356 
Fishery Bulletin 117(4) 
Figure 7 
(A) Visualization of pooled locations of all 8 juvenile smalltooth sawfish (Pristis pectinata) tagged in the Peace River, 
Florida, during the time the acceleration data loggers and acoustic transmitters were deployed (May 2014 through 
November 2015). The color scale denotes the number of detections of transmitters by an array of 35 receivers in the 
river (see Figure 1). Only the first detection per hour for each individual was included. A comparison of (B) the mean 
percentage of time smalltooth sawfish were active and (C) the mean burst probability in mangrove-lined tidal creek 
habitats and habitats in the main stem of the river. Error bars indicate standard errors of the mean. An asterisk indi¬ 
cates significant differences between means (Tukey’s honestly significant different test: P<0.05). 
Huston et al., 2017; Scharer et al., 2017). The results of 
these previous studies indicate that juveniles rest in shal¬ 
low creeks during the day and are active in the main chan¬ 
nel of a river at night. Although data from the use of these 
monitoring techniques describe horizontal movements, it 
can be difficult to use acoustic data to accurately assess 
activity levels because movements are recorded during 
passive acoustic monitoring only when sawfish travel 
between receivers, not when they travel within the area 
covered by an individual receiver. In our study, data from 
ADLs directly confirmed these diel patterns. 
In addition to time of day and temperature, depth 
and tide also significantly affected activity. Smalltooth 
sawfish were more active in the middle range of depths 
observed during ADL deployments and in the lower half 
of the tidal cycle. Decreased activity in the shallowest 
depths likely reflects sawfish refuging in the shallowest 
habitats. The decrease in activity at depths >1 m by fish 
could be a method of predator avoidance, given that bull 
sharks are more common in these deeper areas (Heupel 
et al., 2010), or it could indicate a lie-in-wait hunting 
strategy in deeper water. Alternatively, smalltooth saw¬ 
fish could use deeper areas only for transiting through 
different habitats, as suggested by May et al. (2019), and 
could forage more actively in relatively shallow habitats 
of approximately 1 m. The relationship between activity 
levels and tidal stage observed in our study has not been 
indicated by results of previous studies (e.g., Hollensead 
et al., 2016). In our study, increased activity at lower tide 
heights could have been due to lower refuging potential 
at low tide because some shallow mangrove habitats that 
sawfish use as refuges may not be available at low tide 
(Simpfendorfer et al., 2010), forcing individuals to leave 
refuge locations and become more active as the tide drops. 
This pattern emerged despite a relatively small daily 
change in tide height, with daily tidal variance <1 m. 
Time of day was also an important predictor of burst 
activity, with the greatest number of burst events observed 
during evening and night hours. However, although indi¬ 
viduals generally adhered to these patterns, some forag¬ 
ing events also occurred during the day, timing that has 
been confirmed by using baited-rod-and-reel-derived public 
encounter data (Poulakis and Seitz, 2004), indicating that 
smalltooth sawfish opportunistically feed during the day 
and at night. Time of day was the only factor included in 
the final model describing burst frequency; however, burst 
probability was best described by age class, with YOY 
more likely to burst than individuals age 1 or older. Small¬ 
tooth sawfish are known to double in length by the end of 
their first year (Scharer et al., 2012); therefore, YOY may 
need to feed more often, on smaller prey items, than older 
individuals to achieve these growth rates. In addition, 
YOY individuals could be less experienced feeders and 
are under higher predation pressure in comparison with 
individuals age 1 or older. As a result, there may be more 
unsuccessful feeding attempts and a higher frequency of 
predator escape behavior (or startle responses) in YOY 
fish, contributing to their higher rate of burst activities. 
In addition to differences in burst activity between age 
classes, there were also significant differences between the 
