An inexpensive method for reliable recovery of stream temperature data

Abstract Objective Water temperature is perhaps the single‐most important environmental driver of fish populations. The strong relationship between fish and water temperature allows fisheries managers to make predictions about the influence of temperature on fishes under both current and future climatic conditions. These predictions are more robust if based on year‐round and long‐term data. However, water temperature data are commonly compromised or lost altogether when data‐logging temperature sensors are damaged or go missing. In recognition of the need for reliable ways to collect long‐term, year‐round temperature data, we designed, implemented, and tested a durable but cryptic logger deployment and retrieval system. Methods We used metal housings and stakes to protect and anchor temperature loggers on the streambed and, when necessary, used a metal detector to assist with logger recovery. We then evaluated logger recovery rates across 12 years and 312 deployments at 85 sites in first‐ to ninth‐order Rocky Mountain streams and rivers. Result Although we recovered only 73% of loggers with traditional means of retrieval (e.g., GPS or photo), presumably owing to the inconspicuous nature of our metal housings and streambed anchor stakes, we recovered 96% of loggers when a metal detector was also used. Ordinal and binary logistic regression revealed that a metal detector was especially beneficial when trying to recover loggers from unfamiliar monitoring sites or those deployed for long periods of time (years). Conclusion Our methods could be replicated for a reliable and inexpensive approach to acquiring year‐round stream temperature data.