the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Revealing joint evolutions and causal interactions in complex eco-hydrological systems by a network-based framework
Abstract. Climate change and human activities have evidence to change eco-hydrological systems, yet the complex relationships among ecological (normalized difference vegetation index, gross primary productivity, and water use efficiency) and hydrological variables (runoff, soil water storage, groundwater storage, etc.) remain understudied. This study develops a novel framework based on network analysis alongside satellite data and in-situ observations to delineate the joint evolutions (phenomena) and causal interactions (mechanisms) in complex systems. The former employs correlations and the latter uses physically constrained causality analysis to construct network relationships. This framework is applied to the Yellow River basin, a region undergoing profound eco-hydrological variations. Results suggest that joint evolutions are controlled by compound drivers and direct causality. Different types of network relationships are found, namely, joint evolution with weak causality, joint evolution with high causality, and asynchronous evolution with high causality. The upstream alpine subregions, for example, where the ecological subsystem is more influenced by temperature while the hydrological one is more driven by precipitation, show relatively high synchronization but with weak and lagged causality between two subsystems. On the other hand, eco-hydrological causality can be masked by intensive human activities (revegetation, water withdrawals, and reservoir regulation), leading to distinct evolution trends. Other mechanisms can also be deduced. Reductions in growing season water use efficiency are directly caused by the control of evapotranspiration, and the strength of control decreases with the greening land surface in some subregions. Overall, the proposed framework provides insight into the complex interactions within the eco-hydrological systems for the Yellow River basin and has applicability to broader geographical contexts.
- Preprint
(2189 KB) - Metadata XML
-
Supplement
(967 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on hess-2024-226', Anonymous Referee #1, 26 Aug 2024
General comments:
The main objective of this paper is to provide a new perspective to analyze eco-hydrological systems based on network approaches. The integrated framework characterized the joint evolution and causal interactions in the complex system at the levels of “phenomena” and “mechanisms”, respectively. In particular, I think this study made good attempts to clarify causality between variables of different types (runoff, soil water storage, groundwater storage, normalized difference vegetation index, gross primary productivity, water use efficiency, etc.) by constructing causal networks. The framework was then applied in the Yellow River Basin, China. The results are generally interesting and reasonable. This paper is overall well-structured and well-written.
Despite the proposed framework is promising, the manuscript requires improvements to better illustrate both the methodology and the results sections. In addition, some grammatic errors and figures should be revised. Below are the detailed comments for consideration.
Comments in details:
- Methodology: Theflow chart and a large amount of eco-hydrological variables appear abruptly. Before introducing the flowchartand methods, I suggest adding a concept diagram depicting interactions between the hydrosphere and the biosphere. This diagram should illustrate the eco-hydrological processes in greater detail than Figure 2. Then the authors is suggested to explain why they have chosen these variables (R, TWSA, SMSA, GWSA, NDVI, etc.) for this study.
- Line 152: There are many causal inference methods other than PCMCI, such as Convergent Cross Mapping (CCM) and Granger Causality (GC). Can you briefly explain why PCMCI was usedin this study?
- Line 154: Provide a full name of PC as the term is first appeared.
- Equation (6): The symbol of ⊥ is not clarified.
- Line 190: I think a citation is required for the additive model.
- Section 2.3.3: This section is interesting, but how such possible links(physical constraints) incorporated to the causality algorithm (PCMCI) is not clear enough. More explanation is needed.
- Lines 230-236: The Yellow River Basin is divided into several subregions, but the general conditions of these regions arenot fully introduced.I suggest that more information on this should be presented, in order to help explain the eco-hydrological mechanisms in the following sections and to help the readers easily relate different subregions to their corresponding results.
- Line 242: I would like to change “which is divided”into “with the basin divided”.
- Lines266-268: The ways to calculate surface water storageand soil water storage are not clear. For example, what specific components from GLDAS are included in soil water storage/surface water storage? Please add some details.
- Line 268: Revise “soil moisture storage water storage”.
- Line 287: Change “withdrawals”to “withdrawal”.
- Figures 4 (a)-(h): Labels for horizontal coordinatesare missing.
- Figure 5: I think it is better to present the S values (i.e., the synchronization between the two subsystems) in Figure 5b as well.
- Figure 6: This figure is very important, but some of the lines are not clear enough (especially the dash lines). In addition, the resolution of the figure should be improved.
- Line 336: Change “increase”to “increases”.
- Line 407: Change “insignificant WUE decrease”to “an insignificant WUE decrease”. Change “significant WUE decrease”to “a significant WUE decrease”.
- Line 435: I know that PCMCI can calculate autocorrelationof each variable when performing causality analysis. However, I do not see any results regarding autocorrelation. I suggest to add more details about this or just remove “autocorrelation”in this sentence.
- Line 534: Change “to understand”to “for understanding”.
Citation: https://doi.org/10.5194/hess-2024-226-RC1 - AC1: 'Reply on RC1', Lu Wang, 20 Sep 2024
-
RC2: 'Comment on hess-2024-226', Anonymous Referee #2, 26 Aug 2024
With great interest, I have read and reviewed the manuscript by Wang et al. This manuscript explores the joint evolution and causal interactions within eco-hydrological systems by introducing a comprehensive framework that integrates correlation relationships, causality analysis, together with satellite data and in-situ observations. Eight subregions of the Yellow River Basin (YRB) that I am interested in are used as cases for study. Correlations between ecological and hydrological subsystems are found to be decoupled in downstream areas, with the underlying causes investigated through causality analysis and attributed to various human activities. In addition, factors such as climatic forcing are found to create spurious relationships between eco-hydrological variables.
To my opinion, the study presents a promising framework and provides some interesting insights on eco-hydrological interactions in the YRB. The topic of the paper is timely and relevant to the readership of this journal. My recommendation is to be accepted after the following points are revised.
Major comments:
(1) One critical issue is that the description of some technical terms is difficult to understand, such as modularity and the degree of synchronization (Section 2.2.2). An explanation in the form of a diagram would make the terms clearer. A schematic diagram of the causal discovery process (Section 2.3) is also suggested to improve the readability of the corresponding texts.
(2) Further analysis and discussion of the results would be good (especially in Sections 5.1.1 and 5.1.2). There are many studies focusing on the ecohydrological processes in the YRB, more comparisons with these studies are suggested to increase the reliability of the results. In addition, are there any other cases that illustrate such confounding issues in Section 5.1.1?
(3) Although the authors have stated the importance of proposed approaches, more discussion on this is suggested. Eco-hydrological/hydrological models also analyze eco-hydrological interactions, so what are the advantages of your methods over physically-based models? It is promising that “such findings are important to understand the general watershed functioning and could further guide the development of more accurate and region-specific eco-hydrological models (lines 534-535)”, and I think it would be better to give more explanations.
(4) Occasional grammatical errors should be checked and corrected.
Minor comments:
(1) Lines 24-27 - It would be better to expand the introduction of eco-hydrological systems and internal interactions with more information.
(2) Lines 85-87 - More emphasis should be placed on the reason for using the YRB as the study area.
(3) Line 88 - A short presentation of the structure of the paper would be good here.
(4) Lines 125-126 - I think the threshold has a significant influence on the construction of the network. How would the network and clustered modules change if you used a different threshold?
(5) Figure 2 - In terms of physically possible and plausible links, there is a connection between soil water storage (SMSA) and gross primary productivity (GPP), but not between soil water storage (SMSA) and normalized difference vegetation index (NDVI). Why is this?
(6) Lines 201-202 - “PCMCI tests possible links and provides the final results as a subset of the total possible network……” However, in Figure 6, there are lines between SMSA and NDVI, as well as between GWSA and GPP (although you have defined them as spurious ones). It seems that the links beyond your hypothesis are also tested. Please check if the expression here is correct.
(7) Section 4.1 - The ecohydrological conditions of eight subregions are not clear enough to me. This may hinder the understanding of the underlying mechanisms in the following sections. Apart from trends, I would recommend describing the average conditions of the subregions in brief.
(8) Figure 4 - “A gray box denotes no data ……” However, grey and blue are difficult to distinguish in Figures 4(d) and 4(h). In addition, Figure 4(h) lacks a “)”, and the symbol “*” in Figure 4(i) is difficult to recognize.
(9) Figures 5 and 6 - The resolution needs to be enhanced.
(10) Figure 6 - This figure is interesting and contains a large amount of information. To my best knowledge, the source region (subregion I) has frozen soil, yet temperature does not appear to significantly affect soil moisture. Could you explain this further?
(11) Lines 355-359 - “Instead, increased T (Figure S3) was the dominant factor stimulating GPP……” “Meanwhile, increased P (Figure S3) was the crucial driver of the increases in the hydrological subsystem……” To interpret the mechanisms clearer, I prefer to present the temperature and precipitation time series (or trends) in the main body of the manuscript.
(12) Lines 375-377 - I think “essentially” here is strange. The sentence needs to be rephrased.
(13) Line 420 - I think “modest” here is not appropriate. The sentence needs to be rephrased.
Citation: https://doi.org/10.5194/hess-2024-226-RC2 - AC2: 'Reply on RC2', Lu Wang, 06 Oct 2024
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
410 | 141 | 21 | 572 | 75 | 4 | 7 |
- HTML: 410
- PDF: 141
- XML: 21
- Total: 572
- Supplement: 75
- BibTeX: 4
- EndNote: 7
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1