A Remotely Sensed Framework for Spatially-Detailed Dryland Soil Organic Matter Mapping: Coupled Cross-Wavelet Transform with Fractional Vegetation and Soil-Related Endmember Time Series

Round 1

Reviewer 1 Report

Regarding the Manuscript ID: remotesensing-1615296  Entitled: A remotely sensed framework for spatially-detailed dryland soil organic matter mapping: coupled cross-wavelet transform with fractional vegetation- and soil-related endmember time series

The paper presents a nice and novel topic, using RS to map the soil organic carbon which can absorb many attention. The paper has been nicely written well organized and well-discussed.  I would accept the paper after addressing some minor comments:

• Please the reference (Taybe at al. 2021: https://doi.org/10.3390/rs13112223) which successfully used the machine learning algorithms to predict and investigate the SOC.
• Please simplify Figure 2 as it seems a bit complicated.
• Please improve the discussion section, you should focus on the use of RS methods not only on the use of machine learning algorithms.
• The quality of Figure 10 is not acceptable. What do you want to show?. Why the Fig. 10 g and Fig.10 h show high differences?.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The work is well done, written in an academic way, the topic is important and contributes to the development of knowledge in the field.
However, I have some recommendations: at the end of the Introduction part, taking into account the bibliographic study presented, please specify the research hypothesis on the basis of which you set the objectives.
When describing the area (2.1 Study area), please improve the physical, geographical and pedoclimatic description.    

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript “A remotely sensed framework for spatially detailed dryland soil organic matter mapping: coupled cross-wavelet transform with fractional vegetation- and soil-related endmember time series” submitted by Sun et al. describes a new approach to estimate SOC in dryland using cross-wavelet transform and covariates for mapping integrated with ML approaches.

• General concept comments

The article demonstrated the efficacy and reliability of the proposed framework in predicting SOM in a dryland system. The authors extracted selected features from XWT (13 = 80%) and filtered them as covariates for SOM mapping using ML algorithms (RF and GBRT).

The manuscript sounds scientifically robust and interesting for Remote Sensing readers.

Specific comments – please inform the developer, city and country of statistical software used SPSS (Statistical Product and Service Solutions). Line 226

Author Response

Please see the attachment

Author Response File: Author Response.docx