This is a summary of the three-week international survey of the vaquita refuge in heavily fished waters of the northern Gulf of California of the coast of Mexico's Baja California state. It shows what can be accomplished with a fresh effort in the fall of 2021. The expedition included scientists and conservationists from Mexico, the United States and Canada.

1. 1
SurveyreportforVaquitaPhotographicIdentification Research2019
LorenzoRojas-Bracho1
,BarbaraL. Taylor2
,ArmandoJaramillo-Legorreta3
,PaulaOlson2
,DiegoRuiz4
,Eva
Hidalgo5
, TimGerrodette2
,Annette Henry2
1 PNUD/CONANP, Dirección Regional Península de Baja California y Pacífico Norte,
C/o CICESE, Camper 10, Carretera Ensenada-Tijuana 3918, Zona Playitas, Ensenada,
BC 22860
2 Southwest Fisheries Science Center, 8901 La Jolla Shores Drive, La Jolla, CA 92037
3 CONANP, Dirección Regional Península de Baja California y Pacífico Norte,
C/o CICESE, Camper 10, Carretera Ensenada-Tijuana 3918, Zona Playitas, Ensenada, BC 22860
4 Museo de Ballena y Ciencias del Mar, Navarro s/n entre Altamirano y Gómez Farías Unidad
Cultural Prof. Jesús Castro Agúndez, Zona Central, 23000 La Paz, B.C.S.
5 Sea Shepherd Science Department, Sea Shepherd Conservation Society
Executive Summary
Three weeksof efforttoestimate the number of vaquitasinthe ZeroTolerance Areatookplace
betweenSeptember2and October27, 2019. Twoships(the Narvalandeitherthe Farley Mowator the
Sharpie) usedexperiencedobservers tofindandtrack vaquitaswith methodsdevelopedforthe earlier
efforttocapture vaquitasin2017. There were only 4days whenwindswere low enoughtosightand
track vaquitas,andonly2 photographswere matchedtopreviousefforts. Of the 7 sightings,there were
no photographicmatcheswithinthe 3weeks. All but one sightingcontainedasmallerindividualthat
couldhave beena calf. All individualsappearedinrobusthealth.
Because there were insufficientphotographstomake amark-recapture estimate,the methodof expert
elicitationwasused. Observerswere independentlygivenanumberof questionswiththe goal of
estimatingboththe range andmostlikelyvaluesforthe numberof calvesandthe numberof total
individuals. The meanestimate fornumberof calvesseenwas3witha 63% belief thatthere were at
least3 calves. Clearly,the remainingvaquitasare continuingtoproduce youngatveryhealthylevels,
whichmakesconcernsaboutinbreedingdepressionfarlessthanconcernsaboutmortalityingillnets.
The mean estimate forthe numberof unique vaquitasseeninall 7sightingswas9.7 with63% belief that
there were atleast9. The meannumberof vaquitasestimatedin2018 was 9 individuals(rangingfrom6
to 19). Because there ismuch uncertainty, itisunclearif the populationhascontinued todeclineat
50%/year,but the expectednumberwouldbe 4-5individualsand anabsolute minimumof 6 have
survived. The 2018-2019 totoaba seasonwaspotentiallymore dangerousthaninpreviousyears
because the ships removingtotoabanets hadtoleave early due toviolentattacks byfishermen, and
consequently the netremoval effortswere greatlyreduced. Therefore, havingexpertsbelieve thata
meanof 9.7 vaquitaswere seen in2019, and a 25% chance that there were at least12, is surprising. Itis
possible thatthese remainingvaquitas,several of whomhave netscars,are not a randomselection
followingthe deathsof over98% of the species,butratherare selectedindividualsthatare especially
wary of nets. Guardingthese net-wisevaquitasthatare havingrobustcalvescouldstill save the species.

2. 2
However,manygillnetswere
observedbeingdeployedwithinthe
ZeroTolerance Area(ZTA)
(documentedinAppendix 3andalso
inthe data records). The
photographshowsa vaquitapair
neara panga withgillnetdeployedin
the background. Thisoccurred
withinthe ZTA. Noenforcement
actionswere observed. These
gillnetsare notthose usedfor
totoabaand thus representan
increase inthe amountof illegal
fishing. There is alack of
enforcementevenwithinthe small
area of the ZTA (a rectangle onabout 288 km2
or roughly 12 by24 km,whichisa bit smallerthanthe
urban areaof Tijuana).

3. 3
Main Report
The decline invaquitanumbershasbeenwell documented. The firsteffortto coverthe full vaquita
distributionusedvisual line-transectmethods(Jaramillo-Legorretaetal.1999). Thiseffortnotedthe
difficultyinsightingthisspeciesbecause of small groupsize,inconspicuoussurfacingsandavoidance of
the surveyvessels. Impreciseabundance estimatesraisedconcernsabouttimelydetectionof potential
declinesinabundance (TaylorandGerrodette,1997). Acousticmonitoringmethodswere developedto
increase precisionof estimatingbothabundance andtrendsinabundance (Jaramillo-Legorretaetal.
2017), and a combinationof visual andacousticmethodswere usedtoestimate vaquitaabundance in
2008 (Gerrodette etal.,2011) and2015 (Tayloret al.,2016). Acousticmonitoringindicatedthatthe
vaquitapopulationcontinuedtodecline rapidly,about50%/year,through2018 (Thomasetal.2019).
Recentdevelopments,however,have made bothacousticmonitoringandvisual line-transectmethods
difficult. Fishermenhave begunremovingthe acousticdevices(CPODs) usedtorecordvaquitaclicks.
The data recordedon eachdevice islost,andit isexpensive toreplace the stolenCPODs. Unless
enforcementof the fishingbaniseffective andthe theftof equipment isstopped,acousticmonitoring
cannot collectdataas it has inthe past. Visual line-transectmethodsface adifferentproblem. The
numberof vaquitasisnowso lowthat the numberof sightingswouldnotbe sufficienttoestimatethe
necessaryparameters. If aline-transectsurveywere carriedoututilizing the same shipas inpast
surveys(the David StarrJordan/OceanStarr),anestimate of abundance wouldbe possiblewith
relatively few sightings,because the probabilityof detectionisknownforthisship. However,chartering
thisvessel andhiringexperiencedobserversforthe necessarytime would be expensive,atleast
US$3,000,000 for a survey. Unlesssuchfundsare available,the size of the 2020 vaquitapopulation
cannot be estimatedusing line-transectmethods.
Facedwiththese difficulties,vaquitaresearchershave turnedtophotographicidentification,which
requireshighqualityphotographstoidentifyindividual vaquitas. Photographicidentificationof vaquitas
beganin2008 (Jeffersonetal.2009). Opportunisticeffortsresumed in2017 duringthe VaquitaCPR
effort. InSeptember2018 a dedicatedeffortproducedthe firstevidence thatvaquitascouldcalve
annually(Tayloretal.2019) and showedthat a minimumof 6 healthyanimalsremainedinasmall area
nearSan Felipe,Mexico. Thisminimumabundance estimate was the numberof animalsseen
simultaneouslyandwas influentialinthe abundance estimate forthatyear(Jaramillo-Legorretaetal.
2019).
Two shortefforts focusedonphotographicidentification wereconductedin2019. The firsteffort
(Period1) had sightingsduringthe setupperiodinlate Augustandduringthe surveyperiodfrom
September2to 6. The survey period wastriggeredbyawindow of weatherpredictedtobe goodfor
vaquitasightingswithwindslessthan7knots. The second,andlarger,effortfromOctober15-27 (Period
2) was chosenbecause thattime periodhadcalmwindsinthe past andcalm weatherinlatermonths
was lesscommon. Each periodisdescribedseparately because the effortlevelsdiffered.
Period1 (September2– 6, 2019)
Thiseffortusedexperiencedobservers whocouldgettoSanFelipe quicklytotake advantage of a short
periodof good weather. Duringthisperiod,twovesselswere involved: the Museode Ballena’svessel
Narvaland the SeaShepherdConservationSociety’s (SSCS) vesselSharpie. The Narval housedandfed
mostof the visual observationteamsforbothships. Onthe Sharpie,twopairsof big eyes(Fujinon25x
binoculars) were mountedonthe flyingbridge. A trackingsystemwassetupto be able tomap the
ship’stracksand plotvaquitasightings. The trackingsystemcomprised of a computerwithmonitor,a

4. 4
GPS and digital magneticcompass. The computerran a program specificallycreatedtotrackvaquitas
calledVaquitaWinCruz (availableuponrequestfromBarbaraTaylor). Duringthe setupperiodpriorto
the arrival of the full team,one experiencedobserverworkedwiththe SSCSteamandhadsightingson
August19 and August20. Nophotographswere obtained. The sightingswere describedasone large
and one small vaquita.
On the Narval,the observation teamused handheldbinoculars. There wasalsoanacousticteam that
had alreadyanalyzedacousticdatafromearlieranddeterminedwhere vaquitaswere detected. The
acousticteam provideddatafroma limitednumberof acousticdetection locationsduringthe Period1
effort,aswell aslocationsfromdata collectedpriortothe start of the Period1 effort.
From Sep2-6 personnel includedthe following: LorenzoRojas-Bracho,ArmandoJaramillo-Legorreta,
EdwynaNieto,GustavoCardenas,BarbaraTaylor,Jay Barlow,RobertPitman,PaulaOlson, Sarah
Mesnick, TimGerrodette,EvaHidalgo. The focuswas entirelyonphotographicidentification. A
MexicanCITES permitallowingexportof abiopsytothe San DiegoZoo’sfrozenzoocollectionwasnot
obtained,sonobiopsyeffortwasmade.
Searcheffortwasconcentratedinthe part of the ZeroTolerance Area(ZTA) where vaquitashadbeen
detectedacousticallyinthe previousfew weeks. The acousticsampling gridwas usedasa reference to
navigationandcommunication(Appendices1and2). Effortwasplannedtorun generallyfrom
southeasttonorthweststartingatdawn to keepthe sunto observers’backs. Surveyspeedwas
between5-7knots. Search effortwasalsocarriedout westof the ZTA, inan area closerto landwhere
detectionswere made lastyear, butwhere no acousticeffortwaspossible thisyearbecause of theftof
acousticdevices.
Only2 full daysof effortwithconditionswithlessthan7knotsof windwere possible inPeriod1. On 3
September,avaquitapairwassighted (sighting001) at 9:22am at N31:04.32 W114:38.71, and both
vesselswere broughtwithinamile of the sighting. Two pangasand 1 RHIB were deployed. Vaquitas
were trackedfor64 minutesand goodphotographswere obtained. (Fig1a). Bothanimalswere adult
size.All of the photosfromthissightingare somewhatdistantbutone animal canbe identifiedasan
individualbasedonthe unique shape of itsdorsal fin(individual"A",alsoseenin1band 1d).
Leftand right side photosof "A" were comparedto the photosfromJefferson etal. (2009) and to the
photos from 2017 (Wells, unpub.) and 2018 (Taylor and Rojas-Bracho, unpub). Individual "A" was
matched to individual "D" (Figure 1c and 1e) from September 2018. In 2018 "D" was part of a loose
aggregation of 4-6 vaquitas.

5. 5
Figure 1. a. Pairof vaquitaphotographed03September2019. Individual "A"isonthe right. (Martinez_3994)
b. Individual"A"in2019. (Rosales_3563) c. Individual "D"in2018. (Ortiz_8774)

6. 6
1d Individual "A"in2019. (Taylor_1710) 1e Individual "D"in2018. (AdamU_3754)
The minimumnumberof vaquitaspresentbetweenAug 19and Sept6 was 4 vaquitas:the adultpair
that wasphotographed (03 September) andalikelymotherandcalf seen priortothe Period1 effort
(duringset-uponthe Sharpie) onAugust19 that wasnot photographed.
Period2 (October15-27)
Thiseffortinvolved2full teams (eachwith2 pairsof bigeyes anda trackingsystem) onthe Narvaland
the SSCS Farley Mowat. Because of accumulated lossesof CPODs(52prior to thissurveyperiod) and
ongoingillegalgillnetting,noacousticeffortwaspossible. The same systemof navigational pointsused
inPeriod 1 was usedinPeriod2 (Appendices1and2). The dailysearchingpatternwasthatsame as
describedforPeriod1.
Personnel wereasfollows:
Chief Scientists: LorenzoRojas-Bracho,BarbaraTaylor (Narval),TimGerrodette (Farley Mowat)
Recorders:Sarah Mesnick(Narval),EvaHidalgo(Farley Mowat)
Observers (changingvesselsondifferentdays): JuanCarlosSalinas,ErnestoVazquez,Sergio Martinez,
Hiram Rosales,JayBarlow,PaulaOlson,RobertPitman,ToddPusser,Suzanne Yin,Adam Ü,ChrisHoefer,
NickKellar,TomJefferson
Biopsy specialists: GustavoCardenas,Zach Swain
Windconditionswere worse thananticipated. Duringthe firstweektherewere only2daysof marginal
weather(noBeaufort0,virtuallynoBeaufort1). Despite the poorweather,vaquitawere regularly
sightedbutcouldnotbe successfullytrackedandphotographed. On 17 October,a mother/juvenile-calf
pair wasobservedfor16 minutes(sighting003) fromboth ships. There are distantphotographsand
videoof the pairwitha pangapulling agillnetinthe background. The photographsare notsufficient
qualityforphotographicidentification. Fivehourslateralikelymother/calf pairwassighted(sighting
004) about10 milesfromsighting 003.The small boats were deployed forphoto-IDbutthe animals
couldnot be approachedsuccessfullyforphoto-ID. Observerswhosaw bothsightingsdescribedthe
dorsal finsof the adultsto be different,makingitlikelythatalthoughthe paircouldhave swumthe
distance inthe elapsedtime,itismore likelythannotthatthese were twoseparate mother/calfpairs..
Withthese twosightings(003 and 004), and the two adultsphotographedon 3 September, the absolute
minimumnumberof vaquitaisbetween4and6 and more likelytobe 6.

7. 7
Sighting004 wasof 4 animalson19 October:a likelymother/calf pairandanearbyadultpair. Sighting
005 wasa brief sightingof asingle animal seenabout30 minutes afterleavingsighting004. On 27
October,a group of 4 vaquitas(sighting006),probablyall adults,wasfollowedfor44 minutes,andgood
photographswere obtained of 2animals. A little lessthananhourlaterand 3.2 milesaway,amother-
calf pair was seen(sighting007) but no photographswere obtained. Figure 2showstracklinesand
sightinglocations. Detailsof all 7 sightingsare giveninAppendix4.
Figure 2. Track linesfollowedduringthe surveyonboardFarleyMowat(blue lines) andNarval (redlines).
Sightingsare presentedwithyellowcircles,indicatingnumberinginmaintext.VaquitaRefuge presented
with small broken line. Zero Tolerance Area presented with large broken line.
Pangaswithgillnetsdeployedwereseenwithinthe Zero ToleranceArea(ZTA) oneverydaythatwe
searchedforvaquitas. Numerousphotographs of fishingpangaswere taken, withmetadatathatinclude
date,time, latitude andlongitude. Because ourefforthadtofocuson vaquita,onlysporadiclocations
were givenforpangaswithgillnetsinthe waterinthe regulardatarecordingsystem1
. Inaddition,a
dedicatedeffortinaseparate pangawasmade to documentthe amountof illegal gillnettingwithinthe
ZTA (see Appendix 3). Note thatthe ZTA coordinatesgivenonthe mapinAppendix 1are the official
coodinatesbutalsofall onlocationsof acousticdetectors. DuringPeriod2the Museo de Ballenaplaced
delimitingbuoysthataddeda500m buffer,whichwasagreedwiththe Profeparepresentativeaboard
the Narval,to provide safetyforfuture acousticdevicesdeployedinthe ZTA. These buoyswere placed
so that fishermencouldsee the locationof the ZTA. The area demarkedbythe buoysis239.5 km2
with
a perimeterof 68-9 km and approximate sidesof 23.5x10.9 km.
All 7 sightingscoveringboth Periods1and 2 are giveninAppendix 4. Raw data files(calledDASfiles)are
available fromLorenzoRojas-Brachouponrequestandare alsomaintainedatthe SouthwestFisheries
1 These locations are inthe data filesrecordedbythe trackingsystem andare recordedas‘objects’or
‘Navydolphins’because thesefunctionkeysallowedarapidpositiontobe added. Commentsare
typicallyrecordedafterwardstoclarifythatthese are actuallypangas.

8. 8
Science Center. Giventhatthe acousticdata were heavilycompromisedin2019 due to theftof acoustic
detectors,thisfall effortwaslikelythe bestestimateof the minimumnumberof remainingvaquitas
alive in2019. Photographicidentificationwasinsufficienttouse mark-recapture methodstoestimate
the numberswithinthe small areasurveyed. Therefore,anexpertelicitationwasconductedusing
observersasthe experts.
Expertswere providedwiththe exactinformationinAppendix4and asked to provide theirexpert
opinionona numberof questions(Appendix5). Some logical inconsistencieswere seenafterreviewing
the responses. A secondelicitationiscommonpractice andwasdeemedimportantinthiscase both
because the probabilityproblempresentedtoexpertswascomplex andthe initial elicitation wasgiven
hurriedly toexpertobservers atthe endof the survey. A secondelicitationwaswarrantedtoprovide
the bestestimate of vaquitasremaininginthe ZTA. Thiselicitationusedresponses fromthe first
elicitationtogive the expertssome feedbackthat wouldaide intheirestimationof the total numberof
calvesseenandthe total numberof vaquitasseen. Forexample,sightings 002 and 003 bothhad
animalsthatcouldhave beencalvesandcouldhave beenresightingsof the same pairat differenttimes
duringthe same day. Therefore,dependingonthe probabilitiesthateachsmall individual wasacalf and
that the sightingswere the same ordifferent,there couldbe zero,one ortwocalves. The expertswere
giventhese probabilitiesfromtheiroriginal responses(Appendix 5showsthe probabilitiesforan
anonymous expert,buteachexpertwasonlyprovided theirownprobabilities).
Detailedelicitationresultsare giveninAppendix6. Figure 3 (below) showsthe percentexpertsbelieved
giventheirobservationsforthe numberof calvespresentforall 7 sightings. The meanestimate for
numberof calvesseenwas 3 witha 63% expertconfidence thatthere were atleast3 calves.
Figure 3. Percentbelief forthe numberof calvesin7sightingswhere all efforttookplace withinthe
zerotolerance areafromthe secondelicitation(see Appendices5and6).
The mean estimate forthe numberof vaquitasseeninall 7sightingswas9.7 with63% expert
confidence thatthere were atleast9. Figure 4 showsthat the plausible range isfairlylarge suchthat
there isa 25% belief thatthere were atleast12.
0
5
10
15
20
25
30
35
0 1 2 3 4 5 6
Percent
Number of Vaquita Calves

9. 9
Figure 4. Percentbelief fromexpertelicitation2(Appendices5and6) fornumbersof vaquitasin7
sightingswhere all efforttookplace withinthe zerotolerance area.
The final questionwasoptional asexpertshadtodecide whethertheycouldremembereffortsinboth
2018 and 2019. Five expertsresponded tothe followingquestion:
Comparing2019 to2018 withinthe ZeroTolerance Area,please distribute 10likelihoodpointsto:
a. There were lessthanhalf the numberof vaquitasin2019 comparedwith2018
b. There wasbetweenhalf andthe same numberof vaquitasin2019 comparedwith2018
c. There were aboutthe same numberof vaquitasin2019 comparedwith2018
d. There were more vaquitasin2019 thanin 2018
Roughly80% of the expertsbeliefwenttoband c withabout 10% each to a and d.
Discussion
Overall,the abilitytofindandtrack vaquitasremainedgooddespite the lackof consistent acoustic
effort. The data and anyconclusionsare limited,however, inestimatingnumbersoverashort time
frame onlywithinthe ZTA. Concentratingeffortinthe ZTA is logical givenacousticdataindicatingthat
the last vaquitasare mostreliablyfoundinthissmall area. However,the numbersinthisreportcan
onlyreflectthe numbersinthatareaand therefore maynotaccount forthe full numberof vaquitas
remaining. Thatsaid,recentacousticevidencesuggeststhatthe numbersof vaquitasoutside the ZTA
are likelyverylow.
The poor successat obtaining photographswasdue primarilytomarginal windconditions. Animals
couldbe sightedbutnot trackedfor the prolongedperiodsneededtoobtainphotographsof sufficient
qualitytoidentifyindividuals. Muchmore time wouldbe neededto ensure obtainingasufficient
numberof photographs fora mark-recapture estimateof abundance forthe ZTA. In 2019, 3 weeksof
effortresultedinonly4daysof weatherconducive tosightingvaquitas. We shouldalsostresshere that
the impossibilityof monitoringthe vaquitaswiththe acousticdevicesreducesthe chancesof finding
vaquitassince we have nonear-real-time informationtoguide the vesselstowhere vaquitaswere
acousticallyencounteredandmustrelyondatafrom monthsor evenyearsinthe past.
0
5
10
15
20
25
4 5 6 7 8 9 10 11 12 13 14 15 16
Percent
Number of Vaquita

10. 1
0
The use of two shipswithfull observerteamsworkedwell,butwastwice the costof a single ship. Two
strategiesremainviable thatwere exemplifiedby Period1and 2 of thisstudy:1) use locallyskilled
observers and respondquicklyto apredictedgoodweatherwindow,and2) choose dates inadvance to
bringin skilledobservers butincrease the suvey time toaccountforunknownweather. Eitherstrategy
couldworkdependingonthe amountof fundingavailable andhow manyskilledobserverscanbe found
on shortnotice.
The resultof findingabout3 calvesismost encouraging. Clearly,the remainingvaquitasare continuing
to produce youngat veryhealthylevels,whichmakesconcernsaboutinbreedingdepressionfarless
than concernsaboutmortalityingillnets. The estimatednumbersare alsoencouraging. The mean
numberof vaquitasestimatedin2018 was 9 individuals(range:6-19). The estimatesayearlaterif the
roughly50%/yeardecline persistedwouldbe 4-5. The 2018-2019 totoabaseasonwaspotentiallymore
dangerous tovaquitasurvival thaninpreviousyearsasthe net removal effortswere greatlyreduced.
The shipshad to leave early due toviolentattacks byfishermen. Sohavingexpertsbelieve thatamean
of 9.7 vaquitas (range:4-17) were seen in2019, and a 25% chance that there were at least12, is
surprising. Itispossible thatthese remainingvaquitas,several of whomhave netscars,are not a
randomselectionfollowingthe deathsof over98% of the population,butratherare selectedindividuals
that are especiallywaryof nets. Guardingthese net-wisevaquitasthatare havingrobustcalvescould
still save the species.
Literature Cited
Jaramillo-Legorreta,A.,G.Cardenas-Hinojosa,E.Nieto-Garcia,L.Rojas-Bracho,J.VerHoef,J.Moore,N.
Tregenza,J.Barlow,T. Gerrodette,L.Thomas,and B. Taylor. 2016. Passive acousticmonitoringof the
decline of Mexico’scriticallyendangeredvaquita. ConservationBiology31:183-191. Doi:
10.1111/cobi.12789
Jaramillo-Legorreta,A.M.,G.Cardenas-Hinojosa,E.Nieto-Garcia,L.Rojas-Bracho,L.Thomas,J.M.VerHoef,J.
Moore,B. Taylor,J. Barlow,N.Tregenza. 2019. Decline towardsextinctionof Mexico’svaquitaporpoise
(Phocoenasinus).R.Soc.Opensci.6: 190598.
Jefferson,T.A.,Olson,P.A.,Kieckhefer,T.R.,andL. Rojas-Bracho.2009. Photo-identificationof the
vaquita(Phocoena sinus):the world’smostendangeredcetacean. Lat.Am.J.Aquat.Mamm. 7(1-2):53-
56.
L. Rojas-Bracho,F.M.D.Gulland,C.Smith,B. Taylor,R.S.Wells,P.O.Thomas,B.Bauer,M.P. Heide-
Jørgensen,J.Teilmann,A.Jaramillo-Legorreta,G.Abel,A.J.Read,A.Westgate,K.Colegrove,F.Gomez,
K. Martz, R. Rebolledo,S.Ridgway,T.Rowles,C.E.vanElk,J.Boehm, G. Cardenas-Hinojosa,R.
Constandse,E.Nieto-Garcia,W.Phillips,D.Sabio,R. Sanchez,J.Sweeney,F.Townsend,S.Walker,J.C.
Vivanco. 2019. A fieldefforttocapture criticallyendangeredvaquitas(Phocoena sinus)forprotection
fromentanglementinillegal gillnets.EndangeredSpeciesResearch38:11-27.
Taylor,B.L., L. Rojas-Bracho,J.Moore,A. Jaramillo-Legorreta,J.VerHoef,G.Cardenas-Hinojosa,E.
Nieto-Garcia,J.Barlow,T.Gerrodette,N.Tregenza,L.Thomas,andP.S.Hammond. 2016. Extinctionis
imminentforMexico’sendemicporpoiseunlessfisherybycatchiseliminated.ConservationLetters.doi:
10.111/conl.12331

11. 1
1
Taylor,B.L., Wells,R.S.,Olson,P.A.,Brownell,R.L.Jr.,Gulland,F.M.D.,Read,A.J.,Valverde-Esparza,F.J.,
Ortiz-Garcia,O.H.,Ruiz-Sabio,D.,Jaramillo-Legorreta,A.M.,Nieto-Garcia,E.,Cardenas-Hinojosa,G.,and
Rojas-Bracho,L.2019. Likelyannual calvinginthe vaquita,Phocoenasinus:A new hope? Marine
Mammal Science DOI:10.111/mms.12595
Thomas,L., Jaramillo-Legorreta,A.G.Cardenas-Hinojosa,E.Nieto-Garcia,L.Rojas-Bracho,J.M.Ver
Hoef, J.Moore, B. Taylor,J.Barlow,N. Tregenza. (2017). Last call:Passive acousticmonitoringshows
continuedrapiddeclineof criticallyendangeredvaquita.J.Acoust.Soc.Am.142 (5), November2017
Acknowledgements
We thankwholeheartedlyDiegoRuizforfullyfundingobserversandthe costsof the Narval and to Sea
ShepherdConservationSocietyforprovidingthe secondship,keypersonnel,fundingoperation costsof
supportingsmall vessels andhelpingwithmanyof the logistics. We thank the SouthwestFisheries
Science Centerforloaningthe bigeye binocularsandmuchof the otherequipment. Ourappreciationto
Contralmirante RaymundoSánchez,Comandantedel SectorNaval de SanFelipeforhissupportandTte.
Frag. S.M.A.M.L. Biol. VerónicaAcostaChamorro, Tte. Corb. S.M.A.M.L. Biol CynthiaMargarita
Martínez Romeroand Tte. Corb. C.G.DavidSantanderLópezfor theirparticipationonboardthe Narval.
To A Martín Sau y FranciscoValverde de laReservade laBiósferadel Deltadel Río ColoradoyAltoGolfo
de Californiaformakingall permitsandprocesseseasy,toCapitánFranciscoMelchorLópezandthe
crewof the Narval: EstebanRomero,UlisesOmarHiguera,AurelioSaldaña,VenturaMesa,Salvador
Armenta,SergioLuísMuñoz and EnochEnrique Rizo;to our fisherssupportingwiththierpangas
Chalunga,Gata and Cesar. Our gratitude tothe Sea Shepherdcrew of the Sharpie:CaptainOctavio
Carranza, FrançoisVonSull,MelissaRomao,RicardoPonce,CarlosOlivares,JohanSantenn,Brianna
Peteron,LarryDebeck,ElvenVillecourt,YrjaVan,ErikAlejandro,PaulaGreenandtothe crew of the
FarleyMowat:CaptainMark Gibbs,DavidStovall,CesarLeon,CarlosValencia,Andre Meresiev,De Wet
du Toit,Guy March, FedericoScandizzo,Jessica Banderas,RobertNewbyandGuillherme Pira. We
appreciate the skill andattentionof the observerteam: Jay Barlow,GustavoCárdenas,ChrisHoefer,
Tom Jefferson,NickKellar,SergioMartinez,SarahMesnick, EdwynaNieto,RobertPitman,ToddPusser,
Hiram Rosales,JuanCarlos Salinas,ZachSwain,AdamÜ,ErnestoVazquezandSuzanne Yin.

12. 1
2
Appendix1 Period1map and waypointlocations
Period1 usedthe standard acousticsamplingsiteslistedbelow aswaypointstodirectthe shipfor
photographicidentificationefforts.
ACOUSTIC SAMPLING SITES OF VAQUITA MONITORING PROGRAM
A -114.74090 31.22277
B -114.77486 31.13805
C -114.59526 31.00626
D -114.56131 31.09097

13. 1
3
Site Longitude Latitude
1 -114.42809 31.07543
2 -114.51494 30.97472
3 -114.49796 31.01708
4 -114.48098 31.05943
5 -114.46400 31.10179
6 -114.44703 31.14415
7 -114.43005 31.18651
8 -114.41307 31.22887
9 -114.43201 31.29759
10 -114.44899 31.25523
11 -114.46597 31.21287
12 -114.48294 31.17051
13 -114.49992 31.12815
14 -114.51690 31.08579
15 -114.53388 31.04344
16 -114.55086 31.00108
Site Longitude Latitude
18 -114.58678 31.02744
19 -114.56980 31.06979
20 -114.55282 31.11215
21 -114.53584 31.15451
22 -114.51886 31.19687
23 -114.50188 31.23923
24 -114.48490 31.28159
25 -114.46793 31.32395
26 -114.50384 31.35031
27 -114.53780 31.26559
28 -114.55478 31.22323
29 -114.57176 31.18087
30 -114.58874 31.13851
31 -114.60572 31.09615
32 -114.62269 31.05380
34 -114.65861 31.08015
Site Longitude Latitude
35 -114.64163 31.12251
36 -114.62465 31.16487
37 -114.60768 31.20723
38 -114.59070 31.24959
39 -114.57372 31.29195
40 -114.62662 31.27595
41 -114.64359 31.23359
42 -114.66057 31.19123
43 -114.67755 31.14887
44 -114.69453 31.10651
45 -114.69649 31.21759
46 -114.67951 31.25995
47 -114.66253 31.30231
48 -114.71543 31.28631

14. 1
4
Appendix2 Period2 waypointlocations
More waypointswere addedinOctobertoallow close operationof 2 shipswithfull observercapacity.
These waypointswere usedtosettrackline effortandcommunicate locationsbetweenthe ships.

15. 1
5
Oct 2019 effort
Way point Longitude Latitude Way point Longitude Latitude
1 -114.74090 31.22277 29 -114.67657 31.09333
2 -114.74939 31.20159 30 -114.68506 31.07215
3 -114.75788 31.18041 31 -114.63314 31.14369
4 -114.76637 31.15923 32 -114.64163 31.12251
5 -114.77486 31.13805 33 -114.65012 31.10133
6 -114.72294 31.20959 34 -114.65861 31.08015
7 -114.73143 31.18841 35 -114.66710 31.05898
8 -114.73992 31.16723 36 -114.61518 31.13051
9 -114.74841 31.14605 37 -114.62367 31.10933
10 -114.75690 31.12487 38 -114.63216 31.08815
11 -114.70498 31.19641 39 -114.64065 31.06697
12 -114.71347 31.17523 40 -114.64914 31.04580
13 -114.72196 31.15405 41 -114.59723 31.11733
14 -114.73045 31.13287 42 -114.60572 31.09615
15 -114.73894 31.11169 43 -114.61420 31.07497
16 -114.68702 31.18323 44 -114.62269 31.05380
17 -114.69551 31.16205 45 -114.63118 31.03262
18 -114.70400 31.14087 46 -114.57927 31.10415
19 -114.71249 31.11969 47 -114.58776 31.08297
20 -114.72098 31.09851 48 -114.59625 31.06180
21 -114.66906 31.17005 49 -114.60473 31.04062
22 -114.67755 31.14887 50 -114.61322 31.01944
23 -114.68604 31.12769 51 -114.56131 31.09097
24 -114.69453 31.10651 52 -114.56980 31.06979
25 -114.70302 31.08533 53 -114.57829 31.04862
26 -114.65110 31.15687 54 -114.58678 31.02744
27 -114.65959 31.13569 55 -114.59526 31.00626
28 -114.66808 31.11451

16. 1
6
Appendix3 Panga Survey
PANGAS AND VAQUITA IN THE ZERO TOLERANCE AREA (2019/10/17)
Pangaswere countedinthe southernandwesternpartof the ZTA.The countwas made froma research
panga andwiththe helpof 15x50 binoculars on17 October2019.
To count the numberof pangas we stoppedatthe sitesmarkedwithrednumbersonthe map below
(Figure1).The presence of pangaswassurveyed360degreesaround the researchpanga.The red
numbersindicate the numberof pangas counted. Clamfisher,tourist,militaryortrawler vesselswere
excluded.Itisinferredthatthe distance betweenthe positionswhere vesselswerecounted decreased
the chancesof countingthe same boat inmore than one location. Intotal,87 pangas were observed on
the waterin the ZTA. Inadditiontothe pangas,gillnetswere observed.The lengthof some netswas
estimatedtobe approximately1km.The total time tocomplete the countingfromthe 13 siteswas3
hoursand 42 minutes.
The black numbered circlesinFigure 1indicate the acousticsamplingsiteswe attemptedtomonitor
between SeptemberandDecember.Sadly,30mooringsandacoustic detectors were stolenin
November.The three blackcircleswithagreencircle aroundthemare the onlysiteswhere acoustic
activitywasdetectedbetweenJuneand September,fromthe standard 46 samplinggrid. The solidgreen
circle isthe locationof the confirmed mother/calf pairsightingon17 Oct 2019, duringour photo-id
survey.Fishingactivitywasappreciablymore intense inthe vaquitadetectionzone (darkblueellipse),
while towardsthe coastof Baja California(lightblue ellipse) fishingactivitywassignificantlylower.
IT IS CLEAR THAT THE INCREASED FISHING ACTIVITYIS OPERATING IN THE MOST SENSITIVEAREA FOR
VAQUITA.

17. 1
7
Figure 1.Map showingthe acousticsampling grid(blackcircles) insidethe ZTA (blue rectangular
polygon). Positionwhere fishingvesselswerecountedaroundare markedwithrednumbers,which
indicate the numberof vessels observed.The ellipses depicttwoareasof distributionof fishingvessels.
In blue the areawithhighernumberof vessels andwithcyan the areawithlowernumbers. Belowisa
photoof a typical fishingvessel,withacouple of vaquitasclose by.

18. 1
8
Appendix4 Sightingdetails (evidence giventoexpertspriortothe elicitations)
OBSERVATIONSON VAQUITA SIGHTINGS3 SEPTEMBER, 17, 19, & 27 OCTOBER2019
General informationonvaquitasandporpoise
The average group size of vaquitasinall surveys(1997, 2008, 2015) was 2. Althoughthe winterperiod
remainsmostlyunobserved,neonatevaquitas have been recoveredfromearlytotoabafishingin
February. Thus,calvesinOctoberwouldbe 6-8 monthsold,whichisthoughtto be roughlythe age of
weaning. The individual markedbelowasC18 wasconsideredverylikelytobe a calf.
The redsymbolsare that same pair from2018
recordedina WinCruzscreenshot. The pair
had a meanderingpattern. The circulargrid
(regularlyspaced,concentriccircles) has1nmi
spacing.(Ignore the brightcirclesof differing
sizes.) The total time of the sightingwas1 hour.
Withinthe hour,a straight-line distance
betweenfarthest pointstrackedmeasure 4nmi
apart. Conceivablyif vaquitastravelledina
straightline,theycouldcover8 nmi/hr,butitis
more likelyyouwouldn’texpectthemtomove
more than whatis seenhere (about4nmi).
Notesfor2019
Eventhoughthere were only7 sightings,itmay
be helpful towrite eachdownanddraw lines
connectingvarioussightingsthatcouldbe
duplicatestohelpwithyourfinal estimate of
the plausible numbersseenin2019.

19. 1
9
Summaryof Observationsof the 7sightingsin2019
3 September2019
Sightingnumber001
2 individualsweresightedbyRobertPitmanat09:22. 3 small boatswere launchedbutthe vaquitasdid
not surface as closelytothose vesselsastothe 2 ships(Narval andSharpie) where the bestphotographs
were taken. Trackingwas haltedat10:21 whenitwas determinedthatphotographsof IDqualityhad
beentakenandthe search couldresume fordifferentvaquitas. Thiswasthe onlysightingof the
September2-6period. Thispairwere adultsandone was matchedto a photographfrom2018. The pair
remainedcloselyassociatedthroughoutthe periodbutdidnotsurface withone routinelyinthe ‘calf’
position.
17 October2019
Two sightingsof vaquitawere detectedon17 October2019, occurringapproximately fiveand a half
hours and10 nautical milesapart(sightingsno.002 and no. 003). Both groupsconsistedof a pair of
animals,one largerandone smaller.
PaulaOlson(FarleyMowat):
In sightingno.002 the largeranimal wasadultsize and the smallerwasjuvenile size.Fivesurfacings
were observedovera16 minute period.The pairwere alwayswithinone bodylengthof one another
and the smalleranimal wasalwaysobservedtobe trailingthe larger.Swimdirectionsatthe surface
were South,East,South,South,andSouth.On the fourth surfacing,the pairemergedlessthan30
metersfroma panga. Distantphotoswere obtainedfromthissurfacing.
AdamÜ was onthe Narval andobservedbothpairs(below).Basedonhisdescriptionof the dorsal finof
the adultin the secondpair(sighting no.003), I thinkthe dorsal finsof the adultsmay have been
differentshapes.He describedthe secondadulthashavingaconical,erectdorsal fin.Ihad good,
multiple looksatthe dorsal finof the firstadult;thisadultappearedtohave a more backward-oriented
dorsal finand faintlyfalcate.
Giventhe differentshapesof the dorsal fins,andthe distance betweenthe sightings(althoughcertainly
vaquitacouldtravel tennmilesinfourhours) these mayhave beendifferentpairs.
AdamÜ (Narval):
Sighting002 wasinitiallyseenfromthe FarleyMowat.Once the Narval wasalertedtothe sightingwe
were able tosee the pair of animalsfor4-5 surfacings.The animalswere seenswimmingalonga"slick".
A gillnetpangarapidlytransitingthe areastopped ~30m fromthe animalsandappearedtobe preparing
to settheirnet.The animalssurfacedagain~30m fromthe other side of the panga,at whichpoint
photosand videoswere obtained.There were twoanimalsvisibleoneachsurfacingof thissighting,with
the 2nd animal beingsmallerandpositionedslightlybehindthe firstinthe "calf position".
Sighting003 wasonlyseenforthree surfacings.The firstsurfacingappearedtobe asingle animal andit
was soquickthat I was unable toconfirmthatit was infact a vaquita.I gave the bearingandreticle (3.1
or 3.2) toSarah (data recorder) toinputas an "object"andalertedBobon the 2nd set of bigeyesthatI
had a potential vaquitasighting.We decidedtoslow downandturntowardsthe potential sightingto
confirm.Maybe 15 secondslaterI sawthe animal a secondtime (3.4-5 reticles) andwasable toconfirm
it wasindeedavaquita.~15 secondsafterthatI saw the animal a thirdand final time (3.8reticles) and
noticedthere wasa 2nd animal swimminginthe "calf position"alongsideandslightlybehindthe first

20. 2
0
animal.The 2nd animal'sfinlookedsignificantlysmallerbutsince itwasonthe opposite side of the
largeranimal I didnotget a good viewof the entire body/surfacing.All three surfacingsof the larger
animal gave clearperpendicularviews(headingW) andthe finof the largeranimal wasnoticeablynot
falcate;the trailingedge seemedtohave abitof a convex flare toit insteadof a concave/falcate curve.
We calledthe FarleyMowatoverto assistwithtrackingandlaunchedourtwo small boatsbut
unfortunatelywe wereunabletolocate the animalsafterthe thirdsurfacing.Iwasthe onlypersonto
see thissighting.Ididnotget a goodenoughlookat the animalsinsighting002 to be able to
confirm/denythattheywere the same individualsinsighting003 but PAO'sdescriptionof the first
sightingsuggeststhe adultanimalscouldhave beendifferent.
19 October2019
PaulaOlson(FarleyMowat):
Sightingno.004 was initiallydetectedover2 nmilesfromthe ship;thisdetectionconsistedof asingle
animal witha tall,conical dorsal fin.The animal wasseentwice andwasswimmingsouthwhile atthe
surface.The shipturnedtowardthe sightinglocationandaftermovingsome distance aresightwas
made,approximately0.25nmile fromthe original location.Thistime the groupconsistedof a
mother/calf pair.Bothdorsal finswere slightlyfalcate.The calf wassmallerthanthe juvenile insighting
no.002 from17 October.Distantphotoswere obtainedof the mothercalf pair(AdamÜ). A third
resightwasmade,againinalmostthe same location;thistime the mother/calfpairwasseenaswell asa
secondpair,twoadults,that surfacedseveral bodylengthsfromthe mother/calfpair.The twopairsdid
not surface at the same time.(See Ernesto'scomments,below.) The finalresightingfromthe Farley
Mowat was thatof the mother/calf pair.The mother/calf were alwaysobservedswimmingwithinone
bodylengthof one another.The initial single animal mayhave beenone of the pairof adults.A single
animal wasseenlaterbythe Narval (see below).
ErnestoVazquez#125 (FarleyMowat):
Resightof sightingno.004. Asthe cow /calf was resightedforacouple surfacings,while this2went
downanothercouple vaquitassurfacedsome bodylengthstotheirright,seenjustone time atthe
surface,theydidnotsurface at the same time,neverhada view withthe 4vaquitasat the same time at
surface.The pair that wasleadingthe 4 seemeddifferentin size,one largerthanthe other,butcannot
say that wasanothercow/calf pair,althoughtheywere close toeachotherbya bodylength.
R. Pitman#005 (Narval) seenabout30min afterthe lastresightof sighting004 withinabout1-2 miles.A
single individual seentwice asitrolledoverperpendiculartothe vessel.ItappearedsmallandIsuspect
that an adultwas nearbybutunseen
October27 Sightings006 and 007
Sighting#006
Thissightingisseenfor48 minutes. Ittakes44 minutestoconfirmthatthere are 4 animalsandtheyare
describedasnotsurfacingsuperclose togetherwithone a‘tinybitsmaller’thanthe others. The two
photographedwere bythensomewhatseparate andappearedtobe similarinsize (see belowand

21. 2
1
compare to photoof cow/calf above). Thisphototakenfroma panga estimatedtobe about75-100 m
away.
From Tom
Jefferson:TAJ- I
was onthe
port bigeyes. I
have little to
add,exceptthe
following. On
multiple
surfacingsof
the group,I saw 4 animals,all of adult/juvenile size,butnone lookedtobe a calf of the year. Onone
surfacing,Isaw the 4 animalssurface inquicksuccession,andthenanothersurfacingabout4-5sec
later.That final surfacingcouldpossiblyhave beenafifthanimal,butIthinkitmuch more likelythatit
was one of original 4 surfacinga secondtime. Onlaterresightsof the group,I neversaw any indication
of a fifthanimal.
Adam’sphotofromthisgroup appearsto be a differentanimal to
the pair above. He feelsthisphotomatchestoone he saw in
sighting003 above (17 October,whichhadan adultand a smaller
individual).
Sighting007 seen51 minutesafterthe lastresightof 006 about 3.2 milesaway
From Adam:ACÜ –A mom/calf pairwas trackedfor a handful of surfacings,duringwhichtime TomandI
had multiple goodbroadside looksatthe pairslow-rolling,withthe calf beingnoticeablysmallerthan
mom(Tom says“calf of the year”size”) andsurfacinginthe calf positioneverytime.We launchedthe
small boatsbut were notable to track the pair forclose approaches.
Throughoutthe sightingTomand I feltconfidentthatthe calf in007 was smallerthanthe juvenile we
had seenin006. My photosfrom sighting006 showedthe aforementionedcloselyassociatedpair,butif
the pair in007 were partof 006 itwouldmeantheysplitfromthe otherpairand traveledinastraight
line awayfromthe area theyhad beenconsistentlyoccupyingwhenwe werefollowingthemandaway
fromthe directionof travel theywere swimmingwhenwe firstsaw them.
From Todd: -- I wasable to see several surfacingsof bothanimals,oftentimesnice broadside viewsin
goodlightas theymilledaroundasmall areanear twopangasthat had gillnetsinthe water. Itwas
quite obviousthatitwas a cow/calf pairandthe calf was small,Iwouldestimate aroundhalf the body
lengthof the adult. The dorsal finwasconsiderablysmallerandI realizedinstantlythatitwasa
differentanimal fromthe groupof 3 to 4 (sighting006) I had seenearlierinthe day.

22. 2
2
Afterthe sighting,itwasaskedif itwas possible thatthe cow/calf pairwaspart of the group of 4 seen
earlierinthe morning. WhatI can say for sure isthat the cow calf pairwas not the pairof the animals
we photographedfromthe pangainthe earliersighting. Myestimate fromthatgroupin the bigeyes
before Ijoinedthe pangawas3/4/3. I didnot geta goodlookat the 4th
animal inthe group and cannot
offeranyinsightintothe size of thatindividual. So,Isuppose if that4th
animal wasa calf thenthe
cow/calf observedlaterinthe daymayhave beenpart of that group. However,the waythe calf was
surfacingsoclose to itsmother,I thinkI wouldhave seenthatbehaviorinthe previoussightingasIsaw
several surfacingsequencesfromthe groupof three to four. Despite the oddangle tothe surfacings,all
of those animalsappearedtobe the nearlythe same size. So,mygut feelingisthatthe cow/calf were
not part of the earliergroup.

23. 2
3
Appendix 5ExpertElicitation1
ObserverexpertopiniononvaquitasightingsonSept3,Oct 17, 19, & 27.
The objective of these questionsistoquantifyexpertiseonthe numberof vaquitasseen. Eachobserver
whoworkedonthe Octobereffortshouldreadthe Observationdocument.
1. What isyour name?
2. What datesdidyou work(maximumtime isfromOct14-Oct 28)
3. Assign10 likelihoodpointsto:
a. Sighting2 wasa motherandcalf
b. Sighting2 isnot a motherand calf
4. Assign10 likelihoodpointsto:
a. Sighting3 wasa motherandcalf
b. Sighting3 isnot a motherand calf
5. Assign10 likelihoodpointsto:
a. Sighting2 andsighting3 are separate sightings
b. Sighting3 isa resightof sighting2
6. Assign10 likelihoodpointsto:
a. The small individualinsighting4is a calf
b. The small individualinsighting4is nota calf
7. Assign10 likelihoodpointsto:
a. The individual describedinsighting5is a new sighting
b. Sighting5 isa resightof sighting4
8. Assign10 likelihoodpointsto:
a. The individual insighting5isa calf
b. The individual insighting5isnot a calf
9. Assign10 likelihoodpointsto:
a. The smallerindividual insighting6wasa calf
b. The smallerindividual insighting6wasnot a calf
10. Assign10 likelihoodpointsto:
a. The smallerindividual insighting7wasa calf
b. The smallerindividual insighting7wasnot a calf
11. Assign10 likelihoodpointsto:
a. Sighting7 wasa newsighting
b. Sighting7 wasa partial resightof sighting6
12. Assign100 likelihhoodpointtothe plausible numberof calvesseeninOctober
a. 0 calves(none of the sightingsincludedacalf of the year)
b. 1 calves
c. 2 calves
d. 3 calves
e. 4 calves
f. 5 calves
g. 6 calves(all of the putative sightingswithcalveswere unique calves)
13. What isthe lowestplausiblenumberof vaquitasyoubelievewere seen?(give anumber)
14. What isthe highestplausiblenumberof vaquitasyou believe were seen?(giveanumber)

24. 2
4
15. Assign100 likelihoodpointstothe numberof unique vaquitaindividualsyouthinkwere seenin
thisyears’effort(bothSeptemberandall of OctoberthroughOct 27) (use all 100 points,yourresponse
shouldrange betweenyouranswerstoquestion13and 14)
a. 4
b. 5
c. 6
d. 7
e. 8
f. 9
g. 10
h. 11
i. 12
j. 13
k. 14
l. 15
m. 16
n. 17
o. 18
16. Optional question. Answerif youfeel youhave a goodmemoryforthe 2018 effort. Keepin
mindwhenreadingbelow,thatthere were 2shipswithbigeyes(4big eyesaltogether) in2019 and1
shipwithbigeyes(2bigeyes) in2018. In 2018 there were 2 days of goodeffort with5 sightingsand2
daysof poor to goodeffortwithnosightings. Here isanextractfrom our Reportof the effortto aide
your memory:
A field effortto obtain photographsand biopsiesfromvaquita washeld between September22and 28,
2018. Vaquitasightingsweremadeon September26 and photoswereobtained. Sighting #001 wasa
motherand calf thatwere photographed. Thepair wasobserved surfacing within a body length of one
anotherover30 times. At onepointthe pair came close (20m?) to the panga butthey werecoming
directly towardsthevessel and so no biopsiescould be taken.The motherwasphotographically matched
to the likely motherof V01F from2017. This pair wastracked from9:10 until 10:09 (59 minutes). A
second pair (#002) was sighted at 10:03 and tracked until 10:26 (23 minutes). Observersthoughtthe
second pair weredifferentand of roughly equalsize to one another,which wasconfirmed in photos.
Vaquitasightingswerealso madeon September27 and both photosand videos wereobtained. Sighting
#003 wasa pair seen between 06:39 and 06:51. Small boatswerelaunched butwe were unableto track
or photographthispair. Sighting #004 wasalso a pair whereboatswere launched butwe wereunable
to trackor photograph. Sizesof the individualsof both #003 and #004 weretoo distantto determine
relative sizes. Sighting #005 wasa group of 4 individualsthatincluded a small calf and wastracked for 1
hourand 42 minutes(from11:06 until 12:48). During this time they gotwithin 50m of varioussmall
boatsbutneverwithin biopsy range. Thefourvaquitasevidently then split into two pairs (which is
consistentwithphotographsreviewed later). Oneof thepairs waslast sighted at13:41 for a total
tracing period of 2 hours and 35 minutes.Given the timing and spacing of sightings,it wasconcluded
thatat least 6 vaquitaswereseen during field operations.
September28th
and 29th
had marginalconditionsforvaquita sightingsand therewereno further
opportunitiesforeitherphotographsorbiopsies.

25. 2
5
In 2019 there were 4 daysof goodconditionswith7sightings(the equivalentof 8shipdays).There were
2 daysof marginal andsome relativelygoodconditionswithnovaquitasightings.
Comparing2019 to2018 withinthe Zero Tolerance Area,please distribute 10likelihoodpointsto:
a. There were lessthanhalf the numberof vaquitasin2019 comparedwith2018
b. There wasbetweenhalf andthe same numberof vaquitasin2019 comparedwith2018
c. There were aboutthe same numberof vaquitasin2019 comparedwith2018
d. There were more vaquitasin2019 thanin 2018

26. 2
6
Appendix 5Expertelicitation2
The summaryand questionswere sentindependentlytothe 9 observer/experts. The probabilitiesin
reddifferedforeachobserver(shownisone anonymousobserver) andwere calculatedbasedon
answersgivenbyeachobserverinelicitation1.
Short summaryof an anonymousexpert’s responsestosightings
Sept3
Sighting001 2 adults
Oct 17
Sighting002 2 animals(1possible calf)
Sighting003 2 animals(1 possible calf)
Your probabilitythatsightings2& 3 are separate sightings= .80
Your probabilitiesfor:
0 calves0
1 calf .28
2 calves.72
Oct 19
Sighting004 2 animals(1 possible calf)
Sighting005 1 or 2 animals(couldbe a calf withadultnot seen)
Your probabilitythat4 and5 are separate sightings: .70
Your probabilitiesfor:
0 calves0
1 calf .75
2 calves.25
October27
Sighting006 2 animals(1 possible calf)
Sighting007 2 animals(1 possible calf)
Your probabilitythat6 and7 are separate sightings: .8
Your probabilitiesfor:
0 calves0
1 calf .44
2 calves.56

27. 2
7
1. What isthe lowestplausiblenumberof calvesseeninsightings2-7?(picka number)
2. What isthe highestplausiblenumberof calvesseen insightings2-7?(picka number)
3. What isthe mediannumberof calves(youshouldfeelcomfortable thatif youflippedacoin,the
true numberwouldhave evenchancesof beingabove orbelow thisnumber)
4. Assign100 likelihoodpointtothe plausiblenumber of calvesseeninOctober
a. 0 calves(none of the sightingsincludedacalf of the year)
b. 1 calves
c. 2 calves
d. 3 calves
e. 4 calves
f. 5 calves
g. 6 calves(all of the putative sightingswithcalveswere unique calves)
5. What isthe lowestplausiblenumberof vaquitasyoubelievewere seenduringall 7sightings?
(give anumber)
6. What isthe highestplausiblenumberof vaquitasyoubelieve were seenduringall 7sightings?
(give anumber)
7. What isthe mediannumberof vaquitasyoubelieve were seenduringall 7sightings(givea
number)
8. Assign100 likelihoodpointstothe numberof unique vaquitaindividualsyouthinkwere seenin
lastyear’seffort(bothSeptemberandall of OctoberthroughOct 27). Your responsesshouldbe
assignedtonumbersbetweenyour answersto questions2and3 above. Abouthalf yourpointsshould
be above the median. The likelihoodpointsthatyouassigntothe numbersbelow shouldsumto100.
a. 4
b. 5
c. 6
d. 7
e. 8
f. 9
g. 10
h. 11
i. 12
j. 13
k. 14
l. 15
m. 16
n. 17
o. 18

28. 2
8
Appendix6 ExpertElicitationresults
Questionset1 forindividual sightingswhere 10pointswere assignedandindividual observersare given
anonymousletter‘names’:
1. Sighting2 wasa motherandcalf
2. Sighting3 wasa motherandcalf
3. Sighting2 andsighting3 are separate sightings
4. The small individualinsighting4is a calf
5. The individual describedinsighting5is a new sighting
6. The individual insighting5isa calf
7. The smallerindividual insighting6wasa calf
8. The smallerindividual insighting7wasa calf
9. Sighting7 wasa newsighting
Question number
Expert 1 2 3 4 5 6 7 8 9
A 9 10 8 10 5 5 7 10 8
B 7 8 7 6 2.5 5.5 3 7 8.5
C 8 7 7 9 7 7 2 10 9
D 10 8 9 10 4 5 3 10 8
E 3 6 9 9 8 5 2 10 10
F 7 5.5 6.8 8 5 5.6 5.2 9.5 7.5
G 9 8 7 9 8 5 4 10 10
H 8 7 6 9 5 7 5 9 9
I 7 9 9.5 10 0.3 9.7 2 10 10
Average
probability 0.756 0.761 0.770 0.889 0.498 0.609 0.369 0.950 0.889

29. 2
9
Responsestothe secondelicitationgiventhe responsesabove(see Appendix5). Expertswere askedto
assign100 likelihoodpointsacrossthe range of possible calf numbers
Number of vaquita calves
Expert 0 1 2 3 4 5 6
A 0 0 10 40 40 10 0
B 0 10 45 35 10 0 0
C 0 0 0 0 40 60 0
D 0 0 30 40 20 10 0
E 0 21.05 52.6 15.79 10.5 0 0
F 0 15 58 20 4 2 1
G 0 5 10 40 30 10 5
H 0 0 0 44.09 45.2 9.677 1.075
Average
probability 0 0.06 0.26 0.29 0.25 0.25 0.01
At least this
number 1.0 1.0 0.94 0.68 0.39 0.14 0.01
Responsestoexperts’estimatesof total numberof vaquitasseeninall 7sightingsusing100 likelihood
points.
Experts 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
A 0 0 5 5 5 8 20 20 20 9 5 3 0 0 0
B 0 0 10 10 20 20 20 10 10 0 0 0 0 0 0
C
D 4 10 10 10 30 10 10 6 5 3 2 0 0 0 0
E 0 0 10 20 30 20 10 5 5 0 0 0 0 0 0
F 3 5 16 34 22 14 4 2 0 0 0 0 0 0 0
G 0 0 0 0 0 5 5 5 10 15 20 25 10 5 0
H 0 0 5 12 32 24 16 8 3 0 0 0 0 0 0
I 0 0 3 4 25 35 25 4 4 0 0 0 0 0 0
Average
probability .01 .02 .07 .12 .21 .17 .14 .08 .07 .03 .03 .03 .01 .01 0
At leastthis
number 1 .99 .97 .90 .78 .58 .41 .27 .19 .12 .09 .05 .02 .01 0

30. 3
0
Only6 expertsrespondedtothe final question:
Comparing2019 to2018 withinthe ZeroTolerance Area,please distribute 10likelihoodpointsto:
e. There were lessthanhalf the numberof vaquitasin2019 comparedwith2018
f. There wasbetweenhalf andthe same numberof vaquitasin2019 comparedwith2018
g. There were aboutthe same numberof vaquitasin2019 comparedwith2018
h. There were more vaquitasin2019 thanin 2018
Expert 16a 16b 16c 16d
A 2 5 3 0
B 1 4 5 0
C 0 2 3 5
D 2 2 5 1
E 1 3 5 1
F 0 8 2 0
Average
probability 0.100 0.400 0.383 0.117