Dear authors,
based on the comments by the two referees of your manuscript, as well as on your responses to these reports, you are kindly invited to submit a revised manuscript. Both referees have highlighted the fact that your paper certainly is of interest to the readers of the HESS journal. However, considering the mandatory changes requested by both referees, the revisions on a second version of the manuscript are quite substantial (and can be rated as major). Please pay particular attention to referee 1 comment on high resolution information omitted in the first version of the manuscript, as well as to the comment on the need for a clearly stated hypothesis made by referee 2. All other issues raised by the referees need of course also to be considered.
I am looking forward to the revised version of your manuscript.
Best regards,
Laurent Pfister

Dear authors,
following the recommendations of one referee, I agree that there need to be a few additional edits made to your manuscript. More specifically, you would need to consider the following two aspects in both the discussion and conclusion when comparing ground observed fluxes and RS derived fluxes:
- the different footprints of measurements and RS
- the non-closure of the mesured energy balance and their correction.
As mentioned by the referee, boths points are likely to have a considerable impact on potential differences and biases and need therefore to be considered.
I look forward to receive the final version of your manuscript after consideration of the above mentioned edits.
With best wishes,
Laurent Pfister

Dear colleagues,

I have re-checked the latest version of your manuscript - all major points that had been raised by the referees have been taken into account. However, it appears that there are clearly differences in the quality of different sections of the manuscript. I have highlighted certain parts in the paper and made suggestions for clarification (see the attached pdf file with comments/suggestions embedded). Before the final acceptance of your manuscript, I would strongly suggest that you have the core text reviewed by a native English speaker.
I am looking forward to receiving your paper after what should in principle be a final round of edits.

Best regards,

Laurent Pfister

Dear authors,
albeit your manuscript has been significantly improved, a number of corrections, improvements and clarifications remain to be implemented at this stage. Here below, you find a list with the most important points in this respect:
Page 2 - Line 20: … and TRFA handles them at 30m and 300m resolution, which depends …
Page 2 - Line 22: … delete ‘s’ of ‘understandings’
Page 2 - Line 32: Question – is it really the difference of varied landscapes within mixed pixels? Should it not rather read ‘… and the influence of various landscapes inside mixed pixels.’?
Page 3 - Line 17-18: The sentence ‘Non-linear operation model and surface heterogeneity are main issues of remotely sensed spatial scale error …’ This sentence is not 100% clear – it should be rephrased.
Page 3 - Line 40-42: SUGGESTION: ‘However, in case of mixed pixels surface variables such as land surface temperature are set to singular [SENSE NOT 100% CLEAR] to represent the entire pixel area in ET estimation models.’
Page 4 – Line 37-38: SUGGESTION ‘The scatter plot between LST and NDVI values forms a …’
Page 4 – Line 42: WHAT DO YOU MEAN BY ‘inferiority’ ?
Page 5 – Line 10: WHAT DOES ‘uniformed’ MEAN? COULD IT BE ‘standardised’?
Page 6 – Line 27: SUGGESTION ‘The spatial resolution of LST is significantly increased by …’ SPLIT THE SENTENCE IN TWO.
Page 8 – Lines 4 to 6: this sentence is too long and should somehow be structured in a different way (e.g. by using commas)
Page 8 – Line 8: SUGGESTION ‘In the case of using IPUS, the input for the energy balance model are first retrieved at … and then aggregated to 300 m resolution.’
Page 8 – Line 11-12: SUGGESTION ‘… using the nearest neighbour method and the 30 m resolution inputs are used for estimating ET.’
Page 11 – Line 13: SUGGESTION ‘… is the regression kernel of the temperature sharpening procedure and …’
Page 13 – Line 24: SUGGESTION ‘To control model inputs and analyse error sources, the …’
Page 16 – Line 14: SUGGESTION ‘… including land patches in the Gobi region, barren areas and …’
Page 16-17: SUGGESTION ‘Therefore, the LE of barren areas in the south is higher than the LE of desert areas in the southeast, …’
Page 18 – Line 12: SUGGESTION replace ‘Gobi areas’ by ‘Gobi region’
Page 18 – Line 22: SUGGESTION ‘… were also implemented for comparison purposes.’
Page 19 – Line 2: SUGGESTION ‘… the improvements in accuracy between TRFA and IPUS were relatively larger than those between TSFA and TRFA.’
Page 19 – Line 13: SUGGESTION change the sentence to ‘Variations in landscape characteristics systematically trigger variations in surface variables.’
Page 23 – Line 3: SUGGESTION ‘The land surface of EC15 was uniform and consisted of pure pixels covering maize fields. Consequently, the temperature distribution at 30 m resolution was very homogeneous and the surface temperature variations were comprised within a range of 1.6 K.’
Page 24 – Line 7: SUGGESTION ‘… and the observed ground temperature was …’
Page 25 – Line 8-9: this sentence is difficult to read and should be rephrased.
Page 30 – Line 39-40: SUGGESTION, split sentence ‘At noon, the vegetation temperature in those pixels is lower than that of water bodies.’
Page 31 – Line 1: Moreover, G depends on Rn when using the TRFA method and is nearly identical to the values of G obtained with the TSFA method.’
Page 32 – Line 20: SUGGESTION change to ‘4.4.1. Errors in LST’
Page 32 – Line 26: A sensitivity analysis of LE inferred from LST was also carried out. In order to exclude potential influences of other factors, only homogeneous stations were chosen’. COMMENT: what does homogeneous exactly mean in this context?
Page 34 – Line 18: SUGGESTION ‘In summary, the advantages of the TSFA method are described as follows:’
Page 34 – Line 21: SUGGESTION ‘…, which is consistent with results from previous research’
Page 35 – Line 39-40: Sentence not clear – should be reformulated.
Page 35 – Line 41: set ‘variables’ to singular
Thank you very much for considering these modifications in a corrected version of your manuscript.
Yours sincerely,
Laurent Pfister

Dear authors,
thank you very much for having considered the suggested modifications/edits to your paper. In its current form, the manuscript is now really close to final acceptance. There are only two minor technical corrections that would still need to be made:

Page 3 - Line 40-42: Add 'a' before 'single value' - 'However, in case of mixed pixels, surface variables such as land surface temperature are commonly considered as a single value to represent the entire pixel area in ET estimation models, which results in large errors'.

Section 3.2.2. Change sentence to: 'In this procedure, data were rejected when (1) the sensor had been malfunctioning, (2) …'

After these changes, your manuscript can be accepted and processed for publication in HESS.

Best regards,

Laurent Pfister