Quality assurance and quality control of till compositional data using Python

Compositional data derived from tills play a pivotal role in geoscientific investigations, offering insights into geological processes, environmental conditions, and mineral exploration. However, ensuring the accuracy and reliability of these datasets is paramount for meaningful analyses and informed decision-making. To address this need, this report introduces a comprehensive suite of streamlined Quality Assurance and Quality Control (QA/QC) protocols tailored to till data. Designed to simplify QA/QC workflows—particularly for users with limited experience in coding or statistical analysis—these protocols combine simple procedures with user-friendly Python scripts. The protocols cover essential aspects of data quality assessment, including basic statistics, accuracy, precision, and variability, providing users with plots for visual interpretation and a systematic approach to evaluating the integrity and reliability of their datasets. The manuscript outlines key considerations for meticulous planning, proper data formatting, and the effective use of tools such as Microsoft Excel and Python. By following the outlined procedures and using the provided scripts, users can generate summary statistics, assess measurement accuracy, evaluate differences between routine samples and their analytical duplicates, and determine whether data variability is acceptable for mapping and statistical purposes.