FISHERY BULLETIN: VOL. 75, NO. 1 

 TABLE 4. — Stepwise regression of survival index of Atlantic menhaden on environmental factors. 



TABLE 5. — Regression coefficients between independent en- 

 vironmental variables used in the recruit-environment predic- 

 tive equation for Atlantic menhaden. See description of X's in 

 Table 4. 



Wind-driven transport off Delaware Bay was 

 studied as being representative of menhaden 

 spawning areas in the Middle Atlantic Bight. 

 Because the transport values are produced in a 3° 

 grid by the Pacific Environmental Group, there 

 were no available data for a point located centrally 

 on the continental shelf. Two locations were 

 chosen: one at lat. 39°N, long. 75°W, near the 

 mouth of Delaware Bay, the other at lat. 39°N, 

 long. 72°W, near the outer edge of the continental 

 shelf. The two locations are approximately 260 km 

 apart in an east-west direction, and are felt to be 

 representative of Ekman transport over the broad 

 shelf area near the east-west axis of the Middle 

 Atlantic Bight. 



The entrance of larvae into estuaries of the 

 Middle Atlantic Bight occurs variably from 

 September to June, with peak immigration oc- 

 curring in the winter. Reintjes and Pacheco ( 1966) 

 reported on 6 yr of larval collection at Indian 

 River, Del., and showed high rates of influx from 

 December through March. The peak month varied 

 from year to year, but stayed within the 

 December-March period. Correlation coefficients 

 between summed transport values for 

 November-February (the peak period of larval 

 drift) and the survival index (Table 4) are not as 

 large as those from south of Cape Hatteras, but the 

 effect of transport on survival at the inshore point 

 (lat. 39°N, long. 75°W) is significant at the 0.05 

 level. The transport values from the inshore and 



offshore points account for approximately 27% and 

 12%, respectively, of the total variance in the 

 survival index for Atlantic menhaden. When 

 combined with the transports south of Cape 

 Hatteras, these values for the Middle Atlantic 

 Bight account for an additional 12+% of the re- 

 sidual variance. Correlation coefficients are lower 

 than those found for the South Atlantic Bight, and 

 may be indicative of: 1) major nearshore spawning 

 activities, reducing the need for a suitable 

 transport mechanism; 2) a lower level of spawning 

 in the area; or 3) a lower level of recruits per 

 spawner due to mortalities from other en- 

 vironmental factors in the area. 



The model of circulation off Chesapeake Bay 

 developed by Harrison et al. (1967) and discussed 

 in the Larval Transport section would be ap- 

 propriate if larval menhaden were demersal in 

 nature. However, since larvae are more abundant 

 in the upper water column, we would expect a 

 negative relationship between discharge and 

 survival in the Middle Atlantic Bight because 

 high surface discharge would impede larval 

 entrance into estuaries. Chesapeake Bay was 

 chosen to test that hypothesis because of its im- 

 portance as a major nursery area. Average 

 monthly discharge rates from the Susquehanna, 

 Potomac, and James rivers were used in the test 

 because they constitute over 90% of the total 

 inflow into Chesapeake Bay. Discharge during the 

 third quarter (July-September) of the year pre- 

 ceding the year-class year was chosen because 

 there is a lag time of up to 90 days between stream 

 flow and bay discharge (Harrison et al. 1967). The 

 influence from run-off would be felt at the mouth of 

 the Bay in the October-December period when 

 larvae begin entering in increasing abundance. A 

 correlation between the survival index and 

 discharge rate did not result in a significant 



32 



