ABSTRACT
    1. INTRODUCTION         2. METHODOLOGY
    3. RESULTS      3.1 Comparison of Altimeter and Tide Gauge data
                              3.2 Eddy Formation and Migration
                              3.3 Interannual variation of sea surface height from 1993 to 1999
    4. DISCUSSION                5. CONCLUSION              6. REFERENCE
    APPENDIX

   Eddies formation in 1999 and interannual variation of sea surface height from 1993 to 1999 in the East Sea were  examined using TOPEX/ERS2 and TOPEX/POSEIDON  Altimeter data, respectively. The Eddies were formed in September and continuously developed to become stronger anticyclone Eddies in October, November and December  in 1999 . This finding is a new phenomenon compare to the previous result that Eddies began to form in February and diminished  in May. Interannual Variability of Sea Surface Height in the East Sea showed that Sea Surface Height was higher in 1994, 1998 and 1999, and it was lower in 1993, 1995 and 1996. When the warm water region expands to the North,  the heat content of upper layer increases and sea surface height  rises.  In the contrarily, when cold water region expands to the South,  heat content of upper layer decreases and sea surface height  falls.
 

   1. INTRODUCTION

       Sea level and its horizontal slope and time changes are the surface expression of ocean  processes occurring over large spatial and temporal scales.  Large-scale currents (such as the Kuroshio, extending up to kilometers in depth) and planetary Rossby waves (thousands of km  in horizontal extent) carry the memories of past air-sea exchanges that can affect subsequent weather at great distances from the source of the original air-sea exchange.  Sea level also provides evidence of local heating by solar radiation, the perennial tides, and the addition of  water to the ocean by melting ice caps or glaciers (Ocean ESIP ).

     Altimeters emit a sequence of short pulses at microwave frequencies and then measure the return times to ascertain the instrument-surface distance. The TOPEX/Poseidon altimetric data has been widely used for several applications, such as ocean circulation,  El Nino/La Nina studies, sea level monitoring for environmental safety warning, simulation of larvae dynamics, and so on.  This has been diverging from its first specific mission to study the world's ocean. There are  a number of improvements that has been made on the satellite performances and algorithms  in order to study ocean circulation in the scale of basins and gyres( Fu et al, 1994 ). Nowadays, we can use this altimetric data for smaller scales of physical processes in the ocean.

       East Sea (Sea of Japan) is one of the largest marginal seas. Its mean depth is about 2,700 m. The surface water inflows through a narrow strait with sill depth about 100m and outflows through two narrow straits with sill depths less than 50 m( Tomczak and  Godfrey,1994 ).
 

   [Figure 1]  (a)   Horizontal temperature distribution at 100 m depth. (b)   Vertical  structure of temperature along section 133.8'E, 35.4 - 43'N.  From  Altimetry Plots: East/Japan Sea

   We can see a strong horizontal front is formed in the middle of East Sea(Figure 1). It undulates to the North from 38 degree N at the western part and goes down to 37.5 N, 132E and then moves up to 40N again at the eastern part of the East sea (Figure 1 (a) ). The front is separating the Northern Cold Water Region and the Southern Warm Water Region. A Northward western boundary current ( East Korea Warm Current) flows along the coast of Korea and then meanders along the front. The temperature section shows that the the temperature becomes 1 degree C about  300 m depth and almost homogeneous below the depth ( Figure 1 (b) ).

    Our hypothesis are that Eddy formation and migration can be detect by Sea Surface Height Deviation and Mean Sea Surface Height can be computed from climatological hydrographic data. Objectives are to examine Eddy Formation and migration in 1999 and estimate Interannual Variation of Sea Surface Height from 1993 to 1999  in the East Sea..
 

  2. METHODOLOGY

   2.1 TOPEX/POSEIDON Satellite Sensor

  All data taken by satellite are being processed by six science instruments below:

     1. Operational sensors: Dual-frequency radar altimeter (ALT)/NASA for  minimizing the errors caused by the ionospheric free electron ; TOPEX microwave radiometer (TMR)/NASA for estimating total water vapor content in the atmosphere; Laser retroreflector array (LRA)/NASA and Doppler orbitography and radiopositioning integrated by satellite (DORIS); Dual Doppler tracking system receiver/CNES for precision orbit determination.

     2. Experimental sensors: Single-frequency solid state radar altimeter (SSALT)/CNES and Global Positioning System (GPS) demonstration receiver (GPSDR)/NASA.

A number of improvements of those sensors performance make it possible to study Ocean Circulation in the basin scale such as Gyres and Eddies.

    2.2 Relation between sea surface height, heat content of upper layer, and upper layer thickness

    The East Sea has well defined 2 layers in vertical temperature structure (Figure 1-b).
The sea  surface height is a good measure of heat content of upper ocean and upper layer thickness. Since 8 degree C isotherm is located at the top of thermocline, the depth of 8 degree C can be a upper layer thickness. Warm surface layer has about 150 m layer thickness in Southern part of the East Sea. Lower layer is colder than 1 degree C.

Since heat content is given as Heat content = c x density x Temperature,
heat content of upper 300 m layer is the integration of heat content over depth.

Heat content of upper layer = c x average density x average Temperature x 300 m.

 where c is a specific heat,  4,200 J / ( Kg degree C ).

Average temperature over 300 m can represents the heat content of upper 300 m layer.
 
 

 [Figure 2] a. Comparison  between annual mean adjusted sea level and heat content of upper 300m layer at Ulleung Island (South Western East Sea). b. Comparison between the depth of 8 degree C isotherm and mean adjusted sea level. The depth of 8 degree C isotherm represents a thermocline depth or upper layer thickness. From Choi (1994).
 

   Heat content of upper layer and sea level variation are highly correlated (Figure 2-a). The correlation coefficient is 0.84. When the average temperature increases, the sea level rises. The thickness of upper layer and sea level are linearly correlated (Figure 2-b). When the thermocline goes down, upper layer thickness increases, heat content of upper 300 layer increase, sea surface height increases. When the sea surface height is high, the heat content of upper ocean is high and the upper layer thickness is thick.

   2.3 Geostrophic Calculation

             We used UNESCO Training Module for application of TOPEX/POSEIDON satellite Altimetry to produce maps of eddies and other geostrophic features developed by Donald R. Kobayashi, et al (1999). The method can compute sea level height from measured TOPEX altimeter data. We create a simple fortran program to calculate its East-West and North-South velocity components from where we made our current vector plot using Generic Mapping Tool (GMT).  We are also comparing our result with observational data and sea level height anomaly and its geostrophic current taken from interactive map in the internet produced by Colorado Center for  Astrodynamics Research (CCAR), University of Colorado.

                The dz/dx and dz/dy directional gradients from the resulting sea surface topography are used in the following equations to estimate the u (East-West) and v (North-South) components of the geostrophic current in units of centimeters per second:

                 u = -(g/f)dz/dy       and        v = (g/f)dz/dx

                     where

                      g = 980 cm per second (gravitational acceleration)

                      f = 2 W sin ( phi )

                      W = 7.29x10-5 radians per second (earth angular rotational velocity)

                      phi = latitude
 

3. RESULTS

   3.1 Comparison of Altimeter and Tide Gauge data
 

Ulleung Island is located at the South Eastern part of the East Sea(131'E, 37.5'N).
Because it is a small island and the water depth around it is deeper than 1,000 m,
it is a good place to measure sea level and compare its sea level data from tide gauge with the air-borne altimeter data. We compared the annual mean deviation from 1993 to 1998. They have a good linear relation and their ranges are from -4 cm to +8 cm (Figure 3). Ulleung island has higher sea level in 1995 and 1998.
 

  [Figure 3]. Comparison of Altimeter data and Tide Gauge data
 
 
 

   3.2 Eddy Formation and Migration

        Sea surface height deviation data are downloaded from  Colorado University . After we download data, we examine the sea surface height deviation at the middle of each month.

   [Figure 4] Formation and Migration of eddies at each month in 1999.

January 15, 1999:    We have two  strong warm anticyclonic eddies at 137E, 39N and 134E, 37-38N and a very weak  anticyclonic eddy at 130E, 37-37.5N in the southern part of East Sea.

 February 15, 1999:  The two strong Eddies found in January is decaying but The  weaker Eddy is increasing its intense.

March 15, 1999: We found a new weak cyclonic Eddy at 130E, 39N.

April 15, 1999: The weak cyclonic Eddy found in March is decay.

May 15, 1999: The Anticyclonic Eddies are beginning to form again at 131E, 39N; 137-138E, 39N; and 134E, 37-38N.

June 15, 1999: Depression of the sea level in the eastern part of East sea is increasing and make the strong southward geostrophic current moving farther into western part of the region. The Anticyclonic Eddy of 131E, 39N and 134E, 37-38N are decay, but  Eddy of 137-138E, 39N becomes stronger.
 

July 15, 1999: It has relatively similar pattern with the previous month but here it is  giving an increase in the movement of warm water to the west. All Eddies are decay.

August 15, 1999:  It is beginning to form a large area of Cyclonic Eddy at 132-134E, 39N.

September 15, 1999: The cyclonic Eddy in August is becoming stronger and shifting to 135E, 39N. We found also strong anticyclonic Eddy at 131E, 38N.

October 15, 1999:  Strong anticyclonic Eddies are developed at 131-132E, 39-40N; 133-134E, 37-38N; 136-137E, 38-39N.
 

November 15, 1999: The strong geostrophic pattern of Eddies are similar than previous month but a little bit shifting to the south .

December 15, 1999:  We found more energetic eddies in the southern part of East sea.
 

    3.3 Interannual variation of sea surface height from 1993 to 1999
 

       Sea surface height deviation data are downloaded from  Ocean ESIP . After we got the data, we subset the data for the East Sea using FORTRAN program and plot the data using GMT plotting software. Sea  surface height deviation are computed at 1 x 1 degree intervals with respect to the mean of 4-year period, 1993-1996.
 
 

  [Figure 5] Sea level annual mean deviation at each year from 1993 to 1999.

1993                                              1994

       1999
           
   1999: strongest positive deviation in southern part of the East Sea. It exceeds our color index.
 

    Because the strong horizontal front is formed along  40'N and it confines warm water region  below 40'N, the sea surface height, heat content, and upper layer thickness highly change below 40'N in East Sea.  To summarize our Figure 5 we computed the average of sea surface height deviations in Southern East Sea (35' - 40'N) in each year.
 
 

      We can find that the sea surface height ( or heat content of upper layer ) was higher in 1994, 1998 and 1999, and it was lower in 1993, 1995 and 1996.
 

   4. Discussion

      4.1 Eddy Formation

From historical Eddy formation in the East Sea, we found that The Eddy began to form at February where strong colder northerly winter wind system blew in the region and diminished in May as the onset of weaker summer wind system. In contrast, we found different condition of Eddies formation in 1999. The anticyclonic Eddies already formed in January and became to decay in February, March, April and diminished in May.

We found a new result that has never been observed before of the early anticyclonic Eddy formation  in September and continued until October, November and December in 1999, and even in January 2000. We don't have enough wind data and supporting analysis to explain this event yet and we need to analyze our data after we add Mean Sea Surface Height to the sea surface height deviation.
 

     4. 2 Interannual Sea Surface Height Variation

     What is the  oceanographic process that changes sea surface height ( or heat content of upper layer ) in Southern East Sea? We examined the deviation of sea surface height. we cannot understand the oceanographic process from the deviation. We need mean sea surface height.
 

 We can make the mean sea surface height from long term measurement of altimetry data, numerical model, and observation data. US Navy generates Mean Sea Surface Height from long term altimeter data, climatological data and Numerical model.

      [Figure 6]  (a) Sea Surface Height Deviation (b) Sea Surface Height Residual on April 21, 2000. From  Altimetry Plots: East/Japan Sea
 

  A mean steric heights of surface relative to 500 m were calculated using long term hydrographic observation data from KODC and JODC (Figure 7-a ) and we used the mean steric height as a mean sea surface height (Figure 7-b).
 
 

  [Figure 7] (a) Climatological mean temperature at 100 m depth (b) Estimated Mean Surface Height from climatological density data (c) Actual sea surface height distribution from 1993  to 1999.

     We can notice that the movement of the front in the middle of the East Sea changes the oceanographic conditions such as upper layer thickness, heat content, and sea surface height. When Warm Water Region expanded to the North in 1994, 1998 and 1999, upper layer thickness increased, heat content of upper layer increased, and sea surface height rose. When Cold Water Region expanded to the South in 1996, upper layer thickness decreased, heat content of upper layer decreased, and sea surface height fell.

  We need to examine what cause the interannual variation of sea surface height in Southern East Sea in the future.  There are several factors that can affect sea surface heights of the East Sea : First one is the change of inflow and outflow through the straits. If the amount of warm water inflow through Korea and Sushima straits increase, the sea surface height in the East Sea will increase. Second one is wind. If a strong winter monsoon blow from North West, the warm water region becomes smaller and sea surface height will decrease.  Air-sea interaction is the third one. If the air cools the sea surface in cold winter, the water temperature become lower and sea surface height will decrease.
 

    5. CONCLUSION

       We compared the annual mean deviations of sea level data from tide gauge with the air-borne TOPEX/POSEIDON altimeter data from 1993 to 1998 at Ulleung Island.
They have a good linear relation and their ranges are from -4 cm to +8 cm. Ulleung Island has higher sea level in 1995 and 1998.

   Eddies formation in 1999 was examined using TOPEX/ERS2 Altimeter data. The Eddies were formed in September and continuously developed to stronger anticyclonic Eddies in October, November and December. This is a new result compare to the previous years when Eddies began to form in February and diminished  in May.

   Interannual variation of sea surface height from 1993 to 1999 in the East Sea  was studied by TOPEX/POSEIDON Altimeter data. Interannual Variability of Sea Surface Height in the East Sea showed that Sea Surface Height was higher in 1994, 1998 and 1999, and it was lower in 1993, 1995 and 1996. When the warm water region expands to the North,  heat content of upper layer increases , and sea surface height  rises.  When cold water region expands to the South,  heat content of upper layer decreases , and sea surface height  falls.
 

  Acknowledgment

        We would like to express our sincere thank to Hank Statscewich for helping us to use GMT plotting. Prof. Scott Glenn, Prof. Jim Miller, and Prof. Jennifer Francis for continuing support during the project. All friends that make this project finish.

    6.  REFERENCE

1. Byoung Ju Choi (1994): Relations between Sea Level Variations and Oceanographic Conditions around Ulleung Island, Master Thesis in Seoul National University, 54pp.

2. Delcroix, T., C., Gautier, (1987): Estimates of Heat Content Variations from Sea Level Measurements in the central and western Tropical Pacific from 1979 to 1985, Journal of Physical Oceanography, 17, 725-734.

3.  D.R. Kobayashi, J.J. Polovina (1999): UNESCO Training Module for application of TOPEX/POSEIDON satellite Altimetry to produce maps of eddies and other geostrophic features for fisheries research, National Marine Fisheries Service Honolulu Laboratory, NOAA. (http://www.nmfs.hawaii.edu/eco/unesco1.html).

4. Hahn, S.B., (1991): Tendency of Sea Water Temperature Rise and Mean Sea Level Rise in the Korean Water (unpublished).

5.  Kang, Y.Q. , B.D., Lee, (1985): On the Annual Variation of Mean Sea Level along the coast of Korea, Journal Oceanography Society Korea, 20 (1), 22-30.

6.  Kang, Y. , H., KAng, (1991): Long-Term Fluctuations of Water Temperature in the Upper 200m off the Southeast Coast of Korea, Bull. Korean Fisheries Society, 24 (6), 450-458.

7. L.L. Fu, E.J. Christensen, C.A. Yamarone Jr., et al (1994):   TOPEX/POSEIDON mission overview,
Journal of Geophysical  Research, 99, no. C12, 24369- 24381.

8. Oh, I.S., A.B., Rabinovich, M.S. Park, and R.N., Mansurov, (1993): Seasonal Sea Level
Oscillation in the East Sea, Journal Oceanography Society Korea, 28, 1-16.

9. Tomcazak, M. and J. Godfrey (1994) : Regional Oceanography : an introduction Pergamon, 422pp.

    APPENDIX

 Available TOPEX/POSEIDON DATA

   I.  NOAA Topex/Poseidon Analysis   The joint NASA/CNES satellite altimeter, Topex/Poseidon, has been operating since September 1992.  At NOAA these data are are used to form three different types of sea level time series:

        1. Topex/Poseidon Along-track Deviations : Global Topex/Poseidon sea level deviations (relative to the 3-year mean 1993-95) are computed at 1-degree latitude intervals along the satellite track. These data are available via anonymous ftp as either yearly files or one file for the entire T/P mission.

       2. Topex/Poseidon Gridded Deviations: Sea level deviations, computed along the Topex/Poseidon track as described above, are averaged in 4x1-degree (longitude x latitude) cells to construct regular grids of sea level deviation with respect to the 3-year period, 1993-1995. The data are averaged over monthly and seasonal periods. The gridded data are also available via anonymous ftp for monthly and seasonal periods.

      3. Topex/Poseidon Gridded Anomalies: The gridded deviations described above are converted to anomalies by removing the annual and semiannual harmonics. These maps show interannual changes relative to 1993-95 for monthly, seasonal, and annual time periods. The gridded data are also available via anonymous ftp for monthly, seasonal, and annual periods.

  II.  Ocean ESIP  is a member of NASA's Earth Science Information Partner (ESIP) Federation.

   1.  Uniform grid (1° x 1° x 5-day averages)   SeaSea Surface Height Residuals
   2.  Along-track grid (6.2 km spacing)    SeaSea Surface Height Residuals
 

    Sea Surface Height (SSH): Sea surface height (SSH) is the height of the sea surface above the reference ellipsoid. It is calculated by subtracting the corrected altimeter range (see above) from the altitude.

        SSH = Altitude - Corrected Range

   Residual Sea Surface: The residual sea surface will be defined here as the sea surface height minus the mean sea surface or the geoid and minus known effects, i.e., tides and inverse barometer.

    Residual Height = SSH - Geophysical Surface
                          - Tide Effects
                          - Inverse Barometer

    [ Geophysical Surface = Mean Sea Surface or Geoid ]
 

 III.  University of Texas, Center for Space Research (UTCSR)   TOPEX/POSEIDON Dynamic Ocean Topography and Sea Level Anomalies  (1992-Present)

  1. Gridded and Smoothed Data Files (1x1 deg resolution) sea level anomalies
  2. Sea Level Anomaly Images
 

 IV. Colorado University  Colorado University Near real time data Host

   1.   data host  Near Real Time SSH DATA
   2.  Gostrophic velocity viewer Goestrophic velocity
 

 Application of Altimeter data

 Aplication of TOPEX/POSEIDON Satellite Altimetry to  Simulate dynamics of larvae of the spiny lobster  in the NorthWestern Hawaiian Islands, 1995-96.     In Press, Fisheries Bulletin