the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Enhanced Watershed Modeling by Incorporating Remotely Sensed Evapotranspiration and Leaf Area Index
Abstract. To improve the capacity of watershed modeling, remotely sensed products are frequently used to reduce the uncertainty resulting from data limitations. Although remotely sensed evapotranspiration (RS-ET) products are widely used, vegetation parameters governing spatial and temporal variations in evapotranspiration (ET) are often not constrained by benchmark data. Recently, remotely sensed leaf area index (RS-LAI) products are becoming increasingly available, providing an opportunity to assess and improve simulated vegetation dynamics. The objective of this study is to assess the role of the two remotely sensed products (i.e., RS-ET and RS-LAI) in improving the accuracy of watershed model predictions. Specifically, we investigated the role of RS-ET and RS-LAI products in 1) reducing parameter uncertainty and 2) improving model capacity to predict the spatial distribution of ET and LAI at the sub-watershed level. The watershed-level assessment of the degree of equifinality (denoted as the number of parameter sets that produce equally acceptable model simulations) shows that less than half of the acceptable parameter sets for two constraints (streamflow and RS-ET; 14 parameter sets) are acceptable for three constraints (streamflow, RS-ET, and RS-LAI; six parameter sets). Among those six parameter sets, only three can satisfactorily characterize spatial patterns of ET and LAI at the sub-watershed level. Our results suggest that the use of multiple remotely sensed datasets holds great potential to reduce parameter uncertainty and increase the credibility of watershed modeling, particularly for characterizing spatial variability of hydrologic fluxes that are relevant to agricultural management.
- Preprint
(1160 KB) - Metadata XML
-
Supplement
(307 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on hess-2022-187', Anonymous Referee #1, 08 Aug 2022
The authors apply remotely sensed evapotranspiration and leaf area index in addition to in-situ streamflow to calibrate a SWAT model in Tuckahoe Creek Watershed. The paper is well written, albeit the usage of numerous abbreviations. But, I am kind of skeptical to accept this version of the article.
Â
Major comment:
Novelty: Even though the article is presented well, under the hood, it is a calibration paper constrained with two additional RS products which has been investigated previously by other researchers listed below.
Â
- Parr, D., Wang, G., & Bjerklie, D. (2015). Integrating remote sensing data on evapotranspiration and leaf area index with hydrological modeling: Impacts on model performance and future predictions. Journal of Hydrometeorology, 16(5), 2086-2100.
- Andersen, J., Dybkjaer, G., Jensen, K. H., Refsgaard, J. C. and Rasmussen, K.: Use of remotely sensed precipitation and leaf area index in a distributed hydrological model, J. Hydrol., 264(1–4), 569 34–50, doi:10.1016/S0022-1694(02)00046-X, 2002.
- Jiang, D. and Wang, K.: The role of satellite-based remote sensing in improving simulated streamflow: A review, Water (Switzerland), 1615, doi:10.3390/w11081615, 2019.
Â
Suggestion:
Methodology: 20,000 LHS samples have a wealth of information.
- One way to provide insight would be to see among all the parameters that are being calibrated find the one which has the largest influence on the KGE values in par1 and par2. Investigating why these parameters are influential would a very good insight.
- Also, how to choose between single parameter set which gets best performance compared to a cluster of parameters (close in values) which gives good performance ?
- Is there a relationship (linear/non-linear) between parameter values and KGE ?
These are some of the questions that the authors can address to bring more value to science aspects of the paper.
Citation: https://doi.org/10.5194/hess-2022-187-RC1 - AC1: 'Reply on RC1', Sangchul Lee, 07 Sep 2022
-
AC2: 'Reply on RC1', Sangchul Lee, 07 Sep 2022
Dear reviewer,
Sorry for making you confused. Please see this version of my response to the comments.
We expected that a revised manuscript could be uploaded, so the first version did not include all sentences written in the manuscript on the response document.
This version includes all statements. Please take a look at the attached file.
Sangchul
-
RC2: 'Comment on hess-2022-187', Anonymous Referee #2, 07 Sep 2022
I do not recommend this paper for publication. There is nothing new in this paper. A similar topic has been addressed in several previous studies.
Rajib et al., 2020. https://doi.org/10.3390/rs12132148
Ma et al., 2019. https://doi.org/10.1016/j.jhydrol.2019.01.024
Parr et al., 2015. https://doi.org/10.1175/JHM-D-15-0009.1
Rajib et al., 2018. https://doi.org/10.1016/j.jhydrol.2018.10.024Â
Â
Citation: https://doi.org/10.5194/hess-2022-187-RC2 -
AC3: 'Reply on RC2', Sangchul Lee, 07 Sep 2022
We thank the referee for the valuable comments on our manuscript. The response to the comment is shown below. The line numbers (Line)Â on the manuscript referenced may have changed in the final version of the revised manuscript.
Reviewer #2: I do not recommend this paper for publication. There is nothing new in this paper. A similar topic has been addressed in several previous studies.
Rajib et al., 2020. https://doi.org/10.3390/rs12132148
Ma et al., 2019. https://doi.org/10.1016/j.jhydrol.2019.01.024
Parr et al., 2015. https://doi.org/10.1175/JHM-D-15-0009.1
Rajib et al., 2018. https://doi.org/10.1016/j.jhydrol.2018.10.024Â
We hope our response can well show our novelty.
This study was designed to improve the common SWAT modeling approach that uses only remotely sensed evapotranspiration (RS-ET) by adding remotely sensed leaf area index (RS-LAI). We have illustrated the necessity of this paper with an emphasis on the use of two remotely sensed products (RS-ET and RS-LAI) in the introduction section (Lines 77 – 104). To our knowledge, our manuscript is the first paper that demonstrates the benefits of two remotely sensed products in SWAT modeling on both equifinality reduction and spatial calibration. We have thoroughly compared our papers with four papers that the reviewer listed down (Table 1). Three of four papers (Ma et al., 2019; Rajib et al., 2018; Rajib et al., 2020) only used either RS-ET or RS-LAI. Although the study by Parr et al. (2015) adopted both RS-ET and RS-LAI, a different model (VIC) was used.
Â
Â
Table 1. remotely sensed products and model used in the literature
Literature
Remotely sensed products
Model
Rajib et al., 2020
RS-LAI
SWAT
Ma et al., 2019
RS-LAI
SWAT
Parr et al., 2015
RS-ET; RS-LAI
VIC
Rajib et al., 2018
RS-ET
SWAT
SWAT: Soil and Water Assessment Tool; VIC: Variable Infiltration Capacity
Â
To emphasize our novelty, we have stated the differences between our results and previous studies in Lines 477 – 487:
Lines 477 – 487: Especially, our results provided several insights on the use of two additional RS products although previous studies already reported their advantages (Andersen et al., 2002; Jiang and Wang, 2019; Parr et al., 2015). First, our studies showed the benefits of the two additional RS products for SWAT modeling with an emphasis on equifinality reduction. Especially, SWAT improvements with the single use of RS-ET (Becker et al., 2019; Rajib et al., 2018; Wambura et al., 2018) or RS-LAI (Ma et al., 2019; Rajib et al., 2020) have been well investigated while the simultaneous use of the two RS products has been rarely conducted. In addition, a substantial reduction of model uncertainty was highlighted by the model evaluation at two spatial scales using two RS products. Lastly, this study chose the two RS products frequently used to monitor croplands, and thus our results could inform an improved modeling approach for agricultural watersheds.
Citation: https://doi.org/10.5194/hess-2022-187-AC3
-
AC3: 'Reply on RC2', Sangchul Lee, 07 Sep 2022
Status: closed
-
RC1: 'Comment on hess-2022-187', Anonymous Referee #1, 08 Aug 2022
The authors apply remotely sensed evapotranspiration and leaf area index in addition to in-situ streamflow to calibrate a SWAT model in Tuckahoe Creek Watershed. The paper is well written, albeit the usage of numerous abbreviations. But, I am kind of skeptical to accept this version of the article.
Â
Major comment:
Novelty: Even though the article is presented well, under the hood, it is a calibration paper constrained with two additional RS products which has been investigated previously by other researchers listed below.
Â
- Parr, D., Wang, G., & Bjerklie, D. (2015). Integrating remote sensing data on evapotranspiration and leaf area index with hydrological modeling: Impacts on model performance and future predictions. Journal of Hydrometeorology, 16(5), 2086-2100.
- Andersen, J., Dybkjaer, G., Jensen, K. H., Refsgaard, J. C. and Rasmussen, K.: Use of remotely sensed precipitation and leaf area index in a distributed hydrological model, J. Hydrol., 264(1–4), 569 34–50, doi:10.1016/S0022-1694(02)00046-X, 2002.
- Jiang, D. and Wang, K.: The role of satellite-based remote sensing in improving simulated streamflow: A review, Water (Switzerland), 1615, doi:10.3390/w11081615, 2019.
Â
Suggestion:
Methodology: 20,000 LHS samples have a wealth of information.
- One way to provide insight would be to see among all the parameters that are being calibrated find the one which has the largest influence on the KGE values in par1 and par2. Investigating why these parameters are influential would a very good insight.
- Also, how to choose between single parameter set which gets best performance compared to a cluster of parameters (close in values) which gives good performance ?
- Is there a relationship (linear/non-linear) between parameter values and KGE ?
These are some of the questions that the authors can address to bring more value to science aspects of the paper.
Citation: https://doi.org/10.5194/hess-2022-187-RC1 - AC1: 'Reply on RC1', Sangchul Lee, 07 Sep 2022
-
AC2: 'Reply on RC1', Sangchul Lee, 07 Sep 2022
Dear reviewer,
Sorry for making you confused. Please see this version of my response to the comments.
We expected that a revised manuscript could be uploaded, so the first version did not include all sentences written in the manuscript on the response document.
This version includes all statements. Please take a look at the attached file.
Sangchul
-
RC2: 'Comment on hess-2022-187', Anonymous Referee #2, 07 Sep 2022
I do not recommend this paper for publication. There is nothing new in this paper. A similar topic has been addressed in several previous studies.
Rajib et al., 2020. https://doi.org/10.3390/rs12132148
Ma et al., 2019. https://doi.org/10.1016/j.jhydrol.2019.01.024
Parr et al., 2015. https://doi.org/10.1175/JHM-D-15-0009.1
Rajib et al., 2018. https://doi.org/10.1016/j.jhydrol.2018.10.024Â
Â
Citation: https://doi.org/10.5194/hess-2022-187-RC2 -
AC3: 'Reply on RC2', Sangchul Lee, 07 Sep 2022
We thank the referee for the valuable comments on our manuscript. The response to the comment is shown below. The line numbers (Line)Â on the manuscript referenced may have changed in the final version of the revised manuscript.
Reviewer #2: I do not recommend this paper for publication. There is nothing new in this paper. A similar topic has been addressed in several previous studies.
Rajib et al., 2020. https://doi.org/10.3390/rs12132148
Ma et al., 2019. https://doi.org/10.1016/j.jhydrol.2019.01.024
Parr et al., 2015. https://doi.org/10.1175/JHM-D-15-0009.1
Rajib et al., 2018. https://doi.org/10.1016/j.jhydrol.2018.10.024Â
We hope our response can well show our novelty.
This study was designed to improve the common SWAT modeling approach that uses only remotely sensed evapotranspiration (RS-ET) by adding remotely sensed leaf area index (RS-LAI). We have illustrated the necessity of this paper with an emphasis on the use of two remotely sensed products (RS-ET and RS-LAI) in the introduction section (Lines 77 – 104). To our knowledge, our manuscript is the first paper that demonstrates the benefits of two remotely sensed products in SWAT modeling on both equifinality reduction and spatial calibration. We have thoroughly compared our papers with four papers that the reviewer listed down (Table 1). Three of four papers (Ma et al., 2019; Rajib et al., 2018; Rajib et al., 2020) only used either RS-ET or RS-LAI. Although the study by Parr et al. (2015) adopted both RS-ET and RS-LAI, a different model (VIC) was used.
Â
Â
Table 1. remotely sensed products and model used in the literature
Literature
Remotely sensed products
Model
Rajib et al., 2020
RS-LAI
SWAT
Ma et al., 2019
RS-LAI
SWAT
Parr et al., 2015
RS-ET; RS-LAI
VIC
Rajib et al., 2018
RS-ET
SWAT
SWAT: Soil and Water Assessment Tool; VIC: Variable Infiltration Capacity
Â
To emphasize our novelty, we have stated the differences between our results and previous studies in Lines 477 – 487:
Lines 477 – 487: Especially, our results provided several insights on the use of two additional RS products although previous studies already reported their advantages (Andersen et al., 2002; Jiang and Wang, 2019; Parr et al., 2015). First, our studies showed the benefits of the two additional RS products for SWAT modeling with an emphasis on equifinality reduction. Especially, SWAT improvements with the single use of RS-ET (Becker et al., 2019; Rajib et al., 2018; Wambura et al., 2018) or RS-LAI (Ma et al., 2019; Rajib et al., 2020) have been well investigated while the simultaneous use of the two RS products has been rarely conducted. In addition, a substantial reduction of model uncertainty was highlighted by the model evaluation at two spatial scales using two RS products. Lastly, this study chose the two RS products frequently used to monitor croplands, and thus our results could inform an improved modeling approach for agricultural watersheds.
Citation: https://doi.org/10.5194/hess-2022-187-AC3
-
AC3: 'Reply on RC2', Sangchul Lee, 07 Sep 2022
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
969 | 277 | 42 | 1,288 | 95 | 43 | 40 |
- HTML: 969
- PDF: 277
- XML: 42
- Total: 1,288
- Supplement: 95
- BibTeX: 43
- EndNote: 40
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1