Normalized Temperature Drought Index (NTDI) for Soil Moisture Monitoring Using MODIS and Landsat-8 Data

Round 1

Reviewer 1 Report

The manuscript presented a new drought index called NTDI for soil moisture estimation. This paper is well written with a clear motivation, novel mathematical formulations, comprehensive experiments and analysis, and a profound conclusion. In general, it has convincing novelty with good shape and excellent presentation. I only have some minor issues.

 

#1. As an introduction, I think it would be important to have a state of the art of some important studies using soil moisture or NDVI based indices. Some recent studies are needed.

#2. Concerning the use of MODIS data, the database seems very limited to establish reliable statistics, why limit this analysis to 2019 and not have an analysis of all the data during 20 years of MODIS products? In addition, there are more available in-situ soil moisture measurements on the International Soil Moisture Network (ISMN), which can be used as candidate datasets.

#3. Table 2, for clarity, the authors should highlight those values of R² with statistical significance at 95% level.

#4. We do not see a seasonal context of the proposed index? We know that the behaviors are not the same during the different seasons of the surface measurements (temperature, soil moisture, vegetation cover.)?

#5. Why the choice of linear relations in equations (4) and (5).

#6. In Table 2, the authors give the SM estimated results, but there is no information on how the authors retrieve the soil moisture from the different drought indices, which has to be explained clearly.

#7. In Table 2, Why the PDI retrieved results has very low R² on 254 and 274 days but acceptable RMSE? On Day 075, the NTDI has low R² but better RMSE (R²=0.07, RMSE=0.03) than PDI (R²=0.36, RMSE=0.04). Please check the results again. In addition, from the perspective of RMSE, the PDI achieved comparable results with NTDI, why?

 

Author Response

The response to the comments is attached as the Word file. Thanks.

Author Response File: Author Response.doc

Reviewer 2 Report

The manuscript entitled "Improved Monitoring of Soil Moisture Using Normalized Temperature Drought Index Based on MODIS and Landsat-8 Data in Victoria, Australia" presents a new parameterization scheme for estimating soil moisture using satellite data. The authors propose the use of the Normalized Temperature Drought Index that incorporates normalized land surface temperature and Normalized Difference Vegetation Index to improve the sensitivity of SM monitoring, reduce reliance on land cover types, and capture spatial and temporal changes of SM accurately.

 Overall, the manuscript is well-structured and presents a clear method for the development and evaluation of NTDI. The authors provide a comprehensive analysis of the results obtained from MODIS and Landsat-8 data at crop sites in Victoria. However, there are several areas that require critical evaluation and improvement.

 ·         The manuscript claims to present a new parameterization scheme for SM monitoring using NTDI. While the authors mention that NIR and red reflectance are substituted with LSTnor and NDVI, the novelty of this approach compared to existing methods is not clearly demonstrated therefore, the novelty and significance of the proposed approach should be clearly demonstrated by comparing it with existing methods. The manuscript would benefit from a more comprehensive review of existing literature on SM monitoring and a clear justification of how NTDI differs from or improves upon existing methods.

 ·         Methodology used for developing and evaluating NTDI is explained in detail, but there are some concerns. The authors mention that time series of MODIS and Landsat-8 data were used for evaluation, but details of the data processing steps, such as data preprocessing, interpolation, and accuracy assessment, are lacking. Additionally, the methodology for estimating SM using NTDI is not explained clearly. It would be helpful to provide a step-by-step description of the calculations involved in obtaining NTDI and how it relates to SM.

 ·         The results obtained from the evaluation of NTDI are presented in the manuscript, but there are some issues with the analysis. The authors mention that NTDI has a better fitting relationship with SM compared to other drought indices, but the statistical significance of this difference is not presented. The authors also mention that NTDI has a similar RMSE compared to PDI, but it would be helpful to provide quantitative values for the RMSE and compare them with established thresholds for accuracy in SM monitoring. Additionally, the spatial distribution of the drought indices is presented, but there is a limited interpretation of the results in the context of the study area and its implications for drought monitoring.

·         The manuscript briefly mentions some limitations of the study, such as the effect of topography on LST and the need for further evaluation of NTDI in other regions. However, these limitations are not discussed in detail, and potential sources of error in the NTDI estimation are not addressed. The manuscript would benefit from a more comprehensive discussion of the limitations of the proposed approach and suggestions for future research to address these limitations.

·         The results should be analyzed in more detail, including the statistical significance of the differences between NTDI and other drought indices, and the implications of the results for SM monitoring in the study area.

·         The spatial distribution of drought indices is presented in the manuscript, but the interpretation of the results in the context of the study area and its implications for drought monitoring is limited. A more detailed discussion on the spatial patterns of NTDI and their agreement with known soil moisture patterns or drought events in Victoria would enhance the interpretation of the results and their practical implications for drought monitoring and management.

·         The limitations of the proposed approach should be discussed in more detail, and recommendations for future research should be provided.

·         The conclusion section provides a brief summary of the findings but lacks a critical analysis of the implications of the results and their significance for SM monitoring. The authors mention that further theoretical and field work is needed, but do not provide specific recommendations or directions for future research.

Author Response

The response to the comments is attached as the Word file. Thanks.

Author Response File: Author Response.doc