the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 03 Sep 2024
Technical note: What does the Standardized Streamflow Index actually reflect? Insights and implications for hydrological drought analysis
Abstract. Hydrological drought is one of the main hydroclimatic hazards worldwide, affecting water availability, ecosystems and socioeconomic activities. This phenomenon is commonly characterized by the Standardized Streamflow Index (SSI), which is widely used because of its straightforward formulation and calculation. Nevertheless, there is limited understanding of what the SSI actually reveals about how climate anomalies propagate through the terrestrial water cycle. To find possible explanations, we implemented the SUMMA hydrological model coupled with the mizuRoute routing model in six hydroclimatically different case study basins located on the western slopes of the extratropical Andes, and examined correlations between the SSI (computed from the models for 1, 3 and 6-month time scales) and potential explanatory variables – including precipitation and simulated catchment-scale storages – aggregated at different time scales. Additionally, we analyzed the impacts of adopting commonly used time scales on propagation analyses of specific drought events – from meteorological to soil moisture and hydrological drought – with focus on their duration and intensity. The results reveal that the choice of time scale for the SSI has larger effects on correlations with explanatory variables in rainfall-dominated regimes compared to snowmelt-driven basins, especially when simulated fluxes and storages are aggregated to time scales longer than 9 months. In all the basins analyzed, the strongest relationships (Spearman rank correlation values over 0.7) were obtained when using 6-month aggregations to compute the SSI and 9–12 months to compute the explanatory variables, excepting aquifer storage in snowmelt-driven basins. Finally, the results show that the trajectories of drought propagation obtained with the Standardized Precipitation Index (SPI), the Standardized Soil Moisture Index (SSMI) and the SSI may change drastically with the selection of time scale. Overall, this study highlights the need for caution when selecting standardized drought indices and associated time scales, since their choice impacts event characterizations, monitoring and propagation analyses.
- Preprint
(3380 KB) - Metadata XML
-
Supplement
(1597 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on hess-2024-221', Anonymous Referee #1, 19 Nov 2024
General comment
The paper demonstrates strong scientific significance, high quality, and effective presentation. The authors investigate hydrological droughts, focusing on Standardized Drought Indices (SDI), notably the Standardized Streamflow Index (SSI), as tools for understanding drought dynamics, including frequency, intensity, duration, and propagation. The SUMMA hydrological model and the mizuRoute routing model were calibrated and used to analyze six case study basins in the western extratropical Andes. This analysis explores the relationship between SSI and various explanatory basin-aggregated variables, such as precipitation and catchment storage, across different time scales.
The authors address the common use of SSI to quantify hydrological drought without clearly understanding its effectiveness in capturing the dynamics of drought propagation in basins with diverse hydrological regimes. This key issue is woven throughout the paper, making it enjoyable to read.
Although the authors work with a limited number of basins, their analysis has implications beyond the extratropical Andean region, particularly in operational contexts where regulators and decision-makers rely on simplified and easily accessible drought indices without further analysis or differentiation between hydrological regimes. The discussion section provides a solid overview of the approach's limitations and potential avenues for future research. The figures convey important information, making it easy to grasp the main findings.
I have made a few minor comments that I hope will help strengthen the manuscript:
- While the justification for using a hydrological model as a benchmark for evaluating SDIs over observational data is clear, I believe the paper would benefit from a more detailed discussion of the inherent limitations of using a model to represent the long-term behavior of drought in the chosen basins. Analyzing how the model's process representation might lead to inaccuracies in streamflow reproduction—especially in basins where the modeled minimum streamflow values exceed the observed values—could provide additional insight into the analysis.
- I suggest the authors include more information regarding the model's warm-up period or restrict the time results of their full simulation to account for this warm-up period.
- Lastly, I would appreciate it if the authors could elaborate on how they envision the design of regional analysis frameworks that consider more than just hydrological regimes, including similarities in physical features such as slope, elevation, soil properties, and land cover, among others.
In conclusion, this is a well-crafted paper that, with some minor revisions, should make a valuable contribution to the field.
Specific comments
L235-L245: Can the authors discuss the implications of using a model that overestimates the low flow volumes (exceedance probabilities over 90%) for analyzing hydrological droughts for the Choapa and Claro cases? Additionally, what are the most likely causes of the model's misrepresentation of low flow volumes in those basins? Can some of the selected process parameterizations (i.e., snowmelt) negatively affect the obtained results in this respect? Also, did the authors evaluate the influence of multiple model parameterizations on the obtained results?
L260-L276: Did the authors consider a spin-up period for the model before starting the analysis in 04/1983? If so, please include this information. If not, I'd recommend neglecting the first two simulation years (1983-1984) in the subsequent analysis to minimize the influence of initial conditions in the obtained results.
Technical corrections / Typos
L36: “associated to” replace with “associated with.”
L42: “Despite the drought concept refers” replace with “Despite the drought concept referring.”
L80: “percentile-based thresholds that are commonly” replace with “percentile-based thresholds commonly.”
L92: “What are the effects of different time scales on” replace with “How do different time scales affect”
L94: “towards” replace with “toward”.
L96: “To seek for answers,” replace with “To seek answers,”
L113: “Hereafter, to” replace with “Hereafter,”
L117 & L118: “mean annual temperatures between 9 to 16 °C” and “aridity indices between 0.4 to 3” replace with “mean annual temperatures between 9 and 16 °C” & “aridity indices between 0.4 and 3”.
L349: “drought durations ranging 12.3-12.9 months” replace with “drought durations ranging from 12.3-12.9 months”
L386: “SSI is not as relevant in snowmelt-driven basins, compared to mixed regime and rainfall-dominated catchments.” This statement appears weak. I recommend replacing it with: “SSI is less relevant in snowmelt-driven basins than in mixed regimes and rainfall-dominated catchments.”
Citation: https://doi.org/10.5194/hess-2024-221-RC1 -
RC2: 'Comment on hess-2024-221', Anonymous Referee #2, 27 Nov 2024
Review of the Manuscript: "Technical Note: What Does the Standardized Streamflow Index Actually Reflect? Insights and Implications for Hydrological Drought Analysis"
The manuscript provides a detailed investigation into the Standardized Streamflow Index (SSI), a widely used metric for characterizing hydrological drought. The authors employ the SUMMA hydrological model coupled with the mizuRoute routing model across six diverse basins in the Andes to explore the relationships between SSI and potential explanatory variables. Their analysis extends to the impacts of time-scale selection on drought propagation, offering insights into the complex dynamics of drought characterization. The study emphasizes the importance of cautious selection of drought indices and time scales, highlighting their influence on event characterization, monitoring, and propagation analysis.
While the study is timely and relevant, there are areas where the manuscript could be improved to enhance its clarity, rigor, and accessibility. A thorough revision will make the manuscript more accessible to a broader audience and enhance its scientific impact.
General Comments
- Clarity and Consistency
The manuscript often lacks consistency in the use of acronyms and terminology. For instance, precipitation is referred to as "pp" and "P" (e.g., Figure 2). This inconsistency can be confusing to readers. Ensure all acronyms are concise, clearly defined, and consistently used throughout the text and figures. Additionally, figures should "stand-alone" with comprehensive captions that explain all acronyms and variables.
- Methodology Organization
The methods section is repetitive and lacks a clear structure. For instance, the "Approach" subsection is confusing and does not align with Figure 2 or the steps described later. Additionally, some methodological details are scattered throughout the manuscript or mixed with results (e.g., lines 263–265). Reorganizing this section for clarity and separating methods from results would greatly improve readability.
- Time-Scale Terminology
The manuscript uses "time scales" to refer to different concepts—indices aggregation periods and aggregated flux/storage—without clear differentiation. This ambiguity makes the text hard to follow. Clearly define these terms early in the methods section and ensure consistent usage throughout.
- Depth of Analysis
While the study highlights interesting patterns, some key analyses, such as drought propagation (lines 82–84), are underexplored in the results and discussion. Either expand on this analysis or remove it to maintain focus and coherence.
- Practical Implications
The discussion section, particularly "Implications for operational practices," is engaging and provides valuable insights. Expanding this section with concrete recommendations or case studies would significantly enhance the manuscript's impact.
Specific Comments
- Line 37: The computation indices are not for the variables “precipitation, simulated soil moisture, and simulated streamflow” but rather for the different drought types that use these variables as inputs. Rephrase for clarity and conciseness.
- Lines 42–45: The phrase "the most commonly used types" is vague and should specify types of what (e.g., drought indices). Reformulate for clarity.
- Line 55: The statement "the number 163 journal articles" is irrelevant, the number 163 does not have a meaning by itself. Focus instead on the themes or findings from these articles related to SSI and drought.
- Line 74: Be specific about "even longer" time scales—e.g., 12 or 24 months.
- Line 77: Mention common meteorological drought indices to provide context for readers less familiar with this field.
- Line 88: The phrase "we depart from previous hydrological drought" needs clarification. Specify what you mean by "previous" or cite relevant studies.
- Lines 135: Add the reference to the model used in this study.
- Lines 138–139: Explain how you are comparing the different time scales. What criteria or statistical approaches are being applied?
- Lines 143–145: Define acronyms again for clarity and specify what is meant by "other indices" and "state variables."
- Dataset Description: Add a table summarizing the spatial and temporal resolution of each data product, along with sources and citations. Begin with the variables required for the model, then delve into dataset specifics. Clarify whether streamflow and catchment characteristics were retrieved from CAMELs.
- Model Calibration and Evaluation: Specify the time scale used for model calibration (e.g., daily or monthly). Additionally, explain the number of trials conducted and the rationale for using the objective function proposed by Garcia (line 175).
- Lines 203–205: Clarify why SSI was excluded from specific analyses and define "longer time scales."
- Figures:
- Figure 1: Add yellow catchment indicators to the legend. Standardize the y-axis scale for mm/month and temperature to enable better comparison between catchments.
- Figure 2: Define all acronyms directly in the figure legend (e.g., pp, temp, Kin, SPI, SPEI, etc.). Avoid phrases like "See text for details." The figure caption should be self-contained.
Citation: https://doi.org/10.5194/hess-2024-221-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 410 | 135 | 13 | 558 | 34 | 8 | 7 |
- HTML: 410
- PDF: 135
- XML: 13
- Total: 558
- Supplement: 34
- BibTeX: 8
- EndNote: 7
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1