Abstract In several papers, we have attempted to discover the degree to which winds, river flow, and tides affected the area of a river plume by considering 19 years of daily images of the plume of the Fraser River in the NE Pacific. To estimate tidal effects, we use stepwise procedure in which seasonal river‐flow related variability is first removed, and then a tidal signal is estimated. Cao and Yu noticed that our procedure may have removed a small amount of tidal variability in addition to the river‐flow‐related variability during the first step. While we did not recognize this particular problem, we were aware of the general difficulties in looking for a small “tida” signal buried within a larger “river flow rate” response and implemented two independent methods to evaluate the tidal modulation of plume area; on re‐examination the difference between them is consistent with Cao and Yu's estimate of the bias in the first method. Although we concede that a better statistical estimator could (and perhaps should) be designed for this problem, we believe that our qualitative conclusions remain robust and that, considering other sources of uncertainty, our procedures could reasonably be applied to other locations during an initial analysis.
Li et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: