Multiscale Temporal Irreversibility of Streamflow and Its Stochastic Modelling
Round 1
Reviewer 1 Report
Comments to "Multiscale temporal irreversibility of streamflow and its
stochastic modelling" by Vavoulogiannis, Iliopoulou, Dimitriadis and
Koutsoyiannis
Submitted to Hydrology
March 2021
The topic of the paper is important. The analysis of data is
adequate. The presentation of the AMA model needs better motivation.
My recommendation is that the paper may be accepted after the authors
make significant improvements in the presentation.
I cannot entirely agree with the sentence that "Time's arrow . . . is related
to randomness and uncertainty". There are many examples of reversible
stochastic processes.
It sounds odd to say "real world streamflow data". Are there other
world streamflow data?
The use of first and second scale terms does not seem to be
standard, please provide an explanation or a motivation.
Specific comments:
1. The definition of a random variable, given in p3, sect 3.1, is not
adequate. The authors should either give the correct one or delete it.
2. The authors should review nomenclature everywhere in the
paper. There are symbols used without definition (for example, X in Eq
4, and \nu^#, \xi^# and \gama^# in equation 22), in some cases the
definition is given much later than the first use (for example x_i^{(k)} is used
in Eq 5 but only defined in Eq 19). The way the paper is written
makes it very hard to follow. They are reusing
material from previous papers without due care.
3. In various places, the English is hard to read: P1: "Studies have
shown that the time asymmetry of the streamflow process may be marked for scales up to several days". . ." Herein, we investigate the time
asymmetry of real world hourly streamflow time series from the large
USGS database and find it to be marked up to 4 days". . . p5: "processor
multiscale stochastic representation". P6: "between the three tools". . . "mostly monotonic". . . "The calibration function used both the climacogram and the climacospectrum (empirical and theoretical) gives emphasis to the climacospectrum at the finer scales and to the climacogram at the greater scales". P9: "for the preservation of time irreversibility at two scale". . . "For the Monocacy River after optimization the following sequence is found, shown in Figure 5". . . "Sequence of ?
values for the second case of preserving irreversibility at both
scales". P10: "Downstream stations are suspected to have higher
irreversibility than the upstream ones and this could also be studied"
Author Response
Please see attached.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
I have read with interest your paper on the Multiscale temporal irreversibility of streamflow and its stochastic modelling. The article is well-organized, however, I have a few comments:
- Could you explain what were the reasons of the selection of this particular dataset to investigation?
- How does the fact affect the results that mentioned asymmetric rising and falling parts of a flow hydrograph might be determined by genetically different factors: summer rainfall, spring snowmelt, winter ice dam, summer or autumn hydrological drought recession etc.
- On the page 9, in the last paragraph, there should be references to Fig. 6a and 6b, not to 5a and 5b.
- The last chapter ‘5. Discussion and Conclusions’ rather lacks the discussion and contains summary not conclusions. As a result, it should be entitled: Summary
Author Response
Please see attached.
Author Response File: Author Response.pdf
