Limburg et al : Growth, mortality, and recruitment of larval Morone spp 



89 



In the Chesapeake system, temperature can 

 play a key role in setting year-class strength 

 for striped bass (Rutherford and Houde, 1995; 

 Secor and Houde, 1995), although it is not al- 

 ways the determining factor (e.g. Kellogg et 

 al."^). Complete failures of individual cohorts 

 were associated with rapid temperature drops 

 down to 12°C (Rutherford and Houde, 1995). 

 Dey (1981) hypothesized that early cohorts of 

 Hudson River striped bass were eliminated by 

 a sudden drop in water temperatures in late 

 May 1976, and simulation analysis by Boreman 

 (1983) supported this inference. 



Nevertheless, there was no relationship be- 

 tween temperature and year-class strength for 

 Hudson River moronids over the period 1974- 

 90 (Pace et al., 1993). Examining a 40-yr record 

 of Hudson River temperatures, we found that 

 the likelihood of lethal low temperature events 

 (defined as <12°C) declines rapidly during the 

 latter half of April (Fig. 5). Early spawned co- 

 horts of both moronid species have a risk for 

 low-temperature mortality events, but most co- 

 horts are spawned after early May and are 

 highly unlikely to experience this direct source 

 of mortality (Fig. 5). 



We have demonstrated that, like Chesapeake 

 striped bass growth rates, larval moronid 

 growth rates are linked (albeit weakly) to tem- 

 perature in the Hudson River Nevertheless, 

 differential mortality occurred on the faster- 

 growing, late cohorts of striped bass. We postu- 

 late that this mortality was due to predation. 

 Secor and Houde (1995) also speculated that 

 increased mortality rates observed in late 

 Patuxent cohorts was due to predation. Hudson 

 River moronid larvae that co-occurred with the 

 zooplankton blooms were able to benefit ener- 

 getically from increased food availability in re- 

 lation to prebloom conditions (Limburg et al., 

 1997). Consumption rates continued to be high after 

 the bloom, however, so that energetically the larvae 

 continued to do well. Warmer temperatures after the 

 bloom would have enabled larvae to swim faster and 

 encounter more prey, but they would also potentially 

 have encountered more predators. Thus, we infer that 

 late cohorts suffered differentially greater predation. 



We suggest that larger estuarine nurseries, such 

 as the Hudson River, provide more damping of physi- 

 cal fluctuations in characteristics such as water tem- 

 perature than do smaller estuaries with "flashier" 

 drainage regimes. If this is the case, we might ex- 

 pect Chesapeake Bay moronid stocks to be more at 

 risk for episodic, density-independent mortality 

 events of the type described in Rutherford and Houde 



14 

 12 

 10 

 8 

 6 

 4 

 2 

 



Thermal Risk- 



-D- SB. eggs 



-O- SB.ysl 

 -A- SB.pysl 



\ b ^ 



U-U-U-U'U 



1.0 

 0.8 

 06 

 0.4 

 0.2 

 0.0 



Maris Apr 7 Apr 27 May 17 Jun 6 Jun 26 Jul 16 Aug 5 



a 



c 



14 



12-1 



10 



8- 



6- 



4 



2 



0-1 



-2- 



Maria 



Pj.Q ThermalRisk- 



□ 'b-&'0;A-A-A^ -D-WP.eggs 



/O / O-S^ A _o_vVP.ysl 



/ A / ^ ''• 



p o 



;^-A- WPpysl 



1.0 

 0.8 

 0.6 

 04 

 0.2 

 00 



o 



a 



(D 



3 



ro 



o 



o 



Apr 7 Apr 27 May 17 Jun 6 Jun 26 Jul 16 Aug 5 



Week 



Figure 5 



Probability of water temperature occurring at or below 12°C in 

 Hudson River (based on 40-year database: Poughkeepsie Water 

 Works. Poughkeepsie, NY) compared with means of log-trans- 

 formed, river-wide abundances (1000s) of early life stages of (A) 

 striped bass and (B) white perch. Squares = eggs, circles = yolksac 

 larvae, and triangles = postyolksac larvae. 



(1995). Klauda et al. (1980) noted that striped bass 

 year-class strength, indexed by juvenile abundance, 

 varied about 34-fold in the Hudson River, about 47- 

 fold in Chesapeake Bay, and 160-fold in the Roanoke 

 River (where upstream water temperatures are con- 

 trolled by water regulation). The different lengths of 

 the data sets, as well as the fact that Chesapeake 

 Bay juvenile index reflects a mix offish derived from 

 tributaries and the Bay proper, make this compari- 

 son somewhat problematic to interpret. What is likely 

 is that increased physical variability in smaller estu- 

 aries translate into greater recruitment variability. 



Further, evidence from smaller streams suggests 

 that recruitment improves in years with greater 

 spring runoff (McGovern and Olney, 1996; Kellogg 



