论文题目：天气尺度波强迫包络Rossby孤立子理论与阻塞机理
作者简介:
罗德海，男，1963年10月出生，1999年07月师从中国海洋大学秦曾灏教授，于2002年06月获博士学位。
摘
要
在中高纬度地区，大气环流持续性异常（如阻塞形势）通常与大范围灾害性天气有关。许多研究就已经指出，夏季东亚地区阻塞形势的活动对东亚地区特别是我国的天气以及气候有着重要的影响，由于这个原因,50年代我国气象学家就开始对北半球阻塞形势进行系统和详细的观测研究，发现了许多新的结果。与此同时，国外很多学者对阻塞形势的形成机制和活动规律以及对气候的影响进行了大量的观测和统计研究。取得了许多有意义的结果.
目前阻塞现象已成为大气科学研究中的一个非常重要的问题,并引起了世界大气科学家的广泛重视。Berggren等人(1949)最早观测到天气尺度波在激发阻塞环流中的作用,并发现阻塞流是由气旋和反气旋涡旋组成的弯流.但这种阻塞产生的机理和条件仍不清楚.
Yeh（1949）从能量的观点，首先讨论了阻塞在高纬度地区的移动和维持。对于阻塞这样大振幅的天气系统，仅用线性理论来加以研究是有局限性的。从80年代以来，由于非线性科学的发展,阻塞动力学研究得到了迅速发展，并提出了不少理论。这些理论主要包括两大类：一类是阻塞的全球理论，这种理论主要包括Egger(1978)所提出的缓慢移动的自由波与地形强迫波的非线性相互作用理论；Charney和Devore(1979)的多平衡态理论以及Tung和Lindzen（1979）的共振理论。另一类是阻塞的局地理论，这些理论主要包括
McWilliams（1980）所提出的Modons(偶极子)
理论以及Malguzzi和Malannotte-Rizzoli所提出的KdV型Rossby孤立子理论。
McWilliams(1980)最早用Modons解来描述这种偶极子型阻塞。Modons解在南北方向是反对称的，而在东西方向是对称的。尽管他与实际的观测的偶极子阻塞有一定的一致性,然而不难发现它们之间的差别仍是很大的,因此要想完全解释实际大气中所观测到的偶极子阻塞,必须发展强迫的Modons理论。Haines和Marshall(1987)研究了天气尺度波强迫的Modon解并表明天气尺度波的强迫确实能够维持偶极子阻塞，然而天气尺度波是怎样形成和维持偶极子的？他们的研究不能给出圆满的回答。Malanotte—Rizzoli（1984）指出，McWilliams(1980)所提出的modons解的条件很难满足，为此他们发展了KdV型的Rossby孤立子理论，并试图用他来解释大气中所观测到的偶极子阻塞。尽管这种
KdV孤立子结构的外形与偶极子阻塞相像，然而我们注意到这种孤立子存在严重的缺陷。首先他必须要求满足长波近似，其次他必须要求基本流有很强的水平切变。Butchart等(1989)已经指出这些条件对于偶极子阻塞来说很难满足。
McWilliams(1980)所提出的modons解必须要求位涡与流函数呈线性关系，然而Butchart等(1989)的计算表明在阻塞区域位涡与流函数的线性关系严格来说并不成立。因此不难推断用Modons理论来解释大气中的偶极子阻塞是值得怀疑的。而对于KdV型的Rossby孤立子只要他满足长波近似，因此它的偶极子结构必须限制在窄的β通道内。但是，由于偶极子阻塞有大的经向尺度，因此长波近似严格来讲也不成立。在中高纬度地区。所观测到的偶极子的经向尺度一般为
。在
地区，若取
(有量纲量为
)对于波数1的Rossby波,有
；对于波数2的Rossby波有
;而对于波数3的Rossby波有
。由此我们可以推断，对于波数1，使用长波近似基本上是合理的，而对于
的Rossby波则不能使用长波近似。人们对阻塞进行谱分析发现阻塞主要是波数为
波的超长波（李崇银，2000），因此我们认为用KdV型Rossby孤立子来解释大气中的偶极子阻塞是不合适的。Malguzzi和Malanotte-
Rizzoli(1984)指出，对于KdV型Rossby孤立子，要获得偶极子阻塞结构，必须要求基本流有很强的水平切变，而强的水平切使得基本流容易产生正压不稳定，并导致偶极子阻塞结构的破坏。因此，可以认为KdV型Rossby孤立子理论并不是一个很好的理论模式。考虑到大气中存在大振幅的Rossby波，Benney(1979)和Yamagata(1980)首先导出了大气中非线性Rossby波所满足的的非线性Schrodinger方程，并给出一些简单的孤立子解.然而，他们都没有给出包络Rossby孤立子的流场结构，也没有讨论包络Rossby孤立子的具体应用。罗德海和
纪立人（1989）则首先将包络Rossby孤立子与阻塞形势联系起来，提出偶极子阻塞形成的包络Rossby孤立子理论。
本学位论文发展了作者以前提出的阻塞形成的强迫包络Rossby孤立子理论,并从理论上探讨了地形,热源和斜压性在天气尺度波激发阻塞中的作用,结果发现相对于地形,
热源和斜压性而言,天气尺度波对阻塞的激发似乎起主要作用,而地形和热源起次要作用.
主要结果如下:
(1)
在阻塞形成前期,上游天气尺度波起正反馈的作用,而在阻塞崩溃期间上游天气尺度波对阻塞起负反馈的作用.这种正负反馈决定了阻塞的生命周期。
(2)
先期存在的天气尺度波自身非线性相互作用的行星尺度分量对阻塞的发展起关键作用,而天气尺度波与阻塞行星波之间的相互作用主要控制天气尺度波的变化(即阻塞的弯曲).
(3)
大尺度地形(海陆地形对比)主要决定阻塞的非对称性,并使得阻塞期间分裂的天气尺度波的北支强于南支.从而解释了阻塞期间风暴路径以北支为主.
(4)
在阻塞的形成过程中,阻塞的发展主要由天气尺度波所引起的行星尺度偶极子分量的放大来决定,而与单极子分量关系不大.
(5)
先期存在的天气尺度波所产生的北负南正的涡动强迫是阻塞发展的必要条件,而弱西风的气旋切变是阻塞产生的一个有利的环流背景.
此外,本文还使用数值试验验证了上面的一部分理论结果.可以说明天气尺度波强迫的包络Rossby孤立子理论是一个非常成功的阻塞理论.
Abstract
In the mid-high latitudes, the persistent anomalies
(such as blocking flow) of atmospheric general circulation are usually related
to the drought and flood weathers over a wide area. Many studies have indicated
that the activity of blocking patterns in the summer over eastern Asia has an
important impact on the weather system and short-term climate in our country. In
1950’s, many scientists in our country began to make the detailed and
systematic study of blocking circulation over Northern Hemisphere and obtained
many results. At the same time, many foreign scientists also made numerous
observational and diagnostic studies of blocking patterns, and many important
results were obtained. At present, blocking phenomenon is an important problem
for atmospheric sciences, which has attracted many atmospheric scientists and
was widely investigated. Berggren et al (1949) found the role of transient
synoptic-scale eddies in exciting blocking and observed that blocking flow is a
strong meander westerly flow comprising of several isolated cyclonic and
anticyclonic vortices. But the physical mechanism of such a blocking and its
onset condition remain unclear. Yeh(1949) discussed the maintenance and movement
of blocking over high latitude region using linear energy dispersion. It is
unreasonable to investigate large-amplitude weather system using linear theory.
From 1980s, great progresses of blocking dynamics were gained quickly with the
development of nonlinear theories and many theories were put forward. These
theories are divided into two categories: the first one is the global theory of
blocking, which mainly consists of the theory of nonlinear interaction between
slowly-moving free waves and topography-forced waves by Egger(1978), the
multiple equilibria theory by Charney & Devore(1979) and the resonant theory
by Tung & Lindzen（1979）.
Another is the local theory of blocking, which consists of Modons theory by
McWilliams（1980）
and the KdV-type Rossby soliton
theory by Malguzzi & Malannotte-Rizzoli (1984). McWilliams (1980) first
applied the Modons solution to dipole block. Modons solution is anti-symmetric
in the north-south direction, but symmetric in the west-east direction. Although
it has some similarities with observed dipole blocks, they are quite different.
So the forced Modons theory must be developed in order to fairly explain the
observed blockings in the real atmosphere. Haines and Marshall (1987) studied
eddy-forced Modons solution and found that eddy forcing can indeed maintain
dipole block. But how the synoptic-scale waves reinforce and maintain dipole
block remains unclear. Malguzzi and Malanotte- Rizzoli（1984）pointed
out that the conditions of Modons solution proposed by McWilliams(1980) were not
easily satisfied. For this reason they developed the KdV-type Rossby soliton
theory to explain observed dipole blocks. Although the KdV-type soliton looks
like dipole block, there exist serious drawbacks. First it must satisfy the
long-wave approximation; second, the strong horizontal shear of basic flow must
be required to make the soliton solution similar to observed block. However
Butchart et.al (1989) demonstrated that the above conditions are not easy to be
achieved for observed blocks.
The
Modons solution of vortex pair block proposed by McWilliams (1980) requires that
potential vorticity and streamfunction must be linear relationship. However, the
case calculation of Butchart et al (1989) indicated thatsuch a linear
relationship between potential vorticity and streamfunction is not allowed
strictly. Thus, it can be concluded that using the modons solution to explain
vortex pair block remains questionable. For KdV-type Rossby soliton, the
long-wave approximation requires that dipole structure must be confined within a
narrow channel. It is clear that the dipole structure of KdV-type Rossby soliton
is unreasonable for observed vortex pair blocks having large meridional scale.
In the mid-high latitudes, the meridional scale of observed dipole blocks is
generally
。At
, if
(
in dimensional form), one can obtain
for zonal wavenumber 1,
for zonal wavenumber 2 and
for zonal wavenumber 3。Thus,
we can conclude that using the long-wave approximation in KdV-type soliton is
reasonable for zonal wavenumber 1, but unreasonable for zonal wavenumbers 2~3.
The spectral analysis of blocking cases indicates that blocking flows mostly
consists of zonal wavenumbers 2~3(Li，2000).
It can be, thus, concluded that the KdV-type soliton seems unreasonable for the
explanation of observed dipole blocks. Malguzzi and Malanotte-Rizzoli(1984)
found that the existence of soliton dipole resembling observed vortex pair
blocks must require a strong horizontal shear of basic flow. However, the strong
shearly basic flow easily excites barotropic instability that leads to the break
up of dipole structure. This further indicates that the KdV-type soliton is not
a good model for observed blocking. Considering Rossby waves being large
amplitude, Benney(1979) and Yamagata(1980) obtained nonlinear Schrödinger
equations governed by nonlinear Rossby waves independently and presented simple
soliton solutions. Unfortunately, they did not discuss the horizontal structure
of envelope Rossby soliton and present some application。Luo
and Ji (1989) proposed envelope Rossby soliton theory of blocking by relating
envelope Rossby soliton to atmospheric blockings.
This work is to propose a new transient forced envelope Rossby soliton
theory and to examine the different role of large-scale topography, heating
contrast and baroclinicity in the onset of blocking. It is shown that the
synoptic-scale eddies play a key role in exciting and maintaining blocking
circulation, while topographic effect and heating contrast play a secondary
role.
The main conclusions were drawn as follows:
(1)
The synoptic-scale eddies play a positive feedback role before blocking
is fully established. In contrast, they play a negative feedback role during the
decay stage of blocking.
(2)
The planetary-scale projection of the self-interaction between synoptic-
scale eddies play a key role in the amplification of incipient block, while the
planetary- and synoptic-scale interaction seems to determine the change of
synoptic-scale eddies.
(3)
The topographic effect controls the asymmetry of blocking flow and causes
an important fact that the northern branch of synoptic-scale eddies around
blocking region is stronger than the southern one.
(4)
Blocking onset is strongly associated with the amplification of the
dipole component of blocking flow caused by upstream synoptic- scale eddies. The
monopole one is unimportant.
(5)
The negative/positive dipole eddy forcing induced by upstream
synoptic-scale eddies prior to block onset is a necessary precondition for block
onset, and the cyclonic shear of mean westerly wind is a favorable preblock
enviroment.
Numerical experiment with a barotropic model is carried out to confirm
some of above-mentioned results.
