UC Davis

In an effort to describe water temperatures at scales relevant to organisms living in the lower Yuba River (LYR) in California’s Central Valley, we designed and deployed a wireless sensor network in November 2018 through May 2019. Temperatures were measured along a 3 km study reach, across the channel, and into off-channel areas. To capture diel and seasonal fluctuations, sensors were sampled quarter-hourly for six months.

Chapter 1 describes the wireless sensor design, field deployment, data cleaning, and raw data. The deployment adopted event-based software on MSP430 micro-controllers with 433 MHz radio and minimized the networking duty-cycle. To address link failures, we included network storage. As the network lacked real-time clocks, data were timestamped at the destination, which – with the network design – yielded timestamp inaccuracies that were re-aligned algorithmically. We collected over six months of temperature data from 35 sensors across seven nodes. Of the packets collected, we identified 21% as being incorrectly time-stamped and were able to re-align 41% of these incorrectly timestamped packets.

In chapter 2, we use these data to consider the dominant trends in fluvial temporal dynamics. Generally, river waters warm as they move downstream. We had three research questions: At what temporal scales do fluvial temperatures vary most? Do hyporheically connected side-channels experience warming, or do connections to the mainstem limit these effects? Over short reaches of relatively cold water, is the downstream warming trend still prevalent? We created several Bayesian models, which segmented the river geographically, but were not spatially explicit. We found that models that incorporated the interactive effect of a diel and day-of-sampling effects with random walk priors yielded the best fit. Within the study reach, model results indicated that temporal effects dominated the temperature variance, with diel-effects less important than multi-day effects. We found that even with strong hyporheic connections, the sampled side channel was warmer than the main-channel. We also found only minimal downstream warming within our 3 km study reach.

In chapter 3, we describe the spatial structures within a side channel of the study reach. The primary research question addresses whether the spatial structures of river temperatures (longitudinal and lateral) are static or if they change over time, either by time of day or by day of year. We then question if the time-dependence of these spatial structures is related to river flow rate and/or surface heating, indexed by daily solar radiation. The lateral and longitudinal distances between sensors were used to generate semi-variograms, calculated for specific time-of-day and specific day-of-year time bins. We found that the spatial structures of fluvial temperatures were time varying. Diel changes in lateral structure in temperature corresponded with the diel cycle in solar radiation, while day-to-day and seasonal changes in lateral structure corresponded with changes in river flow rate. Longitudinal structure did not exhibit a coherent diel cycle, but we did observe seasonal changes in semi-variogram slope with marked longitudinal structure during low warm-water flows in November, high cold-water flows in February, and with the return of strong diurnal warming in May.