HIROTA: NATURAL HISTORY OF PLEUROBRACHIA BACHEl IN LA JOLLA BIGHT 



coastal waters off southern California is the po- 

 tential effect on the horizontal, offshore distribu- 

 tion of a species. Stevenson (IGSS)"* summarized 

 some wind data for the Newport Beach to Los 

 Angeles Harbor area, which show that in the sea 

 breeze-land breeze diel variation of wind velocity 

 there is a stronger sea breeze component from the 

 west-northwest quadrant during the afternoons 

 and a more variable and weak wind in the morn- 

 ings. Robert Arthur has suggested that a possible 

 result of the diel variation in wind velocity and the 

 observed pattern of vertical migration in P. 

 hachei may in part account for the maintenance of 

 high ctenophore abundances close to shore. By 

 living in the wind-mixed layer during the day 

 with a westerly-northwesterly sea breeze, 

 ctenophores are moved shoreward. At night 

 through the early morning hours the weaker land 

 breeze moves the surface waters offshore, but by 

 living deeper at night the net offshore movement 

 of ctenophores should be relatively smaller than 

 the shoreward displacement. The frequent strand- 

 ing or grounding of drogues nearshore suggests a 

 net onshore movement of water. One mechanism 

 of horizontal, seaward transport is the rip cur- 

 rents, but these are probably important only a few 

 hundreds of meters seaward of the surf zone and 

 are dependent on the size of sea swells. It is not 

 known how important stranding is as a source of 

 mortality to Pleurobrachia, but in summer 

 months Pelagia (Scyphozoa) are frequently 

 stranded on La Jolla beaches and are broken apart 

 in the surf zone. Other macrozooplankton, such as 

 salps and Velella, periodically occur on the beach 

 and in waters near the shore. It is not clear what 

 cues or mechanisms the ctenophores use to main- 

 tain their distribution to within 1 km of shore 

 without most being washed into the surf and 

 killed. A number of net tows taken near the end of 

 Scripps Institution pier and just seaward of the 

 surf zone indicate absence of P. bachei. 



The estimates of abundance of P. bachei at fixed 

 stations located alongshore 2-3 km from the 

 shoreline are subject to variability in time and 

 space from several causes. At a single station the 

 abundance will be affected by: 1) spatial 

 heterogeneity and patchiness on the scale of 

 100-m horizontal distance and 20- to 50-m depth 

 over the course sampled during a tow, 2) the stage 



••Stevenson. R. E. 1958. An investigation of nearshore ocean 

 currents at Newport Beach, California. UnpubL Rep. to Orange 

 Cty. Sanit. Dist., 108 p. 



of the tides and the tidal current velocity (see 

 Figure 15), and 3) the water temperature 

 stratification and near-surface drift due to the 

 wind. The magnitude of replicate sample error is 

 one-half to twice the mean, and the variability in 

 abundance due to the presumed horizontal motion 

 generated by the tides and wind for the annual 

 average is about the same as replicate sample 

 error. 



For estimates of abundance on a given sampling 

 date at stations 1.6 km from the shore at different 

 locations along the coast, "true spatial" variabil- 

 ity exists in addition to replicate sampling error 

 and aliasing due to physical effects of tides and the 

 wind. It is difficult to sort out quantitatively the 

 separate error components due to physical effects 

 and true spatial effects alone, because the time 

 period for the physical effects to bias sampling 

 (about 6 h) is about the same as that required to 

 move through space and sample different stations. 

 Variability around the mean of all stations at one 

 sampling date includes variations due to replicate 

 sampling error, variations due to physical effects, 

 and variations due to true spatial differences. The 

 relative magnitudes of these components of varia- 

 tion estimated from the 95% confidence limits of 

 the two-way analysis of variance and the regres- 

 sion of abundance on tidal height are: 1) the 95% 

 confidence limits about the mean of all stations at 

 a given time of sampling is the mean multiplied 

 and divided by 6.23, 2) the 95% confidence limits 

 about the mean of replicate samples is the mean 

 multiplied and divided by 2.15, 3) the range of the 

 expected abundance from the regression equation 

 over the observed values of tidal heights is four- 

 fold, or a range of about one-half to twice the 

 overall annual mean, and 4) the residual true spa- 

 tial variation calculated by difference is the mean 

 multiplied and divided by 1.45 (i.e., 6.232 = 2.15^ 

 X 2^ X 1.45^). In terms of the relative contribution 

 of these three components to the total variability, 

 the values are 2.2:1.9:1 for replicate sampling 

 error, physical effects, and true spatial variation, 

 respectively. The relative contribution of replicate 

 sampling error vs. physical plus true spatial vari- 

 ations to the total variability of all stations on one 

 sampling date is 1:1.8. These results from a sam- 

 pling program not designed specifically to sepa- 

 rate each effect suggest that physical effects on 

 sampling bias and the replicate sample error are 

 important relative to real spatial differences of 

 abundance between stations equidistant from 

 shore. A synoptic sampling program with two or 



321 



