Assessing the Impact of T-Mart Adjacency Effect Correction on Turbidity Retrieval from Landsat 8/9 and Sentinel-2 Imagery (Case Study: St. Lawrence River, Canada)
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis study selected the Saint Lawrence River section to evaluate the atmospheric correction algorithm for the adjacent effect. Based on a large amount of field measurement data, a machine model was constructed. The model construction and sample size were relatively sufficient for the research needs. However, there were some doubts regarding the verification process and results.
- From the collected samples, they were mainly located near the riverbank and did not reach the river center. Please provide an explanation in the manuscript for such positioning arrangements.
- The turbidity range of the samples is 3 - 105 NTU. However, the results of the model used for modeling, training and testing (as shown in Figure 4) are all less than 40 NTU. Therefore, the applicability of the model in high turbidity conditions should be explained.
- From the inversion results (Figure 6), it seems that the selected atmospheric correction method has a relatively small impact on the inversion results of L89. Comparing (d) and (f), it appears that the inversion effect without atmospheric correction is better. Please explain this in the manuscript.
- What is the depth of the river? How were the influence of the nearshore shallow water area and the bottom sediment on turbidity considered in the manuscript?
- In addition to the construction of the atmospheric correction model, please also provide a detailed description of the turbidity inversion model in the manuscript, including the process and results.
Comments for author File: 
Comments.pdf
Author Response
Please see attachment
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsAtmospheric Correction (AC), including Adjacency Effect (AE) correction, is a major challenge for inland water quality retrieval using optical satellite data. This study evaluated three image pre-processing options for turbidity retrieval in the St. Lawrence
River using Sentinel-2 (S2) and Landsat 8/9 (L8/9) imagery. They applied the Light Gradient Boosting Machine (LightGBM) model: (1) No pre-processing, i.e. use of Top-of-Atmosphere (TOA) reflectance, (2) AC pre-processing, obtaining water-leaving reflectance (Rw) from AC for the Operational Land Imager lite (ACOLITE)’s Dark Spectrum Fitting (DSF) technique, and (3) AE pre-processing, correcting for the AE using T-Mart before obtaining Rw from DSF. Their results demonstrated that AE pre-processing outperformed the other two options. For L8/9, AE pre-processing reduced the Root Mean Square Error (RMSE) and improved the median symmetric accuracy (ε) by 48.8% and 19.0%, respectively, compared with AC pre-processing, and by 48.5% and 50.7%, respectively, compared with No pre-processing. For S2, AE pre-processing performed better than AC pre-processing and also outperformed No pre-processing, reducing RMSE by 28.4% and ε by 50.8%. However, No pre-processing yielded the lowest absolute symmetric signed percentage bias (|β|) among all pre-processing options. Their analysis also indicated that AE pre-processing yielded superior performance within 0-300 m from shore than other options, where the AE influence is strongest. Turbidity maps generated using AE pre-processing were smoother and less noisy compared to the other pre-processing options, particularly in cloud-adjacent regions. In short, their findings suggest that incorporating AE correction through T-Mart improves the performance of the LightGBM model for turbidity retrieval from both L8/9 and S2 imagery in the St. Lawrence River, compared to the alternative pre-processing options.
The methodology and writing is fine with expected figures. Tables are good to express the meaning. But one point still needs to be improved for adding more references up to or more than 40 references to cover all fields in the study.
Author Response
Please see attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThis paper analyses the turbidity estimation by remote sensed data (sentinel 2 and Landsat) using three different level of pre-processing for atmospheric correction: TOA reflectance (no pre-processing), ACOLITE and AE pre-processing. The abstract is well organised and highlights the significance of this research. References are updated, proper and closely relevant to the study. The article is well structured and fluent and the introduction clearly outlines the research problem. The data (in situ and satellite) used in the study and the methodology applied are well described, results are comprehensively presented and explained in detail.
In my opinion the article can be accepted for publication.
Author Response
Please see attachment
Author Response File: Author Response.pdf
