Experimental Sea level Forecasts
(deviations w. r. t. climatology) for the U.S-affiliated Pacific Islands

(i) Seasonal Sea Level Forecast for AMJ, MJJ, and JJA 2006

Forecasts of the sea level anomalies in the USAPI are presented using CCA statistical model. Locations of all stations are shown in Fig 2. Based on the independent SST values in JFM 2006 (for SST data, see http://iridl.ldeo.columbia.edu/expert/SOURCES/.NOAA/.NCDC/.ERSST/.version2/.SST/), the resulting CCA model was used to forecast the sea level of three consecutive months: Apr-May-Jun (AMJ), May-Jun-Jul (MJJ), and Jun-Jul-Aug (JJA) (Table 1). CCA cross-validation forecast skills for 0, 1, and 2-month leads are presented (Fig. 3 ).

Table 1 : Forecasts of sea level deviation in inches (AMJ: Apr-May-Jun, MJJ: May-Jun-Jul, and JJA: Jun-Jul-Aug)

As in the previous season (JFM), this season (AMJ) also provided very skillful forecasts (Fig. 3). Most of the tide gauge stations show strong skill level (Table 1). Only Guam , Malakal, and Yap of north Pacific displayed a relatively lower skill, which is still, however, reasonably well predicted (skill level of 0.5 or more). The majority of the tide gauge stations in a ll the three consecutive months (AMJ, MJJ, JJA) are very well predicted with a mean skill greater than 0.60 (at 0 to 2-month lead time).

Results of the CCA model forecasts revealed that all the tide gauge stations are likely to experience rise in the forthcoming seasons AMJ, MJJ, and JJA (Table 1). This rising trend has also been observed consistently in all the islands located in the vicinity of north-west, north-east, and South Pacific Ocean . However, it is also worth noting here that the central part of the western Pacific (Pohnpei, Majuro, and Kwajalein ) displayed relatively less positive deviations, as compared to others. The higher positive deviation has been observed in the north-western part of Pacific ( Guam , Malakal, and Yap ). Following this trend, a trend of moderately positive deviations was observed in the central part (Pohnpei, Kapingamarangi, and Majuro). Finally, a higher positive deviation was observed again in the vicinity of South Pacific Ocean (Pago-Pago).

The forecast values of sea-level—which determines rise in most of the tide-gauge stations—is consistent with the La Niña conditions present in the tropical Pacific in the beginning of the year.

(ii) Observed monthly sea level deviation in Jan-Feb-Mar (JFM), 2006

The monthly time series (January to March) for sea level deviations have been taken from the UH Sea Level Center. The full time series (in mm) is available at ftp://ilikai.soest.hawaii.edu/islp/slpp.deviations. Deviations are defined here as the difference between the mean sea level for the given month and the 1975 through 1995 mean sea level value computed at each station.

Table 2 provides the monthly observed sea level deviations (in inches). Table 2 provides the monthly observed sea level deviations (in inches). Except Yap , all other tide gauge stations recorded a trend of positive deviations in most of the months in this quarter (1st quarter of 2006) (Table 2). Among others, Guam recorded considerable rise in this quarter (7.6 inches, 8.3 inches, and 11.1 inches in January, February, and March respectively). After a slight recession in October and November 2005, Guam again started rising from December 2005. Yap, on the other hand, did not record any major changes in this quarter. Pago-Pago in the south Pacific recorded a rise during January. It may be mentioned here that the sea-level variation in the northwestern tropical Pacific islands has been identified to be sensitive to ENSO cycle, with low sea level during El Niño and high sea level during La Niña events. As mentioned, consistent with the La Niña conditions , the sea level has been found to be higher than average during January-February-March.

In the last quarter’s issue, our forecasts provided positive deviations for all the tide gauge stations in JFM (Fig #4 dotted line). Negligible variations were predicted in Malakal and Yap . Real-time observed sea-level data in JFM provided consistency with the forecast values (Fig. 4 ). In all cases, the direction of the deviations—either positive or negative— has been found to match all the forecasts. In addition, the quantitative values of these forecasts were very close to the real-time observed values (with the exception of Guam and Makalal.

Table 2 : Monthly observed sea level deviations in inches
(std deviation in parentheses)