STICr: An open-source package and workflow for processing and analyzing data from Stream Temperature, Intermittency, and Conductivity (STIC) loggers

Abstract

Non-perennial streams constitute over half the world’s stream miles, and require hydrologic characterization to understand their flow regimes and impacts on ecosystems and society. Stream Temperature, Intermittency, and Conductivity (STIC) loggers are a widely used tool for studying non-perennial streams because they provide a relatively inexpensive and robust method for characterizing flow presence or absence. However, raw data downloaded from STIC loggers is not immediately suitable for analysis or integration with other datasets and must be processed to generate a usable dataset including temperature, conductivity, and interpreted classification of “wet” or “dry” readings at each timestep. To facilitate rapid, reproducible, and methodologically consistent analyses with STIC data, we present an open-source package written in the R language (STICr) and associated workflow to provide a standardized framework for tidying and processing data from STIC loggers. STICr features include functions to tidy data, develop and apply calibration curves to convert logger output to specific conductivity, classify data into wet/dry readings, and perform quality checks on resulting output data. Using STICr, we demonstrate a reproducible workflow that serves as a project-wide data pipeline for organizing and processing data from over 200 STIC loggers spanning multiple watersheds, years, and research groups. Given the importance of methodologically consistent inter-site and inter-regional comparison in hydrology, as well as a need for increased computational reproducibility in the discipline, we believe that STICr and the associated reproducible workflow represents an important advance for stream intermittency science.