This repository contains the full HTCondor-enabled codebase for the project described in our paper:

"An L-Moments-Based Hypothesis Test to Identify Homogeneous Storm Transposition Regions"
Benjamin FitzGerald, Daniel Wright, Lei Yan, Alyssa Hendricks Dietrich, and Antonia Sebastian
Submitted to Journal of Hydrology, June 2, 2025
Preprint Link: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5292988

This project uses a four-stage pipeline implemented with HTCondor to create a transposition domain for a watershed of interest using a gridded precipitation dataset. Each stage is run as a distinct Condor job with its own input file, submit file, executable script, and Python code.

The pipeline includes the following stages:

1. Stage 1 – Point Precipitation to Watershed-Averaged Precipitation (1.PP2WAP)
   Transposes the watershed shape across the full domain and computes watershed-averaged precipitation at each location to create a gridded dataset.

2. Stage 2 – Annual Maxima Calculation (2.AMC)
   Calculates annual maxima from the watershed-averaged precipitation data.

3. Stage 3 – L-Moment Calculation (3.LMC)
   For each transposition, computes the composite of annual maxima and calculates L-moments for each grid cell in the composite.

4. Stage 4 – Hypothesis Test and Final Domain Drawing (4.HTDD)
   Performs a hypothesis test comparing the L-moments of transpositions to the original watershed. A final domain is drawn based on statistically similar regions.

Each stage is organized in its own subdirectory under scripts/ and can be submitted independently using condor_submit.

The latest version of SLAM is distributed under the MIT open source license: https://opensource.org/licenses/MIT