[1] Chen H, Nan Z, Zhao L, et al. Noah Modelling of the Permafrost Distribution and Characteristics in the West Kunlun Area, Qinghai-Tibet Plateau, China[J]. Permafrost and Periglacial Processes. 2015. DOI:10.1002/ppp.1841.
Chen_et_al-2015-Permafrost_and_Periglacial_Processes (pdf, ~2.2 MB)
何思为, 南卓铜*, 张凌, 余文君, 2015. 用vic模型模拟黑河上游流域水分和能量能量的时空分布. 冰川冻土, 37(1): 211-225.
Cao Y., Nan Z*., Cheng G., 2015. GRACE gravity satellite observations of terrestrial water storage changes for drought characterization in the arid land of northwestern China. Remote Sensing, 7: 1021-1047.
下载 (from the site of Remote Sensing)
Cao et al. 2015 (pdf)
1. Cao Y, Nan Z*, Hu X. 2014. An estimation of groundwater storage variations from GRACE gravity satellites in the Heihe River Basin, northwestern China. Sciences in Cold and Arid Regions, 6(6): 577-586.
下载（百度云）
名称: Remove Long Path
版本: v1.0 initial release
build: 2014.12.11
南卓铜 Zhuotong Nan ([email protected])
https://nanzt.info
本工具可以移除极长路径。这些路径在Windows资源管理器中无法被删除。目前有一些收费和免费的工具，作者在本机上测试了一个目录深度达1500以上的例子，这些已有工具要不效率低下，要不在运行过程中出现错误(弹出Stop working)。本工具可以有效工作在Windows 32位和64位。
使用方法：
1. 解压缩至一个目录下
2. 在资源管理器里，打开该目录（removelongpath.exe所在目录），按shift的同时点击右键，在弹出菜单里选择“在此打开命令窗口”
3. 在命令窗口里敲入removelongpath.exe，可以看到使用方法提示。
4. 命令格式是 removelongpath [longpath] {time_interval}
[longpath]指定要移除的目录，此目录以下的任何深度的任何子目录和文件均将被删除。如果路径包括空格，请使用双引号。
{time_interval}是删除每个文件或目录的间隔，在Windows下，过快删除将导致系统拒绝后续删除操作。单位是秒，默认是1秒，可以是任何数字甚至小数，比如0.5表示半秒。
例子：removelongpath .temp 0.5
将移除当前目录下的temp目录下的任何子目录和文件。
5. 命令执行时，需要按两次y，由于命令将删除指定下的任何内容，需要格外慎重。除y或Y以外的任何按键，会取消此命令的执行。
Tips:
1. 如果没有删除干净，可以多次重复使用删除。并试着延长设置更长的间隔时间。
当前版本的md5 sum:
4f46c34a349186d1bd8b6cd5085dabdb removelongpath.exe
c4b4409f186da70fcf2bcc60d5f05489 libgcc_s_dw2-1.dll
dbda60d92e774b4acb3b1cd71f909426 mingwm10.dll
db57cd45846d1b6188412ed69f51eec0 QtCore4.dll
下载：removelongpath.b20141211.zip (百度云盘)
如需源代码，请联系作者。
^_^
新版本的qt creator 3.2.2 要求 GDB 7.6以上，且支持python。所以老版本的，也包括从mingw官网里下载过来如GDB 7.6的都不能用。
新版的Qt往往自带qt creator，直接安排就好，但老版本的，比如qt 4.8版本就需要自己配置。很费一翻劲。
qt 4.8要求mingw GCC 4.4.0，现在qt官网上都已经不提供。我以前从nokia ftp上下载来的mingw-gcc440_1.zip，可以用。但里面自带的gdb是6.8版本，不能被新版本的qt creator所用。如果不想用qt creator 3.x，那得倒退到2.0以前的版本吧（我测试过creator 2.5.2版是不能用的）。
然后从非官方的gdb for mingw下载支持python版的gdb。地址是http://sourceforge.net/projects/gdbmingw/files/?source=navbar。我的测试是目前最新的gdb2014-05023.zip就可以用。
将下来的gdb-python解压缩覆盖到 mingw的安装目录，会提示覆盖，覆盖即是。
此外还需要安装 python 2.7，从python官网下载即是，但最好是 2.7.5版本以上，我的是2.7.8。gdb好像不支持python 3.x。
然后在creator里手工配置Tools > Options > Build & Run 里的 Qt versions, Compilers, 和Debuggers，保存后，配置Kits。即可。
一个启示是如果我们想编译qt 4.x for mingw版本，qt creator可以不必是基于mingw，但gdb必须是基于mingw，且是被qt creator识别的支持python版本的。
一种简单有效的MODIS LST内插方法
Zhuotong Nan ([email protected], 南卓铜)
本软件实现了一种新的基于相似性原理的MODIS陆表温度(LST)的插值方法。这种LST数据空间插值方法利用具有相似温度变化特征的已知LST像元集合推算缺失的LST。在青藏高原的案例表明传统的地统计方法基本无法处理大范围连续的数据丢失（比如因为云），而本方法明显优于传统方法，可能较合理的得到缺失LST。此方法尤其适用于大范围山地区域。
This tool implements a new interpolation method for MODIS land surface temperature (LST) following the theory of similarity. It estimates the missing LST pixels by known LST pixel sets which bear similar characteristics of LST variation as the missing pixels. A case study on the Qinghai Tibet plateau has already been carried out, showing its obvious advantages over the traditional geostatistic methods, the latter was unable to do with a large area and temporally continuous data missing situations. This approach is especially good for a study area with a large area and mountainous terrain.
fig 1a, with this approach
fig 1b, with Kriging
Zhuotong Nan (南卓铜, [email protected])
Three files related to stream data are required to run DHSVM. One is the stream network file, which include information for each stream segment. The second is the stream mapping file, which contains stream information on a cell by cell basis. The last one is a look up table file specifying the information of each channel class. In addition, a soil depth grid is also related to the stream data.
In this document, I used the following example files.
The first three data files are in text format which can be opened with any text editors. The last is in ESRI ArcGIS Grid format that can be viewed with ArcMap or any GIS with support of this format.
See the 7-page document for details.
pdf, 186KB
Workflow of Preparing Stream Data for DHSVM
Zhuotong Nan (南卓铜, [email protected])
Preparing stream data for DHSVM is very complicated. Many AML scripts as well as Java are involved. Through the example showed below, I want to examine what the preparation really does and what happens behind the scripts. The scripts need ArcGIS workstation, which is not commonly installed. Thus I wrote this article, in the hope that without installation of ArcGIS workstation, you can see clearly the processes and if you want you can re-implement it with other languages.
I created a folder named “highplandpark” on the desktop, whose actual path is c:usersnztdesktophighlandpark. The contents in this folder include,
Those files can be found here.
In order to repeat the following steps, ArcGIS Desktop and Workstation, Java Runtime Environment JRE, shall be installed as prerequisites. I used version 10.0 for both ArcGIS Desktop and Workstation and JRE 7. My OS is Windows 7 x64.
Generated soil depth grid
Read the 18-page full document, pdf (478KB)
Two attached files