Resume this blog

Time flies. It has been more than three years since I wrote the last blog article here. I find that I have been too much focused on getting something published rather than enjoying the process of learning or coming up something new in my mind. I decide to change that, and a good way is to resume this blog. Basically, I have three purposes here: 1) record ideas that come to my mind when I read or learn something interesting, no matter they are stupid or not; 2) share ideas, if there are actually people reading articles here; and 3) practice english writing. As a foreigner, english is not my mother tongue. A good way, maybe the only way, to improve my english, is to use it. Writing something that people can at least understand is vital. If you find something in this blog that is interesting to you, please don’t hesitate to contact me. Maybe we can get something more interesting done together.

How to speed up the launch of termial under OSX...

Recently I noticed that the launch of the mac osx terminal is getting slow and slpw. After google the problem online, I found that it seems to be a problem of the log files. A solution for that is to type the following command in your terminal:

sudo rm -rf /private/var/log/asl/*.asl

For more information, refer to http://osxdaily.com/2010/05/06/speed-up-a-slow-terminal-by-clearing-log-files/

The effects of Tehuantepec and Papagayo Eddies on the Biological Production

Distribution of chlorophyll in surface ocean observed from space (SeaWiFS, 1997-2000 mean) shows some strong singles in the region where the Tehuantepec and Papagayo eddies occur. It seems that the eddies have some effects on the Biological production. Could we do some little work on that?

Can the low-frequency currents along the EPR result in low-frequency earthquake?

Recently, Dr. Thurnherr and I just published a paper on GRL: Subinertial variability in the deep ocean near the East Pacific Rise between 9° and 10°N (http://www.agu.org/pubs/crossref/2011/2011GL046675.shtml). When I was thinking about the broader impacts of our work, I came to a question that whether the low-frequency variability we observed in the deep ocean can affect the seismicity near the ridge crest. Prof. Tolstoy’s team at Lamont has written several papers mentioning that tides are related with the seismicity over mid-ocean ridges and that is partly due to the tidal lodging effects. From our study we know the low-frequency variability in the deep ocean near the EPR is related with the westward-propagating mesoscale eddies, which can also affect the bottom pressure over the ridge crest. Actually our bottom pressure data over the ridge crest indeed show some low-frequency variability. It seems natural to have this question that whether the mesoscale eddies have the same effects of tides, right? To test this hypothesis, we need to find a huge microearthquake data, and compare with the features of mesoscale eddies revealed in the satellite data.

References:
[1] D. Stroup, D. Bohnenstiehl, M. Tolstoy, F. Waldhauser, and R. Weekly. Pulse of the seafloor: Tidal triggering of microearthquakes at 9◦50′N East Pacific Rise. Geophys. Res. Lett., 34:15, 2007.
[2] D. Stroup, M. Tolstoy, T. Crone, A. Malinverno, D. Bohnenstiehl, and F. Waldhauser. Systematic alongaxis tidal triggering of microearthquakes observed at 9◦50′N East Pacific Rise. Geophys. Res. Lett., 36, 2009.

How did the mixing at the bottom of the mixed layer change when ENSO happens?

When ENSO happens, the present theory and observation suggest that the depth of the mixed layer in the eastern and western Pacific Ocean change and the upwelling will therefore be weakened or strengthened. But how about the mixing processes at the bottom of the mixed layer? Maybe we could use some TOGA data to do some research on that.

Recently, there is paper on the variation of mixing with tropical instability waves and it could be a good reference for this possible work.
[1]        J. Moum, R. Lien, A. Perlin, J. Nash, M. Gregg, and P. Wiles. Sea surface cooling at the Equator by subsurface mixing in tropical instability waves. Nature Geoscience, 266:761–765, 2009.

如何在Fedora 11上安装Intel C++ Compiler 11.0 和 Intel Fortran Compiler 11.0

1. 要在Fedora系统上安装 Intel C++ Compiler 11.0 或 Intel Fortran Compiler 11.0， 需要libstdc++.so.5，而Fedora 11自带的是libstdc++.so.6.0.10，所以我们要做的第一步是下载并安装 libstdc++.so.5。需要注意的是有两个不同版本的libstdc++.so.5（libstdc++.so.5 (libc6,x86-64) ，libstdc++.so.5 (libc6) ）。读者应当根据自己机器内核选择不同的版本。对于32位的版本，读者可以登入root帐号，利用如下命令安装：
yum install libstdc++.so.5
对于64位的版本，可从下面的地址下载相应的库文件并安装：ftp://ftp.muug.mb.ca/mirror/fedora/linux/releases/10/Everything/x86_64/os/Packages/compat-libstdc++-33-3.2.3-64.x86_64.rpm。安装时候，系统可能会提示已经存在更新的版本，在此情况下，可使用 如下命令进行强制安装：
rmp --force compat-libstdc++-33-3.2.3-64.x86_64.rmp

2. Fedora默认的SELINUX的安全模式会阻止Complier的安装，我们可以通过以下的命令解决：
#su
password:
#vim /etc/sysconfig/selinux
将“SELINUX=enforcing”改为“SELINUX=permissive”后保存退出。
#reboot

3. 从intel的官方网站下载l_cproc_p_11.1.059.tgz或l_cprof_p_11.1.059.tgz， 并用以下命令解压：
#tar -xzvf l_cproc_p_11.1.059.tgz
然后cd到该目录底下，执行安装的脚本文件：
#./install.sh
接着一路回车就行了。

4.设置环境变量：网友提供的方法是在非安装目录运行如下命令：
#source /bin/iccvars.sh [ia32|intel64|ia64]
or
#source /bin/iccvars.csh [ia32|intel64|ia64]
最后面的参数视OS的具体情况而定。

不过我实验的结果并不成功，我们可以采用直接添加路径的方法解决这个问题，如果你用的是bash的话，方法是在.bashrc文件中添加如下语句：
PATH=$PATH:$HOME/bin:/opt/intel/Complier/11.1/059/bin/intel64
这里/opt/intel/Complier/11.1/059/bin/intel64是icc和ifort所在的目录，请根据具体情况进行相应的修改。

5. 如果一切顺利的话，你现在应该可以利用icc和ifort进行编译了。Enjoy it～

参考资料： http://blog.sina.com.cn/s/blog_5a074c340100dqro.html
http://www.mdbbs.org/thread-19027-1-1.html

How to change the default term of Octave from Aquaterm to X11

Octave is a great sci software, which is very similar to Matlab. What makes it better than Matlab is the price 0. However, when the Octave is installed on Mac, the default term for figure output is Aquaterm. Although Aquaterm looks pretty good, one big disadvantage is that you can not close the windows by just typing ‘close’ in the terminal and you have to close them manually. Different from Aquaterm, X11 can work with Octave better. You can just use the normal Matlab way (type ‘close’ or ‘close all’) to shut windows down. How to make it? Quit simple, put the following words in ~/.profile, which you can find in your home directory.

export GNUTERM='x11'