STILT is freely available and we encourage others to use it. Kindly keep us informed of how you are using the model and of any publication plans. Please acknowledge the source as a citation. STILT is continuously updated and improved by the development consortium, and in some cases (as when new elements are used for the first time) we may suggest that one or more of the developers be offered participation as authors. If your work directly competes with our analysis, we may ask that we have the opportunity to submit a manuscript before you submit one that uses unpublished features. The software is updated from time to time, and it is your responsibility to ensure that your publication is consistent with the most recent version.
STILT has been compiled to run on UNIX platforms (Mac, Linux). Required software includes
dplyrpackage, for speed and data manipulation
parallelpackage, for parallel computation on a single node
rasterpackage, for raster-based spatial analysis
rslurmpackage, for parallel computation across multiple nodes
uataqpackage, for data manipulation
Two options exist to initialize a new STILT project.
This method uses R to initialize a new project.
stilt_init() in the Utah Atmospheric Trace gas & Air Quality (UATAQ) R package, which includes tools for trace gas data manipulation and analysis, is a wrapper around several system commands that do much of the heavy lifting. The
uataq R package is available on Github and can be installed in R using
if (!require('devtools')) install.packages('devtools') devtools::install_github('benfasoli/uataq')
A new STILT project can then be initialized in plain R code with
The STILT repository is 38 MB and contains source code for the model, website, and documentation.
stiltdirectory which is then renamed
permute.sodynamic link library used to apply gaussian kernels for footprint output
Since the name of the Github Repository is “stilt”, a name other than “stilt” should be chosen for projects. Otherwise, conflicts will arise when attempting to clone the master repository for initializing future projects in the same parent directory.
While the R method is preferable since it streamlines the process of initializing new projects, the same can be accompolished manually. To reproduce the results above,
Clone the repository and set the name of the project
git clone https://github.com/benfasoli/stilt mv stilt myproject
Compile and move the hymodelc executable to the
exe directory using the provided
setup script (or manually)
cd myproject chmod +x setup ./setup
Finally, edit settings in
r/run_stilt.r, being sure to specify the project name and the working directory.