Quantifying stream thermal regimes at multiple scales: Combining thermal infrared imagery and stationary stream temperature data in a novel modeling framework. Issue 1 (7th January 2015)

• Record Type: Journal Article
• Authors: Vatland, Shane J.; Gresswell, Robert E.; Poole, Geoffrey C.
• Is Part Of: Water resources research. Volume 51:Issue 1(2015:Jan.)
• Page Start: 31
• Page End: 46
• Publication Date: 2015-01-07
• Subjects: stream temperature -- thermal infrared -- TIR -- stream thermal regime -- stream temperature model -- spatial variation; Hydrology -- Periodicals; 333.91
• DOI: 10.1002/2014WR015588
• Languages: English
• ISSNs: 0043-1397

Abstract

Abstract: Accurately quantifying stream thermal regimes can be challenging because stream temperatures are often spatially and temporally heterogeneous. In this study, we present a novel modeling framework that combines stream temperature data sets that are continuous in either space or time. Specifically, we merged the fine spatial resolution of thermal infrared (TIR) imagery with hourly data from 10 stationary temperature loggers in a 100 km portion of the Big Hole River, MT, USA. This combination allowed us to estimate summer thermal conditions at a relatively fine spatial resolution (every ∼100 m of stream length) over a large extent of stream (∼100 km of stream) during the warmest part of the summer. Rigorous evaluation, including internal validation, external validation with spatially continuous instream temperature measurements collected from a Langrangian frame of reference, and sensitivity analyses, suggests the model was capable of accurately estimating longitudinal patterns in summer stream temperatures for this system (validation RMSEs < 1°C). Results revealed considerable spatial and temporal heterogeneity in summer stream temperatures and highlighted the value of assessing thermal regimes at relatively fine spatial and temporal scales. Preserving spatial and temporal variability and structure in abiotic stream data provides a critical foundation for understanding the dynamic, multiscale habitat needs of mobile stream organisms. Similarly, enhanced understanding of spatial and temporal variation in dynamic water quality attributes, including temporal sequence and spatial arrangement, can guide strategic placement of monitoring equipment that will subsequently capture variation in environmental conditions directly pertinent to research and management objectives. Key Points: A statistical model using diverse data accurately quantified stream temperature Evaluating stream temperature at fine scales revealed important heterogeneity Temperature data continuous in space and time is critical to guiding monitoring