Calibration of a modified temperature-light intensity logger for quantifying water electrical conductivity

The Stream Temperature, Intermittency, and Conductivity (STIC) electrical conductivity (EC) logger as presented by Chapin et al. (2014) serves as an inexpensive (similar to 50 USD) means to assess relative EC in freshwater environments. This communication demonstrates the calibration of the STIC logger for quantifying EC, and provides examples from a month long field deployment in the High Arctic. Calibration models followed multiple nonlinear regression and produced calibration curves with high coefficient of determination values (R-2=0.995-0.998; n=5). Percent error of mean predicted specific conductance at 25 degrees C (SpC) to known SpC ranged in magnitude from -0.6% to 13% (mean=-1.4%), and mean absolute percent error (MAPE) ranged from 2.1% to 13% (mean=5.3%). Across all tested loggers we found good accuracy and precision, with both error metrics increasing with increasing SpC values. During 10, month-long field deployments, there were no logger failures and full data recovery was achieved. Point SpC measurements at the location of STIC loggers recorded via a more expensive commercial electrical conductivity logger followed similar trends to STIC SpC records, with 1:1.05 and 1:1.08 relationships between the STIC and commercial logger SpC values. These results demonstrate that STIC loggers calibrated to quantify EC are an economical means to increase the spatiotemporal resolution of water quality investigations.