Satellite Tagging, Remote Sensing, and Autonomous Vehicles ...
1
Overview • Northern fur seals (NFS; Callorhinus ursinus) breeding on the Pribilof or Bogoslof Islands in the Bering Sea account for the majority of worldwide pup production in this species 1 . • NFS from these two sites undergo an annual migration of ~8 mo. in late autumn, winter, spring. • Different NFS groups face different pressures: • Reproductive males must acquire energy stores to fast and hold territories on land the following summer; • Pregnant adult females must acquire resources for fetal growth prior to giving birth after the migration; • Newly-weaned pups face winter storms, large waves, declining sea and air temperatures on their first migration, but are of much smaller size than adults and lack experience. • In a typical year, more than half of these pups do not survive 2 . • Over the past two decades, the Alaska Ecosystems Program at the NOAA National Marine Mammal Laboratory has engaged in extensive satellite tracking of Bering Sea NFS in order to better understand habitat requirements and foraging strategies during the migratory period. • Here, key findings from published and ongoing studies are summarized. Methods National Marine Mammal Laboratory researchers equipped 256 NFS (135 pups, 106 adult females, 15 adult males) from the Pribilof and Bogoslof Islands with satellite-telemetery tags capable of providing animal location estimates, and in some cases, dive data. Location data were fit with a switching state-space movement model. Movement and dive behavior were compared to remotely sensed data, atmospheric reanalysis, autonomous in situ ocean sampling, and animal-borne temperature and salinity data. These procedures are described in Sterling et al. 3 and Pelland et al. 4 Behavioral responses to environmental features suggest that the latter influence prey fields, and/or NFS condition, and potentially demography. • Results suggest winds could influence a) pup dispersal patterns and b) availability of resources through forced changes in upper-ocean stratification. These factors may be relevant to unexplained large interannual variations in pup survivorship and are the subject of ongoing study. • Basin-scale interannual variability (e.g. , PDO, ENSO) affects winds, upper-ocean stratification, and ecosystem productivity and plays an unknown role in Bering Sea NFS demography, but is of great significance to San Miguel NFS breeding in the California Current System. • Observations over the past 50 yr indicate changes in character of annual cycle of wave heights in North Pacific Ocean, and a poleward shift in the midlatitude storm track. Ongoing work is investigating the impact of these trends on NFS demography during this period. References: 1 U.S. Department of Commerce (2007). Conservation Plan for the Eastern Pacific Stock of Northern Fur Seal ( Callorhinus ursinus ). Juneau, AK: National Marine Fisheries Service. 2 Lander, R.H. (1979) Fishery Bulletin 77(1):311:314. 3 Sterling, J.T., et al. (2014) PLoS ONE 9(4):e93068 doi:10.1371/journal.pone.0093068. 4 Pelland, N.A. et al. (2014) PLoS ONE 9(8):e101268 doi:10.1371/journal.pone.0101268. Acknowledgements: The authors would like to thank Rod Towell, Jim Thomason, Brian Fadely, Carey Kuhn, Tom Gelatt, and Kate Towell for their field efforts in capturing and instrumenting northern fur seals. Northern fur seal research was performed under National Marine Mammal Laboratory Permit No. 782170800, with additional funding contributions from the North Pacific Research Board, NOAA, the National Cooperative Research Program of the NOAA Fisheries Office of Science and Technology, and the Joint Institute for the Study of the Atmosphere and Ocean at the University of Washington. The recommendations and general content presented in this poster do not necessarily represent the views or official position of the Department of Commerce, the National Oceanic and Atmospheric Administration, or the National Marine Fisheries Service. Figure 1. Migratory tracks of 256 satellite-tagged Bering Sea NFS. Oct-December (top panel), January-March (middle),April-July (bottom). Green contour indicates Transition Zone ChlorophyllFront(TZCF). NFS migrate to different regions depending on their age and/or sex. Noel Pelland, Jeremy Sterling, Devin Johnson, Rolf Ream – National Marine Mammal Laboratory, Alaska Fisheries Science Center, NMFS, NOAA Alan Springer – University of Alaska, Fairbanks Sara Iverson – Dalhousie University Mary-Anne Lea – University of Tasmania Nicholas Bond – Joint Institute for the Study of the Atmosphere and Ocean Charles Eriksen, Craig Lee – University of Washington Satellite tagging, remote sensing, and autonomous vehicles reveal interactions between foraging and migration behaviors of a top marine predator and the environment in the North Pacific Ocean This had been shown previously but with different sexes/ages in separate years. It has probably been known in traditional historical knowledge for millenia. These groups face differing environmental conditions. In particular, most adult females and some pups migrate to productive eastern boundary currents that, compared to the interior North Pacific, have a shallower mixed-layer depth (MLD), fewer high-wind days (Fig. 2), and topographic features where eddy generation is common 4 . Despite this, adults have similar diving behavior. Adult males dived during all times of day and to the base of the mixed layer or below during daytime 3 (e.g. , Fig. 3). Females entering the California Current exhibited similar patterns 4 in a region of shallower MLDs (Fig. 4). Correlation with the MLD implies that upper- ocean stratification affects the depth of NFS prey fields 3,4 . NFS of all groups respond to the wind. −0.1 0 0.1 0.2 0.3 0.4 0.5 0.1 0.05 0 −0.05 0.1 0.05 0 −0.05 0.1 0.05 0 −0.05 0.1 0.05 0 −0.05 0.1 0.05 0 −0.05 0−5 m s −1 5−10 m s −1 10−15 m s −1 15−20 m s −1 20−25 m s −1 Pup mean velocity in downwind/crosswind coordinates Downwind m s −1 Crosswind m s −1 Downwind Crosswind (right of wind) −0.1 0 0.1 0.2 0.3 0.4 0.5 0.1 0.05 0 −0.05 0.1 0.05 0 −0.05 0.1 0.05 0 −0.05 0.1 0.05 0 −0.05 0.1 0.05 0 −0.05 0−5 m s −1 5−10 m s −1 10−15 m s −1 15−20 m s −1 20−25 m s −1 Female mean velocity in downwind/crosswind coordinates Downwind m s −1 Crosswind m s −1 −0.1 0 0.1 0.2 0.3 0.4 0.5 0.1 0.05 0 −0.05 0.1 0.05 0 −0.05 0.1 0.05 0 −0.05 0.1 0.05 0 −0.05 0.1 0.05 0 −0.05 0−5 m s −1 5−10 m s −1 10−15 m s −1 15−20 m s −1 20−25 m s −1 Male mean velocity in downwind/crosswind coordinates Downwind m s −1 Crosswind m s −1 -180 -135 -90 -45 0 45 90 135 180 0 0.01 0.02 0.03 0.04 0.05 0.06 Proportion of Points in 5 ° Bins E N W W S Movement Bearing Histogram, |u 10 | < 3 m s -1 Pups Adult Females At very low winds, adult females show a travel preference to the SE compared to pups (Fig. 6), likely reflecting prior knowledge/experience. High wind was associated with decreasing area-restricted search and increased transit for all groups (Fig. 7). NFS travel correlates with wind direction and speed, being on average downwind and to the right (Fig. 5). Results are consistent with large differences in size and experience between adults and pups, and in size between sexes. NFS mass correlates with maximum dive depth and size with the ability to withstand cold temperatures. Larger adult males: • Have dive depths that correlate with deep, interior MLDs, • Exhibit movements that are least correlated with the wind, and • Initiate consistent resident behavior earliest in the winter (Fig. 7). They are followed by smaller adult females, and later by the still-smaller pups (Fig. 7) . We conclude 3 that: • Adult females are constrained by their physical characteristics to migrate to temperate, productive eastern boundary currents (predictable and accessible prey resources). • Adult males do not face these same constraints, and avoid predation and competition by foraging in the interior North Pacific Ocean and southern Bering Sea. Figure 2. Proportion of high-wind (>11 m s -1 ) 6hr periods in January-March,NCEP Reanalysis 1, 1991-present. Contoursare climatological MLD for these months from Argo array. Figure 3. Example divingrecord from migratory adultmale NFS 3 . (Left) Mean 6hr dive depth vs. time; 6hr periods colored by time of day. Black line shows MLD from animal-borne sensors. (Right)Horizontaltrack. Figure 4. (top) Climatological MLD (dots) and salinity (colors) in Washington coastaltransition zonefrom autonomousunderwater gliders. (bottom) Adult female NFS mean 6hr dive depth (circles) and climatological chlorophyll (colors). Black lines show generalized additive model fit to MLD/diving 4 . Figure 5. Mean NFS travel(vectors),rotated into downwind-crosswind coordinate system based on NCEP Reanalysis 1 winds interpolated to animal location. (left) Pups, (center) adult females, (right) adult males. Mean travel computed in five bins based on wind speed (rows). Whiskers are 95% Student’s-t confidencebounds on mean travel. Figure 6. Histogram of travel bearing for pup (blue) and adult female (red) NFS during Oct.-Jan. at very low wind speeds (<3 m s -1 ), where wind effect on travel is weak. Whiskers are RMS confidence bounds on bin proportion. Figure 7. Ribbon plots show linear mixed-modelbehavioral state response due to days since 1 October (x-axis) and due to wind speed (ribbon colors),for adult males (top), adult females (middle),and pups (bottom). High behavioralstatecorresponds to resident,area-restricted search movement. Resident Transient Resident Transient Resident Transient u v w x y
Transcript of Satellite Tagging, Remote Sensing, and Autonomous Vehicles ...
Overview• Northern fur seals (NFS; Callorhinus ursinus)
breeding on the Pribilof or Bogoslof Islands in the Bering Sea account for the majority of worldwide pup production in this species1.
• NFS from these two sites undergo an annual migration of ~8 mo. in late autumn, winter, spring.
• Different NFS groups face different pressures:• Reproductive males must acquire energy stores
to fast and hold territories on land the following summer;
• Pregnant adult females must acquire resources for fetal growth prior to giving birth after the migration;
• Newly-weaned pups face winter storms, large waves, declining sea and air temperatures on their first migration, but are of much smaller size than adults and lack experience.
• In a typical year, more than half of these pups do not survive2.
• Over the past two decades, the Alaska Ecosystems Program at the NOAA National Marine Mammal Laboratory has engaged in extensive satellite tracking of Bering Sea NFS in order to better understand habitat requirements and foraging strategies during the migratory period.
• Here, key findings from published and ongoing studies are summarized.
MethodsNational Marine Mammal Laboratory researchers
equipped 256 NFS (135 pups, 106 adult females, 15 adult males) from the Pribilof and Bogoslof Islands with satellite-telemetery tags capable of providing animal location estimates, and in some cases, dive data.
Location data were fit with a switching state-space movement model. Movement and dive behavior were compared to remotely sensed data, atmospheric reanalysis, autonomous in situ ocean sampling, and animal-borne temperature and salinity data. These procedures are described in Sterling et al.3 and Pelland et al.4
Behavioral responses to environmental features suggest that the latter influence prey fields, and/or NFS condition, and potentially demography.• Resultssuggestwindscouldinfluencea)pupdispersalpatternsandb)availabilityofresourcesthroughforcedchangesinupper-oceanstratification.Thesefactorsmayberelevanttounexplainedlarge
interannual variationsinpupsurvivorshipandarethesubjectofongoingstudy.• Basin-scaleinterannual variability(e.g.,PDO,ENSO)affectswinds,upper-oceanstratification,andecosystemproductivityandplaysanunknownroleinBeringSeaNFSdemography,butisofgreatsignificance
toSanMiguelNFSbreedingintheCaliforniaCurrentSystem.• Observationsoverthepast50yr indicatechangesincharacterofannualcycleofwaveheightsinNorthPacificOcean,andapoleward shiftinthemidlatitude stormtrack.Ongoingworkisinvestigatingthe
impactofthesetrendsonNFSdemographyduringthisperiod.
References: 1U.S. Department of Commerce (2007). Conservation Plan for the Eastern Pacific Stock of Northern Fur Seal (Callorhinus ursinus). Juneau, AK: National Marine Fisheries Service.2Lander, R.H. (1979) Fishery Bulletin 77(1):311:314.3Sterling, J.T., et al. (2014) PLoS ONE 9(4):e93068 doi:10.1371/journal.pone.0093068.4Pelland, N.A. et al. (2014) PLoS ONE 9(8):e101268 doi:10.1371/journal.pone.0101268.
Acknowledgements: TheauthorswouldliketothankRodTowell,JimThomason,BrianFadely,CareyKuhn,TomGelatt,andKateTowell fortheirfieldeffortsincapturingandinstrumentingnorthern furseals.Northern fursealresearchwasperformedunderNationalMarineMammalLaboratoryPermitNo.782170800,withadditionalfundingcontributionsfromtheNorthPacificResearchBoard,NOAA,theNationalCooperativeResearchProgramoftheNOAAFisheriesOfficeofScienceandTechnology,andtheJointInstitutefortheStudyoftheAtmosphere andOceanattheUniversityofWashington.
TherecommendationsandgeneralcontentpresentedinthisposterdonotnecessarilyrepresenttheviewsorofficialpositionoftheDepartmentofCommerce,theNationalOceanicandAtmosphericAdministration,ortheNationalMarineFisheriesService.
Figure 1. Migratorytracksof256satellite-taggedBeringSeaNFS.Oct-December(toppanel),January-March(middle),April-July(bottom).GreencontourindicatesTransitionZoneChlorophyllFront(TZCF).
NFS migrate to different regions depending on their age and/or sex.
Noel Pelland, Jeremy Sterling, Devin Johnson, Rolf Ream – NationalMarineMammalLaboratory,AlaskaFisheriesScienceCenter,NMFS,NOAAAlan Springer – UniversityofAlaska,FairbanksSara Iverson – DalhousieUniversityMary-Anne Lea – UniversityofTasmania Nicholas Bond – JointInstitutefortheStudyoftheAtmosphereandOcean Charles Eriksen, Craig Lee – UniversityofWashington
Satellite tagging, remote sensing, and autonomous vehicles reveal interactions between foraging and migration behaviors of a top marine predator and the environment in the North Pacific Ocean
Thishadbeenshownpreviouslybutwithdifferentsexes/agesinseparateyears.Ithasprobablybeenknownintraditionalhistoricalknowledgeformillenia.
These groups face differing environmental conditions.
Inparticular,mostadultfemalesandsomepupsmigratetoproductiveeasternboundarycurrentsthat,comparedtotheinteriorNorthPacific,haveashallowermixed-layerdepth(MLD),fewerhigh-winddays(Fig.2),andtopographicfeatureswhereeddygenerationiscommon4.
Despite this, adults have similar diving behavior.
Adultmalesdivedduringalltimesofdayandtothebaseofthemixedlayerorbelowduringdaytime3(e.g.,Fig.3).FemalesenteringtheCaliforniaCurrentexhibitedsimilarpatterns4 inaregionofshallowerMLDs(Fig.4).CorrelationwiththeMLDimpliesthatupper-oceanstratificationaffectsthedepthofNFSpreyfields3,4.
NFS of all groups respond to the wind.
−0.1 0 0.1 0.2 0.3 0.4 0.5 0.1
0.05
0
−0.05
0.1
0.05
0
−0.05
0.1
0.05
0
−0.05
0.1
0.05
0
−0.05
0.1
0.05
0
−0.05
0−5 m s−1
5−10 m s−1
10−15 m s−1
15−20 m s−1
20−25 m s−1
Pup mean velocityin downwind/crosswind coordinates
Downwind m s−1
Cro
ssw
ind
m s−1
Downwind
Crosswind (right of wind)
−0.1 0 0.1 0.2 0.3 0.4 0.5 0.1
0.05
0
−0.05
0.1
0.05
0
−0.05
0.1
0.05
0
−0.05
0.1
0.05
0
−0.05
0.1
0.05
0
−0.05
0−5 m s−1
5−10 m s−1
10−15 m s−1
15−20 m s−1
20−25 m s−1
Female mean velocityin downwind/crosswind coordinates
Downwind m s−1
Cro
ssw
ind
m s−1
−0.1 0 0.1 0.2 0.3 0.4 0.5 0.1
0.05
0
−0.05
0.1
0.05
0
−0.05
0.1
0.05
0
−0.05
0.1
0.05
0
−0.05
0.1
0.05
0
−0.05
0−5 m s−1
5−10 m s−1
10−15 m s−1
15−20 m s−1
20−25 m s−1
Male mean velocityin downwind/crosswind coordinates
Downwind m s−1
Cro
ssw
ind
m s−1
-180 -135 -90 -45 0 45 90 135 1800
0.01
0.02
0.03
0.04
0.05
0.06
Prop
ortio
n of
Poi
nts
in 5°
Bins
E N WW S
Movement Bearing Histogram, |u10| < 3 m s-1
PupsAdult Females
Atverylowwinds,adultfemalesshowatravelpreferencetotheSEcomparedtopups(Fig.6),likelyreflectingpriorknowledge/experience.
Highwindwasassociatedwithdecreasingarea-restrictedsearchandincreasedtransitforallgroups(Fig.7).NFStravelcorrelateswithwinddirectionandspeed,beingonaveragedownwindandtotheright(Fig.5).
Results are consistent with large differences in size and experience between adults and pups, and in size between sexes.
NFSmasscorrelateswithmaximumdivedepthandsizewiththeabilitytowithstandcoldtemperatures.
Largeradultmales:• Havedivedepthsthatcorrelatewithdeep,
interiorMLDs,• Exhibitmovementsthatareleastcorrelatedwith
thewind,and• Initiateconsistentresidentbehaviorearliestinthe
winter(Fig.7).Theyarefollowedbysmalleradultfemales,andlaterbythestill-smallerpups(Fig.7).
Weconclude3 that:• Adultfemalesareconstrainedbytheirphysical
characteristicstomigratetotemperate,productiveeasternboundarycurrents(predictableandaccessiblepreyresources).
• Adultmalesdonotfacethesesameconstraints,andavoidpredationandcompetitionbyforagingintheinteriorNorthPacificOceanandsouthernBeringSea.
Figure 2. Proportionofhigh-wind(>11ms-1)6hrperiodsinJanuary-March,NCEPReanalysis1,1991-present.ContoursareclimatologicalMLDforthesemonthsfromArgoarray.
Figure 3. ExampledivingrecordfrommigratoryadultmaleNFS3.(Left)Mean6hrdivedepthvs.time;6hrperiodscoloredbytimeofday.BlacklineshowsMLDfromanimal-bornesensors.(Right)Horizontaltrack.
Figure 4. (top)ClimatologicalMLD(dots)andsalinity(colors)inWashingtoncoastaltransitionzonefromautonomousunderwatergliders.(bottom)AdultfemaleNFSmean6hrdivedepth(circles)andclimatologicalchlorophyll(colors).BlacklinesshowgeneralizedadditivemodelfittoMLD/diving4.
Figure 5. MeanNFStravel(vectors),rotatedintodownwind-crosswindcoordinatesystembasedonNCEPReanalysis1windsinterpolatedtoanimallocation.(left)Pups,(center)adultfemales,(right)adultmales.Meantravelcomputedinfivebinsbasedonwindspeed(rows).Whiskersare95%Student’s-t confidenceboundsonmeantravel.
Figure 6. Histogramoftravelbearingforpup(blue)andadultfemale(red)NFSduringOct.-Jan.atverylowwindspeeds(<3ms-1),wherewindeffectontravelisweak.WhiskersareRMSconfidenceboundsonbinproportion.
Figure 7. Ribbonplotsshowlinearmixed-modelbehavioralstateresponseduetodayssince1October(x-axis)andduetowindspeed(ribboncolors),foradultmales(top),adultfemales(middle),andpups(bottom).Highbehavioralstatecorrespondstoresident,area-restrictedsearchmovement.
Resident
Transient
Resident
Transient
Resident
Transient
u
v
w
x
y
