The Noah with Multi-Physics options (NoahMP) land surface model

This README describes how to build and run the NoahMP land surface model.

Model developers can also find information about the development instructions and technical documents.

Build

Normally you can build the model in three steps: 1) run configure, 2) check makefile.in and make changes if necessary, and 3) make. A successful build will produce an executable named run/ldas.exe.

Run `configure`

Run configure to select an option that matches your operating system (Linux, Mac OS X Darwin), compiler (GCC/Gfortran, Intel, PGI), and parallel environment (seq for sequential and dm for MPI).

./configure

The script will generate a file named makefile.in.

Notes: If you are going to create model output files that are more than 2Gb, you should consider using NetCDF large file support function. To activate this, one must set the environment variable WRFIO_NCD_LARGE_FILE_SUPPORT before running config. In a bash environment, do

export WRFIO_NCD_LARGE_FILE_SUPPORT=1

Check `makefile.in` and make changes if necessary

makefile.in details the configurations of the compiler and libraries. Check whether they fit you well.

Pay attention to the NETCDF_INC and NETCDF_LIB variables. They specify the locations of the NetCDF development headers and runtime libraries, respectively. Make sure the NetCDF Fortran 90 header netcdf.mod can be found in NETCDF_INC, and the NetCDF Fortran library libnetcdff.[so,dylib,dll] and C library libnetcdf.[so,dylib,dll] can be found in NETCDF_LIB (the library suffix varies with platforms: so for Linux, dylib for Mac OS X, and dll for Windows). If not, locate them and edit the two variables' values accordingly.

Make

If makefile.in fits you, type make at the command prompt:

make

If compilation is successful, an executable named ldas.exe will be created under the run directory.

Run

The NoahMP model uses a namelist file called ldas.namelist as well as some additional parameter files (.TBL files), which are located under the run directory. Users need to copy those files to the directory where the model is going to be executed.

For a NoahMP cold start run (i.e. not from a restart file), the user needs to turn off the flag from_restart=.false. in ldas.namelist and provide an initialization file to the INIT_FILE option.

For running NoahMP from a restart file, the user needs to switch on the flag from_restart=.true. in ldas.namelist and provide an existing restart file to the option RESTART_FILE. Running from a restart condition is common when the land surface has been 'spun-up' by running NoahMP in an offline or 'uncoupled' capacity.

Model Development

This repository adopts the Git Flow [https://nvie.com/posts/a-successful-git-branching-model] branching convention and workflow. You should base your work on the develop or module-* branches and pushes your work back to the feature-* branches.

Normally you should follow the instructions below.

git clone https://github.com/esmwg/ldas.git -o esmwg to clone this repository. The -o flag sets esmwg as the remote name of this GitHub repository.
Optionally, git checkout -b develop esmwg/develop to create a local develop branch that tracks the remote develop branch on GitHub. You may replace develop to module-* if it is the branch that you would like to base your work on. The same applies to the following instructions.
git checkout -b <your_own_branch> develop to create your own branch based on the local develop branch.
Edit and commit on your own branch.
git pull esmwg develop:develop pulls the remote develop updates into the local develop branch. During your work, others may have pushed their work to the GitHub repository.
git merge develop merges the updates into your own branch.
git push eswmg <your_own_branch>:feature-<your_branch_on_GitHub> pushes your work to a GitHub branch prefixed by "feature-". The "feature-" prefix is mandatory. It is also much easier to identify yourself in the branch name, such as "feature-zh-correct-a-bug".
Open the GitHub website, create a Pull Request to merge your own branch into the develop branch or a module- branch.
You can still edit your local branch after the Pull Request. Just push again will automatically update the Pull Request.
If the Pull Request is accepted, delete your own branch using git branch -d <your_own_branch>. If the Pull Request is rebased or squashed remotely, you should force the deletion: git branch -D <your_own_branch>.

Common Git settings are below for your reference.

git config user.name <your_name>
git config user.email <your_email>
git config credential.helper cache

Model Debugging

There are two debugging methods shown as follows. You can combine the two methods or use them separately. Both methods require recompiling.

Print diagnostic information at run time

Set the environment variable HYDRO_D=1 and rebuild the model.

export HYDRO_D=1

A "1" for HYDRO_D results in NoahMP producing some runtime diagnostic information. When HYDRO_D is set to "0 "or not defined, the diagnostic information will not be produced during runtime.

Compile a debuggable executable for use by debuggers

Edit "makefile.in", and append the compiler debug options "-g" to the fortran compiler "F90".

Example

An example can be found in [https://github.com/esmwg/land-val-suite].

The example helps you run your first NoahMP simulation. Use the example as a template of your own simulations.

Porting

The NoahMP model does not presently support OpenMP. The default support platform is Linux with the Intel compiler and Linux with the GFortran compiler. However, NoahMP is fairly easy to port to other systems. The basic steps to do so are as follows:

Provide your own makefile.in under the arch directory.
Edit configure, add your own compling options in the "Compiling Option" section.
Edit configure, add the path of your own makefile.in in the "Compiling Configurations" section.
./configure
make

If there is no error, then users can compile NoahMP on the new platform.

References

Please cite the following papers in your publications:

Niu, G.-Y., Yang, Z.-L., Mitchell, K. E., Chen, F., Ek, M. B., Barlage, M., Kumar, A., Manning, K., Niyogi, D., Rosero, E., Tewari, M., & Xia, Y. (2011). The community Noah land surface model with multiparameterization options (Noah-MP): 1. Model description and evaluation with local-scale measurements. Journal of Geophysical Research: Atmospheres, 116(12), D12109. https://doi.org/10.1029/2010JD015139
Yang, Z.-L., Niu, G.-Y., Mitchell, K. E., Chen, F., Ek, M. B., Barlage, M., M., Lounguevergne, L., Manning, K., Niyogi, D., Tewari, M, & Xia, Y. (2011). The community Noah land surface model with multiparameterization options (Noah-MP): 2. Evaluation over global river basins. Journal of Geophysical Research: Atmospheres, 116(D12), D12110. https://doi.org/10.1029/2010JD015140
Zheng, H., Yang, Z.-L., Lin, P., Wei, J., Wu, W.-Y., Li, L., Zhao, L., & Wang, S. (2019). On the sensitivity of the precipitation partitioning into evapotranspiration and runoff in land surface parameterizations. Water Resources Research, 55(1), 95–111. https://doi.org/10.1029/2017WR022236

JiangJinghong/land