INTRODUCTION OF JAPANESE OCEAN FLUX DATA WITH USE OF REMOTE SENSING
OBSERVATIONS (J-OFURO) VERSION 2
Masahisa Kubota (1),Hiroyuki Tomita (2), Kunio Kutsuwada (1), Masatoshi Akiyama (1), Shinsuke Iwasaki (1)
,(1)
School of Mar.Sci. and Tech., Tokai Univ., 3-20-1, Shimizu, Shizuoka, Shizuka, JAPAN, 424-8610,
Email:Kubota@mercury.oi.u-tokai.ac.jp
(2)
Research Institute for Global Change, Japan Agency for Marine and Earth Science Technology, Yokosuka, Japan,
Email:tomitah@jamstec.go.jp
1.
INTRODUCTION
The ocean actively exchanges heat, water and
momentum with the atmosphere at the ocean surface.
The exchanged heat, water and momentum are
transported by ocean and atmospheric circulations on a
global scale. Since the exchange and transport processes
play important roles in the global climate, the estimation
of the fluxes between the atmosphere and the ocean and
those transports by the ocean and atmospheric
circulations are critical for understanding the mechanism
of the global climate. However, it is difficult to globally
estimate the fluxes and transports by using in situ
observation data such as ship observation data because
they are extremely sparse in time and space. Recently,
we can obtain considerably homogeneous data with high
resolution using analysis and satellite data. Therefore, it
is considered that analysis and satellite data are suitable
for obtaining accurate globally covered fluxes between
the ocean and the atmosphere.
We have constructed ocean surface flux data sets mainly
using satellite data. The data set named Japanese Ocean
Flux data sets with Use of Remote sensing Observations
(J-OFURO) has been provided to scientists in the world
since 2002 [1] and has been used in many research
studies. Recently, new surface flux data sets have been
constructed in J-OFURO, thereby upgrading it to version
2 (J-OFURO2). Version 2 has many improvements over
version 1 for the estimation of surface fluxes. Here, we
introduce the J-OFURO Version 2.
2.
MOMENTUM FLUX
We have constructed gridded products of surface
wind/wind stress over the world ocean using satellite
scatterometer (QuikScat/SeaWinds). Original swath data
by the QuikScat/SeaWinds are referred to [2].
Construction procedures in detail are given by [1] and
[3]. In construction of surface wind/wind-stress vector
products with spatial resolution of 1°x 1° grid over the
almost entire world oceans (60°N-80°S) and daily in
time from the scatterometer data which have
inhomogeneous distribution in space depending upon the
orbital motion of the satellite, we adopt an averaging
method using a weighting function varying with time and
space. The weighting has non-zero values within a radius
of influence around each calculated grid point and
magnitudes of decreasing monotonously with the
distance between the grid point and each datum point and
between maximum (=1.0) on the grid point and zero
along the radius. The radius of influence is a parameter
of depending on the spatial coherency in wind field with
selected time resolution, so generally is taken to be larger
in the zonal direction than in the meridional one. We use
the radii of influence of 300km and 150km for the zonal
and meridional, respectively, directions. The weighting
function also depends upon time and has its Gaussiantype character with a maximum at a noon of each
calculated day and non-zero values for three days
including the previous and subsequent days. By taking
each scatterometer wind as the wind vector at 10-m level
above the sea surface, we calculate wind-stress vector
from the conventional bulk formula. The drag coefficient
depending only on wind speed based on [4] is adopted.
The swath data (level 2B) of the QuikScat/ SeaWinds
based on the new processing software used for the
construction of our new product has been provided by
the National Aeronautics and Space Administration
(NASA) Physical Oceanography Distributed Active
Archive Center (PO.DAAC) at the Jet Propulsion
Laboratory. This data set involves the following three
points: improvement of flagging for rain contamination,
improvement of performance at high wind speeds, and
supply of level 2 data with higher spatial resolution
(12.5km) (See PO.DAAC website). Using this data set,
we reconstruct a new gridded product of wind/windstress over the world ocean. In our new product, we
construct gridded products for additional parameters:
scalar-averaged wind speed and wind-stress magnitudes
as well as the zonal and meridional components of the
wind/wind-stress.
Their reliabilities have been verified by comparisons
with in-situ measurements mainly obtained by moored
buoys on open ocean [5,6], and it has been revealed that
the QuikScat product has relatively high reliability
compared with numerical weather prediction (NWP)
products (National Center for Environmental Prediction
Reanalysis: NRA -1 and 2). Validation study was also
made for the open ocean region of the mid- and highlatitudes where there had been few moored buoys using
the Kuroshio Extension Observatory (KEO) buoy [7,8].
The results reveal that our products have much smaller
mean difference in the study areas than the NWP ones,
meaning higher reliability compared with the NWP
products.
3.
HEAT FLUX
The temporal resolution of heat flux products is
improved from a 3-day mean to a daily mean. Moreover,
the data period is extended from 1992–2000 to 1988–
2005. The radiation data in J-OFURO1 covers the region
of the eastern Indian Ocean and the western and central
Pacific Ocean between 60°N and 60°S, 80°E and 160°W
because we use only the Japanese geostationary satellite
for the estimation of radiation flux. However, in many
cases, scientists require global flux data. Therefore, we
provide the global net heat flux data, which is estimated
by our turbulent heat flux data and International Satellite
Cloud Climatology Project (ISCCP) radiation data, in JOFURO2. However, we have estimated the upward
longwave radiation by using the Merged satellite and insitu data Global Daily (MGD) SST data constructed by
JMA. The MGD SST data set has been constructed by
using infrared radiation SSTs (National Oceanic and
Atmospheric Administration/Advanced Very high
Resolution Radiometer), Microwave SSTs (Aqua/
Advanced
Microwave
Scanning
RadiometerEOS:AMSR-E) and in-situ data [9]. Also not only
turbulent heat flux data but also the related
meteorological variables such as wind speed, saturated
specific humidity and air specific humidity are provided
in J-OFURO2. The information about the related
variables must be useful for most scientists.
There are several major differences between the latent
heat flux (LHF) values of J-OFURO1 and 2. First,
multi-satellite data are used in J-OFURO2, while data
from only one Defense Meteorological Satellite Program
(DMSP)/ Special Sensor Microwave Imager
(SSM/I) sensor has been used for the estimation of
turbulent heat fluxes in J-OFURO1. This is based on the
results provided by [10], who demonstrated a remarkable
reduction in the sampling error by using multi-satellite
data, particularly, in regions with large daily variability.
In J-OFURO2, we have used all available SSM/Is (F08F15), European Remote Sensing satellite (ERS) 1/2,
QuikSCAT, AMSR-E and Tropical Rainfall Measuring
Mission Microwave Imager (TMI) for the estimation of
daily wind speeds. In J-OFURO2, we could recognize
that the usage of multi-satellite data is quite effective,
even if bias adjustments are not applied to each sensor. If
only one satellite is used, the time series shows many
gaps, resulting in less accurate observations. On the other
hand, with multi-satellite data, there are no gaps in the
time series, and the accuracy is considerably higher. On
the other hand, only available SSMIs (F08-F15) are used
for estimation of air specific humidity at present because
it is not easy to estimate accurate air specific humidity
using AMSR-E or TMI data still now. This is a future
issue for the construction of J-OFURO3. Various data
are merged by an optimum interpolation method [11]
into daily mean data with a spatial resolution of 1°.
Second, we changed gridded sea surface temperature
(SST) data from Reynolds SST data [12] to MGD SST
data provided by Japan Meteorological Agency (JMA).
This change is based on the results of [13], wherein five
types of global SST products are compared and
evaluated. Third, the bulk formula suggested by [14] was
also changed to coupled ocean-atmosphere response
experiment (COARE) 3.0 [15]. However, warm layer
and cool skin effects are not included in J-OFURO2. The
estimation method for the sensible heat flux (SHF) has
been drastically changed. In J-OFURO1, we use a
method proposed by [16] for calculating the SHF by
multiplying the LHF by the climatological Bowen ratio
derived from the European Centre for Medium Range
Weather Forecasting (ECMWF) data. However, in JOFURO2, the SHF is estimated from COARE 3.0 using
our merged wind speed, MGD SST data and NRA1 air
temperature data.
The general feature of the global distribution of the JOFURO2 LHF is similar to that of not only J-OFURO1
but also other products such as Goddard satellite-Based
Surface Turbulent Fluxes (GSSTF) 2, NRA1, ECMWF
reanalysis (ERA) 40 and OAFlux. However, the
amplitude in J-OFURO2 is relatively less than that in JOFURO1, particularly in mid-latitudes called as ocean
desert. We have compared the J-OFURO2 LHF with the
in situ LHF data observed by the KEO buoy. In June
2004, the KEO buoy was deployed in the Kuroshio
Extension recirculation gyre at 144.6°E, 32.4°N by
NOAA/ Pacific Marine Environmental Laboratory
(PMEL) to monitor air-sea heat, moisture and
momentum fluxes and upper-ocean temperature and
salinity. Figure 1 shows the daily mean time series of the
J-OFURO2 and the differences between J-OFURO2 and
KEO from June 2004 to December 2007[17]. Overall,
the J-OFURO2 LHF and SHF show good agreement
with the KEO observations. The bias and the root-meansquare (RMS) error for a daily mean value of the JOFURO2 LHF (SHF) are 3 (-4) W m–2 and 40 (11) W
m–2, respectively [17]. References [17] and [18]
showed that the atmospheric reanalysis products such as
NRA1/2 has large bias and RMS. For example, the bias
and the RMS for NRA1 LHF (SHF) are 26 (4) W m-2
and 47 (17) W m-2, respectively. Therefore, the both of
J-OFURO2 LHF and SHF are considerably more
accurate than atmospheric reanalysis products.
It is well known that many famous many warm eddies
exist in the Kuroshio and Kuroshio Extension regions.
Interestingly, we can find a fine-scaled distribution of
LHF associated with that of the sea surface anomaly
observed by the TOPES/Poseidon altimeter. Such
characteristics cannot be found in other LHF fields such
as J-OFURO1, OAFlux and ERA40. This feature
suggests that the ocean plays an active role in the
atmosphere over the meso-scale eddies. As mentioned
earlier, the use of multi-satellite data improves the
accuracy of turbulent heat fluxes. Moreover, the use of
multi-satellite data contributes to the representation of
the fine structure of heat fluxes over the
Kuroshio/Kuroshio Extension Regions.
4.
SUMMARY
We described our new surface flux products, J-
OFURO2 and showed the good accuracy compared
with other products. J-OFURO2 data is available for
scientific use and can be freely obtained from the
website http://dtsv.scc.u-tokai.ac.jp/jofuro/dataset_information.html. J-OFURO2 data are
provided by various formats such as NetCDF,
Binary, ASCII and so on. Also detailed information
about J-OFURO2 can be found in this website.
Figure.1 Daily mean time series of (a) J-OFURO2 and (b) J-OFURO2 minus KEO differences for LHF (black) and,
SHF (red). A positive value indicates an upward heat flux [17].
Figure 2. Monthly-mean values of sea surface height anomaly derived from altimeter data and LHF distribution by JOFURO2, OAFlux and ERA40 in the Kuroshio and Kuroshio Extension regions on November, 2000.
REFERENCES
1.Kubota, M., Iwasaka, N., Kizu, S., Konda, M. & Kutsuwada,
K. (2002). Japanese Ocean Flux Data Sets with Use of
Remote Sensing Observations (J-OFURO), J. Oceanogr.,
58, 213-225.
2. Jet Propulsion Laboratory. (2001) QuikSCAT science data
product user’s manual, overview and geophysical data
products, version 2.2, December 2001, Jet Propulsion
Laboratory, 91 pp.
3. Kutsuwada, K., (1998). Impact of wind/wind stress field in
the North Pacific constructed by ADEOS/NSCAT data, J
Oceanogr. 54, 443- 456.
4. Large, W.G. & Pond, S. (1982). Sensible and latent heat
flux measurements over the ocean, J. Phys. Oceanogr., 12,
464-482.
5. Kasahara, M., Kutsuwada, K., Aoki K., &Takeda S. (2003).
Construction and validation of gridded surface
wind/wind-stress product over the world ocean using
satellite scatterometer data, J. School Mar. Sci. Tech.,
1(1), 79-92. (in Japanese with English abstract)
6. Kutsuwada, K., Kasahara M., & Aoki, K. (2004). Gridded
surface wind-stress product over the world ocean
constructed by satellite scatterometer data and its
comparison with NWP products, Proc. Pan Ocean
Remote-sensing Conference 2004, 68(2), 348-354.
7. Kutsuwada, K, Morimoto N., & Koyama, M. (2006).
Validation study of gridded product of surface wind/
wind-stress derived by satellite scatterometer data in the
western North Pacific using Kuroshio Extension
Observatory Buoy, Proc. Pan Ocean Remote-sensing
Conference 2006.
8. Koyama, M., Kutsuwada, K., &Morimoto, N. (2008). Impact
of Reconstructed Gridded Product of Global Wind/Windstress Field derived by Satellite Scatterometer Data, Proc.
Pan Ocean Remote-sensing Conference 2008.
9. Kurihara, Y., Sakurai T. & Kuragano, T. (2006). Global
daily sea surface temperature analysis using data from
satellite microwave radiometer, satellite infrared
radiometer and in-situ observations, Weath. Bulletin, 73,
special issue, s1–s18. (in Japanese)
10. Tomita, H. & Kubota, M.(2009). Sampling error of daily
mean surface wind speed and air specific humidity due to
sun-synchronous satellite sampling and its reduction by
multi-satellite sampling. Int. J. Remote Sens., (in press )
11.Kako, S. & Kubota, M. (2006). Relationship between El
Nino event and interannual variability of significant wave
height, Atmosphere-Ocean, 44, 377-395.
12.Reynolds, R. W. & Smith, T. M. (1994). Improved global
sea surface temperature analyses using optimum
interpolation. J. Climate, 7, 929-948.
13.Iwasaki, S., Kubota M. & Tomita, H. (2007). Intercomparison and evaluation of global sea surface
temperature products, Int. J. Remote Sens., 29, 62636280.
14.Kondo, J. (1975). Air-sea bulk transfer coefficients in
diabatic conditions. Bound. –Layer Meteor., 9, 91-112.
15.Fairall, C. W., Bradley, E. F. Hare, J. E. Grachev, A. A.
& Edson, J. B. (2003). Bulk parameterization of air sea
fluxes: Updates and verification for the COARE
algorithm, J. Climate, 16, 571–591.
16. Kubota, M. & Mitsumori, S. (1997). Sensible heat flux
estimated by using satellite data over the North Pacific.
In: Remote Sensing of Subtropical Ocean, ed. By C.T.
Liu, Elsevier, 127-136.
17. Tomita, H., Kubota, M., Iwasaki, S., Konda, M., Cronin,
M. F. & Ichikawa H. (2009). An assessment of surface
heat fluxes from J-OFURO2 at the KEO/JKEO sites, J.
Geophys. Res., (in press).
18. Kubota, M., Iwabe, N., Cronin, M.F. & Tomita, H. (2007).
Surface heat fluxes from the NCEP/NCAR and
NCEP/DOE reanalyses at the KEO buoy site, J. Geophys.
Res., 113, C02009, doi:10.1029/2007JC004338.