the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 13 May 2024
Integration of the Vegetation Phenology Module Improves Ecohydrological Simulation by the SWAT-Carbon Model
Abstract. Vegetation phenology and hydrological cycles are closely interacted from leaf and species levels to watershed and global scales. As one of the most sensitive biological indicators of climate change, plant phenology is essential to be simulated accurately in hydrological models. Despite the Soil and Water Assessment Tool (SWAT) has been widely used for estimating hydrological cycles, its lack of integration with the phenology module has led to substantial uncertainties. In this study, we developed a process-based vegetation phenology module and coupled it with the SWAT-Carbon model to investigate the effects of vegetation dynamics on runoff in the upper reaches of Jinsha River watershed in China. The modified SWAT-Carbon model showed reasonable performance in phenology simulation, with root mean square error (RMSE) of 9.89 days for the start-of-season (SOS) and 7.51 days for the end-of-season (EOS). Simulations of both vegetation dynamics and runoff were also substantially improved compared to the original model. Specifically, the simulation of leaf area index significantly improved with the coefficient of determination (R2) increased by 0.62, the Nash–Sutcliffe efficiency (NSE) increased by 2.45, and the absolute percent bias (PBIAS) decreased by 69.0 % on average. Additionally, daily runoff simulation also showed notably improvement, particularly noticeable in June and October, with R2 rising by 0.22 and NSE rising by 0.43 on average. Our findings highlight the importance of integrating vegetation phenology into hydrological models to enhance modeling performance.
- Preprint
(2551 KB) - Metadata XML
-
Supplement
(556 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on hess-2024-75', Anonymous Referee #1, 29 Jun 2024
The paper “Integration of the Vegetation Phenology Module Improves Ecohydrological Simulation by the SWAT-Carbon Model” modifies the SWAT-Carbon model to include a process based method for estimating the start and end of the growing season based on parameterizations of environmental conditions. The paper is logically structured and provides an interesting and relevant analysis. The introduction provides a nice overview of previous research and sets up the study very nicely. The data and methods section describes the details of the experiment very well and is easy to understand. The results present an interesting analysis that compares the two model runs. However, there are two main shortcomings of the paper. First, the paper would benefit from detailed revisions to address several grammatically incorrect and awkward sentences throughout the paper. This would greatly improve the readability and overall quality of the paper. Second, the discussion section is inherently weak and needs extensive revisions. There is little to no discission about the uncertainty of the underlying data sets and models and contextualizing this uncertainty within the analysis. There also seems to be little connection between the discussion and the results presented within the body of the paper. This makes it unclear as to what are the main results from the work. Overall, the paper is interesting and well suited for publication in HESS after some revisions to address these shortcomings. Some specific examples of ways to improve the manuscript are given below.
Revisions:
Line 123: This section needs a few sentences describing how “the current SWAT-Carbon model preforms poorly in estimating vegetation dynamics”. This will then help justify why the authors decide to add the UniChill model and the DM model.
Line 218 – Changing the none growing season LAI to a non-zero value greatly improved the statistics, but this should have been done for the original model too to make it a fair comparison. As is, the statistical improvement between the original and the modified is mostly due to this arbitrary choice of changing the none growing season LAI to a non-zero value and does not capture the improvement in the model’s ability to estimate SOS and EOS. The improvement from the two different changes in the modified model should be carefully analyzed and discussed.
Line 250: Be careful with the wording of “underestimated the future increases in runoff”. You cannot “underestimate” something that is not known. It would be better to say “the original SWAT-Carbon model shows a smaller increase in future runoff compared to the modified SWAT-Carbon”. Then this can be followed up with a discussion about why the original SWAT carbon underestimated historical data and what that may mean about the future.
Line 258: The discussion in section 4.1 is inherently weak. Discussion should focus on contextualizing the results from the study with broader results in the research. Yet, there is little discussion about the results from this paper (lines 266-268) and it mostly reads like an introduction section that provides broad background. The paper would be improved by starting this section about what improvements this work demonstrated and then discussing what they mean in a broader context as well as potential sources of uncertainty.
Line 275: This discussion section also needs revisions. When does the analysis show “a positive correlation between ET and growing season length”? If the figures in supplementary material are so important that they are worth discussing in the discussion section, then they should probably be included in the paper. Again, like the last section, it is difficult to parse out what the authors are discussing as a main finding in the work as most of the discussion section is broad and is only loosely connected with the results.
Line 293 – Again, in this section the authors are primarily focusing on results from the supplementary material (Figure S3) and do not reference much of the results in the paper. This should be better aligned with the main message of the paper and connect the discussion to the main findings.
Line 302 – This paragraph while under the heading of section 4.3, is more general than just section 4.3 so should maybe be part of a new section 5 conclusions where the broader results of the work are summarized. This paragraph is also redundant with other parts of the discussion sections.
Citation: https://doi.org/10.5194/hess-2024-75-RC1 - AC1: 'Reply on RC1', Yongshuo H. Fu, 09 Aug 2024
- AC3: 'Reply on RC1', Yongshuo H. Fu, 07 Nov 2024
-
RC2: 'Comment on hess-2024-75', Anonymous Referee #2, 30 Oct 2024
This study developed a process-based vegetation phenology module and coupled it with the SWAT-Carbon model. This modified model demonstrates improved performance in simulating both vegetation dynamics and runoff in the upper reaches of the Jinsha River watershed, and it is applied to investigate the vegetation effects on runoff. This study shows the strong influence of vegetation phenology on hydrological processes and highlights the importance of integrating a phenology module into hydrological models. The manuscript is well-written and improves the SWAT-Carbon model for historical and future ecohydrological simulation under climate change, though some details need to be further explained and modified.
Some detailed suggestions and comments are listed below:
1. Introduction: The manuscript does not emphasize the significance of choosing the Jinsha River watershed as the study area for applying the modified model. It would be better to clarify the importance of this study area in light of its ecohydrological characteristics.
2. Line 100: Why are these four CMIP6 models chosen for prediction? Some models, such as CanESMs, have coarse spatial resolution especially when applied at the watershed scale. The authors should clarify the considerations for their selection of these climate models.
3. Line 250: Future runoff changes simulated by the modified SWAT-Carbon model could be compared not only with those simulated by the original SWAT-Carbon model but also with runoff directly provided by CMIP6.
4. Figure 2: The information (e.g., abbreviations) in the figures should be clearly interpreted in the figure caption.
5. Figure 6: These times series should include shading to represent the uncertainty, as shown in Figure 5？
6. Abbreviations in the text need to be consistent to enhance readability.
Citation: https://doi.org/10.5194/hess-2024-75-RC2 -
AC2: 'Reply on RC2', Yongshuo H. Fu, 07 Nov 2024
We thank the referee for the supportive and constructive comments, and value their assistance in the acknowledgements. We have revised the manuscript according to the referee’s comments. Please find our point-by-point response to each comment raised in the attached PDF document.
-
AC2: 'Reply on RC2', Yongshuo H. Fu, 07 Nov 2024
Status: closed
-
RC1: 'Comment on hess-2024-75', Anonymous Referee #1, 29 Jun 2024
The paper “Integration of the Vegetation Phenology Module Improves Ecohydrological Simulation by the SWAT-Carbon Model” modifies the SWAT-Carbon model to include a process based method for estimating the start and end of the growing season based on parameterizations of environmental conditions. The paper is logically structured and provides an interesting and relevant analysis. The introduction provides a nice overview of previous research and sets up the study very nicely. The data and methods section describes the details of the experiment very well and is easy to understand. The results present an interesting analysis that compares the two model runs. However, there are two main shortcomings of the paper. First, the paper would benefit from detailed revisions to address several grammatically incorrect and awkward sentences throughout the paper. This would greatly improve the readability and overall quality of the paper. Second, the discussion section is inherently weak and needs extensive revisions. There is little to no discission about the uncertainty of the underlying data sets and models and contextualizing this uncertainty within the analysis. There also seems to be little connection between the discussion and the results presented within the body of the paper. This makes it unclear as to what are the main results from the work. Overall, the paper is interesting and well suited for publication in HESS after some revisions to address these shortcomings. Some specific examples of ways to improve the manuscript are given below.
Revisions:
Line 123: This section needs a few sentences describing how “the current SWAT-Carbon model preforms poorly in estimating vegetation dynamics”. This will then help justify why the authors decide to add the UniChill model and the DM model.
Line 218 – Changing the none growing season LAI to a non-zero value greatly improved the statistics, but this should have been done for the original model too to make it a fair comparison. As is, the statistical improvement between the original and the modified is mostly due to this arbitrary choice of changing the none growing season LAI to a non-zero value and does not capture the improvement in the model’s ability to estimate SOS and EOS. The improvement from the two different changes in the modified model should be carefully analyzed and discussed.
Line 250: Be careful with the wording of “underestimated the future increases in runoff”. You cannot “underestimate” something that is not known. It would be better to say “the original SWAT-Carbon model shows a smaller increase in future runoff compared to the modified SWAT-Carbon”. Then this can be followed up with a discussion about why the original SWAT carbon underestimated historical data and what that may mean about the future.
Line 258: The discussion in section 4.1 is inherently weak. Discussion should focus on contextualizing the results from the study with broader results in the research. Yet, there is little discussion about the results from this paper (lines 266-268) and it mostly reads like an introduction section that provides broad background. The paper would be improved by starting this section about what improvements this work demonstrated and then discussing what they mean in a broader context as well as potential sources of uncertainty.
Line 275: This discussion section also needs revisions. When does the analysis show “a positive correlation between ET and growing season length”? If the figures in supplementary material are so important that they are worth discussing in the discussion section, then they should probably be included in the paper. Again, like the last section, it is difficult to parse out what the authors are discussing as a main finding in the work as most of the discussion section is broad and is only loosely connected with the results.
Line 293 – Again, in this section the authors are primarily focusing on results from the supplementary material (Figure S3) and do not reference much of the results in the paper. This should be better aligned with the main message of the paper and connect the discussion to the main findings.
Line 302 – This paragraph while under the heading of section 4.3, is more general than just section 4.3 so should maybe be part of a new section 5 conclusions where the broader results of the work are summarized. This paragraph is also redundant with other parts of the discussion sections.
Citation: https://doi.org/10.5194/hess-2024-75-RC1 - AC1: 'Reply on RC1', Yongshuo H. Fu, 09 Aug 2024
- AC3: 'Reply on RC1', Yongshuo H. Fu, 07 Nov 2024
-
RC2: 'Comment on hess-2024-75', Anonymous Referee #2, 30 Oct 2024
This study developed a process-based vegetation phenology module and coupled it with the SWAT-Carbon model. This modified model demonstrates improved performance in simulating both vegetation dynamics and runoff in the upper reaches of the Jinsha River watershed, and it is applied to investigate the vegetation effects on runoff. This study shows the strong influence of vegetation phenology on hydrological processes and highlights the importance of integrating a phenology module into hydrological models. The manuscript is well-written and improves the SWAT-Carbon model for historical and future ecohydrological simulation under climate change, though some details need to be further explained and modified.
Some detailed suggestions and comments are listed below:
1. Introduction: The manuscript does not emphasize the significance of choosing the Jinsha River watershed as the study area for applying the modified model. It would be better to clarify the importance of this study area in light of its ecohydrological characteristics.
2. Line 100: Why are these four CMIP6 models chosen for prediction? Some models, such as CanESMs, have coarse spatial resolution especially when applied at the watershed scale. The authors should clarify the considerations for their selection of these climate models.
3. Line 250: Future runoff changes simulated by the modified SWAT-Carbon model could be compared not only with those simulated by the original SWAT-Carbon model but also with runoff directly provided by CMIP6.
4. Figure 2: The information (e.g., abbreviations) in the figures should be clearly interpreted in the figure caption.
5. Figure 6: These times series should include shading to represent the uncertainty, as shown in Figure 5？
6. Abbreviations in the text need to be consistent to enhance readability.
Citation: https://doi.org/10.5194/hess-2024-75-RC2 -
AC2: 'Reply on RC2', Yongshuo H. Fu, 07 Nov 2024
We thank the referee for the supportive and constructive comments, and value their assistance in the acknowledgements. We have revised the manuscript according to the referee’s comments. Please find our point-by-point response to each comment raised in the attached PDF document.
-
AC2: 'Reply on RC2', Yongshuo H. Fu, 07 Nov 2024
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 479 | 178 | 259 | 916 | 51 | 35 | 31 |
- HTML: 479
- PDF: 178
- XML: 259
- Total: 916
- Supplement: 51
- BibTeX: 35
- EndNote: 31
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| China | 1 | 230 | 26 |
| United States of America | 2 | 228 | 26 |
| Germany | 3 | 38 | 4 |
| Romania | 4 | 31 | 3 |
| undefined | 5 | 21 | 2 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 230