACKNOWLEDGEMENTS

We thank E. Dominio, J. Shore, G. Shuman, M. Vekasy, and field biologists for

assistance. Matt Brown (USFWS Midway Atoll NWR Manager), Marilet Zablan (USFWS,

Assistant Field Supervisor for Endangered Species), Gordon Tribble (USGS Pacific Island

Ecosystems Research Center, Director), Holly Freifeld (USFWS, Vertebrate Recovery

Coordinator), John Klavitter (USFWS Midway Atoll NWR Biologist/Assistant Refuge

Manager), and Loyal Mehrhoff (USFWS Pacific Islands Fish and Wildlife Office Field

Supervisor) made this study possible. We thank reviewers R. Camp, J. Hatfield, and A.

McClung for comments and feedback which improved this document. Sarah Nash edited and

formatted this report. This study was funded by U.S. Fish and Wildlife Service‘s Ecological

Service‘s Pacific Islands Ecoregion, Midway Atoll NWR, and U.S. Geological Survey, Pacific

Island Ecosystems Research Center. Any use of trade, product, or firm names in this publication

does not imply endorsement by the U.S. Government.



6

INTRODUCTION

Accurate estimates of population size are often critical to understanding population

dynamics, planning conservation actions, and evaluating responses to management (e.g., Johnson

et al. 2006, King et al. 2006, McCartney et al. 2006). The Laysan Teal (Anas laysanensis) is an

endangered species restricted to approximately 10 km2 on Midway Atoll and Laysan Island, and

is one of the most difficult bird species to monitor effectively in the Papahānaumokuākea Marine

National Monument (Sincock & Kridler 1977). This difficulty is due to the Laysan Teal‘s

tendency to hide in dense vegetation, reluctance to flush, and seasonal changes in observability

(Marshall 1992, Reynolds 2002). Mark-recapture is a useful method for estimating population

abundance for animals such as the Laysan Teal, which are difficult to find, but which can be

individually marked (banded) and resighted in the future.

The U.S. Fish and Wildlife Service (USFWS) Recovery Branch requested marked

population studies in 2008 to determine how best to monitor the species‘ abundance and

population status at Midway Atoll National Wildlife Refuge (hereafter the Refuge). USFWS

Recovery Criteria specified abundance benchmarks and population trends to downlist the species

from endangered to threatened. Monitoring standards required that each population of breeding

adults must be stable or increasing, and monitoring data should demonstrate an average intrinsic

growth rate (λ) not less than 1.0 over a period of ≥ 10 consecutive years to account for

population fluctuations (U.S. Fish and Wildlife Service 2009). Two approaches to calculating λ

are to conduct counts of the population at regular intervals to estimate average annual change or

to quantify demographic parameters such as survival and fecundity. Additionally, the Recovery

Criteria specified that only adult populations be used. Fledglings and juveniles are excluded

because adding these groups tends to overestimate the number and size of breeding populations,

thereby underestimating the risk of extinction (U.S. Fish and Wildlife Service 2009).

Before the marked population studies could begin in 2008, botulism (Clostridium botulinum) type C caused an unusual mortality event for Laysan Teal at Midway (Work et al.

2010). The magnitude of the population decline and the risk to the species was unknown since

no population estimate was determined prior to the die-off. As a result, the Refuge was

interested in detecting emerging catastrophic declines plus annual increases and decreases in

abundance of Laysan Teal (J. Klavitter and M. Brown, USFWS, pers. comm.). The Refuge was

also interested in detecting population responses to management actions, but did not have

specific monitoring standards or adaptive management goals for population responses (Stankey

et al. 2005).

Standards for detecting changes in population abundance should incorporate the

magnitude of change, a time frame for detecting that change, and a measure of statistical

significance (Skalski et al. 2005). An example monitoring standard is "to be 90% likely to detect

a decline that would halve the population in five years." Detecting a change in population

abundance requires the ability to separate a real change in abundance from the uncertainty

inherent in the estimation technique. Collecting data in ways that minimize the sampling error

maximizes the ability to detect a real change in the population abundance (Zar 1996, Seavy &

Reynolds 2009).



7

The unique behavior of Laysan Teal makes surveying the population on Midway Atoll

challenging. The Laysan Teal exhibits nocturnal and crepuscular activity and secretive nesting.

Simple counts would likely underestimate abundance, because the proportion of the total

population observed during any count is unknown and would vary from survey to survey.

Although mark-resight surveys are more complex than simple counts, they are a more

appropriate technique for monitoring this species. Mark-resight methods allow for estimates of

abundance, survival, population growth rate, and other demographic parameters. Likelihood

population models require three years of mark-resight data, but we were able to use this single

year of data from the pilot study to estimate abundance and assess monitoring protocols for the

population at Midway Atoll. We also summarized USGS translocation post-release population

data (Reynolds et al. 2008) for this report to analyze the six year population trend at Midway

Atoll.

METHODS

STUDY SITE

Midway Atoll (28 12‘N and 177 22‘W) is located 1,930 km from Honolulu, Hawaii,

comprised of three islands (Sand, 452 ha; Eastern, 136 ha; and Spit, 6 ha), and protected as a

National Wildlife Refuge (NWR) within the Papahānaumokuākea Marine National Monument.

There are 14 small diverse wetlands on the Refuge totaling about two hectares. In addition to

supporting the re-introduced population of Laysan Teal, Midway Atoll provides breeding habitat

for approximately three million Pacific seabirds.

FIELD METHODS

Population monitoring From November 2008–November 2009, monitoring consisted of weekly atoll-wide

counts of Laysan Teal using two survey methodologies. The first survey methodology was used

to estimate population abundances (resight survey), and the second provided an index count

(called all-wetland counts; Fig. 1a, b) to be validated using marked population methods. Skilled

observers searched wetlands using 10x binoculars or a spotting scope (40-60x) to detect or read

coded leg-bands of all Laysan Teal visible between, on, and around the wetlands (Appendix 3).

Surveys were initiated approximately two hours before sunset or at sunrise—times when the

greatest number of Laysan Teal are concentrated at interior wetlands. The starting location and

direction of surveys were randomly varied to reduce spatial-temporal bias. This protocol was

similar to methods used on Laysan Island as described in Marshall (1992) and Reynolds and

Citta (2007).



8

a)

b)

Figure 1a, b. Survey routes on Sand (a) and Eastern (b) islands. (Full-page versions of these maps are provided in

the Appendix, Attachment 3.)

Sunset Rolando

Monument



9

Prior to this study, the U.S. Geological Survey also assisted the Refuge with post-release

monitoring of translocated birds using radio telemetry, and conducted opportunistic pilot surveys

of Sand Island‘s wetlands in January, February, May, and July–September in 2007. A morning

or evening resight survey was conducted on both Sand and Eastern islands in October and

November in 2007. Simultaneous atoll-wide resight surveys were conducted 1–2 times per

month in January and March 2008. We used these early data and data from this pilot study to

construct a time series of abundance estimates and evaluate the sampling to improve survey

methods. For detailed field methodology refer to the survey protocols (Appendix 3). Survey

protocols were updated based on pilot study results, and may be updated in the future based on

future results.

Capture and marking Laysan Teal were trapped at wetlands, in a variety of terrestrial habitats, and on roads

around Midway Atoll in November 2008, March–May 2009, and September–October 2009.

Most birds were caught at night using a flexible handheld net and headlamps or at crepuscular

periods using noose carpets. A numbered aluminum band was placed on one leg and a plastic

color band with alpha or numeric codes on the other leg of each bird. Nasal or wing marks,

commonly used in waterfowl marking, were inappropriate for this endangered species due to a

high risk of entanglement. All birds were released near their capture site. Prior to this study, all

individuals were marked through 2006, 56 juveniles were marked with permanent bands in 2007,

and 33 birds (6 juveniles, 27 adults) had temporary bands replaced in 2008–2009. The marked

population ranged from 20 birds in 2004 to 300–345 birds by 2009.

STATISTICAL ANALYSES

Lincoln-Petersen style abundance estimates We used a Lincoln-Petersen (LP) framework (Seber 1982) in which a known number of

birds are marked and released to mix freely in the population, so that during a later survey the

known number of marked birds and the ratio of marked to unmarked birds can be used to

estimate the total population abundance. The LP estimators tend to overestimate population

sizes, especially if few birds are seen or less than 20% of the population is marked (Robson &

Regier 1964; as of January 2010, 50–60% of the Laysan Teal at Midway were marked). We

chose this method because there was only one year of pilot study data, the estimator was simple

to calculate, and the same method had also been used on Laysan Island, permitting comparisons

of population abundance (Marshall 1992, Reynolds & Citta 2007, USFWS 2009).

The accuracy and precision of LP estimates depends upon six assumptions about the

animal population and the efficiency of data collection (Skalski et al. 2005):

1. The number of marks in the population is known

2. The population is geographically and demographically closed (no births, deaths,

immigration or emigration)

3. All animals have the same probability of being caught or sighted

4. Marking does not affect future detection

5. Animals do not lose their marks

6. All marks are correctly reported

It is important to note that the methods used to estimate the Laysan Teal population from

this pilot survey violate these assumptions in important ways, but we have devised methods to



10

compensate for violations. Appendices 1 and 2 provide details. Where there was important

uncertainty (i.e., assumption #1), we presented a range of estimates that can be refined with

additional resight data.

Chapman‘s (1951) bias-corrected hypergeometric modification to the LP estimator is:

1)1)(1(ˆ

t

ttt m

nMN

where tN̂ is the population estimate, Mt is the total marked population, nt is the number of

animals counted, and mt is the number of marked animals counted (i.e., resighted), all at a given

time t. A nearly unbiased variance estimator (Seber 1982) is

)2()1(

))()(1)(1()ˆ(

2

tt

ttttttt mm

MnmMnMNVar .

To explore the quality of the data, we determined the total count and marked population

needed for a LP point estimate with 10% or 25% accuracy (assuming a population of 200–500

birds, the approximate range of our estimates)

using standards by Robson and Regier (1964;

Figure 2). Again, mt is the number of birds

resighted during a survey, and Mt is the number of

birds marked in the population (marked estimated

alive, MEA) at the time of the survey, as

estimated from banding and resight data

(Appendix 2). All calculations were carried out

in a Microsoft Excel workbook (Microsoft Corp.

2003; available upon request).

Simple count index of population We assessed the validity of using the total

number of birds observed in the resight surveys

and the all-wetland counts as an index of a more

detailed LP-style population estimate (Appendix

2) with a simple regression model. An index is a

population assessment technique that uses a

relative difference in abundance (e.g., from an

incomplete direct count) to describe changes in a

population (Lancia et al. 1996). Ideally, an index

is a constant ratio that can be calibrated with the

population estimate. Indices are different from

abundance estimators due to their inability to be

converted to absolute animal abundance, typically

because the necessary auxiliary data to make the

conversion are not collected (Skalski et al. 2005).

We describe the uncertainty in using a count

index to predict total Laysan Teal abundance in

more detail later in this paper. Because the

individual surveys varied in meeting accuracy and

Figure 2. Numbers of birds that need to be

marked and examined for marks on

subsequent surveys to estimate population

size within 10% (bold numbers) or 25%

(regular font) accuracy. This figure is

modified from Robson and Regier (1964).

The bold line represents combinations of

numbers of birds marked and numbers of

birds examined for marks that produce

population estimates accurate within 10% of

a population of 500 (line used in this study).

For example, if the population size (N) is

thought to be about 500, then you would

need to mark ~400 birds if your surveys

detected only 50–60 total birds. If your

surveys typically detected ~100 birds, 300

would need to be marked for estimates to be

accurate within 10%.

Number marked Mt



11

precision standards, we used only the higher-quality LP estimates, where accuracy guidelines

(total marks plus birds observed > N̂ ) and precision guidelines (coefficient of variation of N̂ <

0.05) were met.

Using the estimated number of marked birds in the population and the actual number of

marks counted during resight surveys, we estimated the monthly probability of detecting an

individual and derived Horvitz-Thompson estimators based on detection probability (Thompson

1992), both for each month and also for morning vs. evening surveys. We compared those

estimators to LP-style abundance estimators using linear regression to test the strength of the

relationship between the two methods.

Power to detect a decline There are two major sources of variation in the annual estimate of the Laysan Teal on

Midway: actual fluctuation in bird abundance (process variation) and sampling error resulting

from random noise and inaccuracies in field data collection. In order to detect changes in

abundance over time, we would like to calculate a threshold outside which we are confident that

an observed trajectory represents a real trend in the population, beyond the fluctuations

introduced by sampling error.

Intrinsic growth rate We used abundance estimates from post-release radio telemetry and LP estimates from

resight surveys for a non-linear regression using program R (R Development Team 2009) to

estimate the intrinsic growth rate (λ) and model the potential carrying capacity on Midway Atoll.

Using a logistic growth model of the form

where N is the abundance and dtdN is the rate of population growth over time (the derivative of

abundance with respect to time), r is the natural logarithm of the growth rate (λ), adjusted for

density dependent growth, and K is the carrying capacity, or maximum population the

environment can support. Carrying capacity is likely confounded by events such as the botulism

epizootics in 2008 and 2009, and thus carrying capacity estimates are considered preliminary.

RESULTS

Lincoln-Petersen style abundance estimates There were 38 atoll-wide resight surveys between October 2007 and January 2010.

Seventeen of these resight surveys were considered high quality surveys and had enough total

resight observations to produce potentially precise (small uncertainty) and accurate (little or no

bias) estimates (Table 1), including one (23 October 2007) with enough information on known

age birds to produce a separate estimate for the number of juveniles. However, three of these

high-quality surveys (5, 12, and 19 November 2008) were made during the botulism outbreak,

when the mortality rate was high, thus violating the closed population assumption of the

Chapman estimator (i.e., marked birds classified as available to be counted may have been dead)

leading to biased estimates. All surveys in December 2009 did not use the standardized survey

NKNrdt

dN 1



12

protocols, although the observer collected useful resight data for detecting individuals and

calculating median resight intervals (observer on different islands on different days). Estimates

that come closest to meeting LP model assumptions, meet accuracy and precision standards, and

were collected by experienced observers using standardized protocol are highlighted and

considered ―best estimates.‖

Table 1. Mark-resight Lincoln-Petersen estimates of Laysan Teal abundance, meeting the Robson and Regier (1964) standards for accuracy and precision and a

coefficient of variation < 0.05. We also show the results of the 23 October 2007 survey, the only data for 2007, and inferred pre-breeding abundance for

2008. Birds whose banding status could not be determined (e.g., swimming in turbid water) were assumed to be marked or unmarked in the same

proportions as those whose status could be determined. Best estimates—those which come closest to meeting LP model assumptions, while also meeting

accuracy and precision standards, and were collected by experienced observers using standardized protocol (two observers survey Sand and Eastern

simultaneously) —are in bold. The 2010 estimates are considered ―best preliminary estimates.‖ Surveys marked with * were conducted using different

methods, influencing the MEA (estimated marks alive). The 90% quantile vs. median detection intervals increase the time allowed to elapse before an

individual is considered missing (dead) and not available to be surveyed. Additional resights will allow for an improved MEA and LP estimate at the

median detection interval (see Appendix 2).

Resight survey results Using the median detection interval Using the 90th

quantile detection interval

Survey date

Resighted

during

survey

Marked

resighted

Marked

birds

LP

estimate 95% CI

c.v.

estimate

Marked

birds

LP

estimate 95% CI c.v. estimate

23-Oct-07 139 60.0 92 212.3 (189-236) 0.06 94 217 (193-241) 0.06

5-Nov-08 453 109.6 135 557.3 (518-596) 0.04 138 570 (528-611) 0.04

12-Nov-08 350 107.2 149 485.8 (446-526) 0.04 152 495 (454-537) 0.04

19-Nov-08 357 142.5 156 390.5 (376-405) 0.02 159 398 (382-414) 0.02

19-Nov-08 332 121.0 156 427.4 (399-456) 0.03 159 436 (406-465) 0.04

26-Nov-08 285 130.3 163 356.2 (336-376) 0.03 166 363 (342-384) 0.03

17-Dec-08 370 143.9 200 513.7 (479-548) 0.03 205 526 (490-563) 0.04

24-Dec-08 334 123.8 197 530.4 (486-575) 0.04 205 552 (504-600) 0.04

31-Dec-08 369 150.2 196 481.0 (453-509) 0.03 205 503 (471-535) 0.03

5-Oct-09 211 122.2 282 486.1 (444-528) 0.04 293 505 (461-549) 0.04

6-Nov-09 319 148.5 276 592.0 (545-639) 0.04 294 631 (579-682) 0.04

17-Nov-09 274 139.3 269 528.4 (486-571) 0.04 290 570 (522-617) 0.04

1/2-Dec-09* 353 185.0 268 (483-539) 0.03 287 (515-579) 0.03

15/16-Dec-09* 276 160.7 310 (495-568) 0.04 345 (549-635) 0.04

29/30-Dec-09* 278 165.4 272 (429-484) 0.03 340 (531-610) 0.04

8-Jan-10 207 123.2 215 360.6 (334-387) 0.04 322 540 (493-587) 0.04

11-Jan-10 210 136.9 175 268.4 (256-281) 0.02 309 473 (439-508) 0.04 2010 estimates are preliminary.

Reynolds, Brinck, and Laniawe 14 Population monitoring of Laysan Teal at Midway Atoll

Table 1 presents two sets of abundance estimates. The more conservative MEA uses the

median interval between detections when estimating the number of marks available to be

counted. There are also preliminary estimates using the 90th

percentile of the distribution, which

were used for recent survey dates (late 2009 and early 2010), where there has been little time

after banding birds and less post-survey effort to establish which birds were alive on the survey

date. Early (preliminary) estimates can be enhanced with future resights which improve

estimates of MEA. In addition, on the 23 October 2007 survey all birds were of known age,

therefore we were able to produce separate estimates of adults and juveniles: exactly 92 adults

and 155 (95% confidence interval of 141 to 168) juveniles.

Table 2 presents abundance

estimates from the translocation of

Laysan Teal to Midway through

January 2010. Total numbers should

be considered "pre-breeding" since a

combination of juveniles and adults

were used to estimate the 2010 adult

population. This number

overestimates the number of potential

breeders in a year since juvenile

mortality is higher than adult mortality

(Reynolds & Work 2005, Reynolds &

Citta 2007, Reynolds et al. 2008,

Work et al. 2010). Insufficient

resights were made in late 2009, and

only two surveys were conducted in

2010. Missing and variable survey

effort also creates a gap in the time

series for 2008, however, we could

provide the estimate for 2007, which

was based upon radio-tracking data of

adults. We lack the resight data in

2008 to make adult (breeding

population) estimates for 2009 using

LP methods. However, with

additional standardized 2010/2011 resight surveys, pre-breeding adult population size (during

January–March) can be estimated using these methods.

Summary of resight statistics Resights of Laysan Teal were collected using three basic methods: resight surveys

(standardized resight surveys of both islands), opportunistic resights during all-wetland counts,

and incidental or opportunistic resights. All resights plus re-captures while banding were utilized

to calculate the distribution of detection intervals used to estimate the number of MEA birds in

the population available to be counted as described in Appendix 2.

Table 3 details the distribution of all resight observations where location and method

were recorded. Half of pooled resight observations were from resight surveys, and two-thirds

Table 2. Total abundance (all post-fledging) and pre-breeding

estimates for Laysan Teal on Midway Atoll. Before 2008

all adult birds were given radio transmitters, so the

numbers are considered exact. From 2008–2010 the

abundance estimators are Chapman mark-recapture

estimators as described in this document. The 2009 and

2010 estimates are of post-fledging individuals in

December or January, before juvenile mortality is

considered to be over, so are over-estimates of the

breeding adult population. Finally, the 2010 estimate is

based on the number of marks estimated from the 90%

quantile of detection intervals to compensate for the

absence of post-estimate resight effort to establish which

birds were known to be alive at the time of the survey.

Year

Total adults and

juveniles 95% CI Adults

2004 20 4 2005 51 24 2006 104 51 2007 247 (233-260) 92 2008 -- (189-236) 212 2009 481 (453-509) -- 2010 473 (439-508) --


were collected on Sand (vs. Eastern) Island. Incidental sightings were important on Sand Island

and made up nearly one-third of records (Table 3).

Most resighted individuals (66%) were seen on

both islands, while 51 individuals (12%) were seen

only on Eastern Island (Table 4). Incidental resights

contributed to many detections—17% of birds were

never seen during the standardized resight surveys

(which did not survey all wetlands during the pilot

study). Just two percent of color-banded birds were

missed during surveys and could be recorded only

incidentally (Table 5).

Because birds can fly from wetland to wetland

much faster than observers can walk between them, it is possible for a Laysan Teal to be

observed more than once during a resight survey. Identifying individuals via color bands

enabled us to estimate the frequency with which this occurs. On 61% (26 of 38) of surveys

individually marked birds were seen at more than one site, including 16 (42%) surveys where

individuals were seen on multiple islands. Among 38 surveys on Eastern Island, anywhere from

zero to seven marked birds were seen at multiple sites. Marked birds double-counted on

multiple islands and multiple wetland

sites ranged in number from zero to

nine. A maximum of 13% of marked

birds were counted multiple times

(range 0–13%, mean 2.4%, s.d. 2.9%).

Laysan Teal also do not

remain stationary while a wetland is

being surveyed. On 35 of 38 (92%)

surveys birds flew into a wetland

during the survey, with a mean of 13.0

and a maximum of 53 fly-ins during a

survey. Similarly, on 32 of 38 (84%)

surveys birds left the wetland while

being counted, with a mean of 5.0 fly-

Table 3. Distribution of Laysan Teal resights from October 2008 through 11 January 2009, by

island and type of resight effort. The effort proportions differed for each island (Chi-

squared test 2

2 = 586.7, p < 0.0001) likely a result of few incidental resights on Eastern

Island.

Island All-Wetland

Counts Incidental Resight Survey Total

Sand 402 (7%) 1870 (32%) 1577 (27%) 3849 (67%) Eastern 278 (5%) 304 (5%) 1330 (23%) 1912 (33%)

Total 680 (12%) 2174 (38%) 2907 (50%) 5761 (100%)

Table 4. Location of resights of individual

Laysan Teal from October 2008

through 11 January 2010. Two-thirds

of birds were observed on both

islands.

# Laysan Teal

Sand Only 98 (23%) East Only 51 (12%)

Both 286 (66%)

Table 5. Number of individual Laysan Teal re-sighted from

Oct 2008 through Jan 11, 2010, categorized by type of

effort for that individual. All birds were re-sighted at least

once after banding. Some individuals were observed using

two or more methods and a duck that was only seen on an

all-wetland count was also never seen on both re-sight

surveys and incidental re-sights, so percentages do not

total to 100%.

# of individual ducks Only seen Never seen

All wetland counts 5 (1%) 133 (31%) Re-sight surveys 29 (7%) 72 (17%)

Incidental re-sights 9 (2%) 23 (5%)


0

0.1

0.2

0.3

0.4

0.5

0.6

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Fra

ctio

n o

f m

ark

ed

du

cks d

ete

cte

d

Figure 3. Monthly surveys in 2009 reveal the pooled proportion of marked

birds (post-fledglings and adults) detected.

outs per wetland and a maximum of 19 fly-outs per wetland during a single survey.

Simple count index of population There was moderate correlation (r

2 = 0.51) between the number of Laysan Teal counted

during a resight survey and the abundance estimate produced by the survey, and almost no

correlation (r2 = 0.02) between all-wetland counts and the abundance estimate from resight

surveys. With more high-quality resight surveys and accompanying population estimates, the

regression relationship between counts could be refined and, if validated, simple counts of

birds—without distinction between marked and unmarked birds—would be a faster and easier

method to produce an index of the population size. In its current form, however, the regression

model produces excessively wide confidence intervals (CI), e.g., a count of 200 birds predicts a

population estimate of between 138 and 560 birds. If future quality surveys continue to show the

same variability, it will require another 47 resight surveys (or two years at bi-monthly survey

frequency) to calculate abundance estimates (for a total of 65) reducing the 95% CI to a width of

± 100 Laysan Teal.

Figure 3 shows the monthly proportion of marked birds detected during 2009, with most

individuals having fewer total detections from April–June during the birds‘ breeding season.

Most individuals were detected (with some exceptions) from November–February during post-

fledging flocking and adult

courtship flocking. If the

probability of detection is

known then abundance can be

estimated any time of year by

dividing the number detected

by the probability of detections

(Thompson 1992). However,

our results show a weak

relationship between the

Chapman abundance estimates

and estimates based on the

monthly detection probabilities

(linear regression, p = 0.36, r2

= 0.02), meaning Thompson-

style estimators cannot be used

to estimate abundance.

There was only a

single, atoll-wide evening survey meeting both accuracy and precision criteria, so the efficacy of

Horvitz-Thompson estimators based on morning vs. evening detection probabilities cannot be

tested. However across all atoll-wide surveys the detected proportion of marked birds was

different between morning and evening (Chi-squared test, Yates corrected = 31.8, p < 0.0001),

and examination of the residuals indicates that this is due to lower detection rates in the evening

(28.7% vs. 38.6%).


0

100

200

300

400

500

600

700

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

La

ysa

n T

ea

l a

bu

nd

an

ce

on

Mid

wa

y .

Estimate Logistic growth model

Figure 4. Logistic growth curve fit to the Laysan Teal post-fledging

abundance estimates reported in Table 2. The growth curve shown is

consistent with an intrinsic maximum reproduction rate (λ) of 2.1 and

a carrying capacity (K) of 603 birds. The 2007 population estimate

includes adult birds from radio-tracking data in late 2007 and marked

post-fledglings. The 2008 estimate is an estimate of adults only as

there were no resight surveys conducted to estimate the post-breeding

population size in 2008.

Power to detect a decline All power calculations are based on statistical tests using a type I error rate of 0.10.

Based on the abundance estimates for 2009 and 2010, a one-sample t-test detecting a

catastrophic halving of the Laysan Teal population in a single year would be significant 97.5% of

the time. The same one-year change would be 90% likely to detect a 15% decrease. Detecting a

decline over a larger time span is more difficult. Based on the abundance estimates from 2007–

2010, a linear regression estimator would have 74% power to detect a 50% decline over five

years.

Intrinsic growth rate and carrying capacity

Using previously published and current abundance estimates (Tables 1, 2) we used non-

linear regression to fit a logistic growth curve. Figure 4 shows the growth curve fit to the

available abundance estimates. The parameters of the model are consistent with a maximum

intrinsic growth rate of λ = 2.1 [P(λ ≤ 1 = 0.0534)] up to a carrying capacity of 603 birds [P(K ≤

0) = 0.0278, 95% CI = 161–1045]. The model predicts that the maximum growth rate to date

occurred during 2008, but in the past few years the population may have begun to approach the

carrying capacity. Additional resight surveys in 2010 are needed to provide supporting data for

2010 pre-breeding preliminary total estimates. If the population has not reached carrying

capacity and is released from botulism mortality, then the growth rate and carrying capacity

estimates will change with further data.

DISCUSSION

MANAGEMENT GOALS

Refuge managers at

Midway have two non-

overlapping monitoring goals

for the Laysan Teal population.

The first goal is to track the

status and annual trend of the

abundance of Laysan Teal. The

second goal is to detect and

determine the magnitude of any

potentially catastrophic die-offs

such as the Echinuria and

botulism epizootics in 1993 and

2008, respectively, on Laysan

and Midway islands (Work et

al. 2004, 2010).

The first goal requires

effort to enumerate the Laysan

Teal population. However, as

described previously, a simple

count will not necessarily


provide a valid index of the total population of Laysan Teal and has low precision that will vary

throughout the year (Figure 3). Mark-recapture methods use the observed proportions of marked

and unmarked birds to infer the total population size and provide an estimate of true abundance.

Historically, the Laysan Teal has been susceptible to catastrophic declines due to disease

or introduced mammals. The dramatic fluctuations seen in Laysan Island population estimates

(Seavy et al. 2009, USFWS 2009) demonstrate the importance of the Refuge‘s second goal of

continuous monitoring throughout the year to detect the effect of disasters such as botulism

outbreaks or environmental catastrophes such as tsunamis and hurricanes.

In addition to the two Midway Atoll refuge goals, the recovery criteria set forth by the

U.S. Fish and Wildlife Service must be met both to downlist the species designation of Laysan

Teal from endangered to threatened and also to provide benchmarks of recovery and species

status (USFWS 2009). The Revised Laysan Duck Recovery Plan monitoring criteria require

that:

The Laysan Island population remains at roughly 500 birds over a period of at least 15

consecutive years.

At least 1,800 potentially breeding birds exist in a stable or increasing population (for ≥

10 consecutive years) on a combination of predator-free Northwestern Hawaiian Islands

(including Laysan Island and Midway Atoll) and at least one predator-controlled site in

the Main Hawaiian Islands.

In order to remove the species from the endangered species list entirely (i.e., to delist it), Laysan

Teal must be stable or increasing (as monitored over 15 consecutive years) in five or more

populations consisting of at least 3,000 potentially breeding adults (≥ 500 per population). These

populations, existing on a combination of predator-free Northwestern Hawaiian Islands

(including Laysan Island and Midway Atoll) and at least two predator-controlled sites in the

Main Hawaiian Islands, should be self-sustaining, requiring only minimal monitoring and

management of threats (e.g., epizootics, catastrophic declines; USFWS 2009). A new population

viability analysis (PVA) may be needed to evaluate if population sizes listed in the recovery plan

are considered viable in the face of climate change (Reynolds et al., in review).

BEST ESTIMATES

Power analysis A true power analysis requires multiple years of abundance estimates before it is possible

to detect a trend. A minimum time period is at least five years, but that will depend upon the

variation in the estimates (Elzinga et al. 1998). Beyond normal population fluctuations and

sampling variation, the amount of ‗noise‘ will depend upon whether or not the population has

stabilized or is still increasing to carrying capacity and whether or not botulism epizootics

continue to make an impact. With additional yearly estimates to establish population trends, it

will be possible to distinguish between the biological variation and sampling variability and to

calculate the survey precision (White & Burnham 1999). With the current dataset we cannot

distinguish between natural variation and sampling error, so our power calculations are limited to

simple (non-auto correlated) regression (known to under-estimate power; Elzinga et al. 1998) or

a single-point change via t-test.


Mark-recapture likelihood models Abundance and survival of Laysan Teal on Midway Atoll could be estimated using an

open mark-recapture model such as the Cormack-Jolly-Seber (CJS; Seber 1965, 1982; White &

Burnham 1999; Williams et al. 2002). Sampling designs estimate births and deaths and require a

minimum of three intensive capture occasions, but there are drawbacks to this model. CJS study

designs would require skilled endangered species biologists to capture and handle the Laysan

Teal, and it might be undesirable to disturb an endangered species to the extent of attempting to

capture most individuals every year. Finally, annual capture surveys are unable to detect a

catastrophic event as it happens; a massive die-off due to a hurricane or epizootic could only be

detected after the event when the resulting abundance estimate is calculated.

Simple index tracking The minimal effort to detect a catastrophic event such as an epizootic will require regular

efforts to survey the Laysan Teal. Such events can happen quickly, therefore some measure of

abundance is needed a short time before and after the event to pinpoint exact time and magnitude

of the population decline. A count of all wetland habitats could serve as a simple index of

abundance. We know that Laysan Teal behavior dictates that the simple count will naturally

vary throughout the day and year (Figure 4) but a sharp decline of a standardized index count

may indicate a potential die-off and trigger a more intensive survey effort (see Seavy et al. 2009).

Without calibration of an index to an abundance estimator (LP or CJS mark-resight or -

recapture) and a longer time series, a simple index would be unable to provide an estimate of

Laysan Teal abundance or other needed demographic parameters identified in the revised

Recovery Plan (USFWS 2009).

Hybrid methods By using color leg bands to identify individual Laysan Teal, the detection history of each

bird can be used to estimate abundance and other demographic parameters such as intrinsic

growth rate (λ), survival, and detection probability. A CJS model (Seber 1982) uses banding and

re-capture information to estimate abundance, and a multi-state Barker model (Barker 1997) that

includes capture, recapture, resight, and carcass recovery information about marked birds is most

likely to be useful. Simpler models require fewer years of observations (with a minimum of

three years for the CJS model and four years to estimate annual survival, re-capture, and resight

statistics with a Barker model).

Effective population growth rate (λ) can be estimated with a likelihood model. Data must

be collected for at least another three years in order to fit such a model, with at least 20% marked

birds in the population and intensive resight effort—either intensively over a few months every

year or gradually throughout the year (the current protocol). Surveying throughout the year

allows managers to pinpoint population change in time. This is especially useful for tracking

breeding phenology, detecting Refuge-wide disease events, and observing and collecting

carcasses. However, if resight protocols were changed and resight data collection were

concentrated to one time period, November through February would be the best months for

surveying since 2009 surveys had higher detection probabilities and low mortality during these

months (Figure 4).


Mark-recapture model data collection Mark-recapture models are dependent upon exact identification of individuals, so

increasing the accuracy and rate of detection would improve the current protocol. One way to do

this is by using experienced wildlife biologist or bird watchers familiar with the band colors and

codes used at Midway Atoll and capable of reading bands quickly and accurately, and recording

data efficiently. Strategies to improve detection will improve data quality. For example,

perching logs could be placed near or over wetlands encouraging birds to remain out of the water

where their bands can be more easily seen. Observation towers and blinds built near wetlands

provide surveyors with better angles to observe birds without spooking them into the water

where their leg bands become concealed. Improving detection probabilities in these ways may

reduce the incidental resighting effort required, improve estimates, and reduce the number of

surveys needed to validate an index count.

Lincoln-Petersen style estimates Until enough data have been collected to fit mark-recapture likelihood models,

abundance can be estimated via simpler Lincoln-Petersen style mark-recapture estimates, as we

have done in this report. In order for these estimates to be accurate and useful, standardized

resight surveys must detect a large number of birds with known marked or unmarked status—the

greater the number of individuals re-surveyed (resighted), the greater the accuracy and precision

of the abundance estimates (Chapman 1951).

Importantly, LP estimates assume that the number of marked animals is known. As

detailed below (Appendix 1), this assumption was violated, and we have compensated by

estimating the number of marked birds estimated alive (MEA) as described in Appendix 2. An

accurate estimate of the number of marked birds depends upon frequent and regular resights of

marked birds both to confirm that they are still alive and to establish the distribution between

detections for individual birds. The effect of missing or infrequent resight data on population

size estimates can be seen in Table 1 (p. 16) for the 11 Jan 2010 estimate. As explained in

Appendix 2, using the conservative (50th

percentile: 175) versus the less-conservative (90th

percentile: 268) estimate of the number of marked birds changes their corresponding abundance

estimates from 309 to 473. It is clear that less stringent standards for resight data can have an

inflationary effect on population size estimates.

LP-style estimates require that all individuals have an equal probability of detection. For

birds on Midway this is unlikely to be a problem if marked birds are as likely to be detected as

unmarked birds. However, if marked birds, because they are individually distinguishable, are

less likely to be double-counted than unmarked birds, the number of unmarked birds will be

inflated, resulting in a positive bias to the estimate. Observers must strive to count marked and

unmarked birds as carefully as possible, without double-counting either category. Since 17% of

marked individuals were missed during the standardized resight survey (Table 5), an individual‘s

detection might be improved during resight surveys if all wetlands were included in the survey

route, or if incidental resighting effort continued.

In some resight surveys, as many as 20% of birds counted were recorded as having an

unknown banding status because they were swimming with their legs concealed or flew out of

the wetland before they were identified. In this analysis, rather than ignore these birds, we

assume that unknown status birds occur in the same proportions as the known status birds (i.e.,


ratio of marked/unmarked individuals). We recommend that at least 90% of observed birds have

known status, so observers must do their best to determine banding status for all birds.

Experienced observers and the physical interventions described previously (duck perches or

observation towers) will aid in determining marked/unmarked status. Observer experience

improves accuracy by minimizing disturbance, since loafing birds often become swimming birds

or ―fly-outs‖ when disturbed.

A large fraction of the Laysan Teal population (40%) must remain marked (Robson &

Regier 1964) to produce accurate estimates with a Lincoln-Petersen style estimator. At least

twenty percent of the population should be marked for accurate and precise maximum likelihood

(i.e., CJS) models (Robson & Regier 1964). In 2009, 68% of the post-fledging population was

marked, meaning the population could increase by 50% before diluting the current pool of

marked birds. It is not necessary to band new birds in 2010, but it is possible that banding may

be needed by fall 2011 or 2012, to preserve the proportion of marked birds if birth and death

rates are high. However, to maintain the current marked population, birds with worn bands

should have their bands replaced as soon as possible after discovery (with the exception of hens

with ducklings whose bands should be replaced only after fledging young). If a population index

can be validated using data from the marked birds for several years, population trends may be

detected using simple counts, and marking new birds would not be needed.

Indices A simple count of Laysan Teal does not currently serve as a useful index of abundance.

If the current relationship between index and abundance remains the same, then collecting more

data will make the linear regression model more useful. The continuously refined regression

model will allow input of a simple count and produce an interval estimate for the corresponding

LP abundance estimate (the Excel spreadsheet described in the Statistical Analyses section is

available upon request).

MONITORING STANDARDS

Power As mentioned above, the current LP population estimates are highly likely to detect a

catastrophic reduction of half the population within a single year. A more subtle 50% decline

over ten years would be more difficult to discern, detected only 75% of the time.

As monitoring continues and annual population estimates are generated, more

sophisticated power models will account for the effects of temporal correlation, population

growth, and epizootics. These more sophisticated models will allow a calculation of minimum

effort and survey intervals to meet specific monitoring goals. Humbert et al. (2009) recommend

at least a ten-year span of data to properly fit these models.

Continuous resights The estimate of the number of marked birds available using LP methods depends upon

regular resight surveys with a high proportion of marked birds detected and identified. Even

though detectability varies throughout the year, standardized resight effort is needed to establish

which individuals are available to be counted for LP estimates of the population. Regular resight


surveys will also allow regular abundance estimates to be calculated in order to detect and

estimate the magnitude of catastrophic events in the population.

All-wetland counts and incidental resights Since the total birds counted during resight surveys is correlated with population

estimates, it follows that all-wetland counts may be discontinued or ‗merged‘ with resight

surveys, if it is possible to use resight surveys as a population index. If so, it is also important to

maintain the same number of incidental resights or include all of the wetlands in the resight

survey. If this is not possible, then at least some of the discontinued all-wetland survey effort

should be redirected to a less intensive incidental resighting effort, especially if the large

proportion (38%) of incidental resights was due to the intensive resight effort by dedicated staff

during the pilot study.

SUGGESTIONS FOR MONITORING

Continue bi-monthly standardized atoll-wide resight surveys.

Conduct standardized atoll-wide resight surveys in the morning rather than evening.

Resight surveys should be conducted by dedicated staff or trained volunteers skilled at

bird-watching and at data collection. Observers must be familiar with these monitoring

protocols, familiar with the color and banding codes used at Midway Atoll, and will

require several weeks of practice surveys with bird ID confirmation (ID data validation).

Training should be provided by a biologist experienced with endangered Laysan Teal

monitoring at Midway. Data collected by inexperienced observers should not be used to

estimate population sizes since there is a high risk of further violating model assumptions

(see Appendix 1 for violations to assumptions 3, 5, and 6). Misidentified birds also

produce inaccurate MEA estimates and unreliable population estimates.

Worn bands should be replaced and illegible bands should be identified immediately so

that MEA estimates are not biased.

If the population continues to grow or a catastrophic event kills a large number of marked

birds, band additional birds as needed to maintain a 40% marked rate in the population

and maintain the accuracy of LP-style estimates. As few as 20% marked birds is useful

for validating a population index and for precise estimates in some (open) likelihood

models. Most CJS designs require birds to be marked every year. Less than 20% marked

is not useful for accurate or precise estimates, especially for small population sizes like

those of the Laysan Teal.

If survey effort needs to be reduced due to limited skilled staff, cease the all-wetland

count but redirect some of the effort to collect more supplemental/ incidental resights and

modify the resight survey to include all of the wetlands. Incidental resights could be

collected morning or evening.

At least two additional years are needed to validate a survey index from direct counts.

There would be advantages to converting the current data entry/storage platform from the

spreadsheet-organized Microsoft Excel to a relational database, such as Microsoft

Access. In addition to consolidating all the various data (marking, resights, band

recovery, and survey results) in a single location, the relationships can be leveraged to

improve accuracy and efficiency. For example, the process of looking up a bird‘s identity

based on the observed color bands could be automated, or the data entry form could flag


an error if a resight is entered for a bird already listed as dead. Birds found dead should

be updated in a continuously revised resight reference for an accurate MEA.

With a new proposal or scope of work between the USFWS and the USGS, the USGS

could update the 2010 abundance estimate and/or convert Microsoft Excel data files to an

Access database.


REFERENCES

Barker, R.J. 1997. Joint modeling of live-recapture, tag-resight, and tag-recovery data.

Biometrics. 53:666–677.

Chapman, D.G. 1951. Some properties of the hypergeometric distribution with applications to

zoological censuses. Univ. Calif. Publ. Stat. 1:131–160.

Elzinga, C.L., Salzer, D.W. & Willoughby, J.W. 1998. Measuring and monitoring plant populations. BLM Technical Reference 1730-1, Bureau of Land Management, Denver,

CO.

Humbert, J.Y., Mills, L.S., Horne, J.S. & Dennis, B. 2009. A better way to estimate population

trends. Oikos. 118:1940–1946.

Johnson, L., Camp, R.J., Brinck, K.W. & Banko, P.C. 2006. Long-term population monitoring:

lessons learned from an endangered passerine in Hawai'i. Wildlife Society Bulletin. 34:1055–1063.

King, R.B., Queral-Regil, A. & Stanford, K.M. 2006. Population size and recovery criteria of the

threatened Lake Erie watersnake: Integrating multiple methods of population estimation.

Herpetological Monographs. 83–104.

Lancia, R.A., Nichols, J.D., & Pollock, K.H. 1996. Estimating the number of animals in wildlife

populations. In, T.A. Bookhout, ed. Research and Management Techniques for Wildlife and Habitats. Fifth edition, The Wildlife Society, Bethesda, MD.

Lincoln, F.C. 1930. Calculating waterfowl abundance on the basis of banding returns. U.S. Department Agriculture Circ. 118.

Marshall, A.P. 1992. Censusing Laysan ducks Anas laysanensis: a lesson in the pitfalls of

estimating threatened species populations. Bird Conservation International. 2:239–251.

McCartney, J., Armstrong, D.P., Gwynne, D.T., Kelly, C.D. & Barker, R.J. 2006. Estimating

abundance, age structure and sex ratio of a recently discovered New Zealand tusked weta

Motuweta riparia (Orthoptera, Anostostomatidae), using mark-recapture analysis. New Zealand Journal of Ecology. 30:229–235.

Microsoft Corporation. 2003. Microsoft Office Excel 2003 SP3. © 1985–2003 Microsoft

Corporation, Redmond, WA.

R Development Core Team. 2009. R: A language and environment for statistical computing. R

Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org.

Reynolds, M.H. 2002. The foraging ecology, population dynamics and habitat use of the Laysan

teal (Anas laysanensis). PhD, Virginia Polytechnic Institute and State University,

Blacksburg, VA.

Reynolds, M.H. & Citta, J.J. 2007. Post-fledging survival of Laysan ducks. Journal of Wildlife Management. 71:383–388.

Reynolds, M.H., McGowan, C., Converse, S.J., Mattsson, B., Hatfield, J.S., McClung, A.,

Mehrhoff, L., Walters, J.R., & Uyehara, K. in review. Trading off short-term and long-


term risk: minimizing the threat of Laysan Duck extinction from catastrophes and sea-

level rise.

Reynolds, M.H., Seavy, N.E., Vekasy, M.S., Klavitter, J.L., & Laniawe, L.P. 2008.

Translocation and early post-release demography of endangered Laysan teal. Animal Conservation. 11(2):160–168.

Reynolds, M.H., & Work, T.M. 2005. Mortality in the endangered Laysan Teal Anas laysanensis: conservation implications. Wildfowl 55:31–48.

Robson, D.S. & Regier, H.A. 1964. Sample size in Petersen mark-recapture experiments. Trans. Amer. Fish. Soc. 93:215–226.

Seavy, N.E., Reynolds, M.H., Link, W.A. & Hatfield, J.S. 2009. Postcatastrophe population

dynamics and density dependence of an endemic island duck. Journal of Wildlife Management. 73(3):414–418.

Seber, G.A.F. 1965. A note on the multiple recapture census. Biometrika. 52:249–259.

Seber, G.A.F. 1982. The Estimation of Animal Abundance and Related Parameters. MacMillan,

New York, NY.

Sincock, J.L. & Kridler, E. 1977. The extinct and endangered endemic birds of the Northwestern

Hawaiian Islands. 108 pp. Unpublished USFWS report, Portland, OR.

Skalski, J.R., Ryding, K.E. & Millspaugh, J.J. 2005. Wildlife Demography: Analysis of Sex, Age, and Count Data. Elsevier, New York, NY.

Stankey, G.H., Clark, R.N. & Bormann, B.T. 2005. Adaptive management of natural resources:

theory, concepts, and management institutions. 73 p. General Technical Report PNW-

GTR-654, U.S. Department of Agriculture, Forest Service, Pacific Northwest Research

Station, Portland, OR.

Thompson, S.K. 1992. Sampling. John Wiley & Sons, New York, NY.

U.S. Fish and Wildlife Service. 2009. Revised Laysan duck recovery plan. U.S. Fish and

Wildlife Service, Portland, OR.

White, G.C. & Burnham, K.P. 1999. Program mark: Survival estimation from populations of

marked animals. Bird Study. 46 Supplement:120–138.

Williams, B.K., Nichols, J. & Conroy, M.J. 2002. Analysis and Management of Animal Populations. Academic Press, San Diego, CA.

Work, T.M., Klavitter, J.L., Reynolds, M.H. & Blehert, D. 2010. Avian botulism: a case study in

translocated endangered Laysan Ducks (Anas laysanensis) on Midway Atoll. Journal of Wildlife Diseases. 46(2):499–506.

Work, T.M., Meteyer, C.U. & Cole, R.A. 2004. Mortality in Laysan ducks (Anas laysanensis) by

emaciation complicated by Echinuria uncinata on Laysan Island, Hawaii, 1993. Journal of Wildlife Diseases. 40(1):110–114.

Zar, J.H. 1996. Biostatistical Analysis, 3rd edition. Prentice-Hall, London, U.K.


APPENDIX 1. ASSUMPTIONS OF LINCOLN-PETERSEN MODELS AND THEIR

VIOLATIONS IN SURVEYS OF LAYSAN TEAL ON MIDWAY ATOLL

Assumptions of Lincoln-Petersen style estimators (Skalski et al. 2005)

Assumption Status Solution

1. The number of marks in the

population is known Violated Number of marks is estimated

2. The population is closed Violated

Marks are estimated on the date of the

resight survey; survey data is used only

during periods with low mortality and

low/no births

3. All animals have the same probability

of being caught or sighted Not violated

4. Marking does not affect future

detection Not violated

5. Animals do not lose their marks Not violated Birds with temporary bands are considered

to be unmarked

6. All marks are correctly reported Violated

Unknown status birds are assumed to be

marked and unmarked in the same

proportions as known birds.

Assumptions three and four are very likely to be met, but the first two assumptions of LP

style estimators are violated during Midway surveys: the number of marked birds is not known

exactly and the population is not closed between the time of marking and the time of the

secondary survey. We attempt to accommodate these violations by using the frequency of

captures and band resights (which take place during resight surveys) to determine whether or not

a bird is still available to be counted according to criteria and methods described in Appendix 2.

The adult population approaches geographic and demographic closure over the period of several

months after the breeding season and botulism outbreaks. There is no immigration or emigration,

and adult mortality was typically low when birds were radio tracked on Midway (0.08, 95% CI

0.02–0.17; Reynolds et al. 2008) and adult mortality was also typically low on Laysan Island

(0.001–0.208; Reynolds & Citta 2007). The closure assumption is less likely to be violated for

surveys taken during periods when mortality is relatively low. Figure 3 details the number of

marked carcasses recovered, by month, during the course of this study.

Assumption 5 states that birds do not lose their marks and is not violated in the Midway

surveys of Laysan Teal. One hundred five birds were originally given temporary color bands,

intended to fall off after an unknown short length of time. Many of these birds were later re-

captured and given permanent bands. For purposes of the LP population estimates, birds that

have temporary bands are included in the unmarked category. If temporary bands were not


distinguished from permanent bands as described in survey protocols, then the assumption would

be violated.

Assumption 6 is violated in most Midway surveys because the marked/unmarked status

cannot be determined for some birds (e.g., leg bands on swimming birds). We accommodate this

violation by assuming that unknown-status birds are marked or unmarked in the same proportion

as those of known-status birds. This is an assumption; it would be best if there were no unknown

birds, and in surveys where a significant portion (> 10%) of birds are of unknown status the

results should be interpreted carefully and supported with additional surveys.

Only marked, post-fledging birds are used in the LP estimator, and for the purposes of

these survey estimates, young-of-year are included beginning in the month of December when

survival of fledglings may stabilize. Note that this consideration deviates from the usual, normal

definition of an adult, second-year (SY) bird as one that has survived an extra month into the

January following its hatch year. Because of this inability to distinguish age classes, estimates

calculated in December through January are actually of the post-fledging abundance (of the

previous year) and will overestimate the number of breeding adults. Surveys in August–

November are likely to violate closed model assumptions if botulism is ongoing, however, they

are useful for obtaining resights and for establishing a baseline for validating an index from

standardized counts.

0

5

10

15

20

25

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov

Nu

mb

er

of d

uck c

arc

asse

s fo

un

d

Figure 3. Total known mortality of marked Laysan Teal, by month from

4 Dec 2004 to 14 Nov 2009. These numbers only include marked birds

whose bands were recovered from the carcasses. All carcasses recorded

in the month of August were recovered in 2008 and 2009 during

botulism epizootics. The spike in May is due to 11 carcasses recovered

due to an unusual human hazard in 2009 (problem was removed and

unlikely to reoccur); the other two May carcasses were from 2006.

December and February had very low to no mortality detected.


APPENDIX 2. ESTIMATING THE NUMBER OF MARKED LAYSAN TEAL

AVAILABLE TO BE COUNTED: MARKS ESTIMATED ALIVE (MEA)

Lincoln-Petersen abundance estimators were designed for studies with short intervals

between the original marking period and the later re-survey of marked and unmarked animals

and assume that the number of marked animals in the population is known (Lincoln 1930). The

population is assumed to be closed (no mortality, birth, immigration, or emigration) during that

period. Immigration and emigration do not occur on Midway, but mortality does occur so we

must estimate the number of birds still alive to be counted during a resight survey. We refer to

this as the marks estimated alive (MEA).

On Midway Atoll Laysan Teal were marked during the three years prior to this study.

Resighting of individual birds is conducted regularly, and most birds are seen frequently (median

interval eight days). However, if a carcass is not discovered, and a bird is not seen, its status

(dead or alive) is not known and must be estimated. Banding records and resights of individuals

determine the last date a bird was known to be alive. If the last known alive date is earlier than

the date of the survey being analyzed, the possibility exists that the bird has died in the interim

and is not available to be counted during a survey. If a bird is counted as not available for a

survey on date "D," it is possible that it could be resighted later. We would then know, after the

fact, that the bird was available on date D, so our estimate for the number of birds on date D

would retrospectively change very slightly (e.g., one additional marked bird would reduce the

Chapman estimate of abundance by fewer than three birds under current conditions (around 400

birds) with > 50% marked and > 200 birds resighted on each survey).

How long should we wait, after a Laysan Teal is last known to be alive, before it is

counted as missing? That amount of time will depend upon the behavior of the individual, and

we can use the previous record of detections for that bird to determine how long to wait. If a

bird is shy and seldom seen, the intervals among detections will be long, so we will wait a longer

time before declaring it missing. Likewise if an outgoing bird that is seen every day disappears

for a week, we suspect the worst. By looking at each bird‘s individual distribution of intervals

between re-detections and choosing a quantile of the distribution we can determine the maximum

time to wait before a bird is considered missing. In this report we present results using the more

conservative 50% quantile (or median) interval length and the 90% quantile, the interval length

where only ten percent of the observed intervals for that bird are longer.

Because a Laysan Teal that is still living always has a chance of being resighted while

resight effort continues, the further in the past a survey was conducted the more confidence we

have that a bird declared missing is dead, in general giving us greater confidence in population

estimates calculated for older surveys than our confidence in more recent surveys. For example,

using Table 1, the median interval on the 11 Jan 2010 survey of 488 marked birds reports that

only 60 (12%) were known to be dead and 264 (54%) were considered missing. Using the 90th

percentile of resight intervals only 127 (26%) were considered missing; this changes the

estimated abundance from 248 to 152 birds. By contrast, in the 23 Oct 2007 survey, switching

from the median to 90th

percentile of detection intervals changes the number of missing only

slightly from 7 (6%) to 5 (4%) of the total number of marked adults—a negligible amount with


little effect on the population estimate. Additional resight surveys in 2010 will allow us to

increase our confidence in determining the survival status of birds early in the year.

There are two additional issues to be considered when determining the length of time

before a bird is declared missing. Some birds on Midway were given temporary color bands.

Intervals among resights of temporary bands were not used to establish the distribution of

intervals among detections. Because temporary bands are expected to fall off, leaving the bird

unidentifiable by casual resight, a bird might vanish from the resight records when only its bands

were gone, not the individual. This omission is consistent with the established protocol of

identifying and omitting temporary bands during surveys.

In late 2009, some Laysan Teal were marked too recently or resighted too infrequently to

establish a distribution of intervals among detections for that individual bird. For birds with

fewer than ten intervals in their detection history we used the distribution of all intervals from all

birds in the dataset as their detection history. Future resights may increase an individual's

number of detection intervals to ten or more, allowing its individual distribution to be calculated.

This change will not affect the results of historical surveys, except in the case described above

when a bird that was once declared missing is later discovered to have been alive. These cases

should be carefully confirmed with positive field identification.


APPENDIX 3. MIDWAY ATOLL LAYSAN TEAL RESIGHT SURVEY PROTOCOL

INCIDENTAL RESIGHTS

An Incidental Resight is any band combination identified from a Laysan Teal (this protocol also

uses ―Laysan Duck‖ or LADU interchangeably) other than during an official (timed) Resight

Survey. Incidental resight information is valuable data used to determine the identification and

number of marked birds available to be counted. Data can be transcribed into a Laysan Teal

Incidental Resight Data Book. This data book or resighting data should be accessible in the

USFWS office. Include the following data:

Date, Time

Observer

Color Band and Color Band Symbol Right Leg (RL) and Left Leg (LL)

FWS or metal band (Right Leg or Left Leg and number), if not read DNR or ―did not read.‖

Most birds can be identified by their color band, and it is not necessary to read the aluminum

(AL) band unless the color band is missing or too faded.

Age Class of bird, if known (Table 6; L—duckling, AHY—after hatch year, ASY—after

second year)

Sex of bird, if known1

Location of sighting

Comments on band condition (worn or faded), occurrence of ducklings, injuries

OPTIONAL Behavior (was it sleeping, feeding)

OPTIONAL Any associated ducks or mates and the band combination (ID) of the other

duck(s)

1 Female Laysan Teal have a paler bill, often with small ―freckles‖ at the base. The legs are dull orange, and the

plumage may be lighter brown than males. Males have a black saddle on the bill with the sides of the bill deep

green. The legs are often bright orange, and males may have an upturned curled tail feather. The plumage on the

head may be darker brown.


Table 6. Laysan Teal age class characteristics for Midway Atoll (Note Stage Ia–Ib ducklings are

very susceptible to mortality related to human disturbance).

Days Old Plumage Class Description

1 - 6 I a Bright ball of fluff; down bright and new; rounded

body; neck and tail not prominent

7 - 12 I b Fading ball of fluff; down fading

13 – 18 I c Gawky downy; down color and pattern faded; neck

and tail prominent; body long and oval

19 – 25 II a First feathers; feather begins on sides of body

26 – 35 II b

Mostly feathered; side view shows one half side and

flank feathered; down remains on nape of neck, back

and/or rump

36 – 45 II c Last down; little down on rump, neck, or back;

sheaths visible on erupted primaries, no eye ring

46 - 60 III a Feathered flightless = no down visible; flight feather

out of sheath but not fully developed

Flying 52 -63 III b Flies; still with brood

Flying

> 63 – 73 III c

Flies and more independent; small amt. white

plumage over eye; no sexual dimorphism

Hatched that

year > 73 HY / Fledgling

Independent; > 73 days until Jan 1 of following year.

Small; eye ring; glossy plumage; some sexual

dimorphism evident

Hatched

Previous Year SY

Glossy plumage; small eye ring; plumage often

distinguishable from Jan 1 until about March

Generic Adult AHY Any mature adult of unknown age

> 360 days ASY At least 50% plumage on head is white (general

guide-line for adults of unknown age)

Possible Band Numbers, Letters, and Symbols Use the following codes to record band colors and combinations or write out the colors and the

color of symbols on the color band. Please examine the string of sample color bands in the

biology office before attempting to record band resights.

Numbers: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9

Letters: A, C, E, F, H, J, K, M, N, P, R, S, T, U, W, X, Y

Letters with Numbers: C1–C9, X1–X10

Two Letters: WA-WY, XA-XY

Symbols: *, #, $, <, >, ?, \ , /, &, +, -

Grooves: single (e.g., R/WG) or double (e.g., GN/2WG)

Permanent Bands Current permanent color band codes on Midway (Apr 2009):

AL = Federal aluminum

BK = Black with or without white letters, numbers or symbols (BK temporary bands without

numbers or symbols also occur)

BKY = Black with yellow numbers, letters, or symbols

BR = Brown with white numbers, letters, or symbols


B = Blue (royal blue) with white numbers, letters, or symbols (Light Blue (LB) bands or Dark

Blue (DB) bands without symbols are also deployed) – if unfamiliar with codes please

write out the word Blu or Blue to distinguish from Black, or Brown

BY = Blue (royal blue) with yellow numbers, letters, or symbols

GN = Green with white numbers, letters, or symbols (Dark Green (DGN) and Light Green

(LGN) temporary color bands without symbols are also deployed)

GY = Grey with black numbers, letters, or symbols—if unfamiliar with codes write out the word

of the color and do not use the code GR which is confused with green

O = Orange with or without black letters, numbers, or symbols

R = Red with or without white letters, numbers, or symbols (be aware of Purple (PU) and

Fuchsia (FU) temporary bands that can be confused with red)

W = White with or without black letters, numbers, or symbols

WM = White band with mint green letters, numbers, or symbols

Y = Yellow with black numbers, letters, or symbols (Y temporary bands without numbers or

symbols also occur)

Bands with colored grooves or stripes are recorded as the band color, and the number and color

of grooves (e.g., Yellow with 1 Black Groove is coded as Y/BKG; Green with 2 White grooves

is coded as GN/2WG).

Temporary Bands Temporary color bands were deployed on young Laysan Teal in 2007 and rehabilitated birds

during the botulism epizootic in 2009. These typically have no symbol and no AL band. Colors

include all listed above plus: hot pink (HP), fuchsia (FU), purple (PU), light green (LGN), dark

green (DGN), light blue (LB), dark blue (DB). Many have fallen off so you may see a color

band on only one leg. It is important to try and specify temporary bands versus permanent

bands, especially those on birds that may have lost one of their temporary bands. Temporary

bands are coded using lower-case letters followed by ―-temp‖ (e.g., hot pink temporary is coded

as hp-temp; dark green temporary is coded as dgn-temp). These birds should be given

permanent bands to be included in population estimators.

Aluminum Only There are birds that only have an AL band with no color band. Please try to identify the sex of

the birds and read the numbers on these bands for a positive ID. Use a spotting scope to reduce

disturbance to the birds, especially during duckling season. Do not approach the birds closer

than 30 m unless you are doing a resight survey or a botulism carcass search. Never approach a hen with ducklings. Leave the area quickly if a hen with young ducklings approaches you. It is

ok for birds without ducklings to approach observers closer than 30 m especially during the non-

breeding season. Due to possible color band loss or unrecorded removals, there may be more

birds with ―AL only‖ than noted in the Resight Reference (the reference file for all bird bands

used). Always attempt to read all AL only numbers.

Band Discrepancy If there is a band combination that has been resighted that is not in the Resight Reference (RR),

record the bird as observed, make a note of it and try to resight the bird again. Some band

combinations may be incorrect in the RR because they might have been transposed accidentally


during banding. After several sightings of the same combination, the RR should be corrected

(notify [email protected], Leona Laniawe, or John Klavitter). List the date and

person that reported the observation from the field.

mailto:[email protected]


PROCEDURE FOR STANDARDIZED RESIGHT SURVEY OF ALL WETLANDS

The accuracy of a population estimate or population status index will depend largely on the

variation in counts (detection probabilities). Probability of detecting wetland-dependent birds

varies with time of day, diurnal patterns, weather conditions, and possibly seasonally. Thus,

pilot surveys will identify the optimal time to conduct surveys at a site to maximize detection

probability. Frequent surveys (weekly) are recommended to evaluate the proportion of the

population likely to be observed during survey efforts, and confirm the survival and the number

of marked birds in the population. In addition to the Standard Resight Survey, a timed atoll-wide

All-Wetlands Count (this survey will be discontinued beginning in 2010) will help determine the

proportion of birds we might be missing on the Standard Resight Survey. Simultaneous surveys

of both Sand and Eastern islands are needed for a population estimate and for a more complete

count and to reduce the frequency of double counting unmarked birds. The Standard Resight

Survey of all wetlands (Resight Survey, combines the Standard Resight Survey with the atoll-

wide All-Wetlands Count into one survey), should be conducted bi-monthly only by experienced

wildlife observers with a quality zoom spotting scope and good binoculars. If volunteers

participate, volunteers should be accompanied by an experienced Midway bird/wildlife biologist,

or have bird watching experience and undergo numerous training sessions before participating in

time-dependent surveys. All sightings should be confirmed using the Midway Resight Reference

list (page 48, Fig. 7) after returning from the field.

Materials Needed: Sample Color Bands (check these out in the office before conducting any resights)

Garmin GPS for sunrise and sunset times

Time Piece

Tally Counters for counting flock sizes, banded, and unbanded birds

Resight Data Clipboard and Data Sheets on Rite in the Rain® paper

Spotting Scope, Tripod, and 10x40 Binoculars

Pencils

Footwear that can get wet

Survey Map/Survey Route (Each survey is randomized to avoid spatial & temporal bias)

3 hours per Sand Island Resight Survey (15 minute grace period)

2 hours per Eastern Island Resight Survey (15 minute grace period)

Bike or Golf Cart (low density albatross season)

Procedure 1. Check Garmin GPS (accessories menu) for exact sunrise/sunset time before each survey.

Fill in this time on the data sheet before conducting survey. Example data sheets are

presented in Attachment 1. On Sand Island, begin morning surveys at sunrise or evening

surveys 3 hours before sunset. On Eastern Island, begin morning surveys at sunrise or

evening surveys 2 hours before sunset if camping; or 2 hours 30 minutes before sunset if

traveling by boat to allow time to return with sufficient daylight. The allotted time for

the Resight Survey is 2 hours on Eastern Island; 3 hours on Sand Island. Surveys may

not exceed the allotted time by more than 15 minutes—aim to be efficient and finish on

time.


2. Much of the Laysan Teal population resides on Eastern Island. Two observers are

needed to conduct a simultaneous survey with the Sand Island Survey. Simultaneous

atoll-wide surveys are needed to estimate the population size. If sea conditions are

unsuitable, reschedule the Resight Survey. If a survey has been started and cannot be

finished because unsuitable weather conditions arise, the survey must be rescheduled and

repeated in its entirety. Resights read prior to cancellation will be included as incidental

resights. Plan on conducting one weekly Resight Survey and/or atoll-wide All-Wetland

Count (or two Resight Survey and All-Wetland Counts per month). If a Resight Survey

cannot be completed in a week because of persistent unsuitable weather, attempt to

conduct the survey the following week (a count and a resight survey during the same

week). If two skilled resight observers are not available, the survey should be postponed

until a second skilled observer is available. As a last resort, if only a single observer is

available for > two weeks, a single skilled observer should survey Eastern Island on the

first morning and Sand Island the second morning.

3. Check the hard copy of the previous survey‘s Start Locations and Direction of Travel

Table (Attachment 2) and record the date of the survey‘s randomized start. For the Sand

Island Resight Survey, eight random survey numbers have been assigned to specific

starting locations and direction of travel at 13 different wetlands. An example of the

clockwise order of the Resight Survey route is: Catchment Basin → Mauka-Makai, →

Ironwood, → Aviary,→ Rusty Bucket, → Radar Hill Seeps, → Ball Field Seeps, → Fuel

Farm, → Brackish Pond, → Sunrise, → Communications, → Parade, → and R2. For the

Eastern Island Resight Survey, six random survey numbers have been assigned to the

three wetlands using different randomized orders. Refer to the map for survey travel

routes and familiarize yourself with the route in the field before attempting to conduct a

survey (Attachment 3). Follow the route outlined in the prescribed direction.

4. Aim to arrive at the start location five minutes before the official start time. Record

weather variables as soon as you arrive. Review the rain scale (Table 7) and Beaufort

wind scale (Table 8, page 46) if you are not familiar with them. If weather conditions

change, record the range of conditions during most of the survey period. Reschedule the

survey in steady heavy rain, winds ≥ 6 or storm conditions. Use the 15 minute grace

period to stop survey during squalls of short duration.

Table 7. Rain scale for data entry.

Notation Description

0 no rain

1 mist (fog)

2 light drizzle

3 light rain

4 heavy rain

5. Record your initial arrival times at each wetland and then count all post-fledgling birds,

note independent juveniles (HY) if they are distinguishable (June–August).

6. Rapid transitions between wetlands are needed. The amount of time spent at each

wetland will depend on the number of birds using it, your efficiency in reading bands,

your travel time between wetlands, and the albatross density.

7. Approach wetlands with a low profile, slowly and quietly as birds will jump into the

water or hide in dense vegetation if alarmed; this will obscure leg bands. Be sure to


record all birds that emerge, fly out, fly in, or fly by. If birds fly in or out of the

wetland, they should be tallied and included in the total count at that wetland, unless you

are able to observe the birds moving between wetlands (and then you can subtract them

to prevent known double counting). Often, birds that are flying out will be of unknown

band status. Many wetlands will require that you look for birds from two different

vantage points due to vegetation obscurity at loafing sites. Use observation towers and

blinds if available.

8. Determine if each bird is banded, unbanded, or unknown (if legs are concealed while

flying, swimming in turbid water, or too far away). Plan on getting your feet wet to read

band status at some wetlands. Surveyors may also try sitting at the edge of the wetland,

which allows the birds to relax, come out of the water, and walk on the shore where their

bands can be read. Some birds have bands on both legs; others are banded on only one

leg. It works well to count them all first, check for band status, record that, then begin

reading and recording the color bands. Often you will need to count them all again at the

end if birds have emerged during your count. Because of the time constraints of the

survey, you will need to be very efficient. Recruit a qualified scribe if there is an

interested person available that has some wildlife counting experience. Because running

the survey efficiently requires practice, dedicated observers are needed to conduct

surveys.

9. If duckling broods and associated hens are observed, keep your distance and lower your

profile (hide). Take care not to cause brood fragmentation, prolonged alarm response, or

curiosity/distraction from the mother hens. Use a spotting scope and/or visual barriers if

hens with broods are known to occur at a wetland.

10. While en route, record any Laysan Teal you see, as well as those flying by, flying in, or

that flush from the wetlands. Include these resights in your total Post Fledgling Count if

they were sighted during the survey (start and end time limit) or after, but within the time

limit; change the end time and include the birds as survey birds. If the time limit is over

(beyond the grace period) then the birds will be considered incidental resights.

11. Enter survey totals and check resights using the Midway_Resight_Reference.xls (see

Resight Survey Data Entry and Fig. 7). Enter all unique resights with the updated date

and your initials into the Midway_All_Resights.xls (Fig. 8). Enter methods as ―Resight

Survey‖ or ―Incidental Survey.‖ Enter the survey totals in the

Midway_LADU_Survey_Summaries.xls (Fig. 11).


Table 8. Beaufort Scale: Specifications and equivalent wind speeds for sea and land.

Double Counts

Birds may move during the survey. If a unique banded bird is resighted at multiple seeps/sites

during the survey, record both resights (e.g., AL/GNXA resighted at both Brackish and Sunrise).

After the survey, when completing the data form, correct for double counts by including only

one of the two or more resights in the total number of birds. During a Simultaneous Survey,

birds may move between islands. If a unique banded bird is resighted on both Sand and Eastern

Beaufort

Number

Wind Speed Wind

Descriptor Sea Description Land Description

Miles/hr Knots

0 0–1 0–1 Calm Flat. Calm. Smoke rises vertically.

1 1–3 1–3 Light air Ripples without crests. Smoke drift indicates wind direction, still

wind vanes.

2 4–7 4–6 Light breeze Small wavelets. Crests of glassy

appearance, not breaking

Wind felt on exposed skin. Leaves rustle,

vanes begin to move.

3 8–12 7–10 Gentle breeze Large wavelets. Crests begin to

break; scattered whitecaps

Leaves and small twigs constantly moving,

light flags extended.

4 13–18 11–16 Moderate breeze Small waves w/ breaking crests.

Fairly frequent white horses.

Dust and loose paper raised. Small branches

begin to move.

5 19–24 17–21 Fresh breeze

Moderate waves of some length.

Many white horses. Small amounts

of spray.

Branches of a moderate size move. Small

trees in leaf begin to sway.

6 25–31 22–27 Strong breeze

Long waves begin to form. White

foam crests are very frequent. Some

airborne spray is present.

Large branches in motion. Whistling heard

in overhead wires. Umbrella use becomes

difficult. Empty plastic garbage cans tip

over.

7 32–38 28–33

High wind,

moderate gale,

near gale

Sea heaps up. Some foam from

breaking waves is blown into streaks

along wind direction. Moderate

amounts of airborne spray

Whole trees in motion. Effort needed to

walk against the wind.

8 39–46 34–40 Gale, fresh gale

Moderately high waves with

breaking crests forming spindrift.

Well-marked streaks of foam are

blown along wind direction.

Considerable airborne spray

Some twigs broken from trees. Cars veer on

road. Progress on foot is seriously impeded.

9 47–54 41–47 Strong gale

High waves whose crests sometimes

roll over. Dense foam is blown

along wind direction. Large amounts

of airborne spray may begin to

reduce visibility.

Some branches break off trees, and some

small trees blow over.

Construction/temporary signs and

barricades blow over.

10 55–63 48–55 Storm, whole

gale

Very high waves with overhanging

crests. Large patches of foam from

wave crests give the sea a white

appearance. Considerable tumbling

of waves with heavy impact. Large

amounts of airborne spray reduce

visibility.

Trees are broken off or uprooted, saplings

bent and deformed. Poorly attached asphalt

shingles and shingles in poor condition peel

off roofs.

11 64–72 56–63 Violent storm

Exceptionally high waves. Very

large paths of foam driven before

the wind, cover much of the sea

surface. Very large amounts of

airborne spray severely reduce

visibility.

Widespread damage to vegetation. Many

roofing surfaces are damaged; asphalt tiles

that have curled up and/or fractured due to

age may break away completely.

12 73–83 64–71 Hurricane-force

Huge waves. Sea is completely

white with foam and spray. Air is

filled with driving spray, greatly

reducing visibility.

Very widespread damage to vegetation.

Some windows may break; mobile homes

and poorly constructed sheds and barns are

damaged. Debris may be hurled about.


during the survey, record the total number of double-counted unique banded birds identified

atoll-wide on the data sheet.

Temporary Bands On the data sheet, include the number of Laysan Teal with: unique temporary bands (birds that

can be connected to a Link #; temporary bands were seen on BOTH legs), not unique temporary

bands (without Link #, only one leg was seen with a temporary band or both legs were seen with

only one temporary band), and the total number of temporary bands. Correct totals for double

counts; e.g., if bk-temp/bk-temp was recorded at both Catchment and Brackish, include one of

the two detections in the total.

Resight Survey Data Entry All data sheets should be scanned and saved electronically as a backup copy in case the hard

copy goes missing. Data should be entered by the observer as soon as possible after completing

the survey. Timely data entry avoids data backlogs, misplacement of data sheets, and provides

an opportunity for the observer to add data to incomplete data fields and decipher illegible

writing.

There are four Excel files that will be used with Laysan Duck Resight Survey data.

They are:

Midway Resight Reference—for reference only, containing all the band combinations in

use in the Laysan Duck population and data on those individuals (Figs. 7 and 9). Data

entry in this file is not covered here.

Midway All Resights—where all the information on each resighted individual is recorded

during the Resight Survey or incidental resights (Figs. 8 and 10).

Midway LADU Band—used during bird banding. This file can be used as a back-up for

band information if the current Resight Reference file is unavailable. Data entry in this

file is not covered here.

Midway LADU Survey Summaries: Standard Resight Summary tab—overall numbers of

birds for each island and atoll-wide including times of surveys and weather conditions

(Fig. 11).

1. The first step is to enter a resight record for each banded bird that was uniquely identified

by bands during the survey. This will verify that the band was read correctly and the age

and gender were appropriately recorded. To make the entry, open both the ―Midway

Resight Reference‖ (Fig. 7) and ―Midway All Resights‖ files (Fig. 8). Check with the

Laysan Duck data manager to verify that the Resight Reference file is current with all the

latest Laysan Duck banding data updates (newly banded birds, lost bands, changed bands,

unreadable bands, and mortalities). If the Midway Resight Reference is not current, data

entry should be delayed or the data manager should enter the data using the ―Midway

LADU Band‖ file. If the Midway Resight Reference is current, use this file to search for

the identity of the bird observed. To most easily search the Midway Resight Reference,

click on the small arrow located in the lower right corner of the header row cell

containing the field you would like to search under (Fig. 9). This is typically the band

observed on either the right or left leg. If the band was read accurately on the correct leg,

the band will be found. Once found, click on the correct band and the record for the bird


or birds wearing this band will pop up. Use your mouse and highlight the information

contained in cells A to F, copy and paste it into the next available line in the cells under

the columns H to M in the Midway All Resights file (Fig. 10). Next, in the Midway All

Resights file type in all of the other information (date, time, etc.) pertaining to the

observation. Continue this process for all bands that were recorded during the survey.

Once all bands have been entered, save the file with a new name including observer

initials and the current date (e.g., Midway_All_Resights_MHR_19Jul2010.xls).

Figure 7. Laysan Duck Midway Resight Reference Excel file showing the header line and the first nine Laysan

Ducks currently surviving in Midway‘s population. Column A is the Midway Link # which is a unique

identifying number for each Laysan Duck. Columns B through D include the band on the right leg, band on the

left leg, and sex of the bird. Column E is the bird‘s age in 2009. Column F is the most recent aluminum

USFWS band number. Column G gives the origin of the bird, for example whether the bird was a translocated,

or founder, bird. Column H provides notes on the condition of the band, etc. Columns I and J list the location

and date the bird was last seen.


Figure 8. Laysan Duck Midway All Resights file showing the header line and resight entries for Midway‘s population. Column A and B are the date and time of the

observation. Column C lists the observer initials. Columns D and E are the zone and the location where the bird was observed. Column F is the habitat type. Column G

is the unique link # for the bird observed. Columns H and I are the bird‘s bands for the right and left legs. Columns J and K list the sex and age. Column L (USGS #, i.e.,

USFWS number) is the current aluminum band the bird was wearing. Column M is used for recording numbers on the aluminum band that are read by the observer.

Columns N to R are used for recording information about the observed bird‘s mate. Column S is used for comments (this is where any behaviors can be recorded as well).

Column T lists the method of surveying that was used when the bird was observed.


Figure 9. Using the Midway Resight Reference file to verify that a band was read correctly and locating the record for the bird. The small arrow in the lower right corner of

the header row for RLEG was clicked, allowing a search of available bands.

Figure 10. Midway All Resights file with the information pasted in that was copied from the Midway Resight Reference file.


2. After all of the resights have been entered and all the bird bands, ages, and

genders have been confirmed, the next step is to enter the summarized survey

data. Open the ―Midway LADU Survey Summaries‖ file. Click on the ―Standard

Resight Summary‖ tab (Fig. 11). Three separate data lines will be entered

(assuming an atoll-wide survey was completed): one for Sand Island, one for

Eastern Island, and a third Atoll Wide line, which sums the results from Sand and

Eastern. If only one island was surveyed, only one line of data is entered for that

island surveyed. For the island surveyed, the date and island are entered in

columns A and B. In columns C and D, Atoll Wide or One Island and AM or PM

are written. Observer(s) initials are typed in column E. Start and end times and

time of sunset or sunrise are entered in columns F, G, and H, using a 24-hour

clock. Weather variables are recorded in columns I to K. Starting point

(Attachment 2), starting location, and direction of survey (clockwise or counter-

clockwise, entered as Clock or Counter) are entered in columns L to N. Next,

follow the data sheet and enter all the information cell by cell as indicated by the

header row from columns O to BR. If there is only one island to enter, proceed

with saving the file with a new name which includes the observer initials and date

(e.g., Midway_LADU_Survey_Summaries_MHR_19Jul2010.xls). If both islands

were surveyed, continue the data entry for the second island. Once the entry is

complete, add a third row of data which will sum the information for Sand and

Eastern islands. The date will be the same as the Sand Island survey date. Record

Atoll Wide for both columns B and C. Record morning or evening survey

(column D) and observer(s) initials (column E). Sum the survey information from

the two surveys for columns AE to AZ using the computer calculator, not by

hand. Once this is finished, the file is saved as indicated previously.

Figure 11. The Midway LADU Survey Summaries file, Standard Resight Summary tab showing the

headings and cells from columns A to N. Headings and columns are not shown for O to BR.


PROCEDURE FOR PILOT ATOLL-WIDE ALL-WETLANDS COUNT

This survey will be discontinued in 2010. Resight surveys will be modified to

include all wetlands.

1. Twice per month, an atoll-wide All-Wetlands Count index should be conducted at

sunset or sunrise. The count must go no more than 15 minutes past the allotted

time for the data to be usable—aim to be efficient and finish on time.

On Sand Island every permanent and ephemeral wetland known should be

completed in 1 hour 45 minutes. Begin survey at sunrise or 1 hour 45 minutes

before sunset. Wetlands counted include Rusty Bucket Seep, Aviary Seep,

Ironwood Seep, Mauka-Makai, Mauka-Makai Drainage Ditch, Catchment

Basin, R2 Wetland, R2 Treatment Pools, Tarmac Pool, Communications Seep,

Parade Seep, Sunrise Seep, Brackish Pond & overflow areas, Fuel Farm,

Radar Hill Seeps, and Ball Field Seeps.

On Eastern Island: Monument, Rolando, and Sunset Seeps should be

completed in 1 hour 15 minutes. Begin survey at sunrise or 1 hour 15 minutes

before sunset if camping or 1 hour 45 minutes before sunset if traveling by

boat to allow time to return with sufficient daylight.

2. Check the hard copy of the previous surveys‘ Start Locations and Direction of

Travel Table (Attachment 2) and record the date of the survey‘s randomized start.

For the Sand Island All-Wetlands Count, eight survey numbers have been

assigned to specific starting locations and direction of travel at five different

wetlands.

For the Eastern Island All-Wetlands Count use the same series of randomized

starting locations and direction as the Resight Survey. Refer to the map

(Attachment 3) for survey travel routes and familiarize yourself with the

actual route before attempting to conduct a survey.

3. Record the following data: # total ducks, # banded, # unbanded, # unknown,

birds that fly in, fly out, and fly by and any broods/ducklings with age classes, if

known. If birds fly in or out of the wetland, they should be noted and subtracted

from the total count at that wetland.

4. Make special note of birds with temporary bands, or other bands that appear to

be falling off. After the survey has been completed, record the total number of

temporary bands observed on the data sheet.

5. Rapid transitions between wetlands are needed. Typically, there is insufficient

time to read the band combinations. This is an index to see how many birds are

using areas not included on the Standard Resight Survey, and may be a good

indicator of seasonal detection probabilities or flock behavior. A separate data

sheet is used for the All-Wetlands Count (Attachment 1).

6. If you finish the count early and there are band combinations you can read and

record (as incidentals), continue surveying until birds have been identified. End

your count at approximately 1 hour 45 minutes on Sand Island and 1 hour 15

minutes on Eastern Island. If an infrequently used wetland is skipped because of

lack of water or time constraints, mark the wetland ―not checked.‖

7. On both surveys, always be alert for sick birds, bird carcasses, hens with new

duckling broods, or birds with band injuries (i.e., swelling or debris in leg bands).

Note the ID of the injured bird, and seek assistance, if needed, to capture sick or


injured birds. Collect carcasses, or return to collect carcasses as soon as possible.

Avoid disturbance of hens with ducklings by maintaining maximum distance, not

approaching broods, and leaving the wetlands as soon as possible to prevent

brood fragmentation to downy ducklings.

8. Edit, total, and summarize your completed Standard Resight Survey or atoll-wide

All-Wetlands Count on the data sheet. Transcribe incidental resights into the

incidental resight book. Enter count total data within 48 hours of collection.

Atoll-wide All-Wetlands Count Data Entry 1. The first step is to enter all resight records for banded birds that were

opportunistically identified by bands during the survey as described previously

(Figs. 7–10, Incidental resights).

2. After all of the resights have been entered and all the bird bands, ages, and

genders have been confirmed, the next step is to enter the summarized survey

data. Open the ―Midway LADU Survey Summaries‖ file. Click on the

spreadsheet tab ―All Wetland Count Summary‖ (Fig. 12). For each wetland

visited during the survey, one line of data will need to be entered starting with the

date in column A and continuing to the right to include island, survey type (island

or atoll-wide), wetland location, arrival time, if water is present in the wetland,

number of banded birds, number of unbanded, number of unknown, total post

fledglings, number of broods, number of ducklings, number of hatch-year banded,

number of hatch-year unbanded, number of birds flying in, out, and by the

wetland and their totals, and finally comments are entered in column S. Repeat

this data entry for each wetland visited during the All-Wetland Count. The data

entered in the spreadsheet follows the data sheet very closely.

3. After all of the All-Wetland Count Summary data have been entered, next click

on the ―Count Conditions and Totals‖ tab from the Midway LADU Survey

Summaries file (Fig. 13). In the spreadsheet, data are summarized from all the

wetlands visited during the All-Wetland Count. If only one island is visited

during the survey, only one line of data is entered. If both Sand and Eastern

islands are visited, then three lines of data will be entered, one for Sand, one for

Eastern, and one for the Atoll where data from Eastern and Sand are totaled. Start

with Eastern Island and begin entering information from the data sheet including

the date, island, survey type (island or atoll-wide), AM or PM survey, observers,

time start, time end, time of sunrise or sunset, rain, wind, cloud cover, starting

point number, start location, end location, direction of travel, total number banded

birds, total unbanded birds, total unknown birds, total post-fledglings, total

number of temporary bands, total broods, total number of ducklings, total hatch-

year banded and unbanded birds, total number of birds that flew in, out, by, and a

summary total of these, and finally notes in Column AC. Continue this same data

entry for Sand Island. A third line of data will be entered and used to sum the

data from both Eastern and Sand islands. This line of data will be highlighted in

yellow (Fig. 13). Once all of the survey data have been entered, rename the file

with observer initials and the current date (e.g.,

Midway_LADU_Survey_Summaries_MHR_05Jul2010.xls).


Figure 12. The Midway LADU Survey Summary File open to the All Wetland Count Summary tab.


Figure 13. The Count Conditions and Totals tab from the Midway LADU Survey Summaries file.


August 2010 47

ATTACHMENT 1. REVISED DATA SHEETS:

SAND ISLAND LADU Standard Resight Survey of All Wetlands (3 hours)

Date: Observers: Time start: Time end: Sunrise/Sunset: _____/_____

Rain: _(0-4) Wind: __ (Beaufort Scale) Clouds: %

Starting Point #______ Start Location:________________ Direction:_______________

Standard Atoll Wide Protocol □ Opportunistic One Island Survey □

TEMPORARY bands—# of unique:_____ # of not unique:_____ Total: _______

Total # not unique ALUMINUM ONLY: ________

Total # of banded DOUBLE COUNTS—at multiple sites: ______ Atoll Wide (for simul. surveys): _____

Total # of UNIQUE BANDS during Resight Survey (check reference list, correct for dbl counts): _____

Location Ball Field Big / Little

Fuel Farm Brackish

Pond / &

adjacent Sunrise

Communi-

cations Parade

R2 Treatment/

wetland

Arrival Time / / /

Depth / / /

Total Laysan

Ducks Bnd / Bnd Bnd / Bnd Bnd Bnd Bnd /

Unb / Unb Unb / Unb Unb Unb Unb /

Unk / Unk Unk / Unk Unk Unk Unk /

TOTAL / TOTAL TOTAL / TOTAL TOTAL TOTAL TOTAL /

# LADU flew

in/ flew out/

flew by / / / / / / / / / / / / / /

Migrant

waterfowl

Location Catchment

Mauka-

Makai / Ditch

Aviary Ironwood Rusty

Bucket

Radar

Hill 1 / 2

Arrival Time / /

Depth / /

Total Laysan

Ducks Bnd Bnd / Bnd Bnd Bnd Bnd /

Unb Unb / Unb Unb Unb Unb /

Unk Unk / Unk Unk Unk Unk /

TOTAL TOTAL / TOTAL TOTAL TOTAL TOTAL /

# LADU flew

in/ flew out/

flew by / / / / / / / / / / / /

Migrant

waterfowl

Total observed ducks on the move (flew in/flew out/flew by): / /

Post

Fledglings Male Female

Unidentified

Sex, Post

Fledglings

Total

Post

Fledglings

Hatch Years

(HY) Ducklings

Banded

Unbanded

Unknown

Total:

TOTALS

Broods:


August 2010 48

Example of Clockwise wetlands route: Catchment, Mauka-Makai, Aviary, Ironwood, Rusty Bucket, Radar

Hills, Ball Fields, Fuel Farm, Brackish Pond & adjacent wetlands, Sunrise, Communications, Parade, R2.

Resight data sheet

Location RLeg LLeg Band No. Sex Age Paired?

Y/N/U Mate ID Notes

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45


August 2010 49

Midway LADU Sand Island Original All-Wetlands Count (1 hour 45 minutes)

This count will be discontinued in 2010. Date: _______ Observers: Time start: _____ Time end: _____ Sunrise/Sunset: _____/_____

Rain: (0-4) Wind: (Beaufort Scale) Cloud Cover: %

Start location _______________ End Location: _______________ Direction: ________

Standard Atoll Wide Protocol □ Opportunistic Survey □

Total number of birds observed with temporary bands:

Sand Island

Location

Arrival

Time

Water

(Y/N) BND UNB UNK

Total Post

Fledglings

Broods/

Ducklings

(stage if

known)

*HY

Banded

HY

Unbanded

Flew in/

Flew

out/

Flew by

Notes

Radar Hill Seeps / /

Ball Field Big

Ball Field Little

Fuel Farm

Brackish Pond

Brackish Pond

Adjacent Wetland

Brackish Pond

Forest Depression

(overflow)

Sunrise Seep

Communications

Seep

Parade Seep

Tarmac Pool

R2 Treatment

Pools

R2 Wetland

Catchment Basin

Mauka-Makai

Drainage Ditch

Mauka-Makai

Ironwood Seep

Aviary Seep

Rusty Bucket

Seep

Other (describe)

or give GPS

Coordinate

Totals

*HY may not be distinguishable in the field by plumage after September. May-Aug most unbanded birds can be aged as HY or AHY

by eye ring and plumage. All temp banded birds are likely to by ASY. Example of Clockwise wetlands route:

Catchment, Mauka-Makai, Aviary, Ironwood, Rusty Bucket, Radar Hills, Ball Fields, Fuel Farm,

Brackish Pond & adjacent wetlands, Sunrise, Communications, Parade, R2.

Incidental Resights/ Observations:


August 2010 50

EASTERN ISLAND LADU Standard Resight Survey of All Wetlands (2 hours)

Date: Observers: Time start: Time end: Sunrise/Sunset: _____/_____

Rain: _(0-4) Wind: __ (Beaufort Scale) Clouds: %

Starting Point #______ Start Location:________________ Direction:_______________

Standard Atoll Wide Protocol □ Opportunistic One Island Survey □

TEMPORARY bands—# of unique:_____ # of not unique:_____ Total: _______

Total # not unique ALUMINUM ONLY: ________

Total # of banded DOUBLE COUNTS—at multiple sites: ______ Atoll Wide (for simul. surveys): _____

Total # of UNIQUE BANDS during Resight Survey (check reference list, correct for dbl counts): _____

Location RLeg LLeg Band No. Sex Age Paired?

Y/N/U Mate ID Notes

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

Post

Fledglings Male Female

Unidentified

Sex, Post

Fledglings

Total

Post

Fledglings

Hatch Years

(HY) Ducklings

Banded

Unbanded

Unknown

Total:

TOTALS

Broods:

Location MONUMENT SEEP SUNSET SEEP ROLANDO SEEP TOTALS

Arrival Time

Depth

Total Ducks Bnd

Unb

Unk

TOTAL

Bnd

Unb

Unk

TOTAL

Bnd

Unb

Unk

TOTAL

Banded

Unbanded

Unknown

TOTAL

# LADU flew In /

flew out / flew by / / / / / / / /

Migrant Waterfowl


August 2010 51

Midway LADU Eastern Island All-Wetlands Count (1 hour 15 minutes)

This count will be discontinued in 2010. Date: _______ Observers: Time start: ______ Time end: _____Sunrise/Sunset:

______/______

Rain: (0-4) Wind: (Beaufort Scale) Cloud Cover: %

Start location: ____________ End Location: _____________ Direction: ________

Standard Atoll Wide Protocol □ Opportunistic Survey □

Total number of birds observed with temporary bands:

*HY may not be distinguishable in the field by plumage after September. May-Aug most unbanded birds can be aged as HY or AHY

by eye ring and plumage. All temp banded birds are likely to by ASY. Start locations same as for Eastern Resight survey.

Incidental resights at last site if time allows:

Eastern

Island

Location

Arrival

Time

Water

(Y/N) BND UNB UNK

Broods/

Ducklings

(stage if

known)

Total

Post

Fledge

LADU

*HY

Banded

HY

Unbanded

Notes

Monument

Seep

Rolando Seep

Sunset Seep

Other

Totals


August 2010 52

ATTACHMENT 2. MIDWAY ATOLL LAYSAN TEAL SURVEY START LOCATIONS AND

DIRECTION OF TRAVEL TABLES

Eight starting points and directions for Sand Island Standard Resight Surveys of All

Wetlands (use in the following order, then restart):

Starting point

Direction

Date

Date

1 Catchment Clockwise

2 Brackish Pond Counter

3 Catchment Counter

4 Rusty Bucket Counter

5 Fuel Farm Clockwise

6 Ball field Counter

7 Brackish Pond Clockwise

8 Rusty Bucket Clockwise

Example of clockwise wetlands: Catchment, Mauka-Makai, Ironwood, Aviary, Rusty

Bucket, Radar Hills, Ball Fields, Fuel Farm, Brackish Pond, Sunrise, Communications,

Parade, R2.

Six starting points and orders for Eastern Island Resight (and All-Wetlands Count)

(use in the following order, then restart):

Start point

Middle point Finish point

Resight

Survey

Date

Wetland

Count

Date

1 Sunset

Monument Rolando

2 Rolando Monument Sunset

3 Monument Rolando Sunset

4 Sunset Rolando Monument

5 Rolando Sunset Monument

6 Monument Sunset Rolando

Attachment 3. Maps illustrating survey routes on Sand Island & Eastern Island. Original Pilot Study: Sand Island Laysan Teal Standard Resight Survey Route

53

Catchment

Mauka-Makai

Radar Hill 1

Radar Hill 2

Ballfield Little

Ballfield big

Brackish Pond

Sunrise

Attachment 3. Sand Island Laysan Duck All-Wetlands Count Route/Revised All-Wetlands Resight Route

54

Catchment

Mauka-Makai

Ironwood

Aviary

Rusty Bucket

Radar Hill 1

Radar Hill 2

Ball field Little

Ball field Big

Fuel Farm

Dump (if water) Brackish Pond

Brackish Adj.

Wetland

Brackish

Forest Dep.

Sunrise

Communications

Parade Tarmac Pool (if water)

R2 Treatment Pools

R2 Wetland

Sand Island New Resight Survey Route

Attachment 3. Eastern Island Original Laysan Duck Survey Route

55

Monument

Rolando

Sunset

ATTACHMENT 4. MIDWAY ATOLL LAYSAN TEAL RESIGHT REFERENCE,

MIDWAY_RESIGHT_REFERENCE_20AUG2010.XLS

Midway

LINK # RLEG LLEG SEX

2010

AGE USFWS ORIGIN/REHAB

NOTES – MIA ducks require careful confirmation if

seen (check sex, correct code, legs, read AL band if

possible).

1 ON AL M ASY 896-37943 2004 Founder Seen 17-Feb-10

2 AL W8 M ASY 896-37941 2004 Founder Seen 25-Jun-10

3 NONE AL M ASY 805-70841 2004 Founder Seen 02-Jul-10

7 AL GN/2WG M ASY 805-70851 2004 Founder MIA; Last seen 11-Mar-08

8 AL o-temp M ASY 896-37956 2004 Founder Glued Temp; confirm sex duplicate band on female; Seen

21-Mar-09

9 NONE AL M ASY 896-37958 2004 Founder Read band; Confirmed; Seen 28-Jun-10

11 BK AL M ASY 805-70850 2004 Founder Seen 16-Dec-09

12 GY5 AL M ASY 805-70856 2004 Founder MIA; worn color band; Last seen 18-May-07

13 AL WWE M ASY 1056-99915 2004 Founder Was AL/BF; Seen 04-Aug-10

14 lb-temp AL F ASY 995-16185 2004 Founder Seen 23-Jun-10

16 Y2 AL F ASY 805-70858 2004 Founder Seen 14-Jul-10

17 OF AL F ASY 805-70837 2004 Founder Worn color band; Seen 23-Mar-09

19 AL Y/BKG M ASY 805-70861 2004 Founder Seen 07-Jul-10

21 W$ AL F ASY 896-37944 2005 Founder Seen 11-Jan-10

22 AL BKY# F ASY 896-37942 2005 Founder Was O* on Eastern; Seen 20-Apr-10

23 AL Y# F ASY 1035-44062 2005 Founder Seen 17-Feb-10

24 AL W6 M ASY 896-37951 2005 Founder New W6 band given in Oct 09; Seen 09-Mar-10

25 GY+ AL F ASY 1056-99809 2005 Founder Seen 29-Jul-09

26 AL NONE

(?) M ASY 896-37945 2005 Founder

MIA; Lost color band GYK?; Read band; Last seen 23-

Oct-07

27 BY3 AL M ASY 896-37987 2005 Founder Was BY2; Seen 18-Feb-10

28 B? AL M ASY 1056-99808 2005 Founder

MIA since Jan 08;. Could be confused w/ BY? Identify

legs, sex, age, color of symbol to confirm alive. Read AL

band if possible.

32 YR AL M ASY 856-89490 2005 Founder Worn color band; Seen 28-Oct-09

36 B- AL F ASY 1035-44033 2005 Founder Worn color band, white head; Seen 11-Jan-10

37 WT NONE F ASY NONE 2005 Founder Seen 28-Jul-10

38 BYA AL F ASY 1035-44022 2005 Founder MIA; Last seen 29-Nov-08

39 WC AL M ASY 805-70826 2005 Founder Worn color band; Seen 09-Dec-09

40 WMA NONE F ASY NONE 2005 Founder MIA; Was YJ; New band combo Nov 2007; Last seen

08-Nov-07

42 NONE B$ F ASY NONE 2005 Founder MIA; Last seen 03-Oct-08

44 AL DB F ASY 896-37957 F1 2005 Read AL; Permanently applied temp color band; was

BYE; Seen 08-Jun-10

45 AL BYF M ASY 896-37939 F1 2005 Seen 16-Dec-09

46 AL BYJ M ASY 896-37902 F1 2005 Seen 15-Dec-09

Attachment 4. Midway Atoll Laysan Duck Resight Reference as of 20 August 2010

57

Midway


2010




possible).

51 Y4 AL M ASY 896-37905 F1 2005 Seen 23-Jun-10

52 YW AL M ASY 1035-94103 F1 2005 MIA; Last seen 09-Mar-07

53 NONE Y0 F ASY NONE F1 2005 Seen 18-Feb-10

54 AL BKY7 F ASY 896-37950 F1 2005 Was BR7; Seen 13-Jul-10

57 AL GYC M ASY 1035-94106 F1 2006 Color band upside down; Seen 14-Jul-10

59 GYE AL F ASY 1035-94107 F1 2006 MIA; Last seen 04-Oct-07

60 AL BR/2WG F ASY 1035-94109 F1 2006 Seen 10-Feb-10

62 GN1 NONE M ASY NONE F1 2006 Seen 11-Aug-10

63 AL R* F ASY 896-37986 F1 2006 Seen 15-Jul-10

64 AL BR+ F ASY 1035-94113 F1 2006 Seen 04-Feb-10

65 AL W M ASY 1035-94114 F1 2006 MIA; Last seen 07-Nov-07

66 AL NONE F ASY 1035-94115 F1 2006 MIA; Need to read band; Last seen 29-Oct-09

68 AL OE F ASY 1035-94117 F1 2006 MIA; Last seen 06-Oct-06

69 AL r-temp M ASY 1035-94118 F1 2006 Maybe on right leg? Seen 05-Oct-09

70 AL OK F ASY 1035-94119 F1 2006 Seen 11-Mar-08

71 AL OM F ASY 896-37959 F1 2006 Seen 23-Mar-09

72 AL OP U ASY 1035-94121 F1 2006 MIA; Last seen 06-Oct-06

73 YA AL F ASY 1035-94122 F1 2006 YA also on female LL; Seen 09-Nov-09

74 NONE B1 F ASY NONE F1 2006 Seen 10-Feb-10

75 AL YC U ASY 1035-94124 F1 2006 Seen 01-Jul-10

76 AL GNA F ASY 1035-94125 F2 2006 May look like a GN4; Seen 01-Apr-10

77 GNC AL M ASY 896-37906 F2 2006 AL read Nov. 9, 2009; Seen 15-Jul-10

79 GNE AL F ASY 896-37908 F1 2006 MIA; Worn band; Last seen 15-Apr-08

80 GNF AL F ASY 896-37909 F1 2006 Seen 05-Jul-10

81 GNH AL F ASY 896-37910 F1 2006 Lots of white on head; Faded band; Seen 11-Jan-10

82 AL YE F ASY 896-37911 F1 2006 Seen 14-Jul-10

83 AL BY+ M ASY 896-37912 F1 2006 White head; Seen 11-Jan-10

84 AL BY$ M ASY 896-37913 F1 2006 Color band worn; Seen 16-Nov-09

86 AL BY& M ASY 896-37915 F1 2006 Seen 28-Jun-10

89 YF AL M ASY 896-37917 F1 2006 Worn band; Seen 11-Jan-10

90 AL NONE M ASY 896-37919 F1-2 2006 Was GN2; Seen 29-Dec-09

91 AL GNJ M ASY 896-37920 Rehab 2008 Held for 48 hrs during Botulism in 2008; F1-2 2006;

Seen 08-Dec-09

92 BRX AL M ASY 896-37921 F1-2 2006 Seen 26-Jul-10

93 BRK AL F ASY 896-37922 F1 2006 Seen 14-Mar-09

94 NONE AL F ASY 896-37923 F1 2006 MIA; Read band; Broken thumb feather; Last seen 12-

Nov-08

95 O> AL M ASY 896-37924 F1-2 2006 White head; Seen 23-Jun-10


58

Midway


2010




possible).

98 B5 AL F ASY 896-37927 F1-2 2006 Very white head; Band worn; Seen 23-Jun-10

99 B6 AL F ASY 896-37928 F1-2 2006 MIA; Could be confused with BY6; Last seen 21-Mar-08

100 B7 AL F ASY 896-37929 F1-2 2006 Seen 06-Nov-09

101 B9 AL M ASY 896-37930 F1-2 2006 MIA; Last seen 03-Sep-08

103 GNT AL M ASY 896-37932 F1-2 2006 Bill tan at tip; Seen 01-Apr-10

105 AL YK M ASY 896-37934 F1 2006 Worn band; Seen 26-May-10

107 AL YP F ASY 896-37936 F1 2006 Seen 14-Jul-10

109 AL YX M ASY 896-37938 F1 2006 Band worn; Seen 30-Jun-10

110 AL YU F ASY 896-37940 F1 2006 Seen 14-Jul-10

111 AL BYM M ASY 896-37946 F1 2006 Seen 17-Jan-09

112 AL BYP F ASY 896-37947 F1 2006 Seen 23-Jun-10

113 YY AL F ASY 896-37948 F1 2006 MIA; Last seen 01-Mar-08

114 AL GN3 M ASY 896-37949 F2 2006 Seen 16-Jul-10

122 AL GNX$ F ASY 1056-99939 F1-2 2007 Glued temps-high band loss; Seen 28-Apr-09

126 AL WWP F ASY 896-37953 F1-2 2007 Seen 07-Jul-10

127 AL WWK F ASY 896-37954 F1-2 2007 Seen 29-Jun-10

128 AL WWT F ASY 896-37955 F1-2 2007 Seen 02-Jul-10

129 GYT AL M ASY 1036-14852 F1-2 2007 Seen 30-Jun-10

130 GNXU AL M ASY 1056-99978 F1-2 2007 Seen 23-May-09

132 AL WMXJ F ASY 1036-14838 F1-2 2007 Was y-temp/dgn-temp; Seen 18-Feb-09

137 AL WW4 M ASY 1056-99926 F1-2 2007 Was w-temp/y-temp; Seen 13-Jul-10

139 GNX- AL F ASY 1056-99936 F1-2 2007 Was hp-temp/hp-temp; Seen 01-Dec-09

140 YC8 AL M ASY 1056-99941 F1-2 2007 Was y-temp/lgn-temp, then just y-temp; Seen 23-Jun-10

149 AL GY\ F ASY 1036-14811 Rehab 2008 F1-2 2007; Seen 05-Oct-09

150 AL WMK M ASY 896-37591 F1-2 2007 New bands, replaced hp-temp, 0-temp; Seen 08-Oct-09

154 GN? AL F ASY 1036-14849 F1-2 2007 Was w-temp/hp-temp; Seen 27-Jul-09

157 RJ AL F ASY 896-37962 F1-2 2007 Seen 03-Jul-10

158 R0 AL F ASY 896-37963 F1-2 2007 Seen 21-Jul-10

160 AL RR M ASY 896-37965 F1-2 2007 Seen 14-Jul-10

161 AL R8 F ASY 896-37966 F1-2 2007 Seen 21-Jan-10

162 AL R6 F ASY 896-37967 F1-2 2007 Seen 04-Aug-10

163 R/WG AL M ASY 896-37968 F1-2 2007 Seen 23-Feb-10

164 AL R2 M ASY 896-37969 F1-2 2007 Seen 15-Jun-10

165 GYU AL F ASY 896-37970 F1-2 2007 MIA; Last seen 21-Mar-08

166 AL RH F ASY 896-37971 F1-2 2007 Confirm right leg and left leg; Possible recording error;

Seen 10-Aug-10

167 AL RN F? ASY 896-37972 F1-2 2007 MIA; Last seen15-Dec-08


59

Midway


2010




possible).

168 AL RU F ASY 896-37973 F1-2 2007 Seen 04-Feb-10

170 AL R# M ASY 896-37975 F1-2 2007 Seen 22-Jul-10

171 AL R$ M ASY 896-37976 F1-2 2007 Seen 11-Jan-10

172 GYN AL M ASY 896-37977 F1-2 2007 MIA; Last seen 15-Dec-08

173 R+ AL F ASY 896-37978 F1-2 2007 Seen 21-Jul-10

174 RP AL M ASY 896-37979 F1-2 2007 Seen 14-Mar-09

175 RX RW M ASY NONE F1-2 2007 RW upside down? Seen 04-Aug-10

176 R3 AL F ASY 896-37980 F1-2 2007 Seen 04-Aug-10

177 AL RY F ASY 896-37981 F1-2 2007 Seen 04-Feb-09

178 RA AL F ASY 896-37982 F1-2 2007 Seen 11-Jan-10

184 WW1 AL M ASY 1056-99917 F1-2 2007 Seen 28-Oct-09

186 BY? AL M ASY 1056-99948 F1-2 2007 Was dgn-temp/none; Seen 01-Jul-10; Confirm could be

confused with B?

190 AL WM5 M ASY 1056-99969 F1-2 2007 Removed GNXP very worn; Seen 07-Jul-10

192 AL OX0 F ASY 1036-14827 F1-2 2007 Seen 04-Aug-10

194 BRE AL F ASY 1056-99963 F1-2 2007 Glued temps-high band loss; Seen 08-Jan-10

196 AL YX9 M ASY 1056-99935 F1-2 2007 Seen 27-Mar-10

198 AL YC9 F ASY 1056-99992 F1-2 2007 Replaced temp O on LL; None RL; Seen 07-Jul-10

200 AL R> M ASY 896-37982 F1-2 2007 Seen 12-Jul-10

201 R5 AL F ASY 896-37983 F1-2 2007 Seen 14-Jul-10

203 AL BK/2YG F ASY 896-37989 F1-2 2007 Seen 25-Jul-10

205 AL BKY+ F ASY 896-37991 F1-2 2007 Seen 23-Dec-09

206 BKYA AL F ASY 896-37992 F1-2 2007 Seen 17-Feb-10

207 AL W/2MG F ASY 896-37993 F1-2 2007 Seen 30-Mar-10

208 WM# AL M ASY 896-37994 F1-2 2007 Check legs! Same color band on opposite leg of different

bird; Seen 11-Jan-10

209 AL BKY> F ASY 896-37995 F1-2 2007 MIA; Last seen 18-Mar-08

211 AL BKY1 F ASY 896-37996 F1-2 2007 Seen 22-Apr-10

212 WM> AL U ASY 896-37997 F1-2 2007 Seen 20-Apr-10

213 WMJ AL M ASY 896-37998 F1-2 2007 Seen 13-Jul-10

214 WM* AL F ASY 896-37999 F1-2 2007 Seen 14-Jul-10

215 WM/ AL M ASY 896-38000 F1-2 2007 Seen 12-Jul-10

217 AL WM- M ASY 1056-99851 F1-2 2007 Seen 11-Aug-10

218 AL BKYM M ASY 1056-99849 F1-2 2007 Seen 29-Dec-09

220 AL WMM M ASY 1056-99852 F1-2 2007 Seen 04-Feb-09

221 BKYT AL M ASY 1056-99853 F1-2 2007 Faded band; Seen 14-Jul-10

222 BKY- AL F ASY 1056-99855 F1-2 2007 Faded band; Seen 23-Jul-10

223 AL WM0 M ASY 1056-99857 F1-2 2007 Seen 15-Dec-09


60

Midway


2010


NOTES – MIA birds require careful confirmation if


possible).

224 BKYH NONE M ASY NONE F1-2 2007 No band on LL; Worn band; Seen 09-Jul-10

225 AL BKY0 F? ASY 1056-99856 F1-2 2007 MIA; Last seen 12-Nov-08

226 WMH NONE M ASY NONE F1-2 2007 Seen 23-Feb-10

227 AL BKY* F ASY 1035-94130 F1-2 2007 5 aluminum, worn band; Seen 08-Jan-10

228 AL WMU F? ASY 1056-99858 F1-2 2007 MIA; Last seen 10-Mar-08

230 AL WM# M? ASY 1056-99859 F1-2 2007 Seen 14-Jul-10

231 BKY3 AL F ASY 1056-99860 F1-2 2007 Some wear; Seen 09-Jul-10

233 BKY8 NONE F ASY NONE F1-2 2007 Seen 06-Jan-10

235 AL BKYR F ASY 976-20001 Rehab 2008 F1-3; Seen 28-Jul-10

236 AL BKYX U ASY 967-20006 Rehab 2008 F1-3 2008; Seen 04-Aug-10

238 AL WM7 F ASY 967-20002 Rehab 2008 F1-3 2008; Seen 17-Feb-10

239 BKY/ AL F ASY 967-20003 Rehab 2008 F1-2 2007; Seen 05-Jul-10

240 AL R4 F ASY 967-20004 Rehab 2008 F1-3 2008; Seen 28-Jul-10

241 AL WMN F ASY 967-20010 Rehab 2008 F1-3 2008; Faded badly; Seen 14-Jul-10

243 BKY$ AL U ASY 1036-14802 Rehab 2008 F1-3 2008; Seen 03-Sep-09

244 AL BKYY M ASY 1036-14801 Rehab 2008 F1-3 2008; Squinty eye; Seen 28-Sep-09

245 BKY6 AL U ASY 1036-14803 Rehab 2008 F1-3 2008; Seen 05-Oct-09

246 BKY? AL F ASY 1036-14804 Rehab 2008 F1-2 2007; Seen 25-Jun-10

248 BKY+ AL M ASY 1036-14806 Rehab 2008 and

2009 F1-3 2008; Seen 11-Aug-10

249 BK/2YG AL M ASY 1036-14807 Rehab 2008 Duplicate combo on opposite leg; F1-2 2007; Seen 11-

Aug-10

251 R/2WG AL F ASY 1036-14809 Rehab 2008 Only resighted once in field; F1-2 2007; Seen 14-Sep-09

252 AL GY1 F ASY 1036-14810 Rehab 2008 Upside down color band; F1-2 2007; Seen 12-Jul-10

254 WM1 AL M ASY 1036-14812 Rehab 2008 F1-3 2008; Seen 08-Jun-10

255 GY4 AL M ASY 1036-14813 Rehab 2008

F1-3 2008; Needs band confirmation, possible data error;

Band could be GYA; Need to check and read AL; Seen

22-Jan-10

255 GYA AL M ASY 1036-14813 Rehab 2008 F1-3 2008; Band could be GY4; Need to check and read

AL; Seen 25-Jun-10

256 RT AL M ASY 1036-14814 Rehab 2008 F1-3 2008; Seen 13-Jul-10

257 AL WY F ASY 896-37593 Rehab 2008 F1-3 2008; Replaced lb-temp/None with permanent

bands; Seen 14-Jul-10

258 LGN AL F ASY 1036-14815 Rehab 2008 Glued temp; F1-2 2007; Seen 05-May-10

259 FU AL M ASY 1036-14816 Rehab 2008 F1-3 2008; Seen 08-Jan-10

260 Y AL M ASY 1036-14818 Rehab 2008 F1-3 2008; Seen 10-Aug-10

261 OR AL M ASY 1036-14819 Rehab 2008 Replaced unglued o-temp with OR; F1-3 2008; Seen 10-

Aug-10

262 R AL F ASY 1036-14820 Rehab 2008 F1-3 2008; Glued temp; Seen 30-Mar-10

263 DGN AL M ASY 1036-14821 Rehab 2008 F1-3 2008; Glued temp; Cloudy left eye; Seen 29-Dec-09

264 AL FU F ASY 1036-14822 Rehab 2008 Glued temp; F1-2 2007; Seen 17-Nov-09

265 AL O F ASY 1036-14823 F1-2 2007 Confirm sex; Duplicate band on ASY M; Seen 14-Jul-10


61

Midway


2010




possible).

267 AL WMX- F ASY 1056-99877 F1-3 2008 Seen 16-Jul-10

268 GNX0 AL F ASY 1036-14828 F1-3 2008 Seen 27-Jan-10

269 OX1 AL M? ASY 1056-99878 F1-3 2008 Seen 10-Jul-10

270 OX2 AL M ASY 1056-99879 F1-2 2007 Seen with marine debris March 2009; Seen 14-Jul-10

272 GNX2 AL M ASY 1056-99881 F1-3 2008 Seen 14-Jul-10

273 AL WMX0 M ASY 1056-99882 F1-3 2008 Seen 09-Aug-10

274 AL OX3 M ASY 1056-99883 F1-3 2008 Seen 07-Jul-10

275 OX4 AL M ASY 1056-99884 F1-2 2007 Seen 03-Jun-09

276 GNX3 AL M ASY 1056-99885 F1-2 2007 Seen 22-Jul-10

278 GYY AL F ASY 1056-99887 Rehab 2009 Mistakenly reported as carcass; F1-2 2007; Seen 08-Aug-

10

279 AL GY2 M ASY 1056-99888 F1-2 2007 Seen 14-Jul-10

280 GY- AL F ASY 1056-99889 F1-2 2007 Seen 11-Aug-10

281 AL GY3 F ASY 1056-99890 F1-2 2007 Confirm this bird; worn color band looks white?; Seen

18-Feb-10

282 AL WMX1 F ASY 1056-99891 F1-3 2008 Seen 07-Aug-10

283 AL WMX2 M ASY 1056-99892 F1-2 2007 Seen 11-Aug-10

284 AL GY6 M ASY 1056-99893 F1-2 2007 Seen 14-Jul-10

286 AL OX5 F ASY 1056-99895 F1-2 2007 Seen 29-Aug-09

289 GY/BKG AL M ASY 1056-99898 F1-2 2007 Seen 23-Jun-10

290 GNX5 AL F ASY 1056-99899 F1-2 2007 Seen 03-Jul-10

291 AL WMXA M ASY 1036-14832 F1-2 2007 Seen 07-Jul-10

296 AL WMXC M ASY 1036-14833 F1-3 2008 Limp; Seen 23-Jun-10

297 GNX8 AL F ASY 1036-14834 F1-2 2007 Seen 30-Jun-10

299 AL WMXE M ASY 1036-14835 F1-3 2008 Seen 14-Jul-10

300 WMXH AL F ASY 1105-11703 F1-3 2008 New AL band on 03-Oct-09; Seen 12-Dec-09

302 WMXK AL M ASY 1036-14840 F1-3 2008 Seen 21-Apr-09

303 NONE O1 F ASY NONE F1-2 2007 Seen 24-Oct-09

304 WMXN AL F ASY 1036-14842 F1-3 2008 Seen 15-Jul-10

305 WMXP AL M ASY 1036-14843 F1-2 2007 Seen 10-Aug-10

306 OU o-temp F ASY NONE F1-3 2008 Seen 29-Jun-10

307 AL GNN M ASY 1036-14886 F1-2 2007 Seen 11-Jan-10

308 AL O$ F ASY 1036-14887 F1-3 2008 Seen 01-Jun-10

309 GN> db-temp M? ASY NONE F1-3 2008 MIA; Glued Temp; Last seen 26-Nov-08

310 GN+ AL F ASY 1036-14865 F1-3 2008 Seen 25-Jul-10

311 OX9 AL M ASY 1036-14866 F1-2 2007 Seen 09-May-10

312 GN* AL F ASY 1036-14867 F1-2 2007 Seen 12-Mar-09


62

Midway


2010




possible).

313 WM4 AL F ASY 1036-14868 F1-2 2007 Seen 24-Oct-09

314 O? AL M ASY 1036-14869 F1-2 2007 Seen 14-Jul-10

315 AL GNR M ASY 1036-14870 F1-2 2007 Seen 14-Jul-10

316 AL WMX+ F ASY 1036-14871 F1-2 2007 Seen 25-Feb-09

317 AL O# M ASY 1036-14872 F1-2 2007 Band worn; Seen 30-Jun-10

318 OX AL M ASY 1036-14873 F1-3 2008 Lots of white on head; Seen 25-Mar-10

320 AL WM3 F ASY 1056-99859 F1-2 2007 Seen 18-Feb-10

321 AL WWX M ASY 1056-99900 F1-2 2007 Check for limp; Seen 05-Jul-10

322 AL GNXA F ASY 1056-99901 F1-2 2007 Worn color band; Seen 23-Jun-10

323 AL BY2 F ASY 1056-99902 F1-3 2008 Seen 14-Jul-10

324 AL WW+ F ASY 1036-14854 F1-2 2007 Last seen 28-Jul-10

325 GN/WG AL M ASY 1036-14844 F1-3 2007/2008 Seen 11-Jan-10

326 AL GNXC M ASY 1036-14850 F1-3 2008 Seen 02-Dec-09

327 AL GN$ F ASY 1036-14855 F1-2 2007 Seen 06-Jan-10

328 WMXU AL F ASY 1105-11701 F1-3 2008 Seen 02-Apr-09

329 AL R? F ASY 1036-14856 F1-3 2008 Confirm sex of bird; Duplicate combo on male; Seen 17-

Mar-10

330 AL WMX4 F ASY 1036-14857 F1-2 2007 Seen 09-Mar-10

331 YC1 AL M ASY 1036-14845 F1-3 2007/2008 Seen 13-Jul-10

332 GY? AL M ASY 1036-14846 F1-3 2007/2008 Seen 08-Jun-10

333 AL R- F ASY 1036-14858 F1-3 2007/2008 Seen 30-Mar-10

334 AL YC4 M ASY 1036-14859 F1-2 2007 Seen 28-Jul-10

335 WMX5 NONE F ASY NONE F1-2 2007 Seen 11-Jan-10

336 AL OT F ASY 1036-14860 F1-3 2008 Confirmed sex is F; Some green in bill; Seen 15-Jun-10

337 AL YC5 M ASY 1036-14861 F1-2 2007 Seen 07-Jul-10

339 AL O- M ASY 1036-14862 F1-3 2007/2008 Seen 11-Aug-10

340 AL BRT F ASY 1036-14847 F1-3 2007/2008 Seen 14-Jul-10

341 AL WMXY M ASY 1036-14848 F1-2 2007 Seen 07-Jul-10

342 AL WMXT F ASY 1036-14853 F1-3 2007/2008 Seen 08-Sep-09

343 WMXR AL M ASY 1056-99903 F1-3 2007/2008 Seen 05-Jan-10

345 GYH AL M ASY 1056-99905 F1-3 2007/2008 Seen 16-Jul-10

346 WW- AL M ASY 1056-99906 F1-3 2007/2008 Seen 07-Jul-10

347 WW> AL M ASY 1056-99907 F1-3 2007/2008 Seen 10-Aug-10

348 AL WW7 M ASY 1056-99908 F1-3 2007/2008 Seen 14-Jul-10

349 AL WWY M ASY 1056-99909 F1-3 2007/2008 Seen 07-Aug-10

350 AL BR$ F ASY 1056-99910 F1-3 2007/2008 Added BR$ on 29-Oct-09; Prior to this only AL; Seen

11-Jan-10

351 AL YXP F ASY 1056-99911 F1-3 2007/2008 Seen 29-Jun-10


63

Midway


2010




possible).

352 AL WW5 F ASY 1056-99912 F1-3 2007/2008 Seen 22-May-09

353 YXE AL F ASY 1056-99913 F1-3 2007/2008 Seen 16-Apr-10

354 AL WWA M ASY 1056-99914 F1-3 2007/2008 Seen 24-Nov-09


356 YX1 AL M ASY 1056-99919 F1-3 2007/2008 Seen 11-Aug-10

357 GNX> AL F ASY 1056-99920 F1-3 2007/2008 Seen 29-Dec-09

358 AL YX2 M ASY 1056-99921 F1-3 2007/2008 This duck is blind; Seen 09-Dec-09

359 YX3 AL M ASY 1056-99922 F1-3 2007/2008 Confirm band combination; Legs may be reversed; Seen

01-Jun-10

360 AL YX4 M ASY 1056-99923 F1-3 2007/2008 Seen 30-Jun-10

361 YX5 AL M ASY 1056-99924 F1-3 2007/2008 Seen 07-Mar-10

362 WW3 AL M ASY 1056-99925 F1-3 2007/2008 Seen 14-Jul-10




366 AL GNX> F ASY 1105-11702 F1-3 2007/2008 Seen 06-May-09

367 GNX? AL F ASY 1056-99930 F1-3 2007/2008 Badly faded; Seen 20-Apr-10

368 YX6 AL M ASY 1056-99931 F1-3 2007/2008 Seen 23-Feb-10

369 AL GNX+ F ASY 1056-99932 F1-3 2007/2008 Seen 01-Jul-10

371 YX8 AL M ASY 1056-99934 F1-3 2007/2008 Seen 07-Jul-10

372 AL WWJ M ASY 1056-99937 F1-3 2007/2008 Seen 22-Apr-10

373 YX> AL F ASY 1056-99938 F1-3 2007/2008 Seen 08-Sep-09

374 AL YC7 M ASY 1056-99940 F1-3 2007/2008 Seen 28-Apr-09

375 AL YC0 M ASY 1056-99942 F1-3 2007/2008 Seen 11-Aug-10

376 AL YXJ F ASY 1056-99943 F1-3 2007/2008 Seen 11-Jan-10

377 YXR AL M ASY 1056-99944 F1-3 2007/2008 Seen 18-Feb-10

378 BYN AL M ASY 1056-99945 F1-3 2007/2008 Seen 14-Jul-10

379 AL BRC F ASY 1056-99946 F1-3 2007/2008 Seen 12-Sep-09

380 AL O/2BKG M ASY 1056-99947 F1-3 2007/2008 Seen 11-Aug-10

381 AL YX+ M ASY 1056-99849 F1-3 2007/2008 Faded band; Seen 17-Mar-10

382 AL RK F ASY 1056-99850 F1-3 2007/2008 Seen 12-Dec-09

383 AL GNXJ M ASY 1056-99851 F1-3 2007/2008 Very faded band; Read AL band; Seen 25-Nov-09; Needs

to be recaptured and band replaced

384 AL BR- F ASY 1056-99852 F1-3 2007/2008 Seen 05-May-09

385 AL BY- M ASY 1056-99853 F1-3 2007/2008 Could be confused with B-; Seen 11-Jul-10

386 BY5 AL M ASY 1056-99854 F1-3 2007/2008 Could be confused with B5; Seen 13-Nov-09

387 AL BY6 M ASY 1056-99855 F1-3 2007/2008 Confirmed as BY9 put on upside-down; Seen 01-Jul-10

388 OBKG AL M ASY 1056-99856 F1-3 2007/2008 Check bands on correct legs; Seen 14-Jul-10


64

Midway


2010




possible).

389 AL BRA F ASY 1056-99857 F1-3 2007/2008 Seen 18-Feb-10

390 YX- AL M ASY 1056-99858 F1-3 2007/2008 Seen 11-Aug-10

391 AL BRJ F ASY 1056-99859 F1-3 2007/2008 Seen 11-Jan-10

393 BRN AL F ASY 1056-99861 F1-3 2007/2008 Seen 08-Jan-10

394 AL BRR F ASY 1056-99862 F1-3 2007/2008 Seen 14-Jul-10

395 BYW AL M ASY 1056-99864 F1-3 2007/2008 Seen 14-Jul-10

396 AL BRH F ASY 1056-99865 F1-3 2007/2008 Seen 10-Feb-10

397 BY7 AL M ASY 1056-99866 F1-3 2007/2008 Could be confused with B7; Seen 16-Jul-10

398 AL BY* M ASY 1056-99867 F1-3 2007/2008 Could be confused with B*; Seen 10-Aug-10

399 YX$ AL M ASY 1056-99868 F1-3 2007/2008 Seen 04-Aug-10

400 AL YC6 M ASY 1056-99870 F1-3 2007/2008 Seen 21-Jul-10

401 WMX# AL M ASY 1056-99971 F1-3 2007/2008 Seen 09-Jul-10

402 AL GNXH F ASY 1056-99972 F1-3 2007/2008 Worn band; Seen 08-Jun-10

403 AL WWH M ASY 1056-99973 F1-3 2007/2008 Seen 11-Jan-10

404 WW0 AL M ASY 1056-99974 F1-3 2007/2008 White head; Seen 07-Jul-10

405 GN# AL F ASY 1056-99975 F1-3 2007/2008 Seen 07-Jul-10

406 BRP AL F ASY 1056-99976 F1-3 2007/2008 Seen 23-Jun-10

407 AL GNXE M ASY 1056-99977 F1-3 2007/2008 Very worn band; Seen 07-Jul-10

408 AL BR> F ASY 1056-99979 F1-3 2007/2008 Seen 25-May-09

409 GN AL F ASY 1056-99980 F1-3 2007/2008 Replaced GN5 with GN on 01-Oct-09; Seen 17-Feb-10

410 GNXT AL M ASY 1056-99981 F1-3 2007/2008 Band very faded-mostly white; Seen 08-Jun-10

411 AL GNXL U HY 1056-99982 Rehab 2009 MIA; F1-3 2008; Last seen 16-Jun-09

412 AL WMX6 M SY 1056-99983 Rehab 2009 Replaced worn GNXR band with WMX6 on 03-Oct-09;

F1-3 2008; Seen 14-Jul-10

413 AL GNXM F ASY 1056-99984 Rehab 2009 Very faded band; F1-3; Seen 17-Mar-10

415 AL GNW$ U HY 1056-99986 Rehab 2009 Band faded, mostly white; May look like WW$; Seen 09-

May-10

416 AL GY8 F? SY 896-37534 F1-3 2007/2008 Seen 29-Sep-09

417 AL WWC U SY 1056-99987 Rehab 2009 MIA; F1-3 2008; Last seen 26-Aug-09

429 WA AL M ASY 1086-89201 Rehab 2009 F1-3 2008; Seen 15-Jun-10

430 AL GNX M? SY 1056-99989 Rehab 2009 F1-4 2009; Confirm sex; Seen 04-Aug-10

431 AL GNU M SY 1056-99990 Rehab 2009 F1-4 2009; Seen 14-Jul-10

432 AL GN1 M SY 1056-99991 Rehab 2009 F1-4 2009; Seen 10-Aug-10

434 AL BYE M ASY 1056-99993 F1-3 2007/2008 Could be confused with BE; Band worn; Seen 11- Aug-

10

435 AL BX U SY 1056-99994 F1-4 2009 Need to confirm sex of this bird; Could be confused with

BYX; Seen 17-Mar-10

436 AL BP U SY 1056-99995 F1-4 2009 Could be confused with BYP; Seen 07-Jul-10

437 AL BA F SY 1056-99996 F1-4 2009 Could be confused with BYA; Seen 21-Jul-10

438 AL BM F SY 1056-99997 F1-4 2009 Could be confused with BYM; Seen 08-Sep-09


65

Midway


2010




possible).

439 AL BC F SY 1056-99998 F1-4 2009 Could be confused with BYC; Seen 20-Apr-10

440 AL BE M SY 1056-99999 F1-4 2009 Could be confused with BYE; Seen 21-Jul-10

441 AL BF M SY 1056-00000 F1-4 2009 Could be confused with BYF; Seen 04-Jul-10

442 AL BH U SY 896-37501 F1-4 2009 Could be confused with BYH; Seen 23-Jun-10

443 AL BJ F SY 896-37502 F1-4 2009 Seen 08-Jun-10

444 AL BR M ASY 896-37503 F1-3 This is a blue band with a white R; Seen 12-Dec-09

445 AL BR F SY 896-37504 F1-4 2009 This is blue with a white R, not temp brown; Seen 09-

Nov-09

446 AL BT U HY 896-37505 F1-4 2009 Confirm sex; Could be confused with BYT; Seen 07-Jul-

10

447 AL R? M SY 896-37506 F1-4 2009 Confirm sex; Duplicate combo on female; Seen 07-Jul-10

448 AL BW F SY 896-37507 F1-4 2009 Could be confused with BYW; Seen 14-Jul-10

449 AL BY M SY 896-37508 F1-4 2009 Could be confused with BYY; Seen 23-Jul-10

450 AL B2 F ASY 896-37509 F1-3 2008 Could be confused with BY2; Seen 30-Mar-10

451 AL B3 M SY 896-37510 F1-4 2009 Could be confused with BY3; Seen 11-Aug-10

452 AL B4 M ASY 896-37511 F1-3 2008 Could be confused with BY4; Seen 30-Dec-09

453 AL B8 M SY 896-37512 F1-4 2009 Could be confused with BY8; Seen 10-Aug-10

454 AL B0 U SY 896-37513 F1-4 2009 Could be confused with BY0; Seen 07-Jul-10

455 AL 0* M SY 896-37514 F1-4 2009 Recycled combo; Missing right tip of bill; Seen 14-Jul-10

456 AL OA F ASY 896-37515 F1-3 Seen 11-Jan-10

457 AL OC F ASY 896-37516 F1-3 2008 Seen 04-Aug-10

458 AL OJ F? SY 896-37517 F1-4 2009 Confirm sex; Seen 20-Apr-10

459 AL OY F ASY 896-37518 F1-3 Seen 21-Jul-10

460 AL O2 M SY 896-37519 F1-4 2009 Seen 30-Jun-10

461 AL O3 F? SY 896-37520 F1-4 2009 Seen 09-Jul-10

462 AL O5 F SY 896-37521 F1-4 2009 Seen 12-May-10

463 AL O6 M SY 896-37522 F1-4 2009 Seen 14-Jul-10

464 AL O7 F ASY 896-37523 F1-3 2008 Seen 14-Jul-10

465 AL O8 M SY 896-37524 F1-4 2009 MIA; Last seen 19-Sep-09

467 AL GYA F SY 896-37526 F1-4 2009 White head; Seen 27-Apr-10

468 AL GYF M SY 896-37527 Rehab 2009 F1-4 2009; Seen 23-Jun-10

469 AL GYK M ASY 896-37528 F1-4 Seen 30-Mar-10

470 AL GYM M ASY 896-37529 F1-4 Seen 28-Sep-09

471 AL GYR U SY 896-37530 F1-4 2009 Seen 21-Sep-09

472 AL GYW M ASY 896-37531 F1-4 Seen 10-Aug-10

473 AL GY6 M SY 896-37532 F1-4 2009 Seen 27-Mar-10

474 AL GY7 F ASY 896-37533 F1-3 Seen 14-Jul-10

475 GYP AL M ASY 1056-99904 F1-4 Corrected Link #; Seen 16-Apr-10


66

Midway


2010




possible).

476 AL Y- F ASY 1036-14889 F1-3 Seen 22-Jul-10

477 AL Y8 F ASY 1036-14890 Rehab 2009 F1-3; Seen 09-Jul-10

479 AL W> F SY 896-37544 F1-4 2009 Seen 15-Jun-10

480 AL Y> F SY 896-37545 F1-4 2009 Seen 07-Jul-10

482 W? AL F ASY 896-37547 F1-3 Seen 23-Jun-10

483 B AL M SY 896-37549 F1-4 2009 Seen 25-Nov-09

484 W0 AL F SY 896-37550 F1-4 2009 Seen 10-Aug-10

485 Y* AL F ASY 896-37551 F1-3 2008 Seen 07-Jul-10

486 GNXY AL M ASY 896-37552 F1-3 Seen 29-Apr-10

487 AL W- F SY 896-37553 F1-4 2009 Seen 21-Jul-10

488 AL BYC M ASY 896-37554 F1-3 Could be confused with BC; Seen 28-Jul-10

489 AL R/ F SY 896-37555 F1-4 2009 Seen 08-Jan-10

490 AL GN8 M SY 896-37598 F1-4 2009 Seen 05-Jul-10

491 AL WMX8 F ASY 896-37599 F1-3 Seen 04-Aug-10

492 BY1 AL F SY 1105-11704 F1-4 2009 Could be confused with B1; Seen 24-Jul-10

493 AL R1 M ASY 896-37556 F1-3 Seen 14-Jul-10

494 AL W1 F ASY 896-37557 F1-3 Seen 11-Aug-10

495 AL R7 F ASY 1105-11705 F1-3 Seen 14-Jul-10

496 AL BY4 M? SY 896-37558 F1-4 2009 Could be confused with BYA, B4 or BA; Seen 06-Aug-

10

497 AL WMX9 F SY 896-37577 F1-4 2009 Seen 13-Jul-10

498 AL WMX/ M ASY 896-37576 F1-3 Seen 01-Oct-09

499 AL GY0 M ASY 896-37575 F1-3 Seen 17-Feb-10

500 AL RE F ASY 896-37559 F1-3 Seen 18-Feb-10

501 AL W2 M SY 896-37560 F1-4 2009 Seen 25-Jun-10

502 AL RY F SY 896-37562 F1-4 2009 Seen 22-Jul-10

503 AL W* F ASY 896-37561 F1-3 Pukas in webbing; Seen 28-Jul-10

504 AL WN F SY 896-37563 F1-4 2009 Seen 08-Jan-10

505 AL WK F? SY 896-37564 F1-4 2009 Seen 01-Dec-09

506 AL RM M ASY 896-37565 F1-3 Seen 14-Apr-10

507 AL WJ F SY 896-37566 F1-4 2009 Seen 17-Mar-10

508 AL W7 F SY 896-37567 F1-4 2009 Seen 14-Jul-10

509 WH AL M? SY 896-37568 F1-4 2009 Seen 11-Aug-10

510 AL WR M SY 896-37569 F1-4 2009 Seen 06-Jan-10

511 AL WU F SY 896-37570 F1-4 2009 Seen 07-Jul-10

512 WMX? AL F? ASY 896-37571 F1-3 Seen 09-Mar-10

513 AL WMX$ F SY 896-37573 F1-4 2009 Seen 07-Apr-10


67

Midway


2010




possible).

514 WWN AL M ASY 896-37572 F1-3 Seen 30-Jun-10

516 WMX> AL M ASY 896-37574 F1-3 Seen 20-Apr-10

517 RC AL M SY 876-37578 F1-4 2009 Seen 10-Jul-10

518 WMX7 AL F ASY 1105-23801 F1-3 Band upside down (confirm which leg); Seen 13-Jul-10

519 AL YN M SY 896-37535 F1-4 2009 Band looks like YH; Confirmed as YN; Band worn; Seen

04-Aug-10

520 AL YA F ASY 896-37536 F1-3 Seen 09-Aug-10

521 AL Y4 M SY 896-37537 F1-4 2009 Check to make sure not YH; Band worn; Seen 06-Jan-10

522 AL YJ U HY 896-37538 F1-4 2009 Seen 30-Dec-09

523 AL YM F ASY 896-37539 F1-3 2008 Seen 04-Feb-10

524 AL YT M SY 896-37540 F1-4 2009 Seen 23-Jun-10

525 AL Y1 F SY 896-37541 F1-4 2009 Seen 04-Feb-10

526 AL Y3 M SY 896-37542 F1-4 2009 Seen 17-Nov-09

527 AL Y5 F SY 896-37543 F1-4 2009 Seen 14-Jul-10

528 BYG AL U SY 1036-14882 F1-4 2009 AL upside down; Seen 08-Jun-10

529 W3 AL F ASY 1036-14892 F1-3 2008 Seen 06-Aug-10

530 AL OX6 F ASY 896-37579 Rehab 2009 F1-3 2008; Released at Aviary Oct-09; Seen 22-Jul-10

531 AL O/BKG F SY 896-37580 F1-4 2009 Seen 09-Jul-10

532 AL O\ F ASY 896-37581 F1-3 2008 Seen 11-Aug-10

533 AL YC3 F SY 896-37582 F1-4 2009 Seen 13-Jul-10

534 AL YC2 M SY 896-37583 F1-4 2009 Seen 14-Jul-10

535 AL GY9 M ASY 896-37584 F1-3 Seen 27-Mar-10

536 AL GY0 F ASY 896-37585 F1-3 2008 Seen 07-Aug-10

537 AL GY> M ASY 896-37586 F1-3 2008 Seen 30-Mar-10

538 AL BR? F SY 1105-11706 F1-4 2009 Seen 05-Oct-09

540 AL WF M ASY 896-37590 F1-3 Big puka on left foot; Seen 13-Jul-10

Documents

POPULATION ESTIMATES AND MONITORING GUIDELINES FOR ... · POPULATION ESTIMATES AND MONITORING GUIDELINES FOR ENDANGERED LAYSAN TEAL, ANAS LAYSANENSIS, AT MIDWAY ATOLL: PILOT STUDY