Interactive comment on “ The relation between sea ice thickness and freeboard in the Arctic ” by V .

This paper brings together in-situ measurements of sea ice to evaluate the uncertainties in the parameters used to convert satellite measurements of sea ice freeboard to thickness. The paper assesses the contribution from the uncertainty in ice density to the total ice thickness uncertainty. This is a useful exercise and suitable for publication, providing some corrections are made to the manuscript. These are detailed below, but the two most serious issues that need to be addressed are:

Uncertainty of ice density for MY-ice was calculated as an average weighted of uncertainties for the upper and lower layers.Due to lack of measurements these uncertainties were estimated based on scientific judgement, and amount to 50 and 20 kg/m3, respectively.
2) The motivation of this paper is to better constrain the uncertainty in satellite derived sea ice thickness measurements, as described in the abstract.However, the paper contains multiple references to constant freeboard/thickness ratios that are not used by the majority of papers that use altimeter measurements of freeboard to compute thickness (e.g.Laxon, Giles, Kwok, Zwally etc.).Constant freeboard to thickness ratios are not used due to the sensitivity of the ratio to the changes in the snow depth and density, which are accounted for when equation 4 is used, taking estimates of the snow depth and density from climatology, for example.All references to this ratio must be removed and the paper should focus on evaluating the ice density uncertainty and how it affects the total uncertainty in ice thickness using equ.5.
Answer: Constant freeboard/thickness ratios are not used due to their sensitivity to snow loading, and this is particularly relevant to laser altimetry.It is known that snow loading increases from autumn to late winter, and do not change significantly during late winter conditions.Our study also shows that snow climatology data (Warren et al., 1999) are correct for the MY-ice in the Central Arctic, but not for the FY-ice in the Eurasian Arctic shelf seas.Therefore in our paper we use constant ratios only for the late winter period, and these relations are different for the FY and MY-ice.The relations C518 (1), ( 6) and ( 7) cannot be used for example in autumn, when snow loading is different.We agree that in general, the isostatic equilibrium equation should be used, but for some particular cases this equation can be transformed into simple linear dependence between ice thickness and freeboard.
The paper also includes a number of grammatical errors that should be corrected.
Answer: Grammatical errors will be corrected Specific Comments P642, L16: The value given for the freeboard uncertainty is 0.05.Where does this number come from?I suggest taking the FB uncertainty of 0.03 m from Giles et al., (2007).
Equation 11 overestimates ice thickness for the FY-ice, because the estimate of snow depth (0.3 m) from Table 1 in (Giles et al., 2007) is typical for MY-ice in late winter, but not for the FY-ice.If to use other parameters from Table 1 in (Giles et al., 2007) and only change snow depth to 0.05 m (as it was found in Sever data) then the equation is: Hi=9.42Fi+0.15,which corresponds to our equation ( 6).P652, Conclusions: The authors could recommend that different densities for FY and MY ice should be used routinely to calculate sea ice thickness from satellite measurements of freeboard (both in laser and radar measurements).
Answer: We will recommend this Figure 1.The labels on the figures are too small.Answer: We will correct this.
Figure 2: the caption could contain the data sources.Do all the data sources estimate density in the same way?If not, how much of the measured interval is due to the error in the measurement techniques and how much is natural variability?
Answer: We will include data sources in figure caption.This figure is based on the review paper by Timco and Frederking (1996), and several other papers.Analysis of the data presented in Timco and Frederking (1996) shows that the measurements of ice density were mostly conducted using mass/volume technique.Ice densities, measured using this technique, vary substantially.
Figure 3: This is an interesting plot showing how the relative contribution of the errors change as a function of freeboard.It would be easier to read if the colours of the contributions from the errors were the same for FY and MY ice e.g. total error is currently green for FY ice and blue for MY ice, and the freeboard error is blue for FY and green for MY ice.The authors could produce the same plot for laser altimetry, using equation 4 in Giles et al., 2007.Answer: The colours of the contribution from the errors are the same for FY-and MY-ice.It is possible to produce the same plot for laser altimetry.

Figure 4 :
Figure 4: The letters a)-e) in the caption do not relate to the figure Answer: We have deleted the letters a)-e) from the figure caption.Interactive comment on The CryosphereDiscuss., 4, 641, 2010.