Hydrochemical Variability in Southcentral Alaska

Description

This repository contains the code, dataset access instructions, and analysis for the research project:

"Analysis of Hydrochemical Variability in Southcentral Alaska Using Clustering Techniques and GIS"

The study applies clustering (K-Means, DBSCAN) and geospatial analysis (GIS, Kriging interpolation, Folium mapping) to investigate hydrochemical variability across Southcentral Alaska. It identifies distinct geochemical zones and contamination hotspots that can inform water resource management and environmental decision-making.

📖 Project Overview

• Dataset: Hydrochemical and geospatial data from Southcentral Alaska (HydroShare)
• Clustering Techniques:
  • K-Means → 5 distinct geochemical clusters
  • DBSCAN → 2 dense clusters + anomalies (possible contamination hotspots)
• Spatial Analysis:
  • Kriging interpolation for unsampled regions
  • Folium maps for interactive visualization
• Goal: Combine clustering with GIS to reveal geochemical zones and contamination hotspots

📊 Dataset

The dataset is publicly available on HydroShare under CC BY 4.0 license:

Coombs & Carling (2025). Water chemistry from Southcentral Alaska glacial watersheds.

🔗 DOI: 10.4211/hs.68cfd9f523794370bf1b750d48f05a90

Download Instructions

To download the dataset into the data/ folder, run:

python data/download_data.py

🚀 Usage

1. Clone the repository

git clone https://github.com/abhinavflac/alaska-hydrochemical-clustering.git
cd alaska-hydrochemical-clustering

2. Install dependencies

pip install -r requirements.txt

3. Download dataset

python data/download_data.py

4. Run the notebook

jupyter notebook notebooks/research.ipynb

Results Summary

The analysis revealed distinct geochemical patterns and identified regional variations across Southcentral Alaska. K-Means clustering successfully segmented the region into 5 geochemical zones, while DBSCAN helped highlight potential contamination hotspots.

Interactive Map

An interactive Folium map visualizing hydrochemical zones:

• 📍 View Map on Figshare

Or open locally:

results/interactive_map.html

🛠️ Technologies Used

• Python: Core programming language
• Scikit-learn: Machine learning algorithms (K-Means, DBSCAN)
• Pandas & NumPy: Data manipulation and analysis
• Folium: Interactive geospatial visualization
• Kriging: Spatial interpolation techniques
• StandardScaler: Data normalization

📜 License

• Code & Results licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)
  • You may share this work with attribution
  • You may not modify it
  • You may not use it commercially
• Dataset licensed separately under CC BY 4.0 (HydroShare)

Citation

Coombs & Carling (2025). Water chemistry from Southcentral Alaska glacial watersheds. HydroShare. DOI: 10.4211/hs.68cfd9f523794370bf1b750d48f05a90