See the attached pdf. 

Dissipation and related diffusion and mixing in the ocean are important for ocean dynamics (even global circulation) and for the transport of oceanic constituents.  However, the causative processes are on very small scales (down to millimetres or less) as well as patchy and intermittent.  Hence they are difficult to measure and quite impractical to model explicitly.  As a result, empirical relationships and parameterisations are much used and evidence to improve these is valuable.  This manuscript especially concerns the much used factor Γ^T = 0.2 in the Osborn relation for turbulent density diffusivity = Γ^T (dissipation)/(buoyancy frequency^2).  Much evidence is cited that Γ^T varies a lot in space and time.  Moreover, turbulence is not the only agent of mixing; for salt finger mixing the equivalent Γ^F may be negative.  The manuscript provides separate evidence and discussion of the effective diffusivities for temperature and salinity associated with salt fingering.

Data for the study are from the western equatorial Pacific the subtropical NE Atlantic and tropical SW Atlantic.  This is a varied set but I am left uncertain as to how comprehensive or representative it may be of all the possible data that might have been used.  It is certainly sufficient to make the case that there can be improvement by moving to Γ^T other than 0.2 with some suggestion of how to derive improved values.

I found section 2.3 lacking in logical development and am unsure as to its value to the rest of the manuscript (it is rarely referred to).  Might it be replaced by a few literature references?

The English is generally understandable but there is some curious usage that should be corrected by the publisher’s copy-editing of a final manuscript.  The authors unfortunately follow the current “fashion” of misleadingly using “increasing trend” when they mean “positive trend” or simply “increase”.  “Misleading” because the expression implies a change of trend.  [Many but probably not all examples are included in the following “Detailed comments”.]

Detailed comments

Line 104.  “we chose five projects that . .”  Did other projects provide χθ and you chose not to use them, or did you use all the projects providing χθ ?  If the latter, better “we chose all five projects that . .”  to show that you did the best possible.

Table 1.  According to figure 1 NATRE is in the North Atlantic.  (“S” –> “N”)

    “Profile Number” –> “Number of Profiles”

Equations (3).  The sequence from left to right is not logical if Γ is already defined as in line 149.  Moreover from (1) and (2) the second and last terms of (3) are directly equal irrespective of the first and third terms.  Please clarify what is definition, what is derivation, and where the form of the third term comes from (there seems to be an analogy with the right-hand side of (4) for which a reference is cited).

Lines 212-213.  Better “. . divide the number of energetic turbulence patches in each depth bin by the total number of energetic turbulence patches in the whole project; . .” and Line 214 “. . by the total number of patches within the same depth bin . .”?

Section 4.1 is very long and I think would benefit from some sub-headings.

Lines 243-244.  Why are there two NATRE median values of Γ^T for each of energetic turbulence and weak turbulence?

Line 257.  Why “alternately”?  “slightly increasing trend” –> “slight increase with depth”? (unless you mean the trend/rate increases)

Lines 268, 270, 451, 498, 510.  “vertical increase” (or decrease”) is unclear until upwards or downwards is specified.  Also (lines 268, 270, 451) I think you mean “increase” not “increasing trend” (c.f. line 257; does the trend/rate increase?)

Line 272.  I think you mean “. . disagree about whether Γ^T is larger for energetic turbulence or weak turbulence.”

Lines 284, 369, 390, 451, 500.  “increasing trend” –> “increase” (indeed, in line 390 referring to figure 11, the trend is positive but actually decreases for R_ρ > 3).

Lines 285, 290, 397.  “decreasing trend” –> “decrease”

Line 286.  “BBTRES” -> “BBTREs”

Figure 8.  In the figure legend, the red line should be ascribed to Re_b > 160.  The grey dashed line is rather indistinct.

Figure 9.  Please explain (in the caption or against the colour bar) that the colour bar refers to median Γ^T

Line 372.  The “ref”erence needs to be included.

Line 376.  “Note that . . .”  I think this sentence should refer to a labelled formula in section 2.3.

Line 403.  “decreasing rate” –> “rate of decrease”

Lines 419-420.  “increasing rates” –> “increases”?

Line 422.  “vertical decreasing trend and magnitude” –> “decrease downwards”?

Line 423.  “increasing trend” –> “increases”

Line 492.  “Vertically, Γ^T in the western equatorial Pacific presents a weak decreasing trend” –> “Γ^T in the western equatorial Pacific presents a weak decrease downwards”