Impact of 4D-variational assimilation of WOCE hydrography on the meridional circulation of the Indian Ocean

被引:26
作者
Ferron, B
Marotzke, J
机构
[1] CNRS, IFREMER, UBO, UMR 6523,Lab Phys Oceans, Brest, France
[2] Southampton Oceanog Ctr, Sch Ocean & Earth Sci, Southampton, Hants, England
基金
美国国家科学基金会;
关键词
D O I
10.1016/S0967-0645(03)00043-2
中图分类号
P7 [海洋学];
学科分类号
0707 ;
摘要
World Ocean Circulation Experiment (WOCE) hydrographic sections and a sea-surface climatology are combined with a ocean general circulation model through a 4D-variational method to analyze the meridional overturning of the Indian Ocean. The regional model is run with realistic surface forcings over year 1995 for which most of WOCE Indian Ocean sections were made. The assimilation controls the initial temperature and salinity fields, surface forcings and open-boundary velocities, temperature and salinity. When no observations are assimilated, the model shows that the deep (below 1000 m) meridional overturning is weak compared to observation-based estimates. This is a common feature of general circulation models. In contrast, after the assimilation, the model develops a deep overturning of 17 x 10(6) m(3) s(-1) at 32degreesS when a 10 x 10(6) m(3) s(-1) Indonesian Throughflow (ITF) is prescribed. The mass flux of bottom waters that moves northward below 3200 m is balanced by a southward mass flux of deep waters between 1000 and 3200 m. The deep overturning carries 10% of the total southward energy flux of 1.2 PW at 32degreesS. The intensity of this deep overturning changes only by +/-2 x 10(6) m(3) s(-1) When the annual mean ITF is zero or 30 x 10(6) m(3) s(-1). The upper circulation is less constrained by the assimilation because of the large temporal and spatial variability of this ocean and also because of limitations in the representation of the mixed layer physics during the assimilation process. Limitations in the physics of the model also are thought to be the source of the slow erosion of the deep overturning when the model is run for several years from its optimal state. (C) 2003 Elsevier Science Ltd. All rights reserved.
引用
收藏
页码:2005 / 2021
页数:17
相关论文
共 33 条
[11]   Recipes for adjoint code construction [J].
Giering, R ;
Kaminski, T .
ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE, 1998, 24 (04) :437-474
[12]   SOME NUMERICAL EXPERIMENTS WITH VARIABLE-STORAGE QUASI-NEWTON ALGORITHMS [J].
GILBERT, JC ;
LEMARECHAL, C .
MATHEMATICAL PROGRAMMING, 1989, 45 (03) :407-435
[13]   Throughflow within Makassar Strait [J].
Gordon, AL ;
Susanto, RD ;
Ffield, A .
GEOPHYSICAL RESEARCH LETTERS, 1999, 26 (21) :3325-3328
[14]  
LEDIMET FX, 1986, TELLUS A, V38, P97, DOI 10.1111/j.1600-0870.1986.tb00459.x
[15]   Inferring meridional mass and heat transports of the Indian Ocean by fitting a general circulation model to climatological data [J].
Lee, T ;
Marotzke, J .
JOURNAL OF GEOPHYSICAL RESEARCH-OCEANS, 1997, 102 (C5) :10585-10602
[16]  
Lee T, 1998, J PHYS OCEANOGR, V28, P923, DOI 10.1175/1520-0485(1998)028<0923:SCOMOA>2.0.CO
[17]  
2
[18]  
Levitus S., 1994, WORLD OCEAN ATLAS 19, V3
[19]   The global ocean circulation: a hydrographic estimate and regional analysis [J].
Macdonald, AM .
PROGRESS IN OCEANOGRAPHY, 1998, 41 (03) :281-382
[20]   Construction of the adjoint MIT ocean general circulation model and application to Atlantic heat transport sensitivity [J].
Marotzke, J ;
Giering, R ;
Zhang, KQ ;
Stammer, D ;
Hill, C ;
Lee, T .
JOURNAL OF GEOPHYSICAL RESEARCH-OCEANS, 1999, 104 (C12) :29529-29547