••~

Not to be ~ited without prior referenee to the author

International Couneil for theExploration of the Sea

C.M.1991/Mini:6

••

Potential Effeets of Matern~l Contributionon Egg and Larva~Population Dynamics

of Striped Bass: Inte~rated Individual-Based Modeland Directed Field Sampling

.James H. Cowan~ Jr.The University of Maryland System

Center for Environmental.and Estuarine StudiesChesapeake Biological Laboratory

Solomo~s, MD 20688-0038

and

Kenneth A. RoseEnvironmental Sciences DivisionOak Ridge National Laboratory

P.O. Box 2008 'Oak Ridge, TN 37831-6036

ABSTRACT

. We have used,a bioenergetically~driven; individual-based,model (IBM) ofstriped bass as a framework tor synthesizing available information on populationbiology and quantifying, in a relative sense, factors'that potentially affect yearclass success. The IBM has been configured to simulate environmerital conditionsexperienced by several 'striped bass populations; i.e., in the Potomac River, MD;in Hudson River, NY; in the Santee~Cooper River System, SC, and; in the SanJoaquin-Sacramento River System, CA. These sites represent extremes in thegeographie distribution and thus, environmental variability of striped bass .spawning. At 'each Iocatiori, data describing the physio-chemical and biologicalcharacteristics of the spawninq population and riursery area are being collectedarid synthesized by means of a prioritized, directed field sampling program that isorganized by the individual-based recruitment model. Here, we employ the stripedbass IBM configured for the Potomac River, MD trom spawning into the larval periodto evaluate the potential for maternal contribution to affect larva survival andgrowth. Model simulations in which the size distribution and spawning day offemales are altered indicate that larva survival is.enhariced (3.3-fold increase)when a high fraction of~fem~les,in the spawning population are large. Larva stageduration also is less (X = 18.4 d and 22.2 d) when large and,smail females,respectively, are mothers in simulations. Although inconclusive, thesepreliminary results for Potomac River striped bass suggest that the eftects of. ,female size, timing of spawning and maternal contribution on recruitment dynamicspotentially are important and illustrate our approach to the study of re~ruitment

'in striped bass •. We hope to use the model, field collections and managementplternatives that vary from site to site"in an iterative manner tor some time tocome. '

1

r---------------- --- -------

',' ..INTRODUCTION

Recruitment success of populations of anadromous striped bass Morone .'saxatilis varies widely throughout its range and,mrich attention has been focusedon early life dynamics (Setzler-Hamilton et al. 1981; Boreman and Austin 1985;Goodyear 1985; Bulak et al.· 1985; Houde et al~ 1988; Stevens 1988; Secor 1990;Rose and Cowan, submitted; Cowan et al.~submitted). The role of materrialcontribution in recruitment potentially isoigriificant because large females spawngreater numbers of heavier eggs (Rogers and Westin 1981; Zastrow et al. 1989; .Monteleone and Houde 1990). Egg, weight and embryo size also are correlated

,. . . ! ,

(Eldridge et al. 1982; Zastrow et al. 1989; Monteleone and Houde 1990) suggestingthat the rnaternal contribution to egg size may influence survival cf embryos andlarvae through increased eriergy and nutrient.stores (Secor 1990). -

Recerit stridies of rnaternal influences on striped basa.eggs and larvae andpossible effects on recruitmerit produced coritradictory results. In ChesapeakeBay, MD, Monteleone and Houde (1990) showed that small «4 kg) females producedsmaller larvae (20% by weight) than large ()15 kg) females. The small larvae grewslower in length at 19 ± 0.5°C and remairied small (-20% by weight) to 25 days posthatch. Monteleone and Houde concluded that maternal influences on egg quality maybe a factor that affects recruitment potential because slow growth prior to .metamorphosis subjects larvae to high mortality during pre-recruitment. InSantee-Cooperi SCi Secor (1990) reported egg size variabilitY.but no significantrelationship between striped bass female size and egg weight for )3 to 12.2 kgfemales. Large eggs used stores faster but produced larger larvae (16% by weight)than broods of small eggs. Initial eggweight advantages persisted for less thanone week in feeding larvaereared at -20.0o C, although larvae from large eggbroods were somewhat less vulnerable to starvation. Secor concluded thatvariation in egg size could be adaptive to variable feeding situations butcautioned against the assumption that egg investment always is positively relatedto larval fitness and.increased recruitment~

Germane to this apparent contradiction is the anecdotal belief that largefemale striped bass iri some populatiöris spawn at cooler temperatures~ earlier inthe spawning season than do smaller females (Hollis 1967)~ Neither of the stüdiesdescribed above considered the energetic,consequerices of egg and larva size atdifferent temperatures nor was there good agreement in female size (Secor 1990~

3.2 to 12.2 kg; Monteleorie and Houde 1990, <4 arid )15 kg) between experiments.Moreoveri Houde.et al. (1989) showed that early spawned (early April) egg cohorts .in the Potomac River; MD, episodically.are killed by cold weatherevents causingtemperatures to drop below 11°C.

. The stze distribution of striped bass female spawners in the Potomac Riverand upper Chesapeake Bay haschanged dramaticallyover the past ten years, showinga substanti~l.deci~easei~ modallength from 1982 ~ntil 1985.Mod~1 lengths h~~eincreased since 1985, presumably.due to a moratorium on striped bass fishing inChesapeake Bay. Modal lengths of females on the' spawning grounds were -1000 mmstandard length (SL)' in 1982~ -425 mm in 1985 and -750 mm in 1990 (MarylandDepartment öf Natur~i ~es6~rces (~D DNR), Tidewaier Administration) •.

To evaluate the potential for maternal contribution to affect striped basslarva survival and.growth"we use an individual-based population model (IBM) ofstriped bass as a framework for synthesizing available information andquantifying, in a relative sensei their interactive'effects on populationdynamics. Here, we employ an IBM of striped bass in the Potomac River subestuary

2

I1

...•..

•

•

~' •. ~ ... ., .... "; il "··''''l;f.'4,:"~''~.~,.\,.t ,.I~·~.' ..•• ".''\ .• i ,

of Chesapeake Bay from spawnfng into the l~~vai period (-135 days) ~o deseribesame aspects of the model. The IBM also has been configured to simulateenvironmental eonditions experienced by other striped bass populations; i.e., inHudson River; NYi ,in the Santee~Cooper River System; SCi and; ,in,the Sari Joaquin­Saeramento River System, CA. Thesesites represent extremes in the geographiedistribution and thus, environmerital variability of striped bass spawning.

, , At eaeh loeation, data descrfbing the size distribution of female spawnersin the spawning grounds are being colleeted for model input by a direeted fieldsampling program sponsored by the Electric Power Research Institute'slong-termstudy of compensatorymechanisms, in fishpopulations (COMPMECH). Sampling isperform~d in cooperation with John Young and the Consolidated Edison Co., NewYork, Don Stevens and the California Department of Fish,and Game, Bay-DeltaFishery Project, and Jim Bulak and the South Carolina Wildlife and Marine

. Resources Department. Evaluation of the impiications on young-of-the-year 'dynamics of changes in female size distributions (e.g.; due to alternativefisheries management strategies),as weil as other factars that potentiallyinfluenee striped bass reeruitment, is a lang-term goal of this integratedmodeling and field project.

In addition to directed fieid sampling for spawning-related information, wealso are,re-examining historical data and, in same cases; iriitiating riew tieldstudies designed to provide data tor modelcorroboration arid testing~ Corro-'borative data from each site are being collected and cataloged by life stages,cohorts within lifestages,years classesand other appropriate units., P~rametersof interest are tied directly to clearly defined li fe cycl~ units in the IBM andinclude:

(1)· Abundance estimates of striped bass entering and leaving a·particular life stage, cohort, or population unit.

(2) Size-,and age-frequency distributions at the beginning and end ofeach cohort orlife stage. New field sampling that focuses on time­varying size distributions of female spawners on the spawning groundsat each site has been initiated (1990).

4It' Items (1) and (2) permit estimation of age/size and cohort-specific growth aridmortality rates.

(3) Sex ratios and maturation schedules.

(4) Measurements of time-varying physio-chemical and biologicaiparameters (prey levels) in spawning and nursery areas~

(5) Egg size distributions and appropriate measures of egg quaiity.(Field sampling initiated in 1990).

These data, where appropriate, will be used for either model inpllt for hypothesistesting or for corroboration of model output to field observations.

Hypotheses about factars that potentiallyaffect striped bass recruitmentare location-specific. Besides experiencing wide envirorimentalvari~bility;.striped bass populations at the directed samplinq locations alsovary widely inthe level of exploitation and potential, or perhaps need, for management. Some ofthe factars that havc beeri identified for development of recruitment' hypotheses

3

inelude:

1) Overfishi~g, changes in numbers and size distributions of spawningfemales (MD and SC),

2) Changes in spawning and nursery habitat fncluding watertemperature and flow regimes and productivity of food for larvae.andjuveniles (MD, SC, CA, NY),

3) Toxies and pollution, environmental degradation and low dissolvedoxygen .(MD, SC, CA, NY),

4) ~ater ~anag~~~nt and diversion from eritical habit~t (CA), ~nd

•~

5) Introduetion of exoties (CA).

METHODS - •Model Description

The striped bass IBM is deseribed in detail ,in Rose and Cowan (submltted)and will be summarized here only briefly. The model begins with the spawning ofindividualiemales andsimulates thegrowth and.mortality of ,young striped bass asthey develop through the egg, yolk-sac larva (YSL), feeding larva, (LAR) and (JUV)early juvenile stages. The model represents' thesedynamies in a daily time stepin a.single, weIl mixed 1000m x 1000m x 4m eompartment.

Each fe~ale's spawn of eggs is.followed as ari individual entity (i.e.,female brood) through hatehing and the yolk-sac stage~ Then a random subsampIe ofindividual feeding larvae is followed through the remainder of the simulation.Length (L, mm), weight (W, mg dw>~ age (days), and lifestage are ehronicled foreaeh individual. The environmental eonditions in the mixed compartmentconsist ofdaily vater temperature (OC), fraetion of the day there. i8 daylight, and 'density(no. m- 3 or m- Z ) of zooplankton prey types., '

Whenever possible, environmental and biologicai conditions in the model forthis example have been speeified using information derived from sampling in the. ~Potomac River. This ensures that the site-specific environmental and biologicaleonditions, that can co-vary in time and space in the River, are simulated in arealistic manner. For edification; a detailed description of female spawning andmaternal effects on egg size i5 given (from Rose and Cowan; submitted) •. Theremainder of the model descriptionwill be narrative except where more.detail iscritical to the.readers· ünderstanding of this study. Place markers of topieswhere specific data relevant to this study are lacking are denoted in the text.

DaiIv Temperature and Daylight: Daily temperatures from the striped bass spawninggrounds on the Potomac River were used to fit the regression equation of long-termaverage temperature (TOC):

(1) T = 14.79 - 12.02 * cos (0.0172 * day) - 6.60 * sin (0.172'* day), Stochasti~ realizations ~i daily temperatures arourid equ~tion (i> were

simulated by: 1) randomly determining withprobability 0.5,whether· a positive .(above average) or negative (below average) run of temperature occurs; 2) randomlydetermining the magnitude (greatest absolute deviation from t~e ~ong-term a~er~ge)

4

.; '. .,:: ... ,\ '.' ! .

.-

•

•

af the run from,a gamma probability distribution; 3) specifying the duration(number of days),of the run from'the magnitude (duration = 1~33 magnitude 1 • 41 ); 4)linearly interpolating temperatures between the beginning and end of the run suchthat the magnitude (peak)occurs'at the rnidpoint cif the run; arid~ 5) determining'whether aneutral (no deviation from lang-term average temperatures) or positive(if previous run was negative) or negative (if previous run was positive) occurs.These five steps are then repeated until an annual temperature series 1sgenerated. The fraction af the 24-hr dar that is daylight (DL) is:

(2) DL = 0.5i + 0.11 * ccis (0.017~ * (day-113» (Dalton 1987)

Spawninq of ,Females: ,The model simuiation begins with the spawning of 50 females(-0.05; 1000 m- 2 ); all ,are assumed to spawn (i.e., no mortalitY,of females beforespawning). We use a size distribution of spawners similar to the 1987-1988 periodin the Potomac,River aS,baseline (Figure 1). Thus, lengths of baseline spawnersranged from 503.3 to 979.5 mm SL, with a modal of 650~9. Large temale spawnersresembled the1982 Potomac River spawning population ranging in length from 511.1to 1284.6 mm SL, with a mode of 1037;8~ . Spawning temperature for each temale"ranging from -14 to 23°e (Setzler-Hamilton et ale 1980), is generated by a randomdeviate trom a triangular distribution with:

Toe minimum ~ i5.5 - 0.00305 * L,(3) Toe icide = 20.0,- 0.0045 * Lf

Toe maximum = 24.0 - 0.0053 * ~ (MD DNR)

The temperature endpoints of the distribution, when related to temale length(L f , mm SL), result in larger females tending tO,spa~n at lower temperatures(earlier in the spawning season) than smaller females, while smaller females showhigh variability in temperatüre at spawning (Figure 2).

Spawning of aiemaie take~ place on ihe first day: (1) temperature (Tiexceeds the assigned temperature of spawriing; arid (2) there are three successivedays of increasing temperature. Females not spawning on,the first day T exceedsthe assigned spawning temperature, spawn on the next available day that is thethird of three consecutive days of increasing temperature.

, , '

Female weight (W" Kg wet weight (ww» is derived trom:

(4) Wf

=2.27 * 10-8 *'~2.94 (Westin and Rogersi978)

" The,number (~.) and averaged weight (W•• mg dw) cif eggs per temale aredetermined from female weight (kg ww) by:

(5) N. = 218000 * W, - 11700 (Setzler-Hamilton et ale 1980), and

(6) W. - 0.00883 * W, + 0.215 (Zastrow et al; 1989);

Develcipment of ~ arid Yoik-Sac Larvae:. Each female'g spawn ,af eggs is followedas a cohort through the YSIJ stage until the initiation of first feeding; ,development proceeds as ä,functionof,daily.temperature (Boreman 1983)~ Uponinitiation of first feeding,initial weights of LAR are determined from eggweights of eachfemale cohort (Zastrow et al. 1989). ,Development rate of eggs andYSL are not related to egg weight-temperature combinations due toa lac~ ofspecific data; We here denote place marker 1 as a topie for needed research.

Mortality of ~ and Yolk-Sac Larvae: The number ofeggs, then YSL in eachfemale's egg cohort is reduced daily by a fraction dYing. Egg mortality istemperature-dependent and is based on a parabolic relationshipbetween hatchingsuccess and temperature (URI 1976; ,Setzler-Hamilton et al. 1980; Morqan et al.1981)~ When the temperature-dependent mortality rates fall below field estimatesof egg and YSLmortality (Dahlberg·1979; Olney:et al~~ in, press), we simply addconstant mortality terms to reconcile the field data and the model in lieii 01specific information on size-dependent predation and othersources of egg arid YSL

. mortality. We here.denote place marker 2;

Initial Numbers and sizes of First Feedin~ Larvae: A sUb-sample of individualfirst-feeding striped bass larvaeare taken from each female egg/YSL cohort andfollowed in daily time steps through growth and/or mortality~ Larvae from eachcohort are sUb-sampled in proportion to the number surviving in the cohort ,compared to the total riumber of all survivors. For model simulations presentedhere, we follow 20,OOO'first-feeding larvae (LAR). Initial lenqths and weights 01LAR vary for female egg-cohorts as a function of female size; LAR sUb-sampled froma given female egg cohort are identical.

Growth of Feedinq Larvae:Daiiy growth of individual larva beginning with firstfeeding is represented with a difference form of a bioenergetics eqüation:

..

•(7) W =W +p*,..." *A-R't t-l '~ax . tot

where Wt = weight (mg dw) at time t in days,c.ax = maximum consumption rate (mg dw d-l)~

A = assimilation efficiency, and ,Rtot = total metabolic rate (mg dw d- 1 ).

The proportion of C.ax realized by a larva on a given day is:

,(8)

(9)

P = Cr and

C.ax

nCr = I BI 'I< PW1 •i = 1

where Cr = biomass (mg dw) of prey consumed,B1 = number of prey type i eaten, arid

PW l = weight (mg dw) per individual of prey type i.

C.ax is a function of weight and temperature~ P is bounded by.O and 1 because Cris ~O and not allowed to exceed C~ax~ The summation over i prey types incomputing Cr is performed via optimal foraging theory (Krebs 1978; Hughes 1980);Determ~natiori of the number,of each preytype c?nsumed' (B l ) encompasses most ofthe computations of the model and is highly dependenton larva size andcapabilities. Values for C.äx (Cox and Coutant1981; ,Moore 1988; Tuncer 1988;Chesney 1989; Houde,1989; Zastrow and Houde 1989), A,(Ware 1975;~,Govorii et al.,1986; Tuncer 1988; Moore 1988; Zastrow and Houde 1989; Houde 1989; MacKenzie etal. 1990)"andRt~t (Eldridge et al. 1982; Mocire 1988; Zastrow and Houde 1989) arefrom the literature, adjusted for temperature eflects followingHewitt and Johnson(1987) .

6

•. 't....' .

computation of~: pis computed based on encounters of a feedinq iarva with tourzooplankton prey types. ,Bi is genera ted for each prey type and larva using abinomialdistribution, deperident on the number of encounters and the captureprobability. .

Larvae are assumed to encouriter patchy prey during the day. Mean number ofencounters is calculated as the product of search volume (or area for JUV) andprey density, search volume (Si) for LAR is, the volume of a cvlirider with radiusequa1 tothe 1arva's reactive distance (RD.) and height equal~to distance swamduring day1ight hours.

(0) RD. 2 * (Distance] * 10- 6 litersmm3

for LAR

"

Distance swam, RD. and, thus, encounters with pr~y items are related tolarva length~ Larger Iarvae swim faster and have ,a larger RD. (]ower angle of .acuity) for a given prey size (Breck and Gitter 1983). We use a swimming speed of0.63 body 1ength's (BL) S-1 for LAR based on observations of free-swimming 8 mmstriped bass 1arvae with food present at 20.0oC (Bowies et a1. 1976).

•

RD is calcülated based on the angle of acuity (<<), formed between sides of atriangle from the 1arva's eye to the,prey item, with each side of the trianq1ebeing RD. and the,base ofthe triangle oppositethe eye being the height,of thepreY,item. -,is'the minimum'ang1e detectab1e by a 1arva and decreases with 1arva1ength (Guma 1982). Reactive distances specific to striped bass 1arvae, and the~ffect.of turbidity on striped bass RD. have not been determined. We hefE! denoteplace marker 3.

Prey Capture: We use a single "capture probability" term (susceptibility, Baileyand Houde 1989) to encompass all of the events between encounter and ingestion.capture probabiiities are specified for each prey type; and are dependentoneither larva age or 1ength. We make capture success of small.prey eaten justafter first feeding (rotifers and copepod nauplii) a function of age to resemb1e alearning process; regardless of larva size, capture success increases with age.Capture success of 1arger zooplankton (adult copepods and cladocerans) eatenduring the later portion of the LAR stage is a function of 1arva langth toresemble physical development and ability; larger larvae have a greater capturesuccess than smaller larvae regardless of age.

Temporal Dynamics of prey: Densittes or abundance of each prey type (number .liter- 1 for zooplankton, are updated daily based on: (l),number of prey consumedby fish 1arvae (in this case striped bass arid white perch Morone americana): (2)how elose prey densities are to equilibrium densities; arid~--turnover rate.Seasonalequilibrium densities of zooplankton were determined from cumulativedistribution functions based on Potomac River field data (Setzler-Hamilton et ale1981; Houde et ale 1988). Turnover rates are estimated from temperature and preybiomass using a general regression relationship (Plante and Downing 1989)corrected for temperatures in the Potomac River.

Prey consumption Qy Larvae: Zooplankton are consumed by striped bass as weIl asby white perch (Morone americana). Effects of,white perch consuming prey areexplicitly represented in the model hecause: (1) striped bass and white perchlarvae are ecologicallysimilar and share spawning and nursery areas (Setzler­Hamiltöri,et ale 1981; Hjorth 1988); (2)white perch larvae generally outnumberstriped hass (Setz1er-Hamilton et al~' 1980; Houde et a1. 1990); and, (3) adequate

7

information for model configuration with white perch in the Potomac River isavailable (Houde et al. ,1990). White perch are treated as 10 size~cohorts (3-12mm by 1 mm size intervals) whose bioenergetics are assumed to be similar to larval'striped bass. .

Hortality of Feeding Larvae: Hortality of, individual LAR and JUV is a,function oflarva weight and length. Our represent~tion of weight-~ep~rid~rit mcirtalii~ is ,based onlaboratory data at low food~levels without predators (Rogers and West in1979, 1981; Eldridge et al. 1981, 1982; Chesriey 1989) and therefore can he viewedas starvation related mortality. Length-dependent mortality is estimated fromfield da ta (Logan 1985) and represents predation or other losses such as transportand gear avoidance. In the model, a larva dies if its weight (mg dw) beeomes lessthan some minimum.weight at length at time t in days, whieh is a barrier at 0~65of the weight predicted for the larva's length. Larvae are not allowed to loselength.

If a genera ted number from a uniform distribution between 0 and 1 is less than theprobability of the larva dying~ the larva is assumed tO,die~ No explieitinformation is available on the sources of striped bass mortality and its size­dependenee in the Potomae River~ We'here denote plaee marker 4.

Hodel Simulations

Four model simulations were performed to evaluate the relativeeffeets ofmaternal eontribution on recruitment variabilitY,of striped bass in the.potomaeRiver. We compare eombinations ofbaseline and large female spawners with day ofspawning temperature-related or random. Egg produetion" larva mortality rates aridsize distributions of survivors an day 135 are compared between simulations~ Day135 is the earliest day of metamorphosis of larvae. We foeus,on spawning throuighthe larval period under the.premise that reeruitment sueeess is established duringthat time (Houde et al. 1988; Uphoff 1989).

RESULTS AND DISCUSSION

The timing of female spawning, when,related to female,size and watertemperature (i.e., when temperature rule iS,"on~), was similar in pattern forhaseline and large model females (Figure J).' The temperature rule produeed threespawning days by baseline females, with a peak oeeurring an ,day 112 (April 22)when 60% of the females spawned. Large females also spawned on three days blitmast spawning (64%) oeeurred on day 100 (April 10), approximately two weeksearlier than by the baseline females. When the temperaturerule was toggled"oft", model females spawned at raridom with no mare than·16% of the femalesspawning on any one day in the simulation between day 100 and 121 (April 10 ~ May1) •

Diserete peaks in spawning and egg produetion by striped bass aretypieal ofwild populations, Frequently, one to three sharp egg produetionpeaks oeeurduring a spawning season arid aeeount for a signiticant portion of total eggs

'produeed (Johnson and Koo 1975; Combs 1979; Bulak et al., in press).

8

•.

•

•

, Egg production,in the simulations varied as an increasinq function of femalesize. Baseline ,females produced a total of 49,571,000 eggs compared to159,680,000 produced by the large model females (Table 1), qiving a 3.2-foldi~cr~aseiin egg ~rciduction.when:,the 50-fish s~a~ning population contained a higherfraction of larger females. '

, Length-frequency distributions of surviving striped ba~s larvae (Figure 4)and numbers of survivors (Table 1) also differed between model simulations. Ingeneral, simulation results suggest that,progeny from populations containing largefemales, may grow faster on average, resultinq in a higher survival rate (Table 1).A snapshot of the length-frequencies of larvae alive on day 135 (Hay15) in eachsimulation (Figure 4) shows that,larvae produced by large females were larger onthatday. Hean stage duration~ Le., the time in days from first-feedingto 20 mmSL '(~etamor~hosis) also ~a~ shorter, for larvae spawned by large females (18.4 d)compared to 22.2 d for larvae from baselirie mothers. Because the larvae froml~rqe fem~les were larger and growiriq fa~ter, the model predicts lower,mortality(X = 0.167 d- 1

) for these small stri~ed bass as weIl (Table 1, compared to 0.194d- 1 for baseline).

The simulations ftirther suggest that the interactive effects on spawning dayand subsequent larva survival, of female,size and water temperature (i.e., whenthe rule is toggled "on") are complex. Larvae ~urvivingon day 135 from baselinemother~ who spawned at random were larger than those ~roduced by baseline mothersspawning by the temperature rule (Figure 4), despite a small difference in meanspawn date between simulation~. The opposite is true for large females, however,as larvae were larger on day 135 in the simulation when female spawning day wasdetermined by the temperature rule. These large larvae also were olde~ as mearispawn day in simulations with large femalesdiffered by 5 days due to thetemperature rule (Table 1). survivorship increased slightly when females spawnedat random in both,cases (Table 1); but mortality rates iri simulations with largefemales were similar. '

Hodel simulations also indicate that the increased survfvorship when morelarge females are present is not merely a function of increased egg production~

The ratio of riumbers of larva survivors to 'metamorphosis from large and baselinefemales, respectively, was 3.5:1, ,which i~ higher than the 3.2-fold increase inegg production obtained in the model runs. When large arid baseline femalesspawned by the temperature rule, the ratio cif survivors to metamorphosi~ was .3.6:1, further illustrating the complex interaction between female size, spawningtemperature and larva survivorship.

~herever they occur; striped bass populations are year-class dominated, anddepend on astrang recruitment every 5-10 yearn to maintain spawning stock and ,spawning potentials (Van Winkle et al~ 1979; Goodyear 1989). ,Density-independent,environmental factors have most often been implied to cause high mortalities ofstriped bass eggs and influence,growth and survival of larvae,(Ulanowicz,andPolgar 1980; Rogers and Westin 1981; Chesney 1989; U~hoff 1989. Logan (1985) andHoude (1987, 1989) also have examiried da ta on slripedbass growth and havediscussed the implication of variable larva growth rate on recruitment success.Results from.these studies suggest that variability in larva stage duration may beimportant and that significant error in mortality estimates may be gerierated ifconstant stage duration is assumed.

Although these prelimlnary resuits for Potomac,River striped bass areinconclusive, they sho~ that the effects of female size, timing of spawning and

9

maternal contribution on recruitment dyriamics potentially are important aridwarrant further research.

ACKNOWLEDGMENTS

This research fs sponsored,bY the Electric Power Research Institüte underContract No; RP2932-2 (DOE No. ERD-87-672) "with the U.S. Departiiient of Energy,under Contract No. DE-AC05-84)R21400 with Kartin Karietta Energy Systems, Inc.

REFERENCES

...

•

Bailey, K.K. and E.D. Houde.and the recruitmerit problem.

1989. Predation on eggs arid larvae cf marine fishesAdv. Mar. Biol. 25:1-83.

Boremari, J.temperature.

,1983. Simulation of striped bass,egq arid larva development based onTrans. Am. Fish; Soc. 112:286-292.

Boreman, J. and H.M. Austin. 1985.bass stock alorig the Atlantic coast.

Prodüetion arid harvest of anadromous stripedTrans. Am. FiSh. See. 114:3-7.

Bowles, R.R., T.T. Polgar and H.H. Zion. 1976., Effects of vater veolcity on ~activity of,j~venile striped bass. ORNL/TM-5368, Oak Ridge National Laboratory,Oak Ridge, TN.

Breek, J.E. and M~J. Gitter 1983. Effeet of fish size on the reaetive distaneeof bluegill (Lepomis macroehirus) sunfish. Can. J. Fish. Aquat~ Sei. 40:162-167.

Bulak~ J;S., J.S. Crane and D.W. Gates. 1985. Natural reproduction of stripedbass. Spec. Progress Rept. WC-l Santee-Cooper Drainage, S~C. Wildl. Kar~ Res.Dept., Columbia, SC. 98 pp.

Bulak, J.S., J.S. Crane and M;S. Udevitz. In press. Egg production dynamics ofthe striped bass in the Cengaree and Wateree Rivers, South Carolina. Trans. Am.Fish. Soc.

Chesney~ E.J.~ Jr. 1989. Estimating the,food re~~irements.of stri~ed bass"larvae, Korone saxatilis: the effeets of light, t~rbidity and turbulence. Kar.~eol. prog. Sero 53:191-200.

10

"I

I • <. . ~ l •...- ' .',~. ~ .'. :',: ~ ~

•

Combs, D.L. 1979. Striped bass spawninq in the Arkansas River tributary ofKeystone reservoir, Oklahoma. proc. Annu. Conf. SE Assoe. Fish wildl~ Agencies.33:371-383.

Cowan, J.H., Jr., K.A. ROS~~ ·E.R~'·Rutherf~~d and E.D. Houde. Individual-basedpopulation dynamics model of striped bass in the Potomac Rivcr, Maryland. II~Factors affecting recruitment. Submitted to Trans. Am. Fish. Soc.

Cox, D.K. and C.C. Coutant. 1981. Growth dynamics of juvenile striped bassasfunctions of temperature and ration. Trans. Am~ Fish. Soc. 110:226-238.

Dahlberg, K.D~ 1979. A review of.survival rates of fish,eggs and larvae inrelation to impact assessments. Har: Fish. Rev. (Harch):1-12.

Dalton, P. 1987. Ecology of bay anchovy (Anchoa mitchilli) eggs and larvae inthe mid-Chesapeake Bay. Hs. Thesis, Univ. Haryland, College Park. 104 pp .

Eldridge, M.B., ,J.A. Whipple, D. Eng, M.J. Bowers and B.M. Jarvis. 1981. Effectsof food and 'feeding factors on laboratory-reared striped bass larvae. Trans. Am.Fish. Soc. 110:111-120.

,Eldridge, H.B., J.A. Wipple and H.J. Bowers. 1982.striped bass, Morone saxatilis, embryos and·larvae.

Bioenergetics and growth ofFish. Bull. 80:461-474.

•

Goodyear, C.P. 1985. Relationship between reported commercial landings andabundance of young striped bass in Chesapeake Bay, Haryland. 'frans. Am. Fish.Soc. 114: 92-96.

Goodyear, C.P~ 1989. Spawning stock biomass per recruit: the biological basisfor a fisheries management tool~ ICCAT working Document~ SCRS/89/82. 10 pp.

Govoni, J.J., G.W. Boehlert and Y. Watanabe. 1986. The physiology of digestionin fish larvae. Envir. Biol. Fish. 16:59-77.

Guma, S.A. 1982. Retinal development and ·retinomotor responses in perch, Percafluviatilis L. J. Fish~ Biol. 20:611-618 •

Hewett, S.W. 'and B.L. Johnson. 1987. A generalized bioeriergetics model of fishgrowth for microcomputers.WIS-SG-87-245, University of wisconsin Sea GrantInstitute, Hadison, WIe

Hjorth, D.A. 1988. Feeding selection of larval striped bass and white perch inthe Peekskill region cf the Hudson River. pp. 134-147, In:.Smith, C.L. (Ed.),Fisheries research in the Hudson River. State universitY-of New York Press.

Hollis, E.H. 1967. An investigation of striped bass in Maryland.Fish Restoration~ F-3-R. 56 pp~

Federal Aid in

Houde, E.D. 1987~

Fish. Soc. Sympos.Fish early li fe dynamics and recruitment variability.2:17-29.

Am.

Houde, E.D. 1989. Comparative growth, mortality and eriergetics of marine fishlarvae: temperature and implied latitudinal effects. Fish. Bull., U.S. 87:471­49~.

11

Houde, E.D. 1990. Temperature-dependent and size-dependent variability in vitalrates of marine fish larvae. leES, C.H. 1990/L:3. 1~ pp.

Houde, E.D, E.J. chesney, Jr., R.M. Nyman and E.S. Rutherford. 1988. Mortality,growth and growth rate variability of striped bäss"larvae in ChesapeakesUbestuaries. Interim. Rept; to Maryland Dept. Natural Resourees. Ref. No.[UHCEES]CBL 88-96.

Houde, E~D;, E.S. Rutherford, R.H. Nyman, E.J. Chesney, T.A. Newberger, D~H.Monteleone and L.G. Marin. 1990. Egg production and larval dynamics,of stripedbass and white pereh in the Potomae River and upper Chesapeake Bay. Rept. to HDDept. of Natur. Resourees, Contraet No. FI45-88-008.

Houde, E.D., R.H. Nyman and E.S. Rutherford. 1989. Striped bass egg productionand spawning biomass, Potomae River, 1987. Emergeney Striped Bass ProgramWorkshop, Hystic~ CT. 5 pp.

Hughes, R~N. 1980.Mar. Biol. Ann. Rev.

Optimal foraging theory in the marine eontext~

18:423-481.Oceanogr. •

'"

Johnson~ R.K. and T.S.V. Koo. 1975. Produetion and distribution of striped bass(Marone saxatilis) eggs in the Chesapeake and Delaware Canal. Ches. Sei. 16:39­55.

Krebs, J.R; 1978. Optimal foraging: Deeision rules for predators. pp. 23-63,In: Krebs, J.R. and N;B. Davies (Eds.),'Behavioural eeology: An evolutionaryapproach. Sinauer Associates, Sunderland, HA.

Logan, D.T. 1985~ Environmental variation and striped bass population dynamies:a size-dependent mortalit~ model. Estuaiies~ 8(1)~28-38.

~

HacKenzie, B.R., W.C. Leggett and R.H. Peters. 1990. Estimating larval fishingestion rates: Can laboratory derived values be reliably extrapolated to thewild? Mar. Ecol; Prog. Sero 67:209-225.

Honteleone, D.M. and E.D. Houde. 1990. Influence of maternal size on survivaland growth of striped bass, Marone saxatilis, eggs and larvae. J. Exp. Mar. Biol.Ecol. 140:1-11.

Moore, C.H. 1988. Food habits~ population dynamies, and bioenergetics of fourpredatorY,fish species in Smith Mountain Lake, Virginia. Ph.D. Dissertation,Virginia Polytechnical Institute and State University, Blacksbur~, VA.

•Morgan, R.P. 11, V.J. Rasin, Jr. and R.L; Coop. 1981.effects on development of striped bass eggs and larvae.110:95-99.

Temperature and saiinityTrans. Am. Fish. SOC.

Olney, J.E., J.D. Field and J.C. HcGovern. In press. Striped bass egg mortality,production and female biomass in Virginia Rivers, 1980-1989. Trans. Am."Fish.Soc.

Plante, C. and J.A. Downing. 1989. Production of freshwater invertebratepopulations in lakes. Can. J. Fish. Aquat. Sei. 46:1489-1498.

12

..

•

\.' ::.. ~ <.. f .

Rog~rs, B~A. and D.T. Westin. i979. The combined effect cf temperature anddelayed initial feeding on the,survival and growth of larval striped bass (Maronesaxatilis Walbaum). pp~ 234-250, In: Jacoff, F.S~ (ed;), Advances in MarineResearch. EPA-600/9-79~035.

Rogers~ B.A. and D.T. Westin~: '1981~ Laboratory studies on effects of temperatureand delayed initial feedirig on development of striped bass larvae. Trans~ Am.Fish. Soc. 110-100-110.

Rose, K.A. and J.H: Cowari, Jr. Individual-based model of young-of-the-yearstriped bass. I. Model description and baseline simulations. Submitted to Trans.Am. Fish. Soc.

Secor~ D.H. 1990. The early,life history of natural and hatchery-producedstriped bass. Ph:D. thesis, University of South Carolina •. 210 pp.

Setzler-riamiiton, E.~., W.R. Boynton,K.V~WoodandH.H. Zion., 1980. Synopsis ofbiological datei. on striped bass, Morone saxatilis (Walhaum). NOAA Tech. Rept.NMFS Circular 433, FAO Synopsis No. 121.~~

Setzler-Hamilton, E.M., W.A. Boynton"J.A. MihurskY, T.T. Polgar arid K.V. ,Wood.1981. Spatial and temporal distribution of st~iped basi eggs, ,l~r~ae, andjuveriiles in the Potomac estuary~ Trans. Am. Fish. Soc•. 110:121-136.

Stevens, D.E. 1988. Striped bass egg and larva moriitoring, and effects of flowregulation on the larval striped bass food chairi" in the Sacramento-Sari JoaquinEstuary. Final Rept., State Water Resour. Control Bd., CA Dept. Fish Game,Stockton, CA. 120 pp.

Tuneer, H. 1988. Growth, survival, arid energetics of larval and juvenile stripedbass (Marone saxatilis) and its white bass hybrid (M. saxatilis X M. chrysops).M:S. Thesis, Uni~. of Maryland, College Park, MD.

Ulanowicz, R.E. and T.T.,Polgar. ,1980: Influences of anadromous spawningbehavior and optimal environmental coriditions upon striped bass (Marone saxatili~)year-class success. Can~ J. Fish. Aquat. Sei. 37:143-154:

Unive~sity of Rhode Island (URI). 1976. Life stage duratiori studies ori riudsonRiver striped bass, Morone saxatilis, (Waldbimm). Final Report Prepared hy, theApplied Research Group for Consolidated Edison Co. of New York, New York, N.Y.

Uphoff, J.J., Jr. 1989~ Environmental effects on survival of eggs, larvae aridjuveniles of striped bass in the Choptank River, Maryland. Traris. Am. Fish. Soc.118:251-263.

Van Winkle, W., B.L. Kirk arid ~.w. Rust. 1979. Periodicities in Atlantic coaststriped bass (Morone saxatilis) commerical fisheries data. J. Fish. Res~ Bd. Can.36:54-62.

Ware, D.M. 1975~, Relation between egg size~ growtli, and natural mortality oflarval fish. J. Fish. Res. Board Can. 32:2503-2512.

Westin~ D.T. ~nd B.i. ~o~ers. ,,1978 •. SYnopsis.of biologicaldata on striped,bass,Morone saxatilis' (Waldbaum), 1972: Marine Technical Report 67, Univcrsity ofRhode Island, Kingston; R.I.

13

Zastrow, C.E. and E.D. Houde. 1989. Maturity, spawning and fecundity of bayanchovy (Anchoa mitchilli) in mid-Chesapeake Bay. Chapter 4 I~: Houde, E.D., E.J.Chesney, T.A. Newberger, A.V. Vazquez, C.E. Zastrow, L.G. Morin, H.R. Harvey andJ.W. Gooch, Population biology of bay anchovy in mid-Chesapeake Bay. Final Rept.to MD Sea Grant. Ref. No. [UMCEES]CBL 89-141. 211 pp.

Zastrow, C.E., E.D. Houde and E.H. Saunders. 1989. Quality of striped bass(Morone saxatilis) eggs in relation to river source and female weight. Rapp.P.-V. Reun. Cons. int. Explor. Mer 191:34-42.

TABLE 1. Summary of results of individual-based model simulations of striped bassspawning and larva growth and mortality in the Potomac River, MD. Baselinespawners have a modal length of 650 mm compared to 1000 mm·SL for large females.

. . .. '."

'siiR;jÜti~n:::::.Ke~·:-D~y::: ::*~:;\~i':::: ... )':r~:c~iÄn)f:' . Me:ap.:> .'::. Mcrrt.~hty .... . :'. ::crf ::spa".. :n..::... :::::. :: ..:Ep.:~.r"·~'S' u::.·c::~.:.>.··.·.:.':: >Lat)ia~::: ::::>::':: :.. -:.:st:ag:e ·:::Rat~·.(d~l)

. . . '.' \;I.U l".U .. ::Siu;~v.;ivi,:ng:::l::o· .. : lluratior{ ... .. ':.: :-Ket:aiOOr:Iih~s:is'. .' .'. >

BaselineandTemperatureRule

Large andTemperatureRule

Baselineand Random

Large andRandom

109.72

104.60

109.48

109.48

4.96 X 10 7

1.60 X 108

4.96 X 107

1.60 X 10 8

14

0.011~

0.0418

0.0159

0.0530

22.5

18.8

21. 9

17.9

0.198

0.169

0.189

0.164

80 A. BASELINE FEMALES

E- 60Z~

U 400::~Q., 20

0500 600 700 800 900 1000 1100 1200 1300

•80 B. LARGE FEMALES

E- 60Z~

U 400::~Q., 20

0500 600 700 800 900 1000 1100 1200 1300

LENGTH (rnrn)

Figure 1. Length-frequency distributions of female spawners used in individual­based model simulations to evaluate the effect of maternal contribution onrecruitment dynamics of striped bass. Modal length of baseline females was 600 mmSL compared to 1000 mm SL for large spawners.

STRIPED BASS23 temperature mIes for timing of spawning-U

0 21--~~ 19 -;;JE- ~tODE

< 16 • • • 0 •~ -~ 0Q., 0

00

~ 13 ).tI~

~ 0E- 0

10400 500 600 700 800 900 1000 1100 1200

LENGTH (mrn)

F'igure 2. Summary of Potomac River, MD data used to generate the rule relatingsize of female spawners to temperature in simulations employing an individual­based model of striped bass life history.

•~

•.

80A. BASELINE FEMALES

CALENDER DAYS 4 APRIL TO 1 MAY

--------------------- --,

E- 60z~ • BASELINE +TEMPU 40 0 BASEUNE + RANDOM~~Q., 20

0

80

E- 60z~

u 40~~

Q., 20

o

100102105106108110112114116118120122124126

B. LARGE FEMALES

• LARGE + TEMP

o LARGE + RANDOM

100102105106108110112114116118120122124126

SPAWN DAY MIDPOINT

Figure 3. Spawn day-frequency of baseline and large striped bass females inindividual-based mod~l simulations with and without a temperature rule relatingfemale size to day of spawning.

A. Simulation 1 Sur\'ivors SOSO Day 135

f-o 40 40z

30 30;..;B..~lio~ Female.

~Cl:: 20 T~mperaton Ru~ 00 20..,;c. 10 10

0 04 5 6 7 8 9 101112 D 141516171819 20

B. Simulation 2· Survi,,·orsDay 13'S

Ba.~lio~ F~male.T~mperatun Rulr Off

4 5 6 7 8 9 1011121314151617181920

•.

c. Simulation 3 Sur\'ivors 50 D. Simulation 4 Survi\'ors50 Day 135 Day 135

~~

f-o7- La'1!r F~male. 30 La'1!e Female.;..; 30 Spawoioll Rulr 00 Tem....raturr Rulr Off~ 20Cl:: 20~

=- 10 10

0 04 5 6 7 8 9 1011121314151617181920 4 5 6 7 8 9 1011121314151617181920

LENGTH (mm) LE:\GTH (mm) eFigure 4. Length-frequency distributions of larvae alive on day 135 inindividual-based model simulations of striped bass in the Potomac River, MD. Sizeof female spawners (baseline and large) and spawn day were altered in thesimulations.