Haero: A High-Performance Aerosol Library

Haero is a library that contains parameterizations that describe the dynamics of aerosols in the atmosphere. Rather than providing an aerosol package to be coupled in a specific way with a host models, it provides direct access to individual aerosol parameterizations tied to specific governing equations. This low-level approach allows an atmospheric "host model" to use its own coupling and time integration logic with these parameterizations.

The short-term goal of Haero is to provide the capabilities of the MAM4 package to E3SM's state-of-the-science cloud-resolving atmospheric model, SCREAM.

Supported Platforms

You can build and run Haero on a Mac (Intel, not M1) or Linux laptop or workstation. We also support a limited number of platforms on which the Haero model is built and tested:

• NERSC Cori
• Compy, Constance, Deception at PNNL

Required Software

To build Haero, you need:

• CMake v3.12+
• GNU Make
• reliable C and C++ compilers
• a working MPI installation (like OpenMPI or Mpich), if you're interested in multi-node parallelism.

You can obtain all of these freely on the Linux and Mac platforms. On Linux, just use your favorite package manager. On a Mac, you can get the Clang C/C++ compiler by installing XCode, and then use a package manager like Homebrew or MacPorts to get the rest.

For example, to download the relevant software on your Mac using Homebrew, type

brew install cmake openmpi

Building the Model

To configure Haero:

1. Make sure you have the latest versions of all the required submodules:

   git submodule update --init --recursive

2. Create a build directory by running the setup script from the top-level source directory:

   ./setup build

3. Change to your build directory and edit the config.sh file to select configuration options. Then execute config.sh to configure the model. If you're on a machine that requires modules to get access to compilers, etc, use source config.sh to make sure your environment is updated.
4. From the build directory, type make -j to build the library. (If you've configured your build for a GPU, place a number after the -j flag, as in make -j 8).
5. To run tests for the library (and the driver, if configured), type make test.
6. To install the model to the location indicated by PREFIX in your config.sh script, type make install. By default, products are installed in include, lib, bin, and share ѕubdirectories within your build directory.

Making code changes and rebuilding

This project uses build trees that are separate from source trees. This is standard practice in CMake-based build systems, and it allows you to build several different configurations without leaving generated and compiled files all over your source directory. However, you might have to change the way you work in order to be productive in this kind of environment.

When you make a code change, make sure you build from the build directory that you created in step 1 above:

cd /path/to/haero/build
make -j

You can also run tests from this build directory with make test.

This is very different from how some people like to work. One method of making this easier is to use an editor in a dedicated window, and have another window open with a terminal, sitting in your build directory.

The build directory has a structure that mirrors the source directory, and you can type make in any one of its subdirectories to do partial builds. In practice, though, it's safest to always build from the top of the build tree.

Generating Documentation

Documentation for Haero can be built using mkdocs. In order to build and view the documentation, you must download mkdocs and its Material theme:

pip3 install mkdocs mkdocs-material

Then, run mkdocs serve from the root directory of your Haero repo, and point your browser to http://localhost:8000.

At this time, Haero's documentation includes an extensive design document describing the design approach used by Haero, including high-level descriptions of its aerosol parameterizations.

FAQ

Building and Rebuilding

• When I run config.sh, I see an error complaining about a bad fd number! You probably typed sh config.sh to run the configuration script. It's actually a bash script. Just type ./config.sh.
• How do I "reconfigure my build"? If you want to change a compile-time parameter in your model, you must reconfigure and rebuild it. To do this, edit your config.sh and change the parameter as needed. Then rerun it with ./config.sh. After the script finishes, executing you can type make -j to rebuild the model.
• A pull request has the reconfig required label. What does this mean? A pull request with the reconfig required label has made a change to the structure of the config.sh script, so you must rerun setup <build_dir> to regenerate your config.sh script. Once you've regenerated this script, you can reconfigure and build as usual.

Testing

• Where are testing results stored when I type make test? All testing results are logged to Testing/Temporary/LastTest.log within your build directory. A list of tests that failed is also written to LastTestsFailed.log in that same directory.

Source Control and Repository

• Git thinks I have modifications in my submodules? Git submodules are annoying! Help! We agree. These warnings about "dirty" modifications are irritating and useless. You can get rid of them by setting the following config parameter for Git:

  git config --global diff.ignoreSubmodules dirty

• Why must I clone the submodules for libraries that I already have installed locally? Git submodules are annoying! Help! We agree. The submodule mechanism is a leaky abstraction and doesn't allow us to easily select which submodules should be cloned, so we just clone them all to keep things simple. We'll address this issue when a solution becomes more obvious.