As' the results from individual stations become available it will be possible 

 to assess the value of each station for continued operation. Some it 

 may be desirable to eliminate after their characteristics have been demon- 

 strated to be essentially similar to those at companion stations, while in 

 some areas of complicated tidal movement the need for additional stations 

 of intermediate location will be apparent. Gradually there should 

 evolve a scheme of station distribution representative of the varying 

 tidal conditions, encountered in different areas and suitable for long- 

 period operation. 



Continuous records over a period of years are necessary for certain 

 purposes, chief among which are the evaluation of heights and frequency 

 of extreme tides of meteorological or seismic origin, the determination 

 of mean tidal values from short series of observations through comparison 

 with reference stations, and the investigation and correlation of monthly 

 and yearly variations in sea-level. 



Meteorological or seismic disturbances occur at irregular intervals and 

 with varying intensity so that a limited period of tide observations cannot 

 be depended upon to give a satisfactory evaluation of their effects on 

 tidal heights. Results from many such disturbances must be recorded 

 before reUable estimates can be made of the heights and frequencies of 

 extreme high and low waters expected under varying conditions. 



As the Pacific Ocean regions develop, tide data will be needed for the 

 local projects of many localities. Meeting this expected demand will be 

 greatly facilitated by the availability of a network of primary tide statio'ns 

 for which basic tide data have been predetermined. Through comparison 

 ^vith the primary station or stations of similar characteristics, short 

 series of tide observations involving little effort or expense to obtain will 

 then suffice to produce tide data of acceptable precision for most 

 practical purposes. 



For any locality there is a distinquishing seasonal variation in sea- 

 level which is approximately periodic from year to year. The form of 

 this month-to-month variation may be indicated from a single year of 

 observations, but several years of observations are generally required to 

 average out the non-periodic part of the variation. The periodic part 

 of this monthly variation must be reliably established as a prerequisite 

 to determining the harmonic constants of the principal long-period 

 constituents and 'to correcting tidal datum planes to their mean values. 



The cumulative yearly sea-level averages at long-period stations* 

 when referred to fixed datums defined by bench marks, furnish a measure 

 of changes in sea-level relative to land elevations. At any particular 

 station the indicated change may be due to an absolute change in sea- 

 level, an absolute change in land elevation, or to a combination of both. 

 To permit a separation and apportionment of the two movements, 

 long-period tide observations of world-wide distribution are needed for 

 comparative purposes. In view of the extensive world coast-line 

 involved the number and distribution of primary tide stations are 

 presently far from adequate, a situation which it is hoped will ultimately 

 be greatly improved by the programme under discussion. 



Any programme of tidal investigations for an area as extensive as the 

 Pacific Ocean must necessarily be in the nature of a continuing project. 

 Its development will depend on many uncertain factors, and many 

 operational difficulties are expected to be encountered. Already some 

 of the original installations have had to be abandoned due to typhoon 



223 



