|The manuscript has been revised and I appreciate the answers/amendments the authors provide. However, in my opinion, the concerns expressed the first previous review are only considered to a minimal extent, still leaving some important questions. |
The remaining points are:
- Line 83: the PACE borehole revealed some ice fills karst, but this is still not discussed further. Is it justified to neglect this finding?
- In the figures, please state clearly what the different colors and line styles refer to.
- Figure 4,5 and Figure 8: the alpha value differs between the reconstructions using the synthetic and real data, why? As seen in Figure 4 and 5, a variation can lead to significantly different results.
- The caption of Fig. 9 is not clear – where do the temperature ranges come from? As seen from Fig. 14 they probably represent the uncertainty in the temperature measurements.
Consequently, it would be interesting to investigate the effect of this temperature uncertainty on the reconstructed GST history, not only showing it along with the temperature anomalies resulting from the reconstructed GST history.
- Fig. 9 is not cited in the manuscript text.
- Please justify your answer to the comment to line 167: „Numerical methods for the estimation of the regularization parameter, like GCV and L-curve, are not always reliable and in our case they provide an underestimation of the best value.“
- Lines 250 to 253: regarding the uncertainty in the reconstruction (which still needs to be determined in my opinion), is justified to give results in this precision (0.96 °C)?
- As mentioned the thermistor at the critical depth seemed to work properly, but this could have been investigated by varying the data used in the inversion (sensitivity study, uncertainty ranges, see comment above).
- 2D model: what would be the influence of the vertical extension of the model (background gradient…)
- I still consider a 3D model necessary in order to assess the 3D-effect on the temperature distribution, although the authors provide arguments that this would not yield further information.
- Please check again, if “°C” is used for absolute temperatures and “K” for temperature differences (and the gradient, Table 1), this makes the distinction easier.
- Table 2: unit for thermal conductivity is not correct.
To summarize, the authors present interesting and valuable data which “deserves” a thorough investigation, which has not been done yet in my opinion. Therefore, I recommend another revision in order to do fully justice to the available data.