CYCLIC COVARIATION IN THE CALIFORNIA KING SALMON, 



ONCORHYNCHUS TSHAWYTSCHA, SILVER SALMON, 0. KISUTCH, 



AND DUNGENESS CRAB, CANCER MAGISTER, FISHERIES 



Louis W. Botsford, 1 Richard D. Methot, Jr., 2 and James E. Wilen 3 



ABSTRACT 



There are apparent cyclic fluctuations in the catch record of both northern and central California 

 salmon fisheries. They are of the same period and strength as well-known cycles in crab catch but of 

 different phase. Statistical characteristics of this covariation, as reflected in estimates of auto- and 

 cross-correlation functions, change following the decline of the central California Dungeness crab 

 fishery. Analysis of a likely cause of this phenomenon, a greater delay in switching of effort from 

 crab to salmon during years of high crab catch, indicates that this mechanism is not present. Phase 

 differences between salmon and crab cycles imply constraints on remaining potential causes, but a 

 cause of the cyclic covariation has not been established. 



Regular patterns in fishery catch records reflect 

 underlying mechanisms that can provide the 

 basis for broader understanding, better predic- 

 tion, and consequently better management of the 

 fishery. Cyclic fluctuations in the northern Cali- 

 fornia Dungeness crab catch have been a topic of 

 research for the past 10 yr. We document here 

 cyclic fluctuations in the northern California 

 salmon catch (Fig. 1) of the same frequency but 

 different phase. 



Coastwide fluctuations in Dungeness crab 

 catch were originally attributed tooceanograph- 

 ic causes (Anonymous 1965). Peterson (1973) 

 demonstrated a statistical relationship between 

 coastal upwelling and crab catch. Botsford and 

 Wickham (1975) concluded from estimates of the 

 appropriate cross- and auto-correlation func- 

 tions that, while crab catch was indeed cyclic and 

 upwelling was correlated with crab catch after a 

 lag of 1 or 2 yr, upwelling itself was not cyclic, 

 hence was not the source of the cycles. Botsford 

 and Wickham (1978) later showed that interage, 

 density-dependent mortality could be the cause 

 of the observed cycles, and derived new stability 

 results that indicated size-selective fishing could 

 decrease population stability, and thereby in- 



'Department of Wildlife and Fisheries Biology. University 

 of California. Davis, CA 95616. 



2 Bodega Marine Laboratory. P.O. Box 247. Bodega Bay, 

 Calif.; present address: Southwest Fisheries Center La Jolla 

 Laboratory, National Marine Fisheries Service, NOAA, P.O. 

 Box 271, La Jolla. CA 92038. 



3 Division of Environmental Studies, University of Califor- 

 nia, Davis. CA 95616. 



crease the propensity for cyclic fluctuations. 

 They cited two known potential interage, den- 

 sity-dependent mechanisms, cannibalism and an 

 egg-predator worm, and are conducting field 

 samples of these. On the basis of a disparity be- 

 tween the period of observed cycles and the peri- 

 od of cycles produced by a model that included 

 cannibalism, McKelvey et al. (1980) claimed that 

 cannibalism could not cause the cycles. Botsford 

 (1981) pointed out that the observed disparity 

 was not new, but had been noted by Botsford and 

 Wickham (1978), and critically analyzed the rea- 

 soning used by McKelvey etal. (1980) in drawing 

 a new conclusion. In summary, the cause of 

 cycles in the northern California Dungeness crab 



Dungeness crob 



Totol salmon 



Silver salmon 



3 O 

 x 



X 



2 o 



< 



< 

 in 



80 



Manuscript accepted April 1982. 



FISHERY BULLETIN: VOL. 80, NO. 4. 1982. 



Figure 1.— Total landings (kg) in the northern California crab 

 and salmon fisheries for the years 1940-76. 



791 



