[42]

Bird Populations 9:42-75© The Institute for Bird Populations 2009

POWER OF THE MAPS PROGRAM TO DETECTDIFFERENCES AND TRENDS IN SURVIVAL

AND A VISION FOR PROGRAM EXPANSION1

DAVID F. DESANTE2, AND JAMES F. SARACCO

The Institute for Bird PopulationsP.O. Box 1346

Point Reyes Station, CA 94956-1346

Abstract. A major goal of the Monitoring Avian Productivity and Survivorship(MAPS) program is to provide information on spatial and temporal variation in vitalrates of North American landbirds. Here we identify landbird species for which theMAPS program could likely detect differences or trends in adult apparent survivalrates with 80% statistical power (α = 0.10 or 0.20) and 20 years of data based on MAPSsurvival-rate and recapture probability estimates, sample sizes typical of the current(2001) MAPS program, and results of power analyses applied to simulated capture-recapture data. We summarize numbers of species for which the MAPS program couldlikely detect differences or trends in survival at continental, MAPS-regional, and“cluster” (six-station) scales. We suggest that differences or trends in survival of ≤ 25%would be detectable (1 - β = 0.80 and α = 0.20) for 105 species at the continental scale,and for 19 (Alaska/Boreal Canada Region) to 47 species (Northwest Region) at theMAPS-regional scale. A cluster-scale analysis of MAPS stations on Pacific Northwestnational forests suggested that at least 15 species could be effectively monitored at thatscale. We estimate that, through targeted enhancement and expansion of the MAPSprogram (increasing its size by 69%), an additional 41 species could be adequatelymonitored at the continental scale, and 16 (Northwest Region) to 27 (South-centralRegion) additional species could be adequately monitored at MAPS-regional scales. Weprovide lists of species that MAPS currently monitors (and could potentially monitor)effectively at those two spatial scales, show that coverage is widespread for most majornon-grassland habitat types, and suggest that coverage of chaparral/sagebrushshrubsteppe, boreal conifer, and boreal hardwood habitats could be improved throughtargeted enhancement of MAPS.

Key words: Cormack-Jolly-Seber models, demographic monitoring, landbirds, MAPSprogram, power analysis, survival rate estimation, trends.

PODER DEL PROGRAMA MAPS PARA DETECTAR DIFERENCIAS Y TENDENCIAS EN SOBREVIVENCIA Y UNA VISION

PARA LA EXPANSION DEL PROGRAMAResumen. Una meta principal del programa MAPS es aportar información sobre

variación espacial y temporal en tasas vitales en aves terrestres de Norteamérica. Aquíidentificamos especies de aves terrestres para las que el programa MAPS podríadetectar diferencias o tendencias en tasas de sobrevivencia aparente de adultos con unpoder estadístico del 80% (α = 0.10 ó 0.20) y 20 años de datos basados en las estimas detasa de sobrevivencia y probabilidad de recaptura de MAPS, y resultados del análisisde poder aplicado a datos simulados de captura-recaptura. Resumimos las especies

____________________1Submitted 19 March 2009; accepted 5 September 20092Corresponding author: ddesante@birdpop.org

INTRODUCTIONThe Monitoring Avian Productivity andSurvivorship (MAPS) program was establishedin 1989 to monitor primary demographicparameters (vital rates) of landbirds in NorthAmerica (DeSante 1992), to use those data toidentify the proximate demographic cause(s) ofpopulation declines (DeSante et al. 2001, Saraccoet al. 2008), and to formulate and evaluatemanagement actions and conservation strategiesto reverse those declines (DeSante andRosenberg 1998, DeSante et al. 2005). The MAPSprogram consists of a cooperative continental-scale network of mist-netting and bird-bandingstations operated each summer. The highestdensities of stations occur towards the coasts(where human population densities are greater)and within the lower 48 U.S. states (Fig. 1). ManyMAPS stations (227 or about 25% of all stationsthat have ever registered with the program) arelong-running stations, having operated for ≥ 10years. MAPS uses a standardized “constant-effort” protocol to provide estimates or indices ofa broad suite of population parameters (Saraccoet al. 2008, DeSante and Kaschube 2009). MAPScomplements efforts that focus on abundanceand trend estimation (e.g., the North AmericanBreeding Bird Survey [BBS]), and could fill animportant niche in the developing NorthAmerican Coordinated Bird Monitoring (CBM)program (Coordinated Bird Monitoring Working

Group 2004, Bart 2005, Bart and Ralph 2005).Among the principal metrics monitored by

MAPS is the annual apparent survival rate ofadult birds (hereafter “survival”). Survival canbe estimated at spatial scales ranging from singlebanding stations (occasionally), through localclusters of banding stations, to regional andcontinent-wide scales (Rosenberg et al. 1999,2000). Here we identify species for which theMAPS program could likely detect trends ordifferences in adult apparent survival rates with80% statistical power (α = 0.10 or 0.20) and 20years of data based on MAPS survival-rate andrecapture probability estimates, sample sizestypical of the current (2001) program, and resultsof power analyses applied to simulated capture-recapture data (DeSante et al. 2009a). Wesummarize numbers of species for which theprogram could likely detect differences or trendsin survival at continental, MAPS-regional(DeSante et al. 1993), and “cluster” (six-station)scales. We list species that are “adequately” mon-itored across a range of effect sizes at continentaland MAPS-regional scales. In addition toassessing the current program, we suggest astrategy for expanding MAPS, so that it cancontribute optimally to a continental-scale CBMeffort. We report improvements in the ability todetect smaller effect sizes under such expansion,as well as a list of additional species that couldbe monitored with MAPS methodology.

DAVID F. DESANTE AND JAMES F. SARACCO

[43]

para las que el programa MAPS podría detectar diferencias o tendencias ensobrevivencia a escalas continentales, regiones MAPS, y “racimos” de seis estaciones.Sugerimos que las diferencias o tendencias de sobrevivencia de ≤ 25% seríandetectables (1 - β = 0.80 y α = 0.20) para 105 especies a escala continental, y para 19(Región Alaska/Canadá Boreal) a 47 especies (Región Noroeste) a la escala de lasregiones MAPS. Un análisis a la escala de racimos en los bosques nacionales delNoroeste Pacífico sugiere que al menos 15 especies podrían ser monitoreadas a esaescala. Estimamos que, mediante el aumento selectivo y expansión del programaMAPS (aumentando su tamaño un 69%), 41 especies adicionales podrían seradecuadamente monitoreadas a la escala continental, y 16 (Región Noroeste) a 27(Región Centro-sur) especies adicionales podrían ser monitoreadas a escala de lasregiones MAPS. Aportamos listas de especies que el programa MAPS monitoreaefectivamente en la actualidad a esas dos escalas, mostramos que la cobertura esamplia para la mayoría de hábitats no relacionados con praderas, y mostramos que lacobertura de los hábitats estepa arbustiva/chaparral, conífera boreal y latifoliadasboreales podrían ser mejoradas mediante aumentos selectivos de MAPS.

Palabras clave: modelos Cormack-Jolly-Seber, monitoreo demográfico, aves terrestres,programa MAPS, análisis de poder, estimación de tasas de sobrevivencia, tendencias.

METHODSThe MAPS program is a cooperative network ofconstant-effort mist-netting stations operatedduring the breeding season (May to August)across the United States and southern Canada.DeSante and Kaschube (2009) provide anoverview of the design and operation of theprogram; detailed protocols for establishing andoperating MAPS stations are provided inDeSante et al. (2009b) and can be found at 2009MAPS Manual. Over 1,000 MAPS stations havebeen operated for at least one year since theprogram’s inception in 1989 and over 450stations were operated annually during thelatter part of the 10-yr period (1992-2001) used

in the analyses presented here. Stations wereestablished in 20-ha study areas where long-term mist netting was permissible and practical.In general, the locations of MAPS stations werechosen by station operators (often according to ahypothesis-driven strategy) and not by aprobabilistic sampling design, althoughelements of a random sampling strategy weresometimes employed.

We assessed the ability of MAPS to detectwith 80% power (i.e., 1 - β = 0.80) (1) differencesin survival between two populations and (2)trends in survival for a single population. Wereport results for two alpha-levels (0.10 and 0.20)reported by other studies that have evaluated

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[44]

FIGURE 1. MAPS stations (through 2001) and regions: AC = Alaska/Boreal Canada (considered as two regionsby DeSante 1992), NW = Northwest, NC = North-central, NE = Northeast, SW = Southwest, SC = South-central,and SE = Southeast. Stations still operating in 2001 are indicated by open circles; those established andoperated, but that stopped prior to 2001, are indicated by open triangles.

power of monitoring data (e.g., Bart et al. 2004).We considered five effect sizes ranging from 5 to25%. For the two-population scenario (1 above),effect sizes represent differences in survivalbetween two populations that begin with equalpopulation sizes. For the linear decline scenario(2 above), effect sizes represent incrementalproportional changes in survival that wouldhalve the population in the same number ofyears as a population whose survival initiallydeclined by that effect size and then remainedconstant. Our calculations assume initially stablepopulations (i.e., � = 1) and constant recruitmentat a level that balances losses at the start of thestudy. Although the larger effect sizesconsidered here could (potentially) halvepopulations over short time intervals (seeDeSante et al. 2009a), differences in survivalrates of this magnitude are typically observed inMAPS data. For example, the largest effect size(25%) is similar to the mean maximumdifference in survival rates between MAPSregions for 89 species for which we were able toobtain 10-yr time-constant estimates of survivalin multiple regions (mean maximum difference= 23%; DeSante and Kaschube 2006).

We assessed whether a species was capturedin sufficient numbers to reject null hypotheses(no difference between populations or nochange in a population) at program-wide (i.e.,continental) and MAPS-regional scales (see Fig.1 for regional boundaries) based on (1) theircurrent sample sizes (see below for detail), (2)the magnitude of their 10-yr “time-constant”

adult apparent survival and recaptureprobability estimates, and (3) the predictedsample sizes needed to reject hypotheses fromanalyses of 20 yr of data simulated frompopulations with similar survival and recaptureprobability estimates (see DeSante et al. 2009afor more detail). We simulated capture-mark-recapture data sets of various sizes using thedeterministic mode of program GENCAPH1(www.mbr-pwrc.usgs.gov/software.html) for 20capture periods (i.e., years for estimates ofannual survival). For all sets of simulations wechose an initial sample size of 10 individualmarked animals released per year. We repeatedthis process for a series of sample sizes rangingfrom 20-2000 annual releases of marked animals.We input simulated capture histories intoProgram MARK (White and Burnham 1999) andestimated survival with Cormack-Jolly-Seber(CJS) models.

Each species was binned according to itssurvival and recapture probability estimate ateach spatial scale (see Table 1 for bin ranges andfor the number of species in each bin at thecontinent-wide scale). We then determined, foreach species, whether current sample sizes were≥ the number of individuals needed to detect agiven effect size as based on simulation resultsusing the survival and recapture probabilitiesfor the bin in which the species was placed.

We also estimated the numbers of species thatMAPS samples in sufficient numbers to detectdifferences in survival among “clusters” of sixstations on six national forests in Oregon and

DAVID F. DESANTE AND JAMES F. SARACCO

[45]

TABLE 1. Combinations of apparent survival and recapture probabilities within which bird species werebinned according to 10-yr time-constant parameter estimates from the MAPS program. Here we indicatenumbers of species that fell within each of these bins at the program-wide (continental) scale.

Recapture (p)__________________________________________________________________________________Survival 0.05 0.20 0.35 0.50 0.65 0.80(φ) (0.000-0.125) (0.125-0.275) (0.275-0.425) (0.425-0.575) (0.575-0.725) (0.725-0.875) Total

0.20 (0.15-0.25) 1 2 1 1 0 0 5

0.30 (0.25-0.35) 1 6 3 1 1 2 14

0.40 (0.35-0.45) 4 14 5 13 2 1 39

0.50 (0.45-0.55) 3 12 37 19 7 1 79

0.60 (0.55-0.65) 7 7 9 3 1 0 27

0.70 (0.65-0.75) 4 5 4 0 0 0 13

0.80 (0.75-0.85) 0 1 1 0 0 0 2

Total 20 47 60 37 11 4 179

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[46]

Washington. MAPS stations are often set up inthis manner (clusters of six) on particularlandholdings (e.g., a national park, nationalforest, or military installation) for logisticalreasons, as a team of two banders can operatesix stations in each 10-d period, even accountingfor 2-3 days of rain.

Current sample size at each spatial scale (foreach species) was determined by estimating thenumber of marked resident birds of each speciesreleased per year under a stable MAPS programat its current (2001) level of effort. To obtainthese estimates, we first estimated the propor-tions of resident birds among newly-bandedadult birds of unknown residency status (i.e.,birds caught once in the season in which theywere banded or birds caught more than once butnot ≥7 d after banding) in the 10-yr (1992-2001)data set using time-constant ad hoc RobustDesign models (Nott and DeSante 2002, Hines etal. 2003). We multiplied these proportions by thenumbers of unknown-status birds in the data setand then added to these the numbers of knownresidents (i.e., birds recaptured ≥7 d after band-ing in their initial year of capture). Althoughthese totals could simply be divided by 10 (forthe 10 years in the data set) to obtain estimatesof the mean numbers of birds released annuallyby MAPS, such estimates would underestimatecurrent MAPS sample sizes because theprogram grew rapidly during 1992-1997 (from179 to 459 stations). MAPS remained relativelyconstant in size from 1997 through 2001 (with anannual mean of about 478 stations), and wemultiplied our estimated total numbers ofresident birds by an inflation factor to reflect thisstable MAPS program prior to obtaining per-year averages.

We determined inflation factors for eachregion and at the continental scale through atwo-step process. First we calculated theproportion of stations that were usable forsurvival analyses (i.e., that were operated forfour consecutive years) for each of the ten years1992-2001. These proportions increased duringthe first three years 1992-1994 (because allstations that stopped operating during thoseyears could not be included in survivalanalyses), stayed constant during the middleyears 1995-1998 (at the stable proportion ofstations that operated for at least fourconsecutive years), and decreased during the

final three years 1999-2001 (because all stationsthat started operating during any of those threeyears could not be included in survivalanalyses). We then multiplied the number ofstations operated in each of the final three yearsof the 10-yr time period (1999-2001) by the meanproportion of stations usable for survivalanalyses during the middle years (1995-1998) toestimate the number of stations that would beusable for survival analyses during each of thosefinal three years in an on-going MAPS program.

In the second step in our inflation factorcalculations, we summed the number of stationsusable for survival analyses during the stablelatter half of the 10-yr period (i.e., the actualnumbers of usable stations during 1997 and 1998plus the estimated numbers of usable stationsduring 1999, 2000, and 2001), and multiplied thissum by 2 to estimate the number of station-yearsuseable for survival analyses over any 10-yrperiod of the current-sized MAPS Program.Finally, we divided this number by the actualnumber of station-years useable for survivalanalyses during 1992-2001 to provide theappropriate inflation factor. The continental-scale inflation factor was 1.30, while regionalinflation factors ranged from 1.13 forAlaska/Boreal Canada to 1.67 for the Southwest(regional mean = 1.30).

Finally, we classified each species “ade-quately” monitored in each MAPS region intoone or more of 14 major landbird habitat typesfound across North America, north of Mexico(Table 2). We excluded grassland and tundrahabitats because MAPS field protocol (operationof about 10 fixed-location mist nets) does noteffectively sample the landbirds of thosehabitats.

EVALUATION OF AN EXPANDED MAPSPROGRAM

The Northwest Region currently hosts thelargest number of MAPS stations, with 128stations operated annually with data usable forsurvival analyses (i.e., stations operated for ≥4yr). The Southeast, Southwest, and Northeastregions each host about half the number ofstations as the Northwest (70, 66, and 65,respectively). We suggest that 20 additionalstations could be established and maintained inthe Northwest Region, and that the number ofstations in the Southwest, Southeast, and

DAVID F. DESANTE AND JAMES F. SARACCO

[47]

TABLE 2. Major landbird habitat types used in this study.

Habitat type (code) MAPS Regions Description

Scrub/successional/ NW, SW, NC, SC, Includes wet meadows in forested habitats and marshydisturbed (SSD) NE, SE, and areas in shrubsteppe habitats as well as areas modified by

AK&BC human disturbance, but excludes extensive areas of chaparral or shrubsteppe habitat

Pacific Northwest NW Includes forests generally dominated by spruces, cedarsconiferous forest (PNC) hemlocks, and firs, but can include some pines and a

limited hardwood component

Western coniferous NW and SW Includes forests generally dominated by pines, but can forest (WC) include substantial amounts of firs and a limited hardwood

component

Western oak or NW and SW Includes relatively open-canopy oak woodland and, awayjuniper woodland (OJW) from the Pacific slope, open juniper, pinyon-juniper, or

mountain mahogany woodland

Western riparian NW and SW Includes relatively extensive riparian areas generallyforest (WR) dominated by willows and cottonwoods, often embedded

in shrubsteppe or grassland habitat; does not includenarrow riparian strips in forest or woodland habitat

Chaparral/sagebrush NW and SW Includes generally extensive areas dominated by shrub-steppe (CSS) woody shrubs, such as California coastal scrub,

chaparral, and sagebrush shrubsteppe; does not include montane chaparral intermixed with western coniferous forest

Desert scrub (DS) SW Includes generally extensive areas dominated by relatively widely spaced woody or cactus vegetation

Eastern northern NC and NE Includes forests generally dominated by red spruce coniferous forest (NC) and balsam fir, sometimes with a limited jack pine or

hardwood component

Eastern northern NC, SC, NE, and Includes forests generally dominated by maple, beech, andhardwood forest (NH) SE yellow birch, often with a coniferous component of white

pine and hemlock

Eastern southern NC, SC, NE, and Includes both upland and bottomland forests generallyhardwood forest (SH) SE dominated by oak, hickory, maple, cherry, tulip-poplar, or

sweet gum, sometimes with a coniferous component ofupland pine or bottomland bald cypress

Eastern southern SC, NE, and SE Includes southern pine savannah and pine forest, oftenconiferous forest (SC) mixed with a hardwood component of oak or hickory

Southern broadleaf SC Includes live oak forests and woodland, as well as semi-evergreen forest (SBE) tropical hammocks of southern Florida and brushlands of

southern Texas

Boreal coniferous AK&BC Includes both upland and flooded forest dominated byforest (BC) spruces and tamaracks, often with a limited hardwood

component of birches

Boreal hardwood AK&BC Includes relatively extensive closed birch forests as well as forest (BH) open aspen parklands

Northeast regions could be increased to thecurrent effort in the Northwest. The South-central, North-central, and Alaska/BorealCanada regions have fewer stations still, with50, 32, and 22 stations, respectively, operatedannually with data suitable for survivalanalyses. We suggest that the numbers ofstations in these remaining regions, which haveconsiderably smaller human populationdensities, more grassland habitat (unfavorablefor passive mist netting), or both, be increased tothe average current level of coverage in theSouthwest, Northeast, and Southeast (67stations). Such expansion would increase thesize of the MAPS program by 300 stations (a69% increase, or 733 total stations operatedannually for ≥4 yr).

We used two methods to identify species forwhich survival could potentially be monitoredeffectively under such an expanded MAPSprogram. In the first method, which we call theexpanded (non-targeted) program, we assumethat, within each region, the relative geographicand habitat coverage and effectiveness ofcapturing birds in the expanded program wouldremain essentially the same as in the currentprogram. Thus, new “monitorable species” werethose for which expected sample size increases,calculated as proportional increases in annualreleases of residents equal to proportionalincreases in the number of stations, reach a levelthat our power analyses suggest we can detect a≤25% difference in survival betweenpopulations or ≤25% decline in survival with80% power, α = 0.20, and 20 years of data. Wedid not apply this method to the Northwest, aswe believe that the current number anddistribution of stations there is sufficient toeffectively monitor an adequate number ofspecies in the habitats sampled.

In the second method, which we call theexpanded and targeted program, we assumedthat new stations would be sited specifically totarget species currently under-represented in theMAPS database, including species identified asmonitorable in the expanded (but non-targeted)program. For the Northwest Region, targetspecies were selected based on a formal analysisof the MAPS program funded by the U.S.Bureau of Land Management (Pyle et al. 2005).Specifically, we selected Northwest targetspecies based on (1) their identification as focal

or priority species in at least one priority habitatin at least one of 13 published Partners In Flight(PIF) bird conservation plans for the Northwest,or (2) their having negative population trendsaccording to BBS or MAPS data in one or morestates or Bird Conservation Regions (BCRs) inthe Northwest. A few species with positivepopulation trends were also selected forcomparison. Priority habitats were identified byexamining PIF habitat designations forPhysiographic Areas in the Pacific Northwestand selecting those that can be effectivelysampled by MAPS (i.e., forest and scrub areas).For the remaining regions, target species wereselected as those thought to be monitorable,based on their habitats, ecology, and behavior,provided that a sufficient number of MAPSstations were operated.

RESULTS

POWER OF THE CURRENT MAPS PROGRAM

Our estimates of the numbers of resident birdsreleased per year in the MAPS program suggestthat we can detect differences in survivalbetween populations or linear declines insurvival with 80% power with 20 yr of data atthe continental scale for 105 species (Table 3).About one third (36) are Partners In Flight (PIF)Species of Continental Importance (“Watch List”or “Additional Stewardship” species; Rich et al.2004). Current MAPS sample sizes suggest wecan detect (at α = 0.20) effect sizes of 5% for ninespecies, 10% for an additional 38 species, 15%for an additional 33 species, 20% for anadditional 13 species, and 25% for an additional12 species. Increasing α from 0.10 to 0.20resulted in relatively small increases in thenumbers of species for which we can reject nullhypotheses (Figs. 2-3).

MAPS program coverage is greatest in theNorthwest Region, where we can currentlydetect differences in survival between popu-lations or linear declines in survival with 80%power for 47 species (Figs. 2-3, Appendix 1).Five-percent effect sizes (for at least one of thehypothesis tests and α = 0.20) could likely bedetected for five of these species in theNorthwest Region, while larger effect sizes couldlikely be detected for an additional 13 species atthe 10% level, 21 species at the 15% level, six

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[48]

DAVID F. DESANTE AND JAMES F. SARACCO

[49]

TAB

LE

3. E

ffec

t si

zes

(%)

curr

entl

y d

etec

tabl

e by

MA

PS w

ith

20 y

rs o

f d

ata,

and

lik

ely

det

ecta

ble

und

er a

n ex

pand

ed M

APS

pro

gram

, wit

h 80

% p

ower

at

the

prog

ram

-wid

e (i

.e.,

cont

inen

tal)

sca

le.

Two

effe

ct t

ypes

wer

e co

nsid

ered

: (1

) a

two-

popu

lati

on c

ompa

riso

n of

sur

viva

l (2

P),

and

(2)

a p

opul

atio

n w

ith

linea

rly

dec

linin

g su

rviv

al (L

D).

Eff

ect s

izes

are

giv

en fo

r tw

o al

pha-

leve

ls, 0

.10

and

0.2

0.

Cur

rent

Exp

and

ed__

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

____

____

____

____

____

___

α=

0.1

0α

= 0

.20

α=

0.1

0α

= 0

.20

____

____

____

____

____

____

___

____

____

____

____

____

___

Spec

ies

Sta1

Ind

2R

es3

�̂4

p̂52P

6L

D7

2P6

LD

72P

6L

D7

2P6

LD

7

Red

-nap

ed S

apsu

cker

(Sp

hyra

picu

s nu

chal

is)

3248

955

0.41

30.

602

2515

2015

2015

1515

Red

-bre

aste

d S

apsu

cker

(Sph

yrap

icus

rub

er)

4970

177

0.45

80.

369

2015

2015

1515

1510

Lad

der

-bac

ked

Woo

dpe

cker

(Pic

oide

s sc

alar

is)

1698

100.

552

0.29

225

2520

Nut

tall'

s W

ood

peck

er (P

icoi

des

nutt

allii

)16

195

220.

551

0.38

120

2515

2515

2015

Dow

ny W

ood

peck

er (P

icoi

des

pube

scen

s)22

719

1513

70.

500

0.35

415

1515

1015

1010

10H

airy

Woo

dpe

cker

(Pic

oide

s vi

llosu

s)15

463

059

0.66

50.

208

2015

1510

1510

1010

Wes

tern

Woo

d-P

ewee

(Con

topu

s so

rdid

ulus

)73

1456

135

0.48

60.

362

1515

1510

1510

1010

Eas

tern

Woo

d-P

ewee

(Con

topu

s vi

rens

)10

161

644

0.49

40.

278

2025

1525

1520

15A

cad

ian

Flyc

atch

er (E

mpi

dona

x vi

resc

ens)

6622

8518

30.

508

0.51

810

1010

1010

1010

10“T

raill

's”

Flyc

atch

er (E

mpi

dona

x al

noru

m/t

raill

ii)80

3105

198

0.48

50.

491

1010

1010

1010

1010

Lea

st F

lyca

tche

r (E

mpi

dona

x m

inim

us)

2410

4473

0.37

90.

435

2515

2015

2015

1515

Ham

mon

d's

Fly

catc

her

(Em

pido

nax

ham

mon

dii)

5611

5511

40.

452

0.40

320

1515

1515

1010

10D

usky

Fly

catc

her

(Em

pido

nax

ober

hols

eri)

5219

8614

70.

485

0.42

515

1010

1010

1010

10“W

este

rn”

Flyc

atch

er (E

. diff

icili

s/oc

cide

ntal

is)

6925

7316

10.

504

0.28

215

1515

1010

1010

10B

lack

Pho

ebe

(Say

orni

s ni

gric

ans)

1619

512

0.46

00.

481

2020

15A

sh-t

hroa

ted

Fly

catc

her

(Myi

arch

us c

iner

asce

ns)

3572

363

0.66

60.

135

1515

1510

1510

1010

Gre

at C

rest

ed F

lyca

tche

r (M

yiar

chus

cri

nitu

s)96

574

330.

615

0.21

620

2025

15B

row

n-cr

este

d F

lyca

tche

r (M

yiar

chus

tyra

nnul

us)

422

530

0.49

20.

276

2020

2515

2015

Whi

te-e

yed

Vir

eo (V

ireo

gri

seus

)82

2719

267

0.53

70.

509

1010

1010

1010

1010

Bel

l's V

ireo

(Vir

eo b

ellii

)17

562

570.

581

0.40

420

1515

1015

1010

10W

arbl

ing

Vir

eo (V

ireo

gilv

us)

118

5017

379

0.48

30.

428

1010

1010

510

510

Red

-eye

d V

ireo

(Vir

eo o

livac

eus)

159

4798

427

0.59

50.

253

1010

1010

1010

1010

Stel

ler's

Jay

(Cya

noci

tta

stel

leri

)60

303

200.

731

0.18

425

2525

2020

15C

arol

ina

Chi

ckad

ee (P

oeci

le c

arol

inen

sis)

121

1312

950.

499

0.22

920

2515

2515

2015

Bla

ck-c

appe

d C

hick

adee

(Poe

cile

atr

icap

illa)

144

3069

294

0.46

80.

367

1010

1010

1010

1010

Mou

ntai

n C

hick

adee

(Poe

cile

gam

beli)

5111

4593

0.45

20.

385

2015

1515

1515

1510

Che

stnu

t-ba

cked

Chi

ckad

ee (P

oeci

le r

ufes

cens

)51

1043

880.

419

0.39

620

2520

2015

2015

Bor

eal C

hick

adee

(Poe

cile

hud

soni

ca)

1013

219

0.49

20.

365

2520

20Tu

fted

Tit

mou

se (B

aeol

ophu

s bi

colo

r)13

718

4018

90.

491

0.38

615

1010

1010

1010

10B

usht

it (P

salt

ripa

rus

min

imus

)43

1117

152

0.29

50.

146

2525

25W

hite

-bre

aste

d N

utha

tch

(Sit

ta c

arol

inen

sis)

8835

820

0.47

70.

306

2020

20

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[50]

Car

olin

a W

ren

(Thr

yoth

orus

ludo

vici

anus

)12

427

1327

00.

397

0.54

115

1010

1010

1010

10B

ewic

k's

Wre

n (T

hryo

man

es b

ewic

kii)

7216

9518

00.

430

0.51

515

1515

1015

1010

10H

ouse

Wre

n (T

rogl

odyt

es a

edon

)97

2863

267

0.34

10.

420

2520

2015

2015

1515

Win

ter

Wre

n (T

rogl

odyt

es tr

oglo

dyte

s)46

967

780.

376

0.50

625

1520

1520

1515

15A

rcti

c W

arbl

er (P

hyllo

scop

us b

orea

lis)

224

929

0.33

90.

605

2525

2025

20V

eery

(Cat

haru

s fu

sces

cens

)54

2274

290

0.58

10.

566

1010

510

55

55

Gra

y-ch

eeke

d T

hrus

h (C

atha

rus

min

imus

)6

253

310.

459

0.68

320

1520

1520

1515

10Sw

ains

on's

Thr

ush

(Cat

haru

s us

tula

tus)

109

1017

511

130.

581

0.62

45

55

55

55

5H

erm

it T

hrus

h (C

atha

rus

gutt

atus

)75

2170

229

0.47

40.

607

1010

1010

1010

510

Woo

d T

hrus

h (H

yloc

ichl

a m

uste

lina)

128

4973

455

0.44

00.

490

1010

1010

1010

1010

Am

eric

an R

obin

(Tur

dus

mig

rato

rius

)26

978

7468

20.

523

0.27

510

1010

105

105

10V

arie

d T

hrus

h (I

xore

us n

aevi

us)

4049

334

0.47

10.

394

2025

1525

1520

15W

rent

it (C

ham

aea

fasc

iata

)37

1325

145

0.53

40.

627

1010

1010

1010

1010

Gra

y C

atbi

rd (D

umet

ella

car

olin

ensi

s)12

194

4687

40.

511

0.46

45

105

55

55

5B

row

n T

hras

her

(Tox

osto

ma

rufu

m)

5864

348

0.52

20.

292

2515

2515

2015

2015

Lon

g-bi

lled

Thr

ashe

r (T

oxos

tom

a lo

ngir

ostr

e)3

133

140.

628

0.39

620

2025

15B

lue-

win

ged

War

bler

(Ver

miv

ora

pinu

s)35

994

790.

521

0.39

420

1520

1515

1515

10O

rang

e-cr

owne

d W

arbl

er (V

erm

ivor

a ce

lata

)72

3875

268

0.44

10.

435

1510

1010

1010

1010

Nas

hvill

e W

arbl

er (V

erm

ivor

a ru

ficap

illa)

3411

1160

0.33

80.

331

2525

25V

irgi

nia'

s W

arbl

er (V

erm

ivor

a vi

rgin

iae)

1246

330

0.47

30.

339

2020

2515

2515

Yello

w W

arbl

er (D

endr

oica

pet

echi

a)12

390

4981

60.

534

0.47

45

105

105

55

5C

hest

nut-

sid

ed W

arbl

er (D

endr

oica

pen

sylv

anic

a)22

839

880.

431

0.54

520

1520

1515

1515

15M

agno

lia W

arbl

er (D

endr

oica

mag

nolia

)15

514

390.

321

0.76

920

2520

2520

2020

Yello

w-r

umpe

d W

arbl

er (D

endr

oica

cor

onat

a)97

4216

328

0.45

10.

276

1010

1010

1010

1010

Bla

ck-t

hr. G

reen

War

bler

(Den

droi

ca v

iren

s)19

372

390.

396

0.55

720

2520

2520

2015

Tow

nsen

d's

War

bler

(Den

droi

ca to

wns

endi

)27

978

101

0.42

80.

224

2525

2020

Bla

ckpo

ll W

arbl

er (D

endr

oica

str

iata

)7

177

140.

313

0.73

525

2525

Bla

ck-a

nd-w

hite

War

bler

(Mni

otilt

a va

ria)

7511

0591

0.51

80.

298

2015

1515

1515

1510

Am

eric

an R

edst

art (

Seto

phag

a ru

tici

lla)

6032

0426

40.

509

0.33

810

1010

1010

1010

10Pr

otho

nota

ry W

arbl

er (P

roto

nota

ria

citr

ea)

2047

949

0.50

90.

206

2520

20W

orm

-eat

ing

War

bler

(Hel

mit

hero

s ve

rmiv

orus

)30

781

600.

529

0.40

225

1520

1520

1515

15O

venb

ird

(Sei

urus

aur

ocap

illus

)11

238

7332

40.

550

0.43

05

105

55

55

5N

orth

ern

Wat

erth

rush

(Sei

urus

nov

ebor

acen

sis)

2049

137

0.49

80.

550

2515

2015

2015

1515

TAB

LE

3. C

onti

nued

.

Cur

rent

Exp

and

ed__

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

____

____

____

____

____

___

α=

0.1

0α

= 0

.20

α=

0.1

0α

= 0

.20

____

____

____

____

____

____

___

____

____

____

____

____

___

Spec

ies

Sta1

Ind

2R

es3

�̂4

p̂52P

6L

D7

2P6

LD

72P

6L

D7

2P6

LD

7

DAVID F. DESANTE AND JAMES F. SARACCO

[51]

Lou

isia

na W

ater

thru

sh (S

eiur

us m

otac

illa)

3556

150

0.51

40.

583

2015

1510

1510

1510

Ken

tuck

y W

arbl

er (O

poro

rnis

form

osus

)53

1834

197

0.53

90.

585

1010

1010

1010

510

Mou

rnin

g W

arbl

er (O

poro

rnis

phi

lade

lphi

a)7

230

300.

444

0.38

925

2520

Mac

Gill

ivra

y's

War

bler

(Opo

rorn

is to

lmie

i)92

5887

596

0.47

70.

612

510

510

55

55

Com

mon

Yel

low

thro

at (G

eoth

lypi

s tr

icha

s)17

585

2377

70.

481

0.50

05

105

105

55

5H

ood

ed W

arbl

er (W

ilson

ia c

itri

na)

4712

2110

90.

489

0.53

615

1015

1010

1010

10W

ilson

's W

arbl

er (W

ilson

ia p

usill

a)82

9116

593

0.41

60.

530

1010

1010

1010

510

Can

ada

War

bler

(Wils

onia

can

aden

sis)

1031

423

0.44

30.

544

2520

2025

20Ye

llow

-bre

aste

d C

hat (

Icte

ria

vire

ns)

7430

4432

80.

468

0.47

410

1010

1010

105

10Su

mm

er T

anag

er (P

iran

ga r

ubra

)57

566

490.

486

0.41

725

1525

1520

1520

15W

este

rn T

anag

er (P

iran

ga lu

dovi

cian

a)86

1769

120

0.54

20.

141

2520

2515

2015

2015

Oliv

e Sp

arro

w (A

rrem

onop

s ru

fivir

gatu

s)3

208

270.

510

0.73

820

1520

1515

1515

10G

reen

-tai

led

Tow

hee

(Pip

ilo c

hlor

urus

)13

297

280.

541

0.35

520

2015

2515

Spot

ted

Tow

hee

(Pip

ilo m

acul

atus

)78

2125

223

0.48

70.

465

1010

1010

1010

1010

Eas

tern

Tow

hee

(Pip

ilo e

ryth

roph

thal

mus

)97

907

980.

484

0.33

120

1515

1515

1515

10C

alif

orni

a To

whe

e (P

ipilo

cri

ssal

is)

1839

849

0.53

90.

353

2515

2515

2015

2015

Ruf

ous-

crow

ned

Spa

rrow

(Aim

ophi

la r

ufic

eps)

1417

517

0.54

50.

334

2520

20A

mer

ican

Tre

e Sp

arro

w (S

pize

lla a

rbor

ea)

719

917

0.48

30.

500

2020

1525

15C

hipp

ing

Spar

row

(Spi

zella

pas

seri

na)

7713

2612

90.

410

0.23

125

2020

2520

Fiel

d S

parr

ow (S

pize

lla p

usill

a)67

2032

219

0.44

70.

350

2015

1515

1515

1510

Lar

k Sp

arro

w (C

hond

este

s gr

amm

acus

)17

422

170.

522

0.27

525

2020

Sava

nnah

Spa

rrow

(Pas

serc

ulus

san

dwic

hens

is)

1250

443

0.53

60.

360

2025

1525

1520

15Fo

x Sp

arro

w (P

asse

rella

ilia

ca)

4199

192

0.53

40.

502

1515

1510

1510

1010

Song

Spa

rrow

(Mel

ospi

za m

elod

ia)

185

1046

511

600.

465

0.56

05

105

55

55

5L

inco

ln's

Spa

rrow

(Mel

ospi

za li

ncol

nii)

5225

2832

90.

424

0.63

410

1010

1010

1010

10Sw

amp

Spar

row

(Mel

ospi

za g

eorg

iana

)14

332

290.

402

0.74

825

2025

1520

1520

15W

hite

-thr

oate

d S

parr

ow (Z

onot

rich

ia a

lbic

ollis

)19

992

970.

351

0.53

220

1520

1515

1515

10W

hite

-cro

wne

d S

parr

ow (Z

onot

rich

ia le

ucop

hrys

)30

1260

140

0.45

90.

479

1510

1010

1010

1010

Gol

den

-cro

wne

d S

parr

ow (Z

onot

rich

ia a

tric

apill

a)5

279

340.

533

0.49

025

1520

1520

1515

15D

ark-

eyed

Junc

o (J

unco

hye

mal

is)

121

6505

722

0.44

20.

503

1010

1010

510

510

Nor

ther

n C

ard

inal

(Car

dina

lis c

ardi

nalis

)17

760

2763

10.

568

0.37

05

105

55

55

5B

lack

-hea

ded

Gro

sbea

k (P

heuc

ticu

s m

elan

ocep

halu

s)10

437

9629

50.

550

0.30

310

1010

105

105

10L

azul

i Bun

ting

(Pas

seri

na a

moe

na)

4717

1999

0.49

60.

297

2015

1515

1510

1510

TAB

LE

3. C

onti

nued

.

Cur

rent

Exp

and

ed__

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

____

____

____

____

____

___

α=

0.1

0α

= 0

.20

α=

0.1

0α

= 0

.20

____

____

____

____

____

____

___

____

____

____

____

____

___

Spec

ies

Sta1

Ind

2R

es3

�̂4

p̂52P

6L

D7

2P6

LD

72P

6L

D7

2P6

LD

7

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[52]

Ind

igo

Bun

ting

(Pas

seri

na c

yane

a)11

236

9835

20.

490

0.37

110

1010

1010

1010

10Pa

inte

d B

unti

ng (P

asse

rina

cir

is)

3215

9211

50.

558

0.43

910

1010

1010

1010

10R

ed-w

inge

d B

lack

bird

(Age

laiu

s ph

oeni

ceus

)69

1934

960.

612

0.20

020

1520

1515

1515

10B

row

n-he

aded

Cow

bird

(Mol

othr

us a

ter)

210

2023

180

0.47

30.

460

1010

1010

1010

1010

Bal

tim

ore

Ori

ole

(Ict

erus

gal

bula

)43

687

460.

534

0.28

725

1525

1520

1520

15B

ullo

ck's

Ori

ole

(Ict

erus

bul

lock

ii)39

1141

680.

469

0.36

525

1520

1520

1515

15Pu

rple

Fin

ch (C

arpo

dacu

s pu

rpur

eus)

5134

5925

90.

460

0.34

010

1010

1010

1010

10A

mer

ican

Gol

dfi

nch

(Car

duel

is tr

isti

s)13

566

2451

40.

430

0.26

620

1515

1515

1515

101N

umbe

r of

sta

tion

s th

at w

ere

oper

ated

for

at le

ast f

our

cons

ecut

ive

year

s d

urin

g th

e 10

-yr

peri

od 1

992-

2001

at w

hich

(a) a

t lea

st o

ne a

dul

t ind

ivid

ual o

f the

spec

ies

was

cap

ture

d a

nd (b

) the

spe

cies

was

a r

egul

ar o

r us

ual b

reed

er. S

tati

ons

wit

hin

1 km

of e

ach

othe

r w

ere

mer

ged

into

a s

ingl

e "s

uper

-sta

tion

" to

pre

vent

ind

ivid

uals

who

se h

ome

rang

e en

com

pass

ed p

arts

of b

oth

stat

ions

from

bei

ng tr

eate

d a

s tw

o in

div

idua

ls.

2To

tal n

umbe

r of

ind

ivid

ual a

dul

t bir

ds

capt

ured

dur

ing

the

10-y

r pe

riod

199

2-20

01 a

t sta

tion

s w

here

the

spec

ies

was

a r

egul

ar o

r us

ual b

reed

er; t

hus

the

tota

lnu

mbe

r of

cap

ture

his

tori

es u

pon

whi

ch th

e es

tim

ate

of s

urvi

val p

roba

bilit

y w

as b

ased

.3E

stim

ated

num

ber

of r

esid

ent b

ird

s re

leas

ed p

er y

ear

as p

art o

f a s

tabl

e M

APS

pro

gram

at t

he c

urre

nt le

vel o

f eff

ort.

See

text

for

det

ail.

4E

stim

ated

tim

e-co

nsta

nt a

nnua

l app

aren

t sur

viva

l rat

e fr

om M

APS

dat

a po

oled

acr

oss

all s

tati

ons

from

199

2-20

01.

5E

stim

ated

tim

e-co

nsta

nt r

ecap

ture

pro

babi

lity

from

MA

PS d

ata

pool

ed a

cros

s al

l sta

tion

s fr

om 1

992-

2001

.6E

ffec

t siz

e (%

dif

fere

nce)

that

can

be

det

ecte

d w

hen

com

pari

ng th

e ad

ult a

ppar

ent s

urvi

val r

ates

bet

wee

n tw

o po

pula

tion

s, b

ased

on

resu

lts

of s

imul

ated

dat

aan

d c

urre

nt M

APS

sam

ple

size

s an

d p

aram

eter

est

imat

es.

7E

ffec

t siz

e (%

cha

nge)

that

can

be

det

ecte

d fo

r a

popu

lati

on e

xper

ienc

ing

a lin

ear

dec

line

in a

dul

t app

aren

t sur

viva

l, ba

sed

on

resu

lts

of s

imul

ated

dat

a an

dcu

rren

t MA

PS s

ampl

e si

zes

and

par

amet

er e

stim

ates

.

TAB

LE

3. C

onti

nued

.

Cur

rent

Exp

and

ed__

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

____

____

____

____

____

___

α =

0.1

0α

= 0

.20

α=

0.1

0α

= 0

.20

____

____

____

____

____

____

___

____

____

____

____

____

___

Spec

ies

Sta1

Ind

2R

es3

�̂4

p̂52P

6L

D7

2P6

LD

72P

6L

D7

2P6

LD

7

DAVID F. DESANTE AND JAMES F. SARACCO

[53]

FIGURE 2. Numbers of species for which current MAPS sample sizes suggest that we could likely detectdifferences in adult apparent survival (φ) between populations with 20 yr of data.

Num

ber

of s

peci

es

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[54]

FIGURE 3. Numbers of species for which current MAPS sample sizes suggest that we could likely detect lineardeclines in adult apparent survival (φ) with 20 yr of data.

Num

ber

of s

peci

es

species at the 20% level, and two species at the25% level. Scrub/successional/disturbed, PacificNorthwest conifer forest, western conifer forest,and western riparian forest habitats were all wellrepresented as “adequately” monitored with 25,27, 24, and 22 species, respectively (Appendix 1).Western oak or juniper woodland andchaparral/sagebrush shrub-steppe habitats werenot well monitored with only five “adequately”monitored species each.

Lists of currently monitorable species and theeffect sizes that can be detected with 80% powerin the remaining six regions are also showngraphically in Figures 2 and 3 and presented inAppendices 2-7. Numbers of currentlymonitorable species are similar for the Southwestand Northeast regions, with tallies of 42 and 38species, respectively (Appendices 2 and 5).Current sample sizes are sufficient to detect a 5%difference in survival between populations withα = 0.20 and 20 yr of data for two species in theNortheast (Appendix 5), but are insufficient todetect 5% effect sizes for any other species orscenario in either of these two regions. At the10% effect size, nine species are currentlymonitorable in the Southwest Region (Appendix2) and an additional six species are monitorablein the Northeast Region (Appendix 5). Theremaining species are sampled in sufficientnumbers to detect only larger effect sizes.Scrub/successional/disturbed, western coniferforest, and western riparian forest habitats werewell represented in the Southwest Region by 21,18, and 20 “adequately” monitored species,respectively (Appendix 2). Western oak orjuniper woodland habitat had 14 “adequately”monitored species but chaparral/sagebrushshrub-steppe habitat had only five “adequately”monitored species. In the Northeast Region,scrub/successional/disturbed, northern hard-wood, and southern hardwood habitats werewell represented by 17, 18, and 19 “adequately”monitored species, respectively (Appendix 5),while northern conifer habitat had only 11“adequately” monitored species and southernconifer habitat, which has a very limiteddistribution in the Northeast Region, had onlyfour “adequately” monitored species.

Currently monitorable species in the fourremaining MAPS regions ranged from 19 in theAlaska/Boreal Canada Region to 25 in theNorth-central Region (Figs. 2-3; Appendices 3-4

and 6-7). No species was sampled sufficiently todetect 5% effect sizes with 80% power in any ofthese remaining regions. However, 10% effectsizes can likely be detected in the North-central,South-central, Southeast, and Alaska/BorealCanada regions for 3, 6, 9, and 3 speciesrespectively, while 15% effect sizes can likely bedetected for 10, 10, 7, and 7 additional species,respectively. In general, scrub/successional/disturbed, northern hardwood, and southernhardwood habitats were well representedaccording to the extent of their distributions by“adequately” monitored species in each of theNorth-central, South-central, and Southeastregions; both northern conifer and southernconifer habitats were less well represented inthose regions. Scrub/successional/disturbed,boreal conifer, and boreal hardwood habitatswere represented by 11, 9, and 6 “adequately”monitored species, respectively, in theAlaska/Boreal Canada Region, likely substan-tially below the numbers of potentiallymonitorable species.

We were able to estimate survival rates on atleast two of the six Pacific Northwest nationalforests for 21 species. At α = 0.20, we found wecould detect 10% differences in survival for onespecies (Swainson’s Thrush), 15% differences insurvival for five additional species (AmericanRobin, MacGillivray’s Warbler, Wilson’s Warbler,Lincoln’s Sparrow, and Dark-eyed Junco), 20%differences for three additional species (DuskyFlycatcher, Warbling Vireo, and Song Sparrow),and 25% differences for six additional species(Hammond’s Flycatcher, “Western” Flycatcher,Winter Wren, Yellow Warbler, Yellow-rumpedWarbler, and Common Yellowthroat), for a totalof 15 of the 21 species (Fig. 4). Interestingly, themean maximum difference in survival betweenforests for the 13 species with CV(�̂) <20% atboth forests was 14.8%; the mean maximumdifference for all 21 species (regardless of CV(�̂))was 25.5%. It is not surprising, therefore, that wehad 80% power to detect the magnitude of thedifference in survival that actually existedbetween the two national forests for eight of the15 species for which we could detect at least a25% difference in survival

POWER OF AN EXPANDED MAPS PROGRAM

Under the expanded (i.e., non-targeted) MAPSprogram, the number of species whose survival

DAVID F. DESANTE AND JAMES F. SARACCO

[55]

would likely be monitorable at the continentalscale would increase by just 13%, despite anoverall increase in the number of stations of 65%(to 713 stations) (Table 4; new species underexpanded (E) or expanded and targeted (T)scenarios are listed in Table 5). The gain inmonitorable species under the expandedscenario was also small at the MAPS-regionalscale. The number of new monitorable speciesincreased by an average of only 22% over the sixexpanded regions (ranging from 9% in theSouth-central Region to 37% in the Alaska/Boreal Canada Region), despite the number ofstations increasing by an average of 104% overthese regions (ranging from 34% in the South-central Region to 205% in the Alaska/BorealCanada Region). The numbers of speciesmonitorable at smaller effect sizes, however,increased markedly under the expandedprogram for both the continental and regional

scales (Figs. 5-6; changes in monitorable effectsizes for individual species are listed in Table 3for continental and Appendices 1-7 for MAPSregions). For example, the number of species forwhich 5% linear declines would likely bedetectable at the continental scale under theexpanded program increased by 160% (from 5 to13 species). As perhaps expected, the 14monitorable species added at the continentalscale by the expanded but not targeted programscenario (Table 5) were typical of habitatsalready well represented by “adequately”monitored species. Only one boreal forestspecies (Olive-sided Flycatcher) and twowestern oak or juniper woodland species (Oakand Juniper titmice) were added.

Under the expanded and targeted programscenario, survival rates of many additionalspecies could likely be monitored effectively byMAPS (Table 4). At the continental scale, the

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[56]

FIGURE 4. Numbers of species for which we would be able to detect differences in survival between clusters ofsix stations on national forest lands in the Pacific Northwest with 80% power over 20 yr. Calculations assumethat capture rates, proportions of residents captured, and survival and recapture probabilities all remained asdocumented on each national forest during 1992-2001.

Num

ber

of s

peci

es

DAVID F. DESANTE AND JAMES F. SARACCO

[57]

TABLE 4. Numbers of monitorable species in the current MAPS program and expected numbers underexpanded or expanded/targeted MAPS programs.

Region1________________________________________________________

NW SW NC SC NE SE AC Continental

No. of stations2

Current 128 66 32 50 65 70 22 433Expanded 128 128 67 67 128 128 67 713Expanded/targeted 148 128 67 67 128 128 67 733

No. of species3

Current 47 42 25 22 38 21 19 105Expanded 47 50 34 24 47 23 26 119Expanded/targeted 63 65 49 49 62 40 43 146

1 See Fig. 1 for region definitions and boundaries.2 Number of stations operated annually that are operated for at least four consecutive years.3 Number of species for which survival can be monitored effectively (i.e., for which 25% differences in survivalbetween two populations or a “25%” linear decline in survival in a single population can be detected with80% power at alpha = 0.2 with 20 years of data).

expanded and targeted scenario would resultin a 39% increase in monitorable species (from105 to 146 species), and a 23% increase over theexpanded program (from 119 to 146 species).The 27 new monitorable species at thecontinental scale from the expanded andtargeted scenario (Tables 4 and 5) include(asterisks indicate species for which MAPS cancurrently obtain continental-scale survival-rateestimates): (1) five species typical of westernoak or juniper woodland habitat (GrayFlycatcher, *Plumbeous Vireo, *Hutton’s Vireo,*Bridled Titmouse, *Black-throated GrayWarbler); (2) seven species typical of sagebrushshrubsteppe or desert scrub and canyons of theWest and Southwest (*Verdin, Sage Thrasher,Canyon Towhee, Brewer ’s Sparrow, *VesperSparrow, *Black-throated Sparrow, *SageSparrow); (3) 12 species that breed in the borealforest and, for some, also in northern coniferhabitat (*Yellow-bellied Sapsucker, Yellow-bellied Flycatcher, *Blue-headed Vireo,Philadelphia Vireo, *Gray Jay, *Ruby-crownedKinglet, *Bicknell’s Thrush, Tennessee Warbler,*Black-throated Blue Warbler, Palm Warbler,Bay-breasted Warbler, and *Clay-coloredSparrow); and (4) three other species fromcentral and eastern North America (*Blue Jay,*Swainson’s Warbler, and *Dickcissel).

At the regional scale, targeted expansion ofjust 20 stations in the Northwest (a 16%increase in stations) could result in a 34%increase the number of effectively monitored

species there (Tables 4 and 5). Targetedexpansion in the remaining MAPS regionscould increase the number of monitorablespecies by an average of 96% (ranging from55% in the Southwest to 132% in the South-central Region; Tables 4 and 5).

DISCUSSIONAlthough our power analyses demonstrate thedifficulty of detecting small (5-10%) differencesin survival between populations or lineardeclines in survival in a single population givensample sizes currently being obtained by theMAPS program, the typical effect sizes seen inMAPS data are often much larger. Indeed, ourresults suggest that we can detect differencesand declines in survival of the magnitudes ofdifferences typically seen between MAPSregions or between clusters of MAPS stationswith 80% power for 84 and 105 species,respectively, at the continental scale, and for 40and 47 species, respectively, in the NorthwestRegion (the region with the largest number ofMAPS stations). The numbers of adequatelymonitored species are smaller in the remainingsix regions; clearly, expansion of MAPS in thoseregions will increase its usefulness as acomponent of continental scale CBM.

Whether a particular species is captured insufficient numbers to detect differences insurvival between populations or changes insurvival within a given population over time

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[58]

TABLE 5. Additional species that would likely be monitorable with 20 yr of data under an expanded (E) ortargeted and expanded (T) MAPS program at continental and regional scales. (Note, √ indicates alreadymonitorable under the current program.)

Region1_____________________________________

NW SW NC SC NE SE AC Cont2

Yellow-bellied Sapsucker (Sphyrapicus varius) T T TRed-breasted Sapsucker (Sphyrapicus ruber) √ E √Ladder-backed Woodpecker (Picoides scalaris) T √Downy Woodpecker (Picoides pubescens) T √ E T √ √ √Hairy Woodpecker (Picoides villosus) √ E E T T √Olive-sided Flycatcher (Contopus cooperi) √ EWestern Wood-Pewee (Contopus sordidulus) √ √ E E √Eastern Wood-Pewee (Contopus virens) T T E T √Yellow-bellied Flycatcher (Empidonax flaviventris) T TAcadian Flycatcher (Empidonax virescens) √ T √ √Gray Flycatcher (Empidonax wrightii) T T TBlack Phoebe (Sayornis nigricans) E √Eastern Phoebe (Sayornis phoebe) √ EGreat Crested Flycatcher (Myiarchus crinitus) E T T T √Brown-crested Flycatcher (Myiarchus tyrannulus) T √Bell’s Vireo (Vireo belli) T √Plumbeous Vireo (Vireo plumbeus) T TCassin's Vireo (Vireo cassinii) T EBlue-headed Vireo (Vireo solitarius) T T THutton's Vireo (Vireo huttoni) T TWarbling Vireo (Vireo gilvus) √ √ T √Philadelphia Vireo (Vireo philadelphicus) T TRed-eyed Vireo (Vireo olivaceus) T √ √ √ √ √Gray Jay (Perisoreus canadensis) E TBlue Jay (Cyanocitta cristata) T √ T TTree Swallow (Tachycineta bicolor) EBarn Swallow (Hirundo rustica) ECarolina Chickadee (Poecile carolinensis) T √ √ √Chestnut-backed Chickadee (Poecile rufescens) T √ √Bridled Titmouse (Baeolophus wollweberi) E TOak Titmouse (Baeolophus inornatus) √ EJuniper Titmouse (Baeolophus ridgwayi) √ EBlack-crested Titmouse (Baeolophus atricristatus) T EVerdin (Auriparus flaviceps) T TBushtit (Psaltriparus minimus) T √ √White-breasted Nuthatch (Sitta carolinensis) E T T √Carolina Wren (Thryothorus ludovicianus) E √ E √ √House Wren (Troglodytes aedon) √ √ √ E E T √Ruby-crowned Kinglet (Regulus calendula) √ T TBicknell's Thrush (Catharus bicknelli) T TWood Thrush (Hylocichla mustelina) √ T √ √ √American Robin (Turdus migratorius) √ √ √ T √ T E √Sage Thrasher (Oreoscoptus montanus) T T TBrown Thrasher (Toxostoma rufum) T √ T T √Blue-winged Warbler (Vermivora pinus) E √ √ √ √Tennessee Warbler (Vermivora peregrina) T TNashville Warbler (Vermivora ruficapilla) T T T T √Virginia's Warbler (Vermivora virginiae) T √ √Lucy's Warbler (Vermivora luciae) E ENorthern Parula (Parula americana) T E T E

DAVID F. DESANTE AND JAMES F. SARACCO

[59]

Yellow Warbler (Dendroica petechia) √ √ √ √ √ T √ √Magnolia Warbler (Dendroica magnolia) T √ T √Black-throated Blue Warbler (Dendroica caerulescens) E TYellow-rumped Warbler (Dendroica coronata) √ T T √ √ √Black-throated Gray Warbler (Dendroica nigrescens) T T TBlack-throated Green Warbler (Dendroica virens) T √ T √Townsend's Warbler (Dendroica townsendi) √ E √Prairie Warbler (Dendroica discolor) T T EPalm Warbler (Dendroica palmarum) T TBay-breasted Warbler (Dendroica castanea) T TBlackpoll Warbler (Dendroica striata) T E √Black-and-white Warbler (Mniotilta varia) E T √ T T √American Redstart (Setophaga ruticilla) √ √ T √ T √ √Prothonotary Warbler (Protonotaria citrea) T E √Worm-eating Warbler (Helmitheros vermivorus) T √ √ √Swainson's Warbler (Limnothlypis swainsonii) T T TOvenbird (Seiurus aurocapillus) E T √ √ T √Northern Waterthrush (Seiurus noveboracensis) T T E √ √Louisiana Waterthrush (Seiurus motacilla) T √ √ √Kentucky Warbler (Oporornis formosus) √ √ E √ √Mourning Warbler (Oporornis philadelphia) √ T T √MacGillivray's Warbler (Oporornis tolmiei) √ T √Common Yellowthroat (Geothlypis trichas) √ √ √ √ √ √ T √Hooded Warbler (Wilsonia citrina) T √ √ √Canada Warbler (Wilsonia canadensis) T T √ √Summer Tanager (Piranga rubra) √ E √Western Tanager (Piranga ludoviciana) √ E √Eastern Towhee (Pipilo erythrophthalmus) T T √ √ √Canyon Towhee (Pipilo fuscus) T TRufous-crowned Sparrow (Aimophila ruficeps) T T √Chipping Sparrow (Spizella passerina) T T T E T √Clay-colored Sparrow (Spizella pallida) T TBrewer’s Sparrow (Spizella breweri) T T TField Sparrow (Spizella pusilla) √ √ T √Vesper Sparrow (Pooecetes gramineus) T T TLark Sparrow (Chondestes grammacus) E T √Black-throated Sparrow (Amphispiza bilineata) T TSage Sparrow (Amphispiza belli) T T TSavannah Sparrow (Passerculus sandwichensis) √ E √Grasshopper Sparrow (Ammodramus savannarum) E ELincoln's Sparrow (Melospiza lincolnii) √ E E √White-throated Sparrow (Zonotrichia albicollis) √ T √Dark-eyed Junco (Junco hyemalis) √ E √ √Rose-breasted Grosbeak (Pheucticus ludovicianus) T T EBlue Grosbeak (Passerina caerulea) √ T T EDickcissel (Spiza americana) T TBrown-headed Cowbird (Molothrus ater) √ √ √ √ T T √Purple Finch (Carpodacus purpureus) √ √ T √American Goldfinch (Carduelis tristis) √ √ T √ T √1 See Fig. 1 for region definitions.2 Continental scale.

TABLE 5. Continued.

Region1_____________________________________

NW SW NC SC NE SE AC Cont2

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[60]

FIGURE 5. Numbers of new species for which we would likely be able to detect differences in survival (φ)between populations with 20 yr of data under an expanded (but not targeted) MAPS program.

Num

ber

of s

peci

es

DAVID F. DESANTE AND JAMES F. SARACCO

[61]

FIGURE 6. Numbers of new species for which we would likely be able to detect linear declines in survival (φ)with 20 yr of data under an expanded (but not targeted) MAPS program.

Num

ber

of s

peci

es

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[62]

depends to some degree on annual variation insurvival rates. As such, our calculations of thenumbers of species for which survival may beeffectively monitored (i.e., “monitorable species”,see Methods: Evaluation of an Expanded MapsProgram) at particular spatial scales may beslightly overestimated. Nevertheless, we rarelyfind strong evidence of time-dependent survivalin MAPS data. In fact, only 46 (11%) of 411species/region combinations for which we wereable to obtain survival rate estimates with the 10-yr 1992-2001 data set showed strong evidence oftime dependence (i.e., models with temporallyvarying survival were within 2 AIC units oftime-constant models in only 11% of these cases;DeSante and Kaschube 2006). Thus, ourestimates of numbers of currently monitorablespecies are likely reasonable.

Our assessment of program expansionoptions indicates that a focused approachtargeting under-represented species and habitatswill be required to maximize the number ofspecies and habitats adequately monitored byMAPS. We suggest that target species beselected based not only on their habitats andecology (and thus their inherent monitorability),but also on their status as species ofconservation concern or whether they representhabitats of special management concern. This isthe approach that we used to identify new targetspecies in both the Northwest (Pyle et al. 2005)and Northeast (DeSante et al. 2008); similaranalyses will need to be conducted for theremaining regions to best direct respectivesampling strategies. The addition andredistribution of stations in the Northwest, aswell as the undertaking of analyses to guideMAPS in the remaining six regions, wouldclearly be facilitated by the integration of MAPSinto a broader CBM effort.

In order to address finer-scale monitoringneeds, we suggest that MAPS will be mostuseful if sampling is targeted to capture alimited number of focal species and habitats (ormanagement regimes), or possibly by samplingalong a simple habitat gradient. For example, inour cluster-scale analysis of Pacific Northwestnational forests, we were able to comparesurvival rate estimates for 21 species betweenvarious 6-station clusters. About half of thesespecies met our criteria for being “adequatelymonitorable”, and many of these were

monitorable at an effect size that was actuallyseen (based on our estimates) between clustersof stations. Yet, an individual species beingcompared between clusters was often eitherabsent or rare at many of the stations in the twoclusters because stations were set up to sample abroad range of species and habitats. Forexample, the average number of stations atwhich a species was present (as a breedingresident) on the two clusters where it was mostabundant was just 9.1 (out of 12 possiblestations). A more focused sampling strategywould be to select a species (or set of species) ofconservation concern and two habitats (e.g.,three stations in each of two habitat types ineach cluster) or a single habitat gradient. Such ahypothesis-driven design would require onlytwo teams of banders (each operating sixstations) and should be capable of providinginference on habitat factors influencing survivaland reproduction.

The targeted hypothesis-driven samplingstrategy described above lends itself well tomanagement, as it provides testable hypothesesregarding habitat types or characteristics thatpromote better survival (or productivity) inpopulations of species of conservation concern.A similar sampling strategy can be employed tomonitor demographic responses to managementactions (i.e., effectiveness monitoring). Each ofthese targeted hypothesis-driven monitoringefforts directly supports management, has beenreferred to as “management-based” monitoringby the Coordinated Bird Monitoring WorkingGroup (2004) or “targeted” monitoring by theNABCI Monitoring Subcommittee (2006), andwas distinguished in those reports from“surveillance” (Coordinated Bird MonitoringWorking Group 2004) or “broad-scale” (NABCIMonitoring Subcommittee 2006) monitoring, forwhich the goals were, among others, to provideessential information for prioritizing species forconservation actions and to identify emergingconservation issues. We feel that the distinctionbetween “management-based” monitoring and“surveillance” monitoring is somewhat artificial,and suggest that they reflect, to some extent, thespatio-temporal continuum of questions thatmonitoring data can address. What is usuallyconsidered to be “management-based”monitoring typically requires a spatiallyintensive sampling effort that can assess changes

over relatively short time frames, while what isusually considered to be “surveillance”monitoring typically requires a spatiallyextensive sampling effort over a long time scale.

We argue that the MAPS is well-suited tofulfilling both “management-based” and“surveillance” monitoring goals. Indeed, theoptimal sampling design of the overall MAPSprogram can be envisioned as collections oftargeted, hypothesis-driven sampling strategiesat local spatial scales that are integrated intolarger regional sampling schemes. The latter, inturn, can provide much of the necessary effortrequired for “surveillance” monitoring atregional and continental scales. In such anoverall strategy, each local hypothesis-drivensampling scheme of about 12 stations would aimfor the 20-yr time horizon, which according tothe power analyses presented above, wouldlikely provide statistically significant results atthe local scale. Such sets of stations would notneed to be operated indefinitely to achieve thetargeted monitoring objective they weredesigned to address. Moreover, if a number ofsuch hypothesis-driven sampling schemes wereintegrated through a CBM framework, thetermination of a few of the stations, even after asfew as 4-5 yr, would still provide important datafor the broader-scale program, provided thatnew stations were initiated in similar habitatswithin the region of interest. This is, in largepart, how the current MAPS program operates –as a collection of smaller scale studies. Untilnow, however, little direction has been given tothe establishment of new stations in such a wayas to maximize the ability of the program toaddress monitoring questions at broader spatialscales.

We also suggest, however, that collections ofintegrated hypothesis-driven sampling schemesfor target species and habitats involving 20-yrtime horizons, even when fully coordinated atregional and continental scales, may not provideall of the broad-scale, long-term needs of acoordinated demographic monitoring program.Priority species and habitats and specifichypotheses driving any current samplingscheme may need to change as new andunforeseen threats and environmental drivingforces arise (e.g., through large-scale processessuch as climate change, soil and wateracidification, and airborne environmental

contamination) that may affect or otherwiseinteract with efforts to monitor habitat-specificdemographic rates or demographic effects ofhabitat management. We suggest that some levelof spatially extensive, long-term “surveillance”monitoring should be a critical component ofdemographic monitoring efforts, and that somenumber MAPS stations be operated indefinitelyacross the continent. We further suggest thatnational parks, research natural areas, and otherprotected areas might be optimal for thismonitoring, as they tend to be relatively pristineand not subject to complicating effects of on-going land management. It is also possible thatresearchers utilizing protocols other than that ofMAPS to collect capture-recapture data onNorth American landbirds could also contributeto continental-scale demographic monitoring(provided that their within- and between-yeartime periods of data collection are sufficient). Anadditional future direction of MAPS may be thesolicitation of such data and the development ofanalytical methods that integrate them withMAPS data.

Our final discussion addresses the fact thatMAPS stations, in general, are not sitedaccording to a probabilistic sampling strategy.Thus, inferences cannot be made beyond thesample of stations being considered for anyparticular analysis. While it might be useful toprovide inferences regarding demographicparameters for specific geographical areas atvarious scales, the greatest value of MAPS lies inits ability to inform conservation by identifyingdemographic causes of population change,inferring ultimate ecological causes ofpopulation change by relating avian demo-graphic rates to habitat and weather variables,and evaluating the effectiveness of managementstrategies designed to reverse populationdeclines and enhance depressed populations. Tooptimize these processes, it is appropriate to sitestations according to hypothesis-drivensampling strategies that, ideally, incorporate aprobabilistic element as to the exact locations ofthe stations. This, in fact, is the manner in whichmany MAPS stations are sited.

Despite the non-random distribution ofstations, program-wide (continental) MAPSpopulation trends (lambda) have been shown tobe correlated with survey-wide (continental)BBS population trends derived from randomly-

DAVID F. DESANTE AND JAMES F. SARACCO

[63]

sited (albeit roadside) BBS routes (Saracco et al.2008). Interestingly, MAPS trends tended to beslightly more positive than BBS trends, likelyreflecting the fact that MAPS stations oftencannot be operated in poor quality habitats,such as extensive agricultural fields and highlydeveloped or urbanized areas, in contrast to BBSroutes that often traverse those habitats. Thus,the distribution of stations resulting frommultiple collections of hypothesis-drivensampling schemes across the continent may notbe too unrepresentative of the overallenvironment, at least at the program-wide(continental) scale, and the trade-off in theability to make inferences at large geographicscales in favor of inferences regarding theeffectiveness of more local habitat managementactions may not be very severe.

ACKNOWLEDGMENTSThe U.S. Geological Survey, Biological ResourcesDivision (USGS-BRD), Forest and RangelandEcosystem Science Center, Corvallis, OR,provided funding for analyses presented here.Support for MAPS during 1992-2001 (the yearsfor which data were analyzed herein) wasprovided by the National Fish and WildlifeFoundation, USDI Fish and Wildlife Service,USGS-BRD, Cape Cod National Seashore,Denali National Park, Yosemite National Park,Sequoia/Kings Canyon National Park,Shenandoah National Park, USDA ForestService (Regions 1 and 6), Flathead NationalForest, DoD Legacy Resources ManagementProgram, Atlantic Division of the Department ofthe Navy, Army Corps. of Engineers, ArmyGarrisons Ft. Belvoir and Ft. Bragg, AdjutantGeneral’s Department of Texas, San FranciscoState University, Confederated Salish andKootenai Tribes, Nature Reserve of OrangeCounty (CA), Sequoia Natural HistoryAssociation, Yosemite Association, YosemiteFoundation, The Nature Conservancy, and themany members and friends of The Institute forBird Populations. We are greatly indebted to thehundreds of independent cooperators that havecontributed to MAPS over the years. IBP staffbiologists provided rigorous verification andvetting of data. Environmental SystemsResearch Institute, Inc. provided GIS softwaresupport through the ESRI Conservation

Program. We thank J. Bart, D. Lambert, and ananonymous reviewer for helpful comments onearlier versions of this paper. This is Contri-bution No. 370 of The Institute for BirdPopulations.

LITERATURE CITEDBART, J. 2005. Monitoring the abundance of bird

populations. Auk 122:15-25.BART, J., AND C. J. RALPH. 2005. The need for a North

American coordinated bird monitoring program,p. 982-984. In C. J. Ralph and T. R. Rich [eds.], Birdconservation implementation and integration inthe Americas: proceedings of the thirdinternational Partners in Flight conference. USDAForest Service General Technical Report PSW-191.

BART, J., K. P. BURNHAM, E. H. DUNN, C. M. FRANCIS,AND C. JOHN RALPH. 2004. Goals and strategies forestimating trends in landbird abundance. Journalof Wildlife Management 68:611-626.

COORDINATED BIRD MONITORING WORKING GROUP. 2004.Monitoring avian conservation: rationale, design,and coordination. International Association of Fishand Wildlife Agencies.

DESANTE, D. F. 1992. Monitoring Avian Productivityand Survivorship (MAPS): a sharp, rather thanblunt, tool for monitoring and assessing landbirdpopulations, p. 511-521. In D. R. McCullough andR. H. Barrett [eds.], Wildlife 2001: populations.Elsevier Applied Science, London, UK.

DESANTE, D. F., K. M. BURTON, P. VELEZ, D. FROEHLICH,AND D. R. KASCHUBE. 2009b. MAPS Manual, 2009Protocol. The Institute for Bird Populations, PointReyes Station, CA. 78 pp.

DESANTE, D. F., AND D. R. KASCHUBE. 2006. TheMonitoring Avian Productivity and Survivorship(MAPS) program 1999, 2000, and 2001 report. BirdPopulations 7:23-89.

DESANTE, D. F., AND D. R. KASCHUBE. 2009. TheMonitoring Avian Productivity and Survivorship(MAPS) program 2004, 2005, and 2006 report. BirdPopulations 9:86-169.

DESANTE, D. F., D. R. KASCHUBE, J. F. SARACCO, AND J.E. HINES. 2009a. Power to detect differences andtrends in apparent survival rates. Bird Populations9:29-41.

DESANTE, D. F., M. P. NOTT, AND D. R. KASCHUBE. 2005.Monitoring, modeling, and management: whybase avian management on vital rates and howshould it be done? Pp 795-804 in Ralph, C.J., andT.D. Rich (eds.), Bird Conservation Implemen-tation and Integration in the Americas. USDAForest Service Gen. Tech. Rep. PSW-191.

DESANTE, D. F., M. P. NOTT, AND D. R. O’GRADY. 2001.Identifying the proximate demographic cause(s) ofpopulation change by modelling spatial variation

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[64]

in productivity, survivorship, and populationtrends. Ardea 89(special issue):185-207.

DESANTE, D. F., AND D. K. ROSENBERG. 1998. What dowe need to monitor in order to manage landbirds?Pp.93-106 in Marzluff, J. M. and R. Sallabanks(eds.), Avian Conservation: Research andManagement. Island Press, Washington, D.C.

DESANTE, D. F., J. F. SARACCO, P. PYLE, D. R. KASCHUBE,AND M. K. CHAMBERS. 2008. Integrating MAPS intoCoordinated Bird Monitoring in the Northeast (U.S. Fish and Wildlife Service Region 5). TheInstitute for Bird Populations, Point Reyes Station,CA. 99 pp.

DESANTE, D. F., O. E. WILLIAMS, AND K. M. BURTON.1993. The Monitoring Avian Productivity andSurvivorship (MAPS) program: overview andprogress, p. 208-222. In D. M. Finch and P. W.Stangel [eds.], Status and management ofNeotropical migratory birds. USDA Forest ServiceGeneral Technical Report RM-GTR-229.

HINES, H. E., W. L. KENDALL, AND J. D. NICHOLS. 2003.On the use of the robust design with transientcapture-recapture models. Auk 120:1151-1158.

NABCI MONITORING SUBCOMMITTEE. 2006. Oppor-tunities for improving North American avian mon-itoring. Draft interim report. < http://www. nabci-us.org/aboutnabci/avianmonitoringdraft906.pdf >

NOTT, M. P. AND D. F. DESANTE. 2002. Demographicmonitoring and the identification of transients inmark-recapture models. Pages 727-736 in J. M.Scott, P. J. Heglund, and M. L. Morrison, editors.Predicting species occurrences: issues of accuracyand scale. Island Press, Washington, D.C., USA.

PYLE, P., D. F. DESANTE, M. P. NOTT, AND D. R.KASCHUBE. 2005. The MAPS Program in the PacificNorthwest: Current Status and Future Directions.The Institute for Bird Populations, Point ReyesStation, CA. 107 pp.

RICH, T. D., C. J. BEARDMORE, H. BERLANGA, P. J.BLANCHER, M. S. W. BRADSTREET, G. S. BUTCHER, D.W. DEMAREST, E. H. DUNN, W. C. HUNTER, E. E.INEGO-ELIAS, J. A. KENNEDY, A. M. MARTELL, A. O.PANJABI, D. N. PASHLEY, K. V. ROSENBERG, C. M.RUSTAY, J. S. WENDT, AND T. C. WILL. 2004. Partnersin Flight North American landbird conservationplan. Cornell Laboratory of Ornithology. Ithaca,NY.

ROSENBERG, D. K., D. F. DESANTE, K. S. MCKELVEY, AND

J. E. HINES. 1999. Monitoring survival rates ofSwainson’s Thrush Catharus ustulatus at multiplespatial scales. Bird Study 46 (Supplement):198-208.

ROSENBERG, D. K., D. F. DESANTE, AND J. E. HINES. 2000.Monitoring survival rates of landbirds at varyingspatial scales: an application of the MAPS program,p. 178-184. In R. Bonney, D. N. Pashley, R. J. Cooper,and L. Niles [eds.], Strategies for bird conservation:the Partners in Flight planning process. USDAForest Service Proceedings RMRS-P-16.

SARACCO, J. F., D. F. DESANTE, AND D. R. KASCHUBE.2008. Assessing landbird monitoring programsand demographic causes of population trends.Journal of Wildlife Management. 72:1665-1673.

WHITE, G.C. AND K. P. BURNHAM. 1999. ProgramMARK: survival estimation for populations ofmarked animals. Bird Study 46 (Supplement):120-138.

DAVID F. DESANTE AND JAMES F. SARACCO

[65]

APP

EN

DIX

1. E

ffec

t si

zes

(%)

curr

entl

y d

etec

tabl

e by

MA

PS w

ith

80%

pow

er a

nd 2

0 ye

ars

of d

ata

for

47 s

peci

es in

the

Nor

thw

est

Reg

ion.

Tw

o ef

fect

typ

es w

ere

cons

ider

ed: a

tw

o-po

pula

tion

com

pari

son

of s

urvi

val (

2P)

and

a p

opul

atio

n w

ith

linea

rly

dec

linin

g su

rviv

al (

LD

). E

ffec

t si

zes

are

give

n fo

r tw

o al

pha-

leve

ls, 0

.10

and

0.2

0. N

ote

that

we

do

not

prop

ose

an e

xpan

ded

but

not

tar

gete

d p

rogr

am i

n th

e N

orth

wes

t. Se

e Ta

ble

2 fo

r m

ajor

hab

itat

s, T

able

3 f

or f

ield

def

init

ions

, and

Tabl

es 3

and

5 fo

r sc

ient

ific

nam

es.

α=

0.1

0α

= 0

.20

____

____

____

____

____

____

____

Spec

ies

Maj

or h

abit

at(s

)St

aIn

dR

es�̂

p̂2P

LD

2PL

D

Red

-nap

ed S

apsu

cker

PN

C,W

C,W

R28

370

360.

366

0.54

420

20R

ed-b

reas

ted

Sap

suck

er

PNC

,WC

,WR

4767

170

0.44

90.

371

2020

Hai

ry W

ood

peck

er

PNC

,WC

,WO

J,WR

5624

925

0.55

00.

286

2025

15W

este

rn W

ood

-Pew

ee

PNC

,WC

,WR

5410

1586

0.51

30.

336

2015

2015

“Tra

ill's

” Fl

ycat

cher

SS

D,W

R32

1115

770.

524

0.46

020

1515

10H

amm

ond

's F

lyca

tche

r PN

C,W

C53

1142

105

0.45

30.

404

2015

1515

Dus

ky F

lyca

tche

r SS

D,P

NC

,WC

4719

0313

30.

488

0.42

115

1515

10“W

este

rn”

Flyc

atch

er

PNC

,WC

5416

8111

80.

481

0.31

420

1515

15W

arbl

ing

Vir

eo

PNC

,WC

,WR

8635

6030

60.

479

0.42

210

1010

10B

lack

-cap

ped

Chi

ckad

ee

PNC

,WC

,WR

4383

173

0.46

60.

443

2015

1515

Mou

ntai

n C

hick

adee

PN

C,W

C41

891

600.

471

0.41

425

1520

15B

ewic

k's

Wre

n SS

D,C

SS13

168

220.

452

0.46

120

2515

Hou

se W

ren

SSD

,WR

2759

757

0.27

80.

373

25W

inte

r W

ren

PNC

,WC

3792

670

0.37

50.

521

2520

2015

Rub

y-cr

owne

d K

ingl

et

PNC

,WC

1462

850

0.25

50.

275

25V

eery

PNC

,WR

492

120.

714

0.38

625

2520

Swai

nson

's T

hrus

hPN

C,W

C,W

R78

7394

856

0.58

60.

629

55

55

Her

mit

Thr

ush

PNC

,WC

3580

866

0.46

10.

537

2015

1515

Am

eric

an R

obin

SS

D,P

NC

,WC

,WO

J,WR

116

3947

370

0.57

50.

258

1010

1010

Var

ied

Thr

ush

PNC

3038

826

0.49

40.

407

2520

Wre

ntit

SS

D,C

SS21

487

580.

527

0.63

715

1515

10G

ray

Cat

bird

SS

D,W

R12

584

510.

561

0.45

015

1015

10O

rang

e-cr

owne

d W

arbl

erSS

D,W

OJ

4014

6894

0.48

10.

418

2015

1515

Yello

w W

arbl

er

SSD

,WR

4940

8636

30.

571

0.49

35

105

5Ye

llow

-rum

ped

War

bler

PNC

,WC

6029

0621

70.

486

0.20

820

1515

15To

wns

end

's W

arbl

er

PNC

2382

382

0.45

10.

216

2020

Am

eric

an R

edst

art

WR

730

424

0.44

30.

526

2520

Mac

Gill

ivra

y's

War

bler

SS

D,P

NC

,WC

8557

6354

20.

479

0.61

45

105

10

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[66]

DAVID F. DESANTE AND JAMES F. SARACCO

[67]

APP

EN

DIX

1. C

onti

nued

.

α =

0.1

0α

= 0

.20

____

____

____

____

____

____

____

Spec

ies

Maj

or h

abit

at(s

)St

aIn

dR

es�̂

p̂2P

LD

2PL

D

Com

mon

Yel

low

thro

at

SSD

2714

5215

30.

499

0.55

615

1010

10W

ilson

's W

arbl

er

SSD

,PN

C,W

C56

3638

265

0.43

80.

526

1510

1010

Yello

w-b

reas

ted

Cha

t SS

D,W

R15

873

890.

448

0.56

020

1520

15W

este

rn T

anag

erPN

C,W

C73

1386

100

0.53

80.

129

2025

15G

reen

-tai

led

Tow

hee

SSD

,CSS

520

921

0.60

20.

360

2025

15Sp

otte

d T

owhe

eSS

D,W

R,C

SS42

955

880.

476

0.50

815

1515

10Sa

vann

ah S

parr

owSS

D2

321

310.

585

0.28

525

1520

15Fo

x Sp

arro

w

SSD

,PN

C,W

C25

505

440.

553

0.45

115

1015

10So

ng S

parr

ow

SSD

.WR

9858

6864

30.

470

0.59

85

105

5L

inco

ln's

Spa

rrow

SSD

,PN

C,W

C37

2091

267

0.43

20.

631

1010

1010

Whi

te-c

row

ned

Spa

rrow

SSD

1459

861

0.47

40.

544

2015

1515

Dar

k-ey

ed Ju

nco

SSD

,PN

C,W

C83

5217

533

0.46

10.

500

1010

510

Bla

ck-h

ead

ed G

rosb

eak

PNC

,WC

,WO

J,WR

68

2057

155

0.56

80.

280

1010

1010

Laz

uli B

unti

ngSS

D,W

R,C

SS25

1074

620.

552

0.24

525

2025

15R

ed-w

inge

d B

lack

bird

SSD

1879

942

0.78

40.

148

1510

1010

Bro

wn-

head

ed C

owbi

rd

SSD

,WR

6074

868

0.48

60.

485

2015

1515

Bul

lock

's O

riol

e W

OJ,W

R19

497

250.

454

0.42

915

2515

Purp

le F

inch

PN

C,W

C32

2386

155

0.43

40.

364

2015

2015

Am

eric

an G

old

finc

h SS

D,W

R20

1434

108

0.47

00.

328

2015

1515

APP

EN

DIX

2. E

ffec

t si

zes

(%)

curr

entl

y d

etec

tabl

e by

MA

PS w

ith

80%

pow

er a

nd 2

0 yr

s of

dat

a, a

nd li

kely

det

ecta

ble

und

er a

n ex

pand

ed M

APS

pro

gram

, for

42

spec

ies

in th

e So

uthw

est R

egio

n. T

wo

effe

ct ty

pes

wer

e co

nsid

ered

: a tw

o-po

pula

tion

com

pari

son

of s

urvi

val (

2P)

and

a p

opul

atio

n w

ith

linea

rly

dec

linin

g su

rviv

al(L

D).

Eff

ect s

izes

are

giv

en fo

r tw

o al

pha-

leve

ls, 0

.10

and

0.2

0. S

ee T

able

2 fo

r m

ajor

hab

itat

s, T

able

3 fo

r fi

eld

def

init

ions

, and

Tab

les

3 an

d 5

for

scie

ntif

ic n

ames

.

Cur

rent

Exp

and

ed__

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

____

____

____

____

____

____

α =

0.1

0α

= 0

.20

α =

0.1

0α

= 0

.20

____

____

____

____

____

____

____

____

____

____

____

____

____

Spec

ies

Maj

or h

abit

at(s

)St

aIn

dR

es�̂

p̂2P

LD

2PL

D2P

LD

2PL

D

Red

-nap

ed S

apsu

cker

W

C,W

R4

119

230.

494

0.66

525

1520

1520

1515

15N

utta

ll's

Woo

dpe

cker

W

OJ,W

R16

195

280.

551

0.38

125

1520

1520

1515

15D

owny

Woo

dpe

cker

SS

D,W

OJ,W

R20

217

240.

605

0.43

120

1520

1515

1015

10O

live-

sid

ed F

lyca

tche

rW

C2

553

0.87

00.

724

2520

2015

1510

1010

Wes

tern

Woo

d-P

ewee

WC

,WO

J,WR

1527

827

0.38

40.

446

2520

2520

2515

“Wes

tern

” Fl

ycat

cher

W

C15

892

430.

595

0.18

325

2520

2515

2015

Ash

-thr

oate

d F

lyca

tche

r W

OJ,C

SS31

676

800.

659

0.13

715

1515

1010

1010

10B

ell's

Vir

eo

WR

511

215

0.48

20.

596

2015

2515

2015

War

blin

g V

ireo

W

C,W

R15

1253

520.

539

0.46

020

1520

1515

1015

10St

elle

r's Ja

y W

C9

9813

0.76

20.

234

2520

1515

15B

lack

-cap

ped

Chi

ckad

ee

WC

,WR

711

017

0.36

60.

600

2520

2025

15M

ount

ain

Chi

ckad

ee

WC

1025

439

0.40

80.

292

2525

2025

20C

hest

nut-

back

ed C

hick

adee

W

C6

259

410.

501

0.51

225

1520

1515

1515

10O

ak T

itm

ouse

W

OJ

1115

318

0.51

40.

358

2520

2515

Juni

per

Titm

ouse

W

OJ

444

90.

655

0.40

925

2515

2015

Bus

htit

SS

D,W

OJ,C

SS30

975

171

0.30

00.

152

2525

20B

ewic

k's

Wre

n SS

D,C

SS38

1047

137

0.42

40.

538

1515

1515

1510

1010

Hou

se W

ren

SSD

,WO

J,WR

2794

111

30.

388

0.43

920

1515

1515

1015

10Sw

ains

on's

Thr

ush

WC

,WR

818

1912

10.

619

0.57

810

1010

105

105

5H

erm

it T

hrus

hW

C9

410

750.

450

0.40

025

1520

1515

1515

10A

mer

ican

Rob

in

SSD

,WC

,WO

J,WR

2770

690

0.52

90.

309

2015

1515

1510

1510

Wre

ntit

SS

D,C

SS16

838

106

0.54

00.

618

1510

1010

1010

1010

Ora

nge-

crow

ned

War

bler

SSD

,WO

J15

1076

790.

413

0.31

820

2520

2015

2015

Vir

gini

a's

War

bler

SS

D,W

OJ

928

323

0.48

40.

395

2520

1525

15Ye

llow

War

bler

SS

D,W

R18

1097

110

0.47

90.

526

1510

1510

1010

1010

Com

mon

Yel

low

thro

at

SSD

2216

2417

80.

525

0.42

610

1010

1010

1010

10W

ilson

's W

arbl

erSS

D,W

C9

2609

124

0.44

60.

510

2015

1515

1510

1010

Yello

w-b

reas

ted

Cha

t SS

D,W

R17

750

108

0.52

20.

521

1510

1510

1010

1010

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[68]

Sum

mer

Tan

ager

W

R6

125

230.

506

0.45

120

2515

2015

2015

Spot

ted

Tow

hee

SSD

,WO

J,WR

,CSS

3611

7016

50.

496

0.43

010

1010

1010

1010

10C

alif

orni

a To

whe

e SS

D17

392

620.

536

0.35

625

1520

1520

1515

10Fo

x Sp

arro

w

SSD

,WC

395

110.

519

0.55

120

2025

15So

ng S

parr

ow

SSD

,WR

2621

0128

80.

520

0.47

910

1010

1010

105

10L

inco

ln's

Spa

rrow

SSD

,WC

311

218

0.43

70.

876

2025

2020

1520

15D

ark-

eyed

Junc

oSS

D,W

C8

336

490.

387

0.56

020

2515

2015

2015

Bla

ck-h

ead

ed G

rosb

eak

WC

,WO

J,WR

3516

8415

70.

526

0.34

515

1515

1010

1010

10B

lue

Gro

sbea

k W

R11

187

170.

409

0.40

825

2020

Laz

uli B

unti

ng

SSD

,WR

,CSS

2164

341

0.37

20.

473

2025

2025

1520

15R

ed-w

inge

d B

lack

bird

SS

D9

232

410.

880

0.03

720

1515

Bro

wn-

head

ed C

owbi

rd

SSD

,WR

3027

538

0.46

30.

568

2515

2015

2015

1510

Bul

lock

's O

riol

e W

OJ,W

R18

566

460.

458

0.37

715

2515

2015

2015

Purp

le F

inch

W

C7

932

960.

531

0.30

420

1515

1515

1010

10

DAVID F. DESANTE AND JAMES F. SARACCO

[69]

APP

EN

DIX

2. C

onti

nued

.

Cur

rent

Exp

and

ed__

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

____

____

____

____

____

____

α =

0.1

0α

= 0

.20

α =

0.1

0α

= 0

.20

____

____

____

____

____

____

____

____

____

____

____

____

____

Spec

ies

Maj

or h

abit

at(s

)St

aIn

dR

es�̂

p̂2P

LD

2PL

D2P

LD

2PL

D

APP

END

IX 3

. Eff

ect

size

s (%

) cu

rren

tly

det

ecta

ble

by M

APS

wit

h 80

% p

ower

and

20

yrs

of d

ata,

and

like

ly d

etec

tabl

e un

der

an

expa

nded

MA

PS p

rogr

am, f

or 2

5sp

ecie

s in

the

Nor

th-c

entr

al R

egio

n.Tw

o ef

fect

type

s w

ere

cons

ider

ed: a

two-

popu

latio

n co

mpa

riso

n of

sur

viva

l (2P

) and

a p

opul

atio

n w

ith li

near

ly d

eclin

ing

surv

ival

(LD

). Ef

fect

siz

es a

re g

iven

for

two

alph

a-le

vels

, 0.1

0 an

d 0.

20. S

ee T

able

2 fo

r m

ajor

hab

itats

, Tab

le 3

for

field

def

initi

ons,

and

Tab

les

3 an

d 5

for

scie

ntifi

c na

mes

.

Cur

rent

Exp

and

ed__

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

____

____

____

____

____

____

α =

0.1

0α

= 0

.20

α =

0.1

0α

= 0

.20

____

____

____

____

____

____

____

____

____

____

____

____

____

Spec

ies

Maj

or h

abit

at(s

)St

aIn

dR

es�̂

p̂2P

LD

2PL

D2P

LD

2PL

D

“Tra

ill's

” Fl

ycat

cher

SSD

,NC

,NH

1264

949

0.48

40.

509

2015

2015

1510

1510

Lea

st F

lyca

tche

rN

H12

847

680.

368

0.43

125

2020

1515

1515

15R

ed-e

yed

Vir

eoN

H,S

H24

631

420.

534

0.39

120

2515

2015

2015

Bla

ck-c

appe

d C

hick

adee

NC

,NH

2865

450

0.41

70.

394

2520

2520

2015

Tuft

ed T

itm

ouse

NH

,SH

1016

317

0.55

50.

407

2020

2015

2015

Hou

se W

ren

SSD

1889

283

0.32

20.

440

2025

2020

2020

15V

eery

NH

939

045

0.57

50.

586

1510

1010

1010

1010

Woo

d T

hrus

hN

H,S

H12

263

310.

453

0.31

720

2025

1520

15A

mer

ican

Rob

inSS

D,N

C,N

H,S

H26

787

390.

380

0.42

325

2520

2520

Gra

y C

atbi

rdSS

D,N

H,S

H23

2196

196

0.49

70.

478

1010

1010

1010

1010

Yello

w W

arbl

erSS

D15

1336

123

0.54

80.

398

1515

1510

1010

1010

Che

stnu

t-si

ded

War

bler

SSD

,NH

438

042

0.37

80.

572

2025

2020

1520

15A

mer

ican

Red

star

tN

H,S

H11

605

440.

454

0.28

220

2515

2015

1515

Ken

tuck

y W

arbl

er

SH2

8812

0.61

00.

698

1525

1520

1515

10M

ourn

ing

War

bler

SSD

,NC

312

215

0.42

10.

608

2525

2025

20C

omm

on Y

ello

wth

roat

SSD

2412

9911

70.

433

0.49

820

1515

1515

1010

10Fi

eld

Spa

rrow

SSD

859

361

0.42

60.

392

2020

2515

2015

Song

Spa

rrow

SS

D22

1137

117

0.44

40.

517

2015

1515

1510

1010

Swam

p Sp

arro

w

SSD

620

114

0.38

60.

776

2520

2520

2515

Whi

te-t

hroa

ted

Spa

rrow

N

C3

309

330.

383

0.62

120

2515

2015

2015

Nor

ther

n C

ard

inal

SS

D,N

H,S

H18

603

500.

539

0.35

525

1525

1520

1515

15In

dig

o B

unti

ngSS

D16

793

700.

518

0.31

225

1520

1515

1515

10B

row

n-he

aded

Cow

bird

SS

D22

219

180.

487

0.41

125

2025

15B

alti

mor

e O

riol

e N

H,S

H14

309

290.

615

0.18

425

2520

2515

Am

eric

an G

old

finc

h SS

D21

1635

164

0.34

70.

360

2025

2020

2015

15

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[70]

DAVID F. DESANTE AND JAMES F. SARACCO

[71]

APP

END

IX 4

. Eff

ect

size

s (%

) cu

rren

tly

det

ecta

ble

by M

APS

wit

h 80

% p

ower

and

20

yrs

of d

ata,

and

like

ly d

etec

tabl

e un

der

an

expa

nded

MA

PS p

rogr

am, f

or 2

2sp

ecie

s in

the

Sout

h-ce

ntra

l Reg

ion.

Tw

o ef

fect

type

s w

ere

cons

ider

ed: a

two-

popu

latio

n co

mpa

riso

n of

sur

viva

l (2P

) and

a p

opul

atio

n w

ith li

near

ly d

eclin

ing

surv

ival

(LD

). Ef

fect

siz

es a

re g

iven

for

two

alph

a-le

vels

, 0.1

0 an

d 0.

20. S

ee T

able

2 fo

r m

ajor

hab

itats

, Tab

le 3

for

field

def

initi

ons,

and

Tab

les

3 an

d 5

for

scie

ntifi

c na

mes

.

Cur

rent

Exp

and

ed__

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

____

____

____

____

____

____

α =

0.1

0α

= 0

.20

α =

0.1

0α

= 0

.20

____

____

____

____

____

____

____

____

____

____

____

____

____

Spec

ies

Maj

or h

abit

at(s

)St

aIn

dR

es�̂

p̂2P

LD

2PL

D2P

LD

2PL

D

Aca

dia

n Fl

ycat

cher

SH6

349

270.

583

0.47

820

1520

1520

1515

10W

hite

-eye

d V

ireo

SSD

,SH

2413

8011

70.

601

0.49

810

1010

1010

1010

10B

ell's

Vir

eoSS

D12

450

410.

593

0.37

720

1520

1520

1515

15R

ed-e

yed

Vir

eoN

H,S

H15

328

300.

575

0.22

125

2525

20Tu

fted

Tit

mou

seN

H,S

H24

318

360.

486

0.29

620

2515

2515

2515

Car

olin

a W

ren

SH,S

C31

888

790.

459

0.51

715

1515

1015

1015

10B

ewic

k's

Wre

nSS

D21

480

440.

431

0.49

820

2520

2520

2515

Gra

y C

atbi

rdSS

D,N

H,S

H9

741

570.

581

0.46

715

1015

1015

1010

10B

row

n T

hras

her

SSD

1528

815

0.40

80.

572

2525

25L

ong-

bille

d T

hras

her

SSD

,SB

E3

133

120.

628

0.39

625

2520

Blu

e-w

inge

d W

arbl

erSS

D,S

H4

244

180.

555

0.49

025

1520

1520

1520

15Ye

llow

War

bler

SSD

310

313

0.35

30.

515

2525

Ken

tuck

y W

arbl

erSH

1145

741

0.60

80.

535

1515

1510

1510

1510

Com

mon

Yel

low

thro

atSS

D14

440

360.

471

0.47

825

1520

1520

1520

15Ye

llow

-bre

aste

d C

hat

SSD

865

172

0.52

10.

391

2515

2015

2015

1515

Sum

mer

Tan

ager

SH,S

C19

223

170.

531

0.37

725

2520

Oliv

e Sp

arro

wSB

E3

208

240.

510

0.73

820

1520

1520

1515

15Fi

eld

Spa

rrow

SSD

3111

0810

80.

482

0.34

320

1515

1515

1515

10N

orth

ern

Car

din

alSS

D,N

H,S

H,S

C48

2475

245

0.58

10.

356

1010

1010

1010

1010

Ind

igo

Bun

ting

SSD

2311

8111

10.

459

0.42

220

1515

1515

1515

10Pa

inte

d B

unti

ngSS

D31

1591

103

0.55

80.

439

1010

1010

1010

1010

Bro

wn-

head

ed C

owbi

rdSS

D39

477

400.

474

0.36

420

2515

2515

2015

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[72]

APP

END

IX 5

. Eff

ect

size

s (%

) cu

rren

tly

det

ecta

ble

by M

APS

wit

h 80

% p

ower

and

20

yrs

of d

ata,

and

like

ly d

etec

tabl

e un

der

an

expa

nded

MA

PS p

rogr

am, f

or 3

8sp

ecie

s in

the

Nor

thea

st R

egio

n.Tw

o ef

fect

typ

es w

ere

cons

ider

ed: a

tw

o-po

pula

tion

com

pari

son

of s

urvi

val (

2P)

and

a po

pula

tion

with

line

arly

dec

linin

g su

rviv

al(L

D).

Effe

ct s

izes

are

giv

en fo

r tw

o al

pha-

leve

ls, 0

.10

and

0.20

. See

Tab

le 2

for

maj

or h

abita

ts, T

able

3 fo

r fie

ld d

efin

ition

s, a

nd T

able

s 3

and

5 fo

r sc

ient

ific

nam

es.

Cur

rent

Exp

and

ed__

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

____

____

____

____

____

____

α =

0.1

0α

= 0

.20

α =

0.1

0α

= 0

.20

____

____

____

____

____

____

____

____

____

____

____

____

____

Spec

ies

Maj

or h

abit

at(s

)St

aIn

dR

es�̂

p̂2P

LD

2PL

D2P

LD

2PL

D

Dow

ny W

ood

peck

erN

H,S

H49

414

280.

445

0.52

525

2025

2025

15“T

raill

's”

Flyc

atch

erSS

D,N

C,N

H17

636

300.

502

0.54

125

1525

1520

1515

15E

aste

rn P

hoeb

eSS

D,N

H,S

H24

238

110.

571

0.40

525

2025

15W

hite

-eye

d V

ireo

SSD

,SH

1329

537

0.47

90.

396

2025

1525

1520

15R

ed-e

yed

Vir

eoN

H,S

H55

1459

130

0.58

20.

295

1010

1010

1010

1010

Blu

e Ja

yN

H,S

N43

268

70.

869

0.16

425

2015

1015

Car

olin

a C

hick

adee

SH19

187

90.

484

0.44

925

2020

Bla

ck-c

appe

d C

hick

adee

NC

,NH

5010

9911

40.

514

0.29

420

1515

1515

1010

10Tu

fted

Tit

mou

seN

H,S

H36

411

510.

393

0.29

825

2025

2020

15V

eery

NH

4117

9224

40.

577

0.57

410

105

105

55

5Sw

ains

on's

Thr

ush

NC

691

130.

621

0.68

625

1525

1520

1515

10H

erm

it T

hrus

hN

C,N

H21

328

470.

457

0.63

120

1515

1515

1010

10W

ood

Thr

ush

NH

,SH

5319

8117

70.

424

0.40

320

1515

1515

1515

10A

mer

ican

Rob

inSS

D,N

C,N

H,S

H,S

C54

1479

115

0.42

90.

329

2520

2015

2015

1515

Gra

y C

atbi

rdSS

D,N

H,S

H50

4849

495

0.51

50.

460

1010

510

510

55

Blu

e-w

inge

d W

arbl

erSS

D,S

H18

352

260.

442

0.38

725

2520

Yello

w W

arbl

erSS

D28

1268

108

0.50

90.

475

1510

1510

1010

1010

Che

stnu

t-si

ded

War

bler

SSD

,NH

1845

947

0.47

50.

527

2015

2015

1515

1510

Mag

nolia

War

bler

NC

1140

131

0.34

60.

738

2520

2520

2020

Yello

w-r

umpe

d W

arbl

erN

C11

292

310.

460

0.50

125

1520

1520

1515

15B

lack

-thr

. Gre

en W

arbl

erN

C,S

C17

337

390.

402

0.58

925

2025

1520

1515

15B

lack

-and

-whi

te W

arbl

erN

C,N

H,S

H40

667

720.

502

0.30

125

1520

1515

1515

10A

mer

ican

Red

star

tN

H,S

H36

1943

171

0.52

50.

331

1510

1510

1010

1010

Wor

m-e

atin

g W

arbl

erSH

1240

532

0.52

00.

395

2020

2515

2015

Ove

nbir

dN

C,N

H,S

H51

1752

166

0.55

70.

425

1010

1010

510

55

Lou

isia

na W

ater

thru

shSH

1118

513

0.47

70.

745

2025

1520

1520

15C

omm

on Y

ello

wth

roat

SSD

4522

6320

90.

502

0.49

310

1010

1010

1010

10H

ood

ed W

arbl

erSH

1549

945

0.46

00.

610

2015

1515

1510

1010

DAVID F. DESANTE AND JAMES F. SARACCO

[73]

APP

EN

DIX

5. C

onti

nued

.

Cur

rent

Exp

and

ed__

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

____

____

____

____

____

____

α =

0.1

0α

= 0

.20

α =

0.1

0α

= 0

.20

____

____

____

____

____

____

____

____

____

____

____

____

____

Spec

ies

Maj

or h

abit

at(s

)St

aIn

dR

es�̂

p̂2P

LD

2PL

D2P

LD

2PL

D

Yello

w-b

reas

ted

Cha

tSS

D7

214

220.

501

0.35

625

2020

2515

Eas

tern

Tow

hee

SSD

,SC

3554

361

0.48

90.

331

2515

2015

2015

1510

Song

Spa

rrow

SSD

3711

9211

70.

336

0.51

125

2020

2020

1515

15Sw

amp

Spar

row

SSD

813

115

0.42

20.

721

2525

2025

20W

hite

-thr

oate

d S

parr

owN

C14

537

580.

289

0.58

320

2520

2020

2015

Nor

ther

n C

ard

inal

SSD

,NH

,SH

,SC

4182

676

0.61

80.

365

1515

1510

1010

1010

Ind

igo

Bun

ting

SSD

2650

441

0.46

50.

543

2515

2015

1515

1510

Red

-win

ged

Bla

ckbi

rdSS

D20

471

240.

587

0.32

020

2515

2015

1515

Bal

tim

ore

Ori

ole

NH

,SH

2230

518

0.41

80.

476

2520

20A

mer

ican

Gol

dfi

nch

SSD

3917

8010

20.

442

0.20

925

2520

2520

MAPS POWER ANALYSIS AND PROGRAM EXPANSION

[74]

APP

END

IX 6

. Eff

ect

size

s (%

) cu

rren

tly

det

ecta

ble

by M

APS

wit

h 80

% p

ower

and

20

yrs

of d

ata,

and

like

ly d

etec

tabl

e un

der

an

expa

nded

MA

PS p

rogr

am, f

or 2

1sp

ecie

s in

the

Sou

thea

st R

egio

n. T

wo

effe

ct t

ypes

wer

e co

nsid

ered

: a t

wo-

popu

latio

n co

mpa

riso

n of

sur

viva

l (2P

) an

d a

popu

latio

n w

ith li

near

ly d

eclin

ing

surv

ival

(LD

). Ef

fect

siz

es a

re g

iven

for

two

alph

a-le

vels

, 0.1

0 an

d 0.

20. S

ee T

able

2 fo

r m

ajor

hab

itats

, Tab

le 3

for

field

def

initi

ons,

and

Tab

les

3 an

d 5

for

scie

ntifi

c na

mes

.

Cur

rent

Exp

and

ed__

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

____

____

____

____

____

____

α =

0.1

0α

= 0

.20

α =

0.1

0α

= 0

.20

____

____

____

____

____

____

____

____

____

____

____

____

____

Spec

ies

Maj

or h

abit

at(s

)St

aIn

dR

es�̂

p̂2P

LD

2PL

D2P

LD

2PL

D

Dow

ny W

ood

peck

erN

H,S

H61

381

160.

620

0.34

425

2025

1520

15A

cad

ian

Flyc

atch

erSH

4717

8615

00.

483

0.55

615

1010

1010

1010

10W

hite

-eye

d V

ireo

SSD

,SH

4310

1210

30.

465

0.56

115

1015

1010

1010

10R

ed-e

yed

Vir

eoN

H,S

H56

2230

224

0.62

00.

212

1510

1010

1010

1010

Car

olin

a C

hick

adee

SH65

569

440.

493

0.28

220

2515

2015

2015

Tuft

ed T

itm

ouse

NH

,SH

6794

893

0.51

10.

434

1515

1510

1010

1010

Car

olin

a W

ren

SH,S

C65

1367

149

0.36

90.

582

1510

1510

1010

1010

Woo

d T

hrus

hN

H,S

H56

2635

258

0.45

50.

552

1010

1010

1515

1515

Gra

y C

atbi

rdSS

D,N

H,S

H25

1054

860.

428

0.47

720

1520

1510

105

10B

lue-

win

ged

War

bler

SSD

,SH

928

631

0.55

80.

279

2515

2015

2015

1510

Wor

m-e

atin

g W

arbl

erSH

1630

928

0.52

90.

404

2020

2515

2015

Ove

nbir

dN

H,S

H46

1588

136

0.52

80.

480

1510

1010

2015

1515

Lou

isia

na W

ater

thru

shSH

1929

531

0.52

00.

516

2515

2515

1010

1010

Ken

tuck

y W

arbl

erSH

3512

3213

70.

507

0.60

010

1010

1010

1010

10C

omm

on Y

ello

wth

roat

SSD

4114

2211

60.

439

0.50

320

1515

1515

1015

10H

ood

ed W

arbl

erSH

2961

959

0.51

40.

525

2015

1515

1515

1510

Yello

w-b

reas

ted

Cha

tSS

D24

524

650.

292

0.32

325

2520

Eas

tern

Tow

hee

SSD

,SC

4527

931

0.47

20.

344

2020

2515

2015

Song

Spa

rrow

SSD

216

715

0.41

00.

520

2525

20N

orth

ern

Car

din

alSS

D,N

H,S

H,S

C69

2105

237

0.54

30.

392

1510

1010

1010

1010

Ind

igo

Bun

ting

SSD

4712

2012

00.

521

0.30

620

1515

1515

1010

10

DAVID F. DESANTE AND JAMES F. SARACCO

[75]

APP

END

IX 7

. Eff

ect

size

s (%

) cu

rren

tly

det

ecta

ble

by M

APS

wit

h 80

% p

ower

and

20

yrs

of d

ata,

and

like

ly d

etec

tabl

e un

der

an

expa

nded

MA

PS p

rogr

am, f

or 1

9sp

ecie

s in

the

Ala

ska/

Bor

eal C

anad

a R

egio

n.Tw

o ef

fect

type

s w

ere

cons

ider

ed: a

two-

popu

latio

n co

mpa

riso

n of

sur

viva

l (2P

) and

a p

opul

atio

n w

ith li

near

ly d

eclin

ing

surv

ival

(LD

). Ef

fect

siz

es a

re g

iven

for

two

alph

a-le

vels

, 0.1

0 an

d 0.

20. S

ee T

able

2 fo

r m

ajor

hab

itats

, Tab

le 3

for

field

def

initi

ons,

and

Tab

les

3 an

d 5

for

scie

ntifi

c na

mes

.

Cur

rent

Exp

and

ed__

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

____

___

____

____

____

____

____

____

α =

0.1

0α

= 0

.20

α =

0.1

0α

= 0

.20

____

____

____

____

____

____

____

____

____

____

____

____

____

Spec

ies

Maj

or h

abit

at(s

)St

aIn

dR

es�̂

p̂2P

LD

2PL

D2P

LD

2PL

D

“Tra

ill's

” Fl

ycat

cher

SSD

1358

428

0.38

30.

506

2520

2015

2015

Bla

ck-c

appe

d C

hick

adee

BH

1024

428

0.42

60.

401

2520

2520

Bor

eal C

hick

adee

BC

1013

216

0.49

20.

365

2525

1525

15A

rcti

c W

arbl

erSS

D2

249

250.

339

0.60

525

2525

2025

20G

ray-

chee

ked

Thr

ush

SSD

,BC

625

327

0.45

90.

683

2515

2015

1510

1510

Swai

nson

's T

hrus

hB

C16

837

760.

456

0.59

015

1015

1010

1010

10H

erm

it T

hrus

hB

C,B

H9

611

570.

499

0.76

615

1015

1010

1010

10O

rang

e-cr

owne

d W

arbl

erSS

D16

1324

960.

413

0.50

420

1520

1515

1010

10Ye

llow

War

bler

SSD

911

1480

0.44

50.

456

2515

2015

1510

1010

Yello

w-r

umpe

d W

arbl

erB

C,B

H16

680

480.

367

0.43

220

2520

1515

1515

Am

eric

an R

edst

art

BH

225

621

0.57

10.

308

2025

1515

1515

10N

orth

ern

Wat

erth

rush

SSD

,BC

921

315

0.52

80.

725

1525

1515

1515

10W

ilson

's W

arbl

erSS

D,B

C15

2861

182

0.36

60.

565

1515

1510

1010

1010

Can

ada

War

bler

BH

214

914

0.46

20.

475

2025

2515

2015

Am

eric

an T

ree

Spar

row

SSD

719

915

0.48

30.

500

2025

2015

2015

Fox

Spar

row

SSD

1339

131

0.51

80.

556

2515

2015

1510

1510

Whi

te-c

row

ned

Spa

rrow

SSD

1362

665

0.43

50.

412

2020

2015

1515

Gol

den

-cro

wne

d S

parr

owSS

D,B

C5

279

300.

533

0.49

025

1525

1515

1515

10D

ark-

eyed

Junc

oB

C,B

H15

630

670.

309

0.63

620

2520

1515

1515