the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 20 Nov 2024
A Novel Framework for Analyzing Rainy Season Dynamics in semi-arid environments: A case study in the Peruvian Rio Santa Basin
Abstract. In semi-arid regions, the timing and duration of the rainy season determine plant water availability, which directly impacts food security. Rainy season metrics, which aim to define and, in some cases, predict the onset and end of rains can support agricultural planning, such as scheduling planting dates and managing water resources. However, these metrics based on precipitation time series do not always accurately reflect plant water availability, and the variety of available metrics can complicate the selection of the most suitable one. This study demonstrates that rainy season metrics are more useful for agricultural purposes when their parameters are calibrated using local vegetation data. Furthermore, a metric's ability to capture observed vegetation variability can indicate its applicability over larger spatial or temporal scales. We test this hypothesis in the semi-arid Rio Santa basin in the Peruvian Andes by evaluating seven common rainy season metrics, both calibrated and uncalibrated, against land surface phenology data obtained from 18 years of satellite-derived Normalised Difference Vegetation Index (NDVI) data. Additionally, we introduce a new bucket-type metric that incorporates a simplified water balance, considering both accumulation and storage. To test the robustness of the metrics under future climate scenarios, we examine the sensitivity of these metrics to variations in rainfall intensity and frequency using statistically downscaled CMIP5 rainfall data for historical (1981–2018) and future (2019–2100) periods under RCP 4.5 and 8.5 scenarios. Our results show that calibrating metrics using vegetation data improves their consistency in capturing the start and end dates of the rainy season. The newly introduced bucket metric outperforms the other metrics in both accuracy and robustness. However, some established metrics exhibit sensitivities that raise concerns about their applicability under potential shifts in rainfall patterns due to climate change. Overall, CMIP5 projections reveal no consistent trends in rainy season onset and only a slight delay in rainy season end, with inter-annual variability and ensemble spread being the dominant factors. Our findings highlight the importance of calibrating metrics and stress-testing them across various climate conditions to ensure their agricultural relevance. The framework introduced here can easily be adapted for application in other semi-arid regions.
- Preprint
(14487 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-3263', Jingwei Zhou, 18 Dec 2024
This manuscript presents a case study in understanding rainy season dynamics in semi-arid regions. The study is intended to develop and validate a new framework for calibrating rainy season metrics. The paper is generally clear.
Although CMIP5 data seem to be applicable in your research, I will still suggest you using the newer CMIP6 data, which are accessible in many data portals provided by organizations such as ESGF Copernicus. CMIP5 data has been more than ten years since developed, while CMIP6 incorporate numerous updates and incorporate more features, let alone it may address the issues you do mentioned in your text (lines 345-350). In addition, the implications or significance for this novel method in evaluating rainy season and to be applied in future research could be discussed further. A brief comment on the potential transferability of this framework to other semi-arid regions would be valuable as well.
Overall, I will give this paper the suggestion of moderate revisions
Specific Comments:
Line 17: I suggest change it to “only a slight delay in rainy season end, but no consistent trends in rainy season onset,” to stress the later parts with the bad results
Lines 40-45: The review of existing metrics could be more comprehensive, particularly regarding their applications in similar geographical contexts. Some results from the former study can be put here.
Line 123: maybe you can mention why you choose these two scenarios
Lines 254-269: I think most of these can be moved into methods, just leave some brief descriptions here in the results
Line 320: "Feb."
Lines 345-367: Most of these can be moved into conclusions part. They are discussions from my perspective
Figure 1: you could move the index plot of Rio Santa Basin with NDVI a little further away from the basic topography map. The Basin looks like it’s a big part located within the eastern part of the map
Figures 2 and 3: you could add a legend indicating different data sources and also different methods
Figures 6 and 7: maybe add a legend showing different data sources and different scenarios
Some paragraphs have indentation while others don’t, please keep them consistent
Citation: https://doi.org/10.5194/egusphere-2024-3263-RC1 -
RC2: 'Comment on egusphere-2024-3263', Anonymous Referee #2, 27 Dec 2024
The manuscript presents an interesting and well-structured approach to understanding rainy season dynamics in semi-arid regions. It contributes meaningfully to hydrology and earth system sciences by introducing a novel framework for calibrating rainy season metrics with vegetation data, addressing their applicability under current and future climate scenarios. However, there are areas where improvements can enhance the manuscript’s scientific rigor and presentation quality.
While the manuscript provides valuable insights using CMIP5 data, incorporating CMIP6 would likely enhance the study due to historical forcings, and more comprehensive future scenarios. As noted by another reviewer, it would be appropriate to perform the study with CMIP6 to align with current advancements in climate modeling. Otherwise, the authors should provide a solid justification for relying on CMIP5, addressing how its limitations might affect the results and conclusions of the study.
Here are some areas of improvement:
Line 14: Introducing the bucket-type metric is a significant contribution but could be more prominently emphasized earlier in the abstract. Currently, it feels buried in the middle.
Line 15: The term "sensitivities" might need elaboration—does it refer to responsiveness, instability, or another issue?
Lines 40–45: The discussion about the lack of strategies to validate rainy season metrics based on independent data is important but somewhat abrupt. Also, the emphasis on uncertainties in precipitation measurements is valid but could be expanded with specific examples of how these uncertainties affect the metrics or decision-making.
For example, tie the need for validation directly to the challenges of applying metrics in real-world scenarios. Add a sentence elaborating on the practical implications of these uncertainties, especially for agricultural or water management applications.
Lines 57: The transition to the Rio Santa Basin context is slightly abrupt, and the text does not adequately establish why this region is particularly suitable for testing the proposed framework.
Line 75: The phrase “and two other precipitation datasets for comparison” is vague and could lead to confusion. It is unclear what “other” refers to—whether additional datasets are used for validation, alternative sources of precipitation data, or datasets of different spatial/temporal resolution.
Lines 78: While using NDVI for calibration is well-justified, it might help to briefly highlight why NDVI is a robust proxy for water availability in this region compared to alternatives.
Lines 95: The final objective includes exploring past and future changes in rainy season dynamics, but it does not specify the importance of these changes for the broader context of climate change. It should emphasize how understanding these changes can inform adaptation strategies in similar semi-arid regions.
Figure 1: The climograph at the bottom is helpful but not explicitly referenced in the text. Discuss the precipitation seasonality and temperature trends shown in the climograph and connect them to the rainy season dynamics discussed in the study.
Recommendation: Add a sentence or two explicitly linking the Rio Santa basin's unique hydroclimatic and socioeconomic characteristics to the study's focus on rainy season metrics. Reference specific features of Figure 1 (e.g., NDVI, climograph) in the study area description to better integrate the figure with the text.
Line 112 (Data): The datasets have varying spatial resolutions. How do you account for or interpret these differences in your study?"
Line 150-160: The differentiation between onset-only metrics and metrics that address the onset and end of the rainy season is crucial but is introduced abruptly.
Recommendation: Include a transitional sentence to guide the reader through this distinction (e.g., “While some metrics are focused exclusively on the onset of the rainy season, others provide a more comprehensive approach by also addressing the season’s end”).
Line 179: The formula is clear, but the conditional structure might be challenging for some readers to interpret.
Recommendation: Add a sentence explaining the formula in simpler terms: “The bucket water content is updated daily based on precipitation input and constant evapotranspiration. It is constrained between a minimum and maximum value, ensuring realistic water balance limits.”
Section 3.3 Past and Future: Transitioning from historical trends to future projections feels abrupt. Add a bridging sentence to guide readers, e.g., “Having established the variability in historical trends, we now turn to the projected changes in rainy season metrics up to 2100.”
Figures 6, 7, A7, and A8 are referenced, but their key findings are not fully summarized in the text.
Line 361: While uncertainties are acknowledged (e.g., in evapotranspiration rates), the potential impact on results is not fully explored. Provide more detail on how these uncertainties might affect the interpretation of trends.
The conclusion summarizes key findings well but could improve with clearer transitions, a detailed discussion of the bucket metric's strengths, and actionable practitioner recommendations.
- While ENSO is mentioned, its significance to the study's findings could be elaborated further. Expand on how ENSO-related precipitation anomalies might influence the calibration and robustness of rainy season metrics.
- The flow between discussions of calibration, the bucket metric, and future projections could be smoother. Use transitional phrases to connect paragraphs, e.g., "Building on these findings, we introduced the bucket metric to address limitations in existing methods."
- The conclusions acknowledge uncertainty in projections but do not fully explore its implications for the study’s findings. Discuss how these uncertainties might affect the interpretation of trends or the applicability of metrics in other regions.
- The reliance on vegetation proxies is discussed, but alternative calibration strategies are only briefly mentioned. Elaborate on potential alternative data sources, such as soil moisture or runoff measurements, and their advantages or challenges.
- The conclusion mentions practitioner needs but could provide more actionable recommendations. Include a sentence like, “Practitioners can use the bucket metric to better predict water availability in semi-arid regions, particularly during ENSO-driven precipitation anomalies.”
- The call for more robust climate models is appropriate but could be tied more explicitly to the study’s findings. Link this need to specific challenges identified in the study, e.g., “The limitations of CMIP5 models, such as their inability to resolve ENSO impacts at finer scales, underscore the importance of developing high-resolution convection-permitting models.”
Citation: https://doi.org/10.5194/egusphere-2024-3263-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 150 | 34 | 10 | 194 | 4 | 4 |
- HTML: 150
- PDF: 34
- XML: 10
- Total: 194
- BibTeX: 4
- EndNote: 4
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1