Earth-surf-dynam-discuss.net

Earth Surf. Dynam. Discuss., 1, C267–C271, 2013 Earth Surface
www.earth-surf-dynam-discuss.net/1/C267/2013/ Earth Surface
Author(s) 2013. This work is distributed under the Creative Commons Attribute 3.0 License.
Interactive comment on “Seasonal logging,
process response, and geomorphic work” by
C. H. Mohr et al.
Anonymous Referee #1
General CommentsThis manuscript uses a Quantile Regression Forests statistical tool to model sedimentconcentrations and yields from logged watersheds in Chile. The authors find that theQRF model outperforms the more traditional sediment rating curve approach to mod-eling sediment yields, in that sedigraphs predicted by the QRF model more closelyagree with measured values than sedigraphs predicted by the rating curve approach.
Given the promise shown by the QRF method, this work will likely garner the interestof scientists and land managers that are engaged in the prediction of sediment yieldsfrom a wide range of landscapes. Overall, I think this is interesting work that appearsto be methodologically sound, but I do have several general comments/questions:The rating curve method uses only one variable (discharge) to predict sediment con- centrations. In contrast, there are 21 variables in the QRF model and it would beuseful to know if pared-down versions of the model perform as well as the full model,with respect to the predictions from the rating curve method. In other words, with onlydischarge data as an input, does the model still outperform the rating curve method?Comparison of modeled and measured annual sediment fluxes are presented in Sup-plementary Table 3, but little of the manuscript text is devoted to discussing the differ-ences. It would be beneficial for the authors to more fully explore/explain these resultsand to explain (if possible) the underlying differences between the modeled and mea-sured values.
There should be more information about the timing of logging and post-logging treat-ments, as the introduction seems to make the case for a need to assess the impact ofclear-cutting, but the discussion indicates that decreased suspended sediment yieldsfollowing dry season logging may be due to replanting.
The implications of this study for geomorphic work do not seem to be as clear asthose presented in the manuscript. The highest discharges measured during the studymay be extreme with respect to the other values in the dataset, but a longer-termrecord is not presented (for either discharge or precipitation) that demonstrates thatthe discharges are extreme with respect to annual exceedence probabilities. Morecontext is needed to demonstrate that these findings differ in a substantial way from,for example, the view put forth by Wolman and Miller (1960).
Specific commentsPage 313Lines 1-2: Avoid leading off the manuscript by bringing up a discussion about man-made forests, especially because this topic is not addressed again in the paper.
Line 8: I don’t believe the road-related landslides documented by Montgomery et al.,were triggered by the failure of road cuts, but the change in upstream drainage area caused by the construction of the roads.
Lines 20-25: It is unclear why re-planting specifically, requires a technique capableof dealing with few samples collected under varying conditions. Put another way, thetools introduced in this paper can likely be put to a much broader use than for assessingsediment yields from re-planted clearcuts in Chile, as a small number of samples, highvariance, and changing environmental conditions are inherent to a broader range ofscenarios where sediment yields need to be quantified. It would be worthwhile topresent a broader utility of the techniques developed as part of this study.
Page 316Line 7: SSC measurement were not made every three minutes, but every 30-60 min-utes, which was not the impression one gets from reading lines 10-11 of the Abstract,which states sediment concentrations were measured every three minutes. I’d encour-age you to report the data that were collected in an un-ambiguous manner.
Page 321Lines 11-12: As presented, it is unclear why the bulk sediment flux measurements areminimum values.
Page 322I’m not convinced that the sediment transport events that were measured are extreme.
These data are not put into the context of a longer record, but the recurrence intervalsseem to be < 1 yr. More context is needed to demonstrate that these results differ in asubstantial way from the view put forth by Wolman and Miller (1960). SupplementaryTable 3The results presented in Supplementary Table 3 contain information that most readerswill want to know: how do total sediment yields predicted by the QRF model compareto those predicted by the sediment rating curve method, and how do both model pre-dictions compare with measured data. These data need to be more fully integrated with the main text of the manuscript. Currently, the only table in the manuscript presents thenumber of samples, whereas the comparison of the model predictions are much moreimportant.
Editorial commentsPage 313Line 10: I suggest revising this sentence, as it is not clear what is meant by “the long-term decay of soil conservation functions”.
Page 315Line 18: Suggest changing “gauges” to “weirs”.
Line 22: Suggest inserting “rain gauge” following “bucket”.
Line 22-24: Suggest revising to: “A Wilcox rank sum test was used to assess whetherhourly rainfall intensities differed significantly (p≤0.05) between each year.”Line 24: It is unclear what “bulk monitoring data” refers to, total sediment yield?Page 316Line 20: replace “larger” with “longer”Lines 22-24: It is unclear what information is trying to be conveyed by this sentence.
Line 25: It would be useful to explicitly define what an “integrated sample” is, as thismay clear up the somewhat confusing text in this paragraph.
Page 323End of line 12: it is unclear what is meant by “they”.
Figure 2 captionChange “base” to “basis”.
Figure 7It is difficult to see red crosses in many of the panels.
Interactive comment on Earth Surf. Dynam. Discuss., 1, 311, 2013.

Source: http://www.earth-surf-dynam-discuss.net/1/C267/2013/esurfd-1-C267-2013-print.pdf