Lauth et al,: Timing and duration of mating and brooding periods of Pleurogrommus monopterygius 



567 



a. c 



^ > 



o QJ 



>. CD 



3 S 

 O" 03 



a, sz 



that some males brood eggs at nesting 

 sites until January. 



There may be several reasons for the 

 2-month difference in the duration of 

 the reproductive cycle between Alaska 

 (7+ months) and the western Pacific 

 Ocean (5+ months; Gorbunova, 1962; 

 Zolotov, 1993). First, all the western 

 Pacific Ocean nesting sites were lo- 

 cated in coastal areas in water depths 

 <32 m where water temperatures were 

 likely higher and the length of the 

 brooding period shorter than those in 

 Alaskan waters. Second, no distinction 

 was made between the onset of the 

 nesting and spawning periods; that is, 

 Gorbunova (1962) and Zolotov (1993) 

 assumed both periods commenced in 

 June. In Alaska, males aggregate 

 and prepare nesting territories 1-2 

 months before spawning. Without a 

 direct means for observing males at 

 a nesting site with such tools as ar- 

 chival tags and time-lapse cameras, it 

 is difficult to determine when males 

 first arrive. Assuming a 1-month nest 

 establishment period during May and 

 longer incubation times for the western Pacific Ocean 

 population, the duration of the reproductive cycle for 

 Atka mackerel is similar to the duration of the cycle 

 for other parts of the North Pacific Ocean. 



The timing of the Atka mackerel reproductive cycle 

 is unusual because fishes in temperate and subarc- 

 tic waters typically synchronize the hatching time of 

 larvae with periods of high zooplankton abundance in 

 the spring or early summer (Gushing, 1990; Haldor- 

 son et al., 1993). The possible adaptive significance of 

 hatching during the late fall and early winter may be 

 predator avoidance, dispersal, or availability of specific 

 prey. Atka mackerel larvae are neustonic after hatch- 

 ing (Kendall and Dunn, 1985) and have large mouths 

 (Gorbunova, 1962) capable of feeding on larger plank- 

 tonic prey. 



The temporal, geographic, and bathymetric distribu- 

 tion of sampling for this study was relatively limited 

 given the temporally and spatially complex marine envi- 

 ronment that Atka mackerel inhabit (Ladd et al., 2005; 

 Ohshima et al., 2005). Large-scale climate changes to 

 the ecosystem (Rodionov et al., 2005) can cause altera- 

 tions in recruitment dynamics that may affect popula- 

 tion structure (Bailey, 2000; Ciannelli et al., 2005) 

 and ultimately change the timing of the reproductive 

 cycle (Hutchings and Myers, 1994; Wieland et al., 2000; 

 Ojaveer and Kalejs, 2005). We did not investigate how 

 water temperature or other environmental factors vary 

 over larger scales, or how such variability influences the 

 timing of the Atka mackerel's reproductive cycle. This 

 study does, however, provide a framework for process- 

 oriented investigations of stock dynamics, recruitment, 

 and distribution of Atka mackerel populations. Time of 



0,25 

 0.20 

 0.15 

 0.10 



B " 0.05 



ir 



0.00 



Jun Jul Aug Sep Oct Nov Dec Jan Feb 



Montfi 



Figure 6 



Plot of incubation model results for Atka mackerel (Pleurogrammus monop- 

 terygius) egg masses collected from Amchitka, Tanaga, and Seguam islands 

 combined. The y-axis is the relative frequency of egg mass spawning (solid 

 line) and hatching events (dashed line) by week for pooled low and high 

 average daily water temperatures. The horizontal line at top represents 

 the duration of the mating and brooding cycle and identifies the peri- 

 ods of nest establishment (dotted line), spawning (solid black line), and 

 hatching (dashed line). 



hatching, geographic location, and hydrographic fea- 

 tures can all affect the dispersal of larvae and play a 

 significant role, along with biological processes, in de- 

 termining recruitment success and population structure 

 (Napp et al., 2000). 



For unknown reasons, nest fidelity varied among 

 the tagged male Atka mackerel. Given a late July 

 onset of spawning and relatively low ambient wa- 

 ter temperatures at nesting sites, it is unlikely that 

 males are able to successfully tend, from start to end, 

 more than one brood of egg masses per season. Nests 

 are commonly composed of eggs from multiple females 

 (Lauth et al., in press) deposited over the 3-month 

 spawning period and by the time the first batch of 

 eggs hatches, the spawning period is nearing comple- 

 tion or has ended. 



It appears that none of the four archival tags con- 

 tained data from a nesting male for the duration of 

 one complete mating and brooding cycle. All four males 

 were captured, tagged, and released while the breed- 

 ing season was underway in late July. The first male 

 (Fig. 2A) did not exhibit nesting behavior until June 

 of the following year. The other three males (Fig. 2, 

 B-D) resumed nesting behavior shortly after being re- 

 leased; and their behavior may indicate that they were 

 defending nesting territories before they were captured 

 and tagged. Two of the males ended nesting behavior 

 after one month, and the third appeared to abandon 

 one nesting territory and establish a second when it 

 abruptly started exhibiting nesting behavior at a shal- 

 lower depth on 14 September 2000 (Fig. 2B). It appears 

 that the male that changed depths may have estab- 

 lished a new nest, spawned, and successfully brooded 



