Pacific ENSO Update

1st Quarter, 2005 Vol. 11 No. 1

 

SPECIAL SECTION: ENSO and Sea-Level Variability (3):
climatology of annual cycle

(Guam, cnmi, Marshalls, palau, FSMand American Samoa)

This is the third article of a series by Dr. Rashed Chowdhury on predicting sea level variations for the US Affiliated Pacific Islands. The first article focused on the historical differences in sea level and the ENSO cycle, the second issue focused on the physical mechanics behind these differences. This issue focuses on the annual sea level cycle.
The climatology and sea level of a region are intimately related. The following section presents a brief examination of this relationship in the U.S-affiliated Pacific Islands. Fifty years of sea level data have been taken from the ‘University of Hawaii Sea Level Center’ (UHSLC) for this analysis. In this analysis, six stations were observed: Guam, Saipan (CNMI), Malakal (Palau), Kwajalein (Marshalls), Yap (FSM), and Pago Pago (ASamoa).
The observed value of long-term monthly sea level variability (Fig. 2 – solid lines) of most of the northern Pacific Islands, by and large, displayed a strong annual cycle (Fig. 2a, 2b, 2c). The sea level of these islands varies slightly from one island to another and are significantly correlated to each other, which mean that variation (rise/fall) of sea level in one island is closely related to the variation of sea level (rise/fall) of the other island (Table 2). In Guam, a gradual increase of sea level from January to July has been observed (Fig. 2a). Soon after the peak in July, a gradual recession starts, which extends up to December. Saipan and Malakal, Palau also experienced similar peak and recession. Yap, on the other hand, experienced higher sea level in boreal summer (June-August) because of its closer proximity to the central Pacific. This is how Yap is different from the other three north Pacific Islands. Kwajalein, on the other hand, displayed a peak in April, and then followed by intermittent fluctuations afterward (Fig. 2e). After October, the sea level recorded a sharp drop in the following few months. Pago Pago, the lone south Pacific station, tended to show several peaks (two of which are major) in the annual cycle (Fig. 2f). The first major cycle indicated a gradual rise of sea level from January and a peak is observed in March. A second major cycle indicated an abrupt rise from June with a peak in July. Slow and intermittent recessions followed in the later part of the year.

Table 2: Correlation coefficient of sea level variation for five USAPI stations
Guam Palau Saipan Kwajalein Yap Pago Pago
Guam 1.000
Palau 0.642** 1.000
Saipan 0.743** 0.680** 1.000
Kwajalein 0.763** 0.777** 0.661** 1.000
Yap 0.771** 0.952** 0.808** 0.681** 1.000
Pago Pago 0.348 0.001 0.217 0.14 0.086 1.000
** Correlation is significant at 0.01 level, * Correlation is significant at 0.05 level

Note: Correlation is a statistical technique which can show whether and how strongly pairs of variables (here sea level of each of the stations) are related. A correlation coefficient of 1.00 or -1.00 is a perfect relationship between two variables. The closer the correlation coefficient is to zero, the less relationship there is between the two variables. In this example sea level variations in Palau and Yap are very closely related (correlation coefficient = .952) where as the sea level variations between Palau and Saipan are less closely related (correlation coefficient = .680)
Significance levels show how likely a result is due to chance. The most common level, used to mean something is good enough to be believed, is “0.01” or “0.05” meaning that the finding has a one percent (0.01) or five percent (0.05) chance of not being true. In our data, all north Pacific stations were significant at the .01 level.

While a qualitative variation of the climatology of annual cycle is identifiable from the monthly average sea level data records (as discussed before) (Figure 2 – solid line), it is the harmonic analysis that can give a picture of quantitative variation in these data (Figure 2 – dashed line). Therefore, to quantitatively evaluate the importance of the annual cycle from these data, harmonic analysis has been performed. Harmonic analysis consists of representing the fluctuations or variation in a time series as having arisen from the adding together of a series of sine and cosine functions. These trigonometric functions are “harmonic” in the sense that they are chosen to have frequencies exhibiting integer multiples of the “fundamental” frequency determined by the sample size of the data series. For example, a common physical analogy is the musical sound produced by a vibrating string, where the pitch is determined by the fundamental frequency, but the aesthetic quality of the sound depends on also on the relative contributions of the higher (1st and 2nd) harmonics. The 1st harmonics represents the annual cycle and explains the maximum variances.
The first harmonic, here in this case, explained about a considerable percentage of variance of the sea level variability in the north Pacific Islands (Fig. 2a-2e). The first harmonic for all islands explained variances of 64-88%. For the western most islands in the north Pacific (Guam, Saipan, Palau and Yap), maximum rise of sea level occurs in summer months (June to August). The annual cycle is relatively weak (though still explaining over 40% of the variance for sea level) in Kwajalein (Fig. 2e), and the annual cycle is extremely weak in Pago Pago (only 2% variance) (Fig. 2f). However, a second harmonic, which represents the semiannual cycle, explains 3-17% of variances (not reported here). This component adds considerably to the variance of Kwajalein (17%) and Pago Pago (11%).