SPLINTER is a generalized additive model (GAM) framework with outlier detection and removal for time series analysis of environmental monitoring data for chemical contaminants.
The SPLINTER analyzes time series data using:
- GAM decomposition
- Change point detection
- Outlier identification and removal
Outputs can be saved automatically to a timestamped results directory.
requirements.txt– Python package dependencies for pip installationsplinter.yml– Conda environment fileSPLINTER_UserGuidelines.docx– detailed user instructionsSPLINTER_V012.ipynb– notebook for running SPLINTER on user datasets, including an example analysis ofPAHSPLINTER_V012_batch.py– batch-mode execution script for processing multiple datasetssplinter_function.py– main SPLINTER functions
We recommend using Visual Studio Code as the IDE for running SPLINTER. To set up the model environment and install all necessary libraries and packages, you can use either of the following approaches:
-
Option 1: pip
For users with an existing Python environment:
pip install -r requirements.txt
- Tested with Python 3.13
- Required packages are listed in
requirements.txt
-
Option 2: Conda
Recommended for managing multiple project environments.
Create a new environment from the provided YML file:
conda env create -f splinter.yml
- Environment file:
splinter.yml
- Environment file:
Both options allow users to run the Jupyter Notebook files interactively within Visual Studio Code.
Detailed instructions for each execution step are provided in:
SPLINTER was developed and is maintained by Matthew MacLeod, Stockholm University, with contributions from:
The code was developed as a resource for the Arctic Monitoring and Assessment Programme (AMAP).