Abstract: I monitored recruitment of the barnacles Balanus glandula and Chthamalus spp. to the intertidal zone at three sites in central California. The sampling period was April 27th-May 28th 1998, during the strong 1997-8 El Nino. The results showed that one large pulse of recruitment occurred at Waddell Beach during a relaxation of coastal upwelling in April. Two smaller recruitment events occurred at Asilomar beach during two periods of relaxation in May. There was a significant correlation of recruitment with sea surface temperature at two moored buoys (r’=.593, r’=.604) and salinity (r’=-593). Buoy observations of north wind velocity and sea surface temperature showed that the two periods of reduced upwelling in May totaled nine days. Only five months during April through June 1988-1997 had nine or more days without upwelling-favorable north winds. This suggests that one effect of the El Nino on the central California coast may have been to decrease upwelling favorable winds. Recruitment of barnacle larvae is thought to occur during periods of decreased northerly winds and the subsequent relaxation of upwelling. However, I could not conclusively say that recruitment at Waddell Beach or Pescadero Beach was significantly greater in May 1998 than in April-June 1996-7. Introduction Along the central coast of California, the dominant animals of the upper intertidal zone are the acorn barnacles Balanus glandula, Chthamalus dalli, and Chthamalus fissus. B. glandula and Chthamalus spp. are cross-fertilizing hermaphrodites (Light 1975). Adults brood nauplii larvae within their shell, and then release the nauplii into the top meter of the water column. Äfter about 14 days, the nauplii metamorphose into cyprid larvae. Äfter another 14 day period, cyprids settle onto rocky substrate and metamorphose into adults. Primarily because of mortality suffered during the pelagic larval phases, recruitment of larvae to the shore is highly variable. Recruitment is influenced both by biotic factors (Gaines and Roughgarden 1987) and oceanographic factors such as surface currents driven by coastal upwelling (Roughgarden et al 1988, Roughgarden et al. 1991, Shkedy and Roughgarden 1997, Farrell et al. 1991 ) . Upwelling along the central California coast is particularly apparent in the spring, when strong northerly winds drive surface water directly offshore by Ekman transport (Gross 1987). In typical years, cold, saline upwelled water flows from below the thermocline to replace the warmer water pushed offshore by Ekman transport. Rosenfeld et al (1994) found the frontal boundary between cold upwelled water and warmer oceanic water to be located 15-20 km off Santa Cruz during a period of upwelling in May 1989 Periods of decreased north wind cause reduced Ekman transport. This decreases the force trapping coastal water against the offshore front. During this relaxation of north wind, warm oceanic water moves rapidly towards shore. At the shore, cold saline upwelled water is replaced by warmer less saline oceanic or California Current water. During one of these relaxation events in May 1989, Rosenfeld et al. (1994) found sea surface temperatures to increase above 13.5° and salinity to decrease to 33.4 psu. Rosenfeld et al. (1994) used several methods to detect relaxation, including moored offshore buoys and shore stations with sensors for sea surface temperature and salinity. Advanced Very High Resolution Radiometry (AVHRR) satellite imaging of sea surface temperature images was also used by Rosenfeld et al (1994) and Miller (1992) to track the shoreward movement of warm oceanic water during these periods of relaxation. Farrell (1991) and Miller (1992) measured the north component of wind velocity, assuming that a decrease in north wind would lead to decreased Ekman transport and a resulting relaxation event. Miller (1992) and others (Roughgarden et al. 1988, Farrell et al. 1991, Roughgarden et al. 1991) hypothesized that during periods of upwelling barnacle larvae are carried offshore by surface currents. Larvae collect at the frontal boundary between cold upwelled water and warmer oceanic or California current water. During relaxation of upwelling, larvae are transported to shore and recruit into adult populations. Shkedy and Roughgarden (1997) found a close correlation between sea surface temperature and recruitment. A polynomial regression of sea surface temperature explained 47.48% of the variance in recruitment. The effects on recruitment of an El Ninno such as the one seen in 1997-1998 have not been determined. Shkedy and Roughgarden (1997) found that overall recruitment in the 1992 El Ninno was not only lower than predicted by SST, but also lower than in the non- El Ninno years 1989 and 1990. However, during the 1982-83 El Ninno, Roughgarden et al. (1988) found unusually high recruitment to the intertidal zone in a year with low upwelling indices and decreased northerly winds. Most recently, Connolly and Roughgarden (1998) found that recruitment for May-September of 1997 was significantly higher than during the same period in 1996, and that in four of nine study sites on the central California coast recruitment increased by an order of magnitude in 1997 compared to 1996. They suggested that increased recruitment could have been due to three factors: reduced offshore transport of larvae due to a weakening of northerly winds caused by a strengthened Aleutian low pressure system in the North Pacific; an eastward shift in the California current that trapped larvae near shore; or reduced offshore transport of water caused by northward-propagating coastally-trapped waves. May of 1998 offered an opportunity to further investigate the effect of El Ninno on recruitment. The 1997-1998 El Ninno first appeared in the tropics in March 1997 (UCSD CoastWatch, El Ninno Advisory, March 1997). By May sea temperatures were 3-4 degrees above normal off central California. Upwelling in May, August, and November was at least one standard deviation below the historical mean for these months. Sea surface temperature anomalies persisted at least through April 1998 and were predicted to continue until June or July (UCSD CoastWatch, El Ninno Advisory, 1998). Upwelling index was normal at 36°N latitude for January and February of 1998, but .5 standard deviation below the mean for April. Since the El Nino continued through the spring of 1998, 1 was interested in determining its influence on coastal upwelling, and any resulting influence on larval recruitment. There were three questions that I attempted to address. First, 1 wanted to determine whether recruitment during the 1998 El Ninno occurred during relaxations of upwelling, as suggested by Roughgarden et al. (1988) for non- El Nino years. Secondly, I wanted to know whether there was an increase in the frequency or duration of relaxation events in May 1998 compared to typical conditions. Finally, if recruitment occurred during relaxation events, and if there were an increase in relaxation during May 1998 compared to typical years, I wanted to test whether overall recruitment was significantly higher in May 1998 than during non -El Ninno years. Methods and Materials Study Sites I monitored recruitment at three open-coast sites in central California: Pescadero Beach (36° 14'N, 122° 24' W, hereafter Pescadero), Waddell Beach (37° 04 N, 122° 15'W), and Asilomar Beach (36 ° 37'N, 121° 56’, hereafter Asilomar). Pescadero and Waddell Beach are between San Francisco and the northern edge of Monterey Bay, and Asilomar is located just south of Monterey Bay (Fig. 1). These sites have similar intertidal barnacle communities of B. glandula, Chthamalus spp., and Tetraclita squamosa rubescens. All three sites contain communities with high abundances of the mussels Mytilus californianus, and algae species including Endocladia muricata, Mastocarpus papillatus, and Prionitis lanceolata. The three sites receive direct wave impact, but swells and surf at Pescadero and Asilomar typically were higher than at Waddell Beach. Pescadero has sandstone substrate, at Waddell Beach the substrate is siltstone or mudstone, and at Asilomar the substrate is a granite/feldspar mix. Rosenfeld et al. (1994) have shown that Anno Nuevo, located south of Pescadero and north of Waddell Beach, is a region of intense upwelling. AVHRR images of sea surface temperature for two cloud-free days on April 26th and 27th (UCSD Coastwatch 1998) showed cold upwelled water from Änno Nuevo touching the shore at both Pescadero and Waddell Beach. The bulk of this water billowed offshore and southward in a plume 40 km long. Asilomar is not reported to lie within an upwelling plume by Rosenfeld et al (1994). This is supported by satellite sea surface temperature images for the period from April 27th-May 28th 1998 (Figs. 2-4). Miller (1992) noted that there is a general gradient of increasing upwelling from San Francisco to Pt. Sur (south of Asilomar), possibly due to the greater size of the Pt. Sur plume compared to the Ano Nuevo plume. Correlating Recruitment and Relaxation of Upwelling I monitored barnacle recruitment to the high intertidal zone at Pescadero, Waddell Beach, and Asilomar. I sampled recruitment on two day intervals at Asilomar, and every four days at Waddell Beach and Pescadero. Sampling began during the last week of April, 1998 and continued through May 28th, 1998. At each site, I used wall anchors and screws to attach six to seven recruitment plates to the rocks in the upper intertidal zone. Recruitment plates were made of 5x1Ocm 7 mm thick acrylic. Each plate was covered with gray Safety Walk tape (3M Co., St. Paul MN), which was replaced after each sampling interval. I placed the plates between .75 and 2.5m above mean lower low water (MLLW), approximately in the middle of the barnacle-Endocladia zone. Üpper intertidal zone plates were intended for comparison with data from Farrell (1991). I removed recruitment plates from the rocks and scored them for barnacle cyprids and juveniles, using a Wild M5 dissecting scope at 25x magnification. I identified Balanus glandula to the species level, and Chthamalus to the genus level, following the identification methods of Miller (1992), and Connolly (pers. comm.). I attempted to identify periods of relaxation during May 1998 that might co¬ occur with recruitment pulses. To detect relaxation events, l'used surface buoys, shore stations, and AVHRR, similar to the methods used by Rosenfeld et al. (1994), Miller (1992), and Farrell et al. (1991). Sea surface temperature, salinity, and wind velocity data were made available by the National Marine Fisheries Service from the Granite Canyon Fisheries CTD and surface meteorological station (1998) (hereafter GC). Additional wind velocity and sea surface temperature was used from the Monterey Bay Aquarium Research Institute buoy MI (MBARI 1998) (hereafter Ml). I analyzed satellite sea surface temperature (SST) maps for cloud-free days from NOAA 12 and NOAA 14 Advanced Very High Resolution Radiometry (AVHRR), made available by the UCSD Coastwatch Program (1988) and Monterey Bay Aquarium Research Institute (1998). I expected that buoy and shore wind sensors would represent sea and wind conditions at my three study sites. If periods of low north wind stress led to relaxation events, I expected that first the offshore buoy in Monterey Bay (Ml) and then the near shore buoy (GC) would record temperatures above 13.5°C due to oceanic water collapsing back onto the shore. To test for relationships between recruitment and signals of relaxation, 1 ran Pearson correlations of recruitment at each site with SST and salinity at Granite Canyon (REINAS 1998) and SST at buoy MI (REINAS 1998) (Sokal and Rohlf 1995). A cross correlation function was run to determine the time lag of SST and salinity around recruitment that would maximize the correlation coefficient. I analyzed all data using SPSS Systat v. 6.01. Frequency and Duration of Relaxation Events I compared the frequency and duration of relaxation events in May 1998 to relaxations typical of the last 10 years. Mean daily north wind strength from 1988-1997 were available from NÖAA buoy No. 46042, located between Monterey and Santa Cruz (NOAA 1998). These wind data allowed me to define relaxation using a parameter based on north wind strength. Since north winds drive upwelling, I arbitrarily defined a relaxation event (cessation of upwelling) as any period with no north wind component. I used MATLAB Student v.4 (The Mathworks, 1994) to indicate which days had no north wind component during April-June 1988-1997. I processed data from April -June, rather than May only, because Connolly and Roughgarden (1998) showed the delivery of larvae to be highly variable between months at Pescadero and Waddell Beach. Including April and June also allowed me to compare May 1998 to three months with similar upwelling-season oceanography. The pattern of relaxations (defined by a decrease in north wind) in April-June 1988-1997 was qualitatively compared to the relaxation events indicated by wind sensors at MI during May 1998. Wind measurements at MI were taken on approximately 30 minute intervals, and mean daily velocity was not available. I defined a relaxation at MI to be any day(s) during which no wind measurement contained a north wind component. The periods of relaxation indicated by lack of north wind stress at Ml were confirmed with SST data at Ml and GC. Testing for Changes in Monthly Recruitment To test for an increase in recruitment in 1998 over 1996 and 1997 I installed six recruitment plates in the mid intertidal zone at Pescadero and Waddell Beach. The plates, installation methods, and species identification methods were similar to those used for the high intertidal zone (see above). I placed mid intertidal zone plates between .75 and 1.9 m above MLLW, which was within one-half meter of the lower limit of the mussel beds. I installed the plates on April 27th. I collected the Waddell Beach plates on May 28th. 1 collected and replaced the Pescadero plates on May 15th due to partial fouling of the plates by algae. I recovered the second set of plates at Pescadero on May 28th. For both Pescadero and Waddell Beach, I compared May 1998 recruitment in the mid intertidal zone to the recruitment found by Connolly and Roughgarden (1998) in April-June 1996 and 1997. April-June 1996 was during a non-El Nino year, and April-June 1997 was at the beginning of the strong 1997-8 El Nino. The comparison used a Model I single factor ANOVA . Data were log transformed to account for heterogeneity of the variances. Post hoc tests were conducted using Tukey's test with a Bonferonni correction (Sokal and Rohlf, 1995). I used SPSS Systat v. 6.01 for all statistical analysis. Barnacle Reproductive Output To determine whether B. glandula and Chthamalus spp. were reproductive during the study period, I collected between 147 and 223 barnacles from the substrate at each site. In all cases barnacles were collected on at least three separate days from at least three rocks per day. Chthamalus spp. were approximately 3 mm in diameter. B. glandula from Waddell Beach and Asilomar were from 4-10 mm, and those from Pescadero were from 4-8mm. I then used a Wild M5 dissecting scope at 50x magnification to identify the barnacles and to determine whether they had larvae brooded within their shells. Results Correlating Recruitment and Relaxation of Upwelling Recruitment monitoring on two and four day intervals showed three recruitment pulses (Fig.5). The largest pulse occurred on April 27th at Waddell Beach, with an average of over 100 barnacles plate"2 days". This pulse was recorded on plates set out for two days in the mid intertidal during preliminary sampling. Tobserved two smaller pulses of recruitment in the high intertidal at Asilomar. These pulses averaged less than 12 barnacles plate“ 2 days 1, on the 1 1th and 17th of April. All three pulses of recruitment occurred during or just after periods of sea surface warming and decreased north wind, suggesting that recruitment occurred during relaxations of upwelling (Fig 6). Sea surface temperature from MI was above 12.5°C on April 27th, and above 13.5°C two days before recruitment occurred on the IIth, and on the 17th. Data from GC (REINAS 1998) also show periods of elevated SST from three to five days before the recruitment events. One to four days before the May 1Ith and 17th recruitment events, wind at Ml was from the east, with no north wind component. AVHRR images confirmed the April 27th and May 17th periods of relaxation. obtained twelve cloud-free or partially cloud-free AVHRR SST images from UCSD Coastwatch (1998) and Monterey Bay Aquarium Research Institute (1998) for the period from April 25th to May 22nd. On April 25th there were strong, broad plumes of cold water at Pt. Sur and Anno Nuevo, extending about 40 km southwest of each point. The Ano Nuevo plume swept southward across the mouth of Monterey Bay. These plumes decreased on April 26th and 27th, but appeared to be moderate in size again by May 9th (Fig. 2). Two images from May 15th and one from May 17th show that the Ano Nuevo and Pt. Sur plumes were greatly reduced on these days. Warm water above 13° C was up against most of the coast from Pt. Sur to San Francisco, and the Änno Nuevo plume did not extend down to the mouth of Monterey Bay (Fig. 3). On May 19th through 22nd, the Pt. Sur and Anno Nuevo plumes were again -40 km in length, and the Ano Nuevo plume sealed off the mouth of Monterey Bay (Fig. 4). Recruitment pulses at Asilomar were well correlated with SST and salinity at GC and MI (Fig.7), further suggesting that recruitment occurred during relaxation. The best Pearson correlation between recruitment and SST at Ml and GC occurred when recruitment was paired with SST from three days earlier (Ml: r’=.593, n-31; GC: r’=.604, n=32) (Sokal and Rohlf, 1995). The best negative correlation between recruitment and salinity at GC occurred with recruitment lagged four days after salinity (r’=-.593, n=32). There was no significant and meaningful correlation between SST or salinity and May recruitment at Pescadero or Waddell Beach. This may be explained by the fact that MI and GC were closer to Asilomar than to Waddell Creek or Pescadero, or the general lack of recruitment to the high intertidal at those sites during May. Frequency and Duration of Relaxation Events Using wind data from Ml, I identified nine days in May 1998 with no north wind component. These days occurred in a four day and five day period (May 6th-lOth, and 16th-19th). These were likely periods of relaxation, and they preceded or overlapped with periods of increased SST at Ml and GC (Fig. 6). Analysis of wind data from NOAA buoy 46042 in April-June 1988-1997 showed that it is not unusual to have two relaxation events per month, but that it is unusual to total nine or more days of relaxation in a month (Table 1). Only five of 30 months had nine or more days without north wind component in daily wind velocity. Furthermore, only one of those months (May 1993) had more than four consecutive days of relaxation. Most relaxation events (10 of 13) had a duration of less than four days. Changes in Monthly Recruitment Recruitment to the mid intertidal at Waddell Beach was 654+212.1 (mean + s.d.) recruits plate month“. Monthly recruitment to Pescadero, estimated by summing the recruitment in two 14 day sampling intervals, was 101.9-54.5 recruits plate month". May 1998 recruitment at Waddell Beach and Pescadero was not conclusively greater than recruitment during April-June 1996 or 1997. At Waddell Beach, May 1998 had significantly greater recruitment than only three of the six months in April-June 1996-7 (Figure 8, Table 2). At Pescadero, recruitment in May 1998 was significantly greater than only two of six months in April-May 1996-7 (Figure 9, Table 3). Overall, out of six comparisons with the non-El Ninno year 1996, recruitment in May 1998 was greater in only two cases. Of six comparisons with the early-El Ninno year 1997, recruitment in May 1998 was greater in only three cases. It can not be conclusively shown that during the El Ninno recruitment to Pescadero and Waddell Beach increased above recruitment in a non-El Ninno or early El Ninno year. Barnacle Reproductive Output B. glandula and Chthamalus spp. were found to be reproductive during the study period. Between 13% and 32% of B. glandula and 52-66% of Chthamalus spp. were brooding larvae when I removed them from the rocks at each site (Fig.10). Only B. glandula at Pescadero were brooding at levels lower than found by Hines in May 1973-4 (1978) or Miller in 1989-90 (1992). Discussion: The close correlation that I found between recruitment and sea surface temperature and salinity suggests that recruitment during May 1998 occurred during periods of relaxation. The three instances of co-occurring recruitment and relaxation that I observed suggest that El Ninno recruitment occurs during relaxation, as proposed by Roughgarden et al. (1988) for non-El Nino years. My two-day interval plates at Asilomar were placed at the exact location used by Roughgarden and Shkedy (1997), who found that during the 1991 El Nino SST at GC did not accurately predict recruitment. Contrary to this, SST at MI and GC during May 1998 explained 59.3% and 60.4% of the variance in recruitment. This disparity could be explained by the presence of a deeper mixed layer in 1991, which would result in upwelling of warm water from above the thermocline. This would tend to cancel out the cold water signal that characterizes upwelling when the mixed layer is shallower. This hypothesis could be tested by comparing CTD casts in 1991 to those from 1998. Connolly and Roughgarden (1998) have suggested that the 1997-8 El Nino caused increased recruitment to the central California coast, possibly by decreasing the prevalence of upwelling-favorable northerly winds . My data from buoy 46042 show that northerly winds were less frequent than normal in May 1998. An analysis of north wind component patterns for the entire upwelling season in 1997 and 1998 could further test this hypothesis of decreased upwelling-favorable winds. Recruitment at Pescadero and Waddell Beach in 1998 was not significantly greater than in the non-El Ninno year 1996. Recruitment at Asilomar was lower in 1998 than in the non-El Ninno period April-October 1988 (.0157 recruits plate ’day“ for 1998 vs. approximately .053 recruits plate" day" for 1988) (Farrell 1991). It is possible that recruitment in central California was actually higher in 1998 than in non El Nino years, but that my limited sampling time or limited number of sites prevented me from detecting this trend. Alternatively, the increase in relaxation during May of 1998 may have been countered by a shortage of food availability for adults or nauplii, leading to decreased reproductive output or increased larval mortality. The effect of either would be a loss of recruitment that could lead to lower overall abundances of adult barnacles. Glynn (1988) noted declines in barnacle populations in Peru during severe nutrient depletion in the 1982-3 El Ninno. Since I found that adult barnacles at my sites were brooding at rates similar to those seen in non-El Ninno years, larval mortality seems more likely than adult reproductive failure. Lang and Marcy (1982) found that nauplii of B. improvisus completely deprived of food died within seven days. Nauplii deprived of food for even only one day showed impaired swimming. Glynn (1988) noted that several researcher: found that nitrogen depletion severely affected kelp late in the 1982-3 El Nino. If nitrogen availability decreased as the 1997-8 El Ninno progressed, and if this led to decreased food availability for barnacles, it could explain the higher recruitment in 1997 compared to 1998. One possibility for testing the hypothesis of increased larval mortality would be to compare plankton samples from the top 1 m in 1998 to those from 1997. If larval mortality was higher in 1998, fewer late-stage nauplii should be present. Plankton samples may be available from the California Cooperative Ocean Fisheries Investigation (CalCOFI), which conducted surface plankton tows in April and May 1998 off Monterey. The increased recruitment seen along the central California coast in 1997 compared to 1996 could also have been due to factors unrelated to the El Nino. Gaines and Roughgarden (1987) suggested that rockfish predation is a major source of mortality for barnacle larvae. If yearly fluctuations in rockfish populations are fairly uniform along the central coast, it is possible that a strong rockfish year class in 1996 caused decreased barnacle recruitment below typical levels. Alternatively, coastal storms or currents unrelated to upwelling may have trapped larvae offshore or limited food resources in 1996. Thus, determinants of barnacle population size other than upwelling and El Ninno should be considered when investigating yearly variation in recruitment. Several inconsistencies exist in my results and procedure. Most apparent is the discrepancy between daily recruitment in the mid intertidal and high intertidal (Fig.11). Although settlement varies with tidal height (Grossberg 1982), the difference I observed is extreme, and the relative amount of recruitment between sites varies with tidal height. One partial explanation is that the high intertidal zone plates at Pescadero were 1.9 m above MLLW, and may not have been submerged on all tidal cycles during my sampling period, leading to low values for high intertidal recruitment at that site. Three more inconsistencies should be noted. Correlations between recruitment and SST and salinity were run for May, but did not include days in April and the recruitment pulse at Pescadero. Secondly, when daily wind data at buoy 46042 become available for May 1998, these data (rather than data from Ml) should be used to compare 1998 north wind patterns to 1988-97 wind data from buoy 46042. Finally, recruitment at Pescadero in the mid intertidal zone may have been underestimated. Simply adding recruitment during two consecutive 14 day intervals may underestimate monthly recruitment. Recruits appeared to settle near older barnacles on the plates, and fresh plates like the ones set out on May 15th probably discourage recruitment. A major limitation of this study was that I was able to collect data only for May, although I attempted to make generalizations and comparisons to the entire April-June upwelling season. More data from June 1998 would help determine whether recruitment in 1998 is actually higher than usual, and whether this is true across sites. Monitoring recruitment in June might also further strengthen the correlation between recruitment and relaxation of upwelling. Monitoring relaxation of upwelling in June could support the suggestion that the duration of relaxation events has increased in 1998 compared to non-El Ninno years Each of the three relaxation events delivered larvae to only one of the three sites I studied. Future work could attempt to identify why certain relaxation events deliver larvae to certain sites. One possibility is that the amount of north wind varies between sites, and therefore leads to site-to-site differences in the strength of upwelling. If accurate data on daily wind velocity near each site are available, it may be possible to calculate the amount of upwelling or relaxation occurring at that site. J. Paduan (pers. comm.) has suggested that accurate measurements of alongshore wind stress, derived from measurements of north wind velocity, could be used to predict the relaxation of upwelling and the arrival of warm water to shore. This might help determine whether it is differences in upwelling that causes differences in recruitment between sites, or whether other oceanographic or biotic factors are at work. Acknowledgments J. Watanabe served as my mentor for this project. His daily encouragement and questions, field assistance during dawn trips to Pescadero, and patient editing and explanations of SYSTAT made research seem human, approachable, and understandable S. Connolly introduced me to the study of barnacles in 1996, and continued to help me with all aspects of this work, from identifying cyprids to editing parts of the manuscript. He also made his recruitment data from two years of research available for my use. J. Paduan of the Naval Postgraduate School provided expertise on the oceanography of upwelling and frontal movement. F. Schwing from the NOAA Pacific Fisheries Environmental Laboratory offered valuable direction and resources linking oceanography and recruitment. R. Polk's assistance measuring tidal heights was greatly appreciated, as was K. Ludwig's help identifying substrate types. K. Gray, A. Fong, and D. Hansen from the California Dept. of Parks and Recreation provided assistance with collection permits. J. Carpenter's knowledge of Asilomar Beach helped uncover previous study sites. Literature Cited Connolly, S., and J. Roughgarden. (1998). Increased recruitment of northeast Pacific barnacles during the 1997-98 El Ninno. Submitted paper. Farrell, T., D. Bracher, and J. Roughgarden. (1991). Cross-shelf transport causes recruitment to intertidal populations in central California. Limnology and Oceanography 36: 279-288. Lang, W., and M. Marcy (1982). Some effects of early starvation on the survival and development of barnacle nauplii, Balanus improvisus. J. Experimental Marine Biology and Ecology 60: 63-70. Gaines, S., and J. Roughgarden. (1987). Fish in offshore kelp forests affect recruitment to intertidal barnacle populations. Science 235: 479. Glynn, P. (1988). El Nino-Southern Oscillation 1982-1983: Nearshore Population, Community, and Ecosystem Responses. Pages 309-346 in R. Johnston, P. Frank, C. Michener (eds). Annual Review of Ecology and Systematics, Vol. 19, Annual Reviews Inc., Palo Alto, CA. Gross, M. (1987). Oceanography: A View of the Earth. Prentice Hall, Inc., Englewood Cliffs, NJ. Grossberg, R. (1982). Intertidal zonation of barnacles: the influence of planktonic zonation of larvae on vertical distribution of adults. Ecology 63: 894-899. Hines, A. (1978) Reproduction in three species of intertidal barnacles from central California. Biological Bulletin, 154: 262-281. Light, S. (1975). Light’s Manual. University of California Press, Berkeley and Los Angeles, CA. Miller, K. (1992). The biology of barnacles. Ph.D. Thesis, Stanford University, Stanford, CA. Monterey Bay Aquarium Research Institute. (1998) Monterey Bay SST Images [web pagel. Available: http://www.mbari.org [Accessed May 19981. NOAA National Marine Fisheries Service Pacific Fisheries Environmental Group. (1998). Upwelling Indices and NOAA Moored Buoys, [webpage). Available: http://www.pfeg.noaa.gov/ [Accessed May 19981. REINAS.(1998). Granite Canyon NOAA Fisheries CTD, [web pagel. Available: http://sapphire.cse.ucsc.edu/cgi-bin/reinaslite-met-query [Accessed May 1998). Rosenfeld, L., F. Schwing, N. Garfield, and D. Tracy. (1994). Bifurcated flow from an upwelling center: a cold water source for Monterey Bay. Continental Shelf Research 14:9 pp. 931-964. Roughgarden, J., S. Gaines, and H. Possingham (1988). Recruitment dynamics in complex life cycles. Science 241 pp. 1460-1466. Roughgarden, J., J. Pennington, D. Stoner, S. Alexander, and K. Miller (199 1). Collisions of upwelling fronts with the intertidal zone: the cause of recruitment pulses in barnacle populations of central California. Acta Oecologia 12:1 pp. 35-41. Shkedy, Y., and J. Roughgarden (1997). Barnacle larval recruitment and population dynamics predicted from coastal upwelling. Oikos 80 pp. 487-498. Sokal, R., and F. Rohlf (1995). Biometry. W.H. Freeman and Co., New York, NY. UCSD CoastWatch Program, West Coast Regional Node. (1998). El Nino Watch and AVHRR SST Images I web page). Av. ailable: http:/coastwatchwc.ucsd.edu/cwatch.html [Accessed May 1998). Figure Legend Fig. 1. chart of the central California Coast. Recruitment sites at Asilomar and Pescadero are labeled. The Waddell Beach site is between Davenport and Anno Nuevo. NOAA buoy 46042 is labelled. The MBARI MI buoy is directly inshore of buoy 46042, at the mouth of Monterey Bay. The Granite Canyon CTD and shore meteorological station are south of Point Sur. From Miller (1992). Fig.2. AVHRR SST images from May 9th. The image is of the central California coast, roughly from Pt. Reyes to Pt. Piedras Blancas. Blue indicates cold upwelled water, and green is warmer oceanic and California current water. Black areas to to the west of the coast are areas covered by clouds. May 9th is a period of strong upwelling, with plumes visible at Pt. Sur, Anno Nuevo, and Pt. Reyes. Upwelling plumes generally extend southwest from the headlands.Note the Anno Nuevo plume extending across the mouth of Monterey Bay. Fig.3. AVHRR SST images of the April 15th-17th 1998 period of relaxation of upwelling. Warm water above 13°C has reached the coast. Upwelling plumes at Anno Nuevo and Pt. Sur have disappeared, and the plume at Pt. Reyes is greatly reduced. Fig.4. AVHRR SST image during May 22nd, 1998. After the May 15th-17th relaxation event, upwelling increased. The plumes of upwelled water are once again visible as blue bands extending up to 40 km from the headlands. Fig.5. April 27th-May28th 1998 recruitment to Asilomar, Pescadero, and Waddell Creek measured on 2 day intervals (Asilomar) and 4-day intervals (Pescadero and Waddell Creek). The data point for Pescadero on April 27th is from the mid intertidal, all other data are from the high intertidal. The y-axis uses a log scale, mean number of recruits of B. glandula and Chthamalus and represents plate per day. Average recruitment during May only was 3.02 recruits spp. per plate day at Asilomar, 38 recruits plate day at Waddell Beach, and 07 recruits plate day at Pescadero. Fig.6. Oceanographic conditions April 27th-May 28th, combined with recruitment at Asilomar, Pescadero, and Waddell Beach. "Salt" is salinity at GC, which varies between 30.4 and 30.7 psu. "GC" is SST at Granite Canyon. "Ml" is SST at the MBARI MI buoy. SST varies from 10 to 14° C. Fig.7. Asilomar high intertidal recruitment, SST at GC and Ml, and salinity. The correlation is for recruitment Maylst through May 28th. Salinity: r-=-.593; GC SST: r’=.604; MI SST: r’=.593. Fig.8. Monthly recruitment at Waddell Beach, April-June 1996-7, and May 1998. Recruitment in May 1998 significanty greater than in April 1997, June 1996, and May 1997. Note that for April and June 1997, recruitment at these sites did not increase above 1996 levels, suggesting that increases to recruitment occured primarily later in that year after the El Nino developed. 1996-7 data from Connolly and Roughgarden (1998). Fig.9. Monthly recruitment at Pescadero, April-June 1996-7, and May 1998. Recruitment in May 1998 significanty greater than April 1996 and June 1997. Here, similar to Waddell Beach, recruitment in April, May, and June 1997 was not higher than during those three months in 1996. 1996-7 data from Connolly and Roughgarden (1998). Fig. 10. Fraction of adult B. glandula and Chthamalus spp. brooding at each of three sites. N°146 at each site. I collected barnacles in late May 1998. At Pescadero fewer B. glandula appeared to be reproductively active than at other sites. Fig.11. Daily mean recruitment in the high and mid intertidal at three sites. This graph illustrates inconsistencies in the data. At Asilomar, the high intertidal had the highest recruitment, while at the other two sites the mid intertidal zone had higher recruitment. Asilomar had the most recruiment in the high intertidal, but Waddell Beach had the most in the mid intertidal. Fig. 1. PT. REYES V SAN FRANCISCO BAY Ocean Beach Montara + 46012 + Pescadero Ano Nuevo Davenport* Capitola Santa Cruz MONTEREY 46042 + Moss Landing BAY Asilomar / Hopkins Monastery Soberanes Pt. Sur 4 46028 + Nighttime SST selection for Central California high resolution 05/09/98 12:00 GMT Fig.3. Daytime SST selection for Central California high resolution 05/15/98 22:00 GMT Fig.4. Nighttime SST selection for Central California high resolution 05/22/98 11:00 GMT Fig.5. Recruitment to the High Intertidal May 1998 100.0 10.0 AS¬ 1.0 -WB- -P. 0.1 10 20 30 40 Days (O=Apr. 27th 1998) EAS » WB • PB Fig.6. Recruitment, SST, and Salinity Apr. 27th-May 28th 100.0 — 10.0 8 0.1 40 30 20 10 Day (O=May 27th 1998) WE AS Fig.7. Asilomar High Intertidal Recruitment, SST, and Salinity 2 35 30 2 26 20 SALT GC 10 MI 8 10 12 Recruits Per Plate Per Day C 3 5 Mean Recruits Per Plate Per Month kkkaaaa- — T 8 — — — S 0 O O Q U 0 D 2 S X — ——— Becruts Per Plate Per Month — — — — — 0 9 Q D 0 2 O Asilomar Balanus Asilomar Chthamalus Waddell Beach Balanus Waddell Beach Chthamalus Pescadero Balanus Pescadero Chthamalus fraction brooding — NoOo . 3 Recruits per plate per day lable WWind Relaxation Events at Buoy 46042 Days listed contain no north wind component in mean daily wind velocity. Bold values indicate months with greater than 8 days of wind relaxation. June Year April May (dates) (dates) (dates) 5th. 8th. 141. 1. 66 8h, 151. 20 104 131 1988 27th 17th-19th 218t. 19th 20th, 24 27th.2 41. 5th 13t 151h 10h, 126, 166. 1989 ond h 29th.30 Z2-2Z 2nd 6h 3rd. 4th 8th. 10th 4h. 8h. 141 161 1990 12t-14½, 164, 27h.281, 19h 218 6 2nd. 1991 18. 194.215 271-28 126.141, 176 66. 7h. 126. 131, 2 1992 14½, 164 191h.218 181 104, 126.141h 166 1993 18120th. 318 24 61.76, 176. 318 15h.16h 2370 1994 154, 164.30 6h 7h 131 18. 9h. 154. 161 1995 97th 30 Z 8h. 9h. 151. 17 14h.184, 215,261 26h.271 1996 19th, 24 31. 8h. 9th. 141h 134, 184 19h. 1997 6h, 181.218 15th, 30 6h. 106, 166.19 NA 1998 NA (MBARI MI Buoy Table 2. Waddell Beach Recruitment in May 1998 vs. April-June 1996 and 1997 Bold values are from months with recruitment significantly less than May 1998. 1996-7 data from S. Connolly and J. Roughgarden (in prep., 1998) Single factor ANOVA of log transformed data w w w df MS Source 200155 31.593 11.065 Among 34 350 Withi g Post Hoc Comparisons Tukey's test, Bonferonni Adjustment May 1997 June 1997 April 1997 June 1996 April May May May May May 19962 1996 5 1998 1998 1998 1998 May May 1998 1998 0035 2.001*k* 2.001** 005** 1.00 (NS) 104 QS) Table 3. Pescadero Beach Recruitment in May 1998 vs. April-June 1996 and 1997 Bold values are from months with recruitment significantly less than May 1998. 1996-7 data from S. Connolly and J. Roughgarden (in prep.) Single factor ANOVA m g w w MS df Source P ... . 001*** 12.731 4.660 Among 35 366 Within Post Hoc Comparisons Tukey's test, Bonferonni Adjustment June 1997 May 1997 April 1997 June May April May May May 19962 1996 1996 2 1998 1998 1998 May May May 1998 1998 1998 2.001*4* 082 (NS) 1.00 (NS) 1.00 (S) .016* 1 (NS) P