Coral 18O Record of SST at Ishigaki Island

The Porites australiensis colony (IU96-07) in Urasoko Bay, Ishigaki Island, is only 300 m from a long-term daily temperature monitoring station (Yaeyama Station, Japan Sea-Farming Association). Therefore, this coral specimen was considered a good material for the calibration of the skeletal 818Oc response to changes in SST (Figs. 3 and 4). Suzuki et al. (1999) proposed the following skeletal 818Oc-SST relationship (Fig. 5):

The seasonal variation of SST around Ishigaki Island is about 10 °C, and the seasonal salinity change is less than 0.5. Watanabe and Oba (1997) estimated the S18Ow variation corresponding to the seasonal salinity change to be 0.1%. Therefore, the effects of seasonal changes in SST on coral S18Oc are far greater than the effect of seasonal changes in salinity on S18Ow; thus, the effect of changes in S18Ow on the 818Oc-SST calibration are negligible.

1996 1997 1998 1999 2000 2001 2002 2003

1998

1999

2000

2001 1998 Year AD

1999

2000

2001

Figure 3: Comparison of satellite-derived SST (IGOSS NMC SST; Reynolds and Smith, 1994) and in situ SST measurements at five coral coring sites: Ishigaki Island (Shiraho reef), Bicol (off Sorsogon City), Chuuk Atoll (East Channel), Pohnpei Island (Main Channel), and Majuro Atoll (Central Shoal). Gray lines and black lines with circle lines indicate weekly means of satellite-derived and in situ SSTs, respectively. The 10-day mean record at Ishigaki Port derived from a daily observation at 10:00 a.m. local time (black line with open circles) is also shown (Ishigaki-jima Meteorological Observatory, the Japan Meteorological Agency; JMA). At the low latidude Micro-nesian and Philippine coral sites, in situ monitoring of SST showed no differences with IGOSS SSTs, which indicates the accuracy of both measures. However, at the higher latitude of Ishigaki Island, wintertime IGOSS NMC SSTs showed relatively large differences of up to 2 °C with in situ SSTs as a result of strong surface cooling.

32

30

28

■ 26

L 24

22

20

18

o

32

S

30

r

er

« !

24

22

32

30

28

26

24

22

1999

2000

2001 1998 Year AD

1999

2000

2001

Figure 3: Comparison of satellite-derived SST (IGOSS NMC SST; Reynolds and Smith, 1994) and in situ SST measurements at five coral coring sites: Ishigaki Island (Shiraho reef), Bicol (off Sorsogon City), Chuuk Atoll (East Channel), Pohnpei Island (Main Channel), and Majuro Atoll (Central Shoal). Gray lines and black lines with circle lines indicate weekly means of satellite-derived and in situ SSTs, respectively. The 10-day mean record at Ishigaki Port derived from a daily observation at 10:00 a.m. local time (black line with open circles) is also shown (Ishigaki-jima Meteorological Observatory, the Japan Meteorological Agency; JMA). At the low latidude Micro-nesian and Philippine coral sites, in situ monitoring of SST showed no differences with IGOSS SSTs, which indicates the accuracy of both measures. However, at the higher latitude of Ishigaki Island, wintertime IGOSS NMC SSTs showed relatively large differences of up to 2 °C with in situ SSTs as a result of strong surface cooling.

Distance along growth axis (cm) 10 5 0

Warm / Wet

Cool / Dry

Distance along growth axis (cm) 10 5 0

Warm / Wet

Cool / Dry

Figure 4: Coral d18O records for the western Pacific region covering about 20 years (locations of coral cores are shown in Fig. 1). Black triangles mark positions of growth perturbations in Ishigaki Island corals identified by abrupt shifts in 818O profiles. Dates of coral collection are shown with arrows. Details of cores (locality, coral species name, time-span of records) are given in the text.

Figure 4: Coral d18O records for the western Pacific region covering about 20 years (locations of coral cores are shown in Fig. 1). Black triangles mark positions of growth perturbations in Ishigaki Island corals identified by abrupt shifts in 818O profiles. Dates of coral collection are shown with arrows. Details of cores (locality, coral species name, time-span of records) are given in the text.

Suzuki et al. (1999) pointed out that the S^O./SST slope in equation (1) is close to those reported earlier for Porites spp. corals from other regions of the Pacific, but differs from that reported by Mitsuguchi et al. (1996) for Porites lutea from the east coast of Ishigaki Island (Fig. 5). Suzuki et al. (1999) suggested that the lower frequency of microsampling by Mitsuguchi et al.

McConnaughey (1989)

Mitsuguchi et al. (1996)

Mitsuguchi et al. (1996)

22 24 26 28 30 32 Seawater Temperature (°C)

Figure 5: Comparison of the temperature dependence of coral S18Oc. Bold lines indicate Porites spp. coral specimens from Ishigaki Island, southern Ryukyus. Specimen details (sample i.d.,818Oc-SST slope, coral species, locality, original reference) are as follows: IS91-06 (-0.140% °C-1, P. lutea, Shiraho, this study), IY99-01 (-0.148% °C-1, P. lutea, Yasura-zaki, this study), IU96-07 (-0.165% °C-1, P. australiensis, Urasoko, Suzuki et al., 1999), and a Porites colony (0.134% °C-1, P. lutea, Yasura-zaki, Mitsuguchi et al., 1996). The negative displacement of calibration lines for the Ishigaki corals is due to the differences in mean S18Ow between Ishigaki Island and the other regions (McConnaughey, 1989; Leder et al., 1991; Gagan et al., 1998; Suzuki et al., 1999, 2001).

22 24 26 28 30 32 Seawater Temperature (°C)

Figure 5: Comparison of the temperature dependence of coral S18Oc. Bold lines indicate Porites spp. coral specimens from Ishigaki Island, southern Ryukyus. Specimen details (sample i.d.,818Oc-SST slope, coral species, locality, original reference) are as follows: IS91-06 (-0.140% °C-1, P. lutea, Shiraho, this study), IY99-01 (-0.148% °C-1, P. lutea, Yasura-zaki, this study), IU96-07 (-0.165% °C-1, P. australiensis, Urasoko, Suzuki et al., 1999), and a Porites colony (0.134% °C-1, P. lutea, Yasura-zaki, Mitsuguchi et al., 1996). The negative displacement of calibration lines for the Ishigaki corals is due to the differences in mean S18Ow between Ishigaki Island and the other regions (McConnaughey, 1989; Leder et al., 1991; Gagan et al., 1998; Suzuki et al., 1999, 2001).

(1996) is a possible cause of the difference in the 818Oc/SST slope relationship between those of Mitsuguchi et al. (1996) and that of Suzuki et al. (1999), because a lower sample frequency would be expected to result in the attenuation of seasonal S18Oc signals.

Through multiple deployments of temperature loggers at the coral sites, we found that IGOSS SSTs generally showed excellent agreement with in situ temperature records in low-latitude tropical regions, including the Micronesian islands and Bicol (Fig. 3). However, wintertime IGOSS NMC SSTs showed a relatively large difference of up to 2°C with in situ SSTs monitored in moats (shallow lagoons) in the subtropical Shiraho and Yasura-zaki areas of Ishigaki Island, suggesting that the utility of satellite-based SSTs for comparison with high-latitude coral records may be limited. On the other hand, SST records from Ishigaki Port collected by the Japan

Meteorological Agency (JMA) showed good agreement with in situ SSTs at coral sites around Ishigaki Island. Thus, we analyzed two coral specimens from the Shiraho and Yasura-zaki coral reefs to calibrate the S18Oc signal against JMA SSTs collected in Ishigaki Port (Fig. 4). The skeletal 818Oc-SST relationships obtained for colonies IS91-06 (Shoraho reef) and IY99-01 (Yasura-zaki channel) are shown by equations (2) and (3), respectively:

818Oc(%„) - -0.999 - 0.148T (°C) (R2 - 0.95) (3)

These slopes fall between -0.134% °C-1 and -0.165% °C-1 reported by Mitsuguchi et al. (1996) and Suzuki et al. (1999), respectively. Because the microsampling frequency of these two specimens (IS91-06 and IY99-01) was comparable to that of IU96-07 (Suzuki et al., 1999), the smaller slopes obtained for IS91-06 and IY99-01 cannot be attributed to signal attenuation caused by a lower frequency of microsampling. This relatively large variation in the SST-S18Oc relationship found at a single island may be related to biological and genetic factors rather than to a difference in microenvironments. Further research, including examination of interspecies differences in the genus Porites, is needed to determine the cause of the variation (e.g., Suzuki et al., in press). Keeping these caveats relating to the S18Oc-SST calibration in mind, we believe that the similarity between the skeletal S18Oc profiles and the in situ SST time series (Fig. 6) indicates that coral S18Oc records can be used as a good proxy for SST in reef settings where the influence of SSS variation on S18Oc is negligible.

Was this article helpful?

0 0

Post a comment