the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Refining Remote Sensing precipitation Datasets in the South Pacific: An Adaptive Multi-Method Approach for Calibrating the TRMM Product
Abstract. Calibration techniques are gaining popularity in climate research for refining numerical model outputs, favored for their relative simplicity and fitness-for-purpose in many climate impact applications. Their range of applicability goes beyond numerical model outputs and can be applied to calibrate remote sensing datasets that can exhibit important biases as compared to in situ meteorological observations. This study presents an adaptive calibration approach specifically designed for calibrating the Tropical Rainfall Measuring Mission (TRMM) precipitation product across multiple stations in the South Pacific. The methodology involves the daily classification of the target series into five distinct Weather Types (WTs) capturing the diverse spatio-temporal precipitation patterns in the region. Various quantile mapping (QM) techniques, including empirical (eQM), parametric (pQM), and Generalized Pareto Distribution (gpQM), as well as an ordinary scaling, are applied for each WT. We perform a comprehensive validation by evaluating 10 specific precipitation-related indices that hold significance in impact studies, which are then combined into a single Ranking Framework (RF) score, which offers a comprehensive evaluation of the performance of each calibration method for every Weather Type (WT). These indices are assigned user-defined weights, allowing for a customized assessment of their relative importance to the overall RF score. Our 'adaptive' approach selects the best performing method for each WT based on the RF score, yielding an optimally calibrated series.
Our findings indicate that the adaptive calibration methodology surpasses standard and weather-type conditioned methods based on a single technique, yielding more accurate calibrated series in terms of mean a extreme precipitation indices consistently across locations. Moreover, this methodology provides the flexibility to customize the calibration process based on user preferences, thereby allowing for specific indices, such as extreme rainfall indicators, to be assigned higher weights. This ability enables the calibration to effectively address the influence of intense rainfall events on the overall distribution. Furthermore, the proposed adaptive method is highly versatile and can be applied to different scenarios, datasets, and regions, provided that a prior weather typing exists to capture the pertinent processes related to regional precipitation patterns. Open-source code and illustrative examples are freely accessible to facilitate the application of the method.
- Preprint
(1364 KB) - Metadata XML
-
Supplement
(702 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2023-1402', Mingxing Li, 14 Nov 2023
The manuscript by Mirones et. al., focuses on the methodology for calibrating the TRMM precipitation data. The authors used Multi-method approach for data from five observational stations. The major concerns from my perspective include: while the results showed better effects of the adaptive calibration method based on five station, the limited representation of the five stations is not enough to support the generalization of method to entire low-middle latitudes. Second, what are the deficiencies of TRMM precipitation data, and how the adaptive method resoled the problem targeted to those deficiencies. Third, how the method performs over the entire low and middle latitudes.
A minor commend:
Fig. 3, Do the horizontal axes represent RF scores?
Citation: https://doi.org/10.5194/egusphere-2023-1402-RC1 -
AC1: 'Reply on RC1', Oscar Mirones, 15 Dec 2023
Iam pleased to provide detailed responses to all the comments raised. Please find attached a document containing a point-by-point response to your feedback, including a "diff" file with the description of the changes made in the manuscript to address your concerns.
-
AC1: 'Reply on RC1', Oscar Mirones, 15 Dec 2023
-
RC2: 'Comment on egusphere-2023-1402', Anonymous Referee #2, 17 Jan 2024
An “adaptive calibration method”, which is based on five weather types to take advantage of the complementary strengths for four methods (i.e., scaling, empirical quantile mapping (eQM), parametric quantile mapping (pQM), and generalized Pareto distribution quantile mapping (gpQM)), was proposed in this paper for calibrating the TMPA product. The validation of the “adaptive calibration method” was executed according to comparison with eQM, pQM, gpQM, scaling, and unweighted method. Although TMPA is a good precipitation product and has been widely used in various fields, it has stopped updating and its successor IMERG has higher spatiotemporal resolution and accuracy. I believe that choosing TMPA as a calibration object does not make much sense. In addition, the primary drawback of this manuscript is that the “adaptive calibration approach” was not clearly explained in the manuscript, such as how to classify the weather types, why selecting those 10 metrics and how to obtain their optimal weights. More importantly, the novelty of the manuscript was not well represented. Therefore, I cannot suggest the acceptance of this manuscript at the current stage.
Major comments:
- The TMPA product was used as a calibration object. The reasons for choosing TMPA should be provided. In particular, its successor. i.e., IMERG, has been released to the public and is better than TMPA in multiple aspects such as resolution, covering period, and quality. Moreover, the TMPA products have stopped updating. I am confused that this study selected TMPA instead of IMERG.
- Methodology: the method section is one of the important content, but the current descriptions are too concise. Based on five weather types, the paper proposed an “adaptive calibration approach” to take advantage of the complementary strengths for four methods and further improve the accuracy of the TMPA product. However, the weighting method was not described well, especially its standards and rules. In addition, could the weighting method used in this paper provide optimal weights? In other words, whether this weighting method can maximize the advantages of different error adjustment methods. In addition, the classification of weather types was not provided. I understand the authors gave the related literature, but this is not enough. A detailed classification scheme should be provided as it is an important component for the “adaptive calibration approach”.
- Results and discussion: The analysis has many subjective descriptions, which is not recommended for scientific writing. I suggest the analysis provides some specific metric values and strengthens in-depth interpretations on the results.
Other comments
- Lines 36-37: this conclusion is true only in some areas, not all regions. I suggest revising this sentence to avoid misleading the readers.
- The differences between the four methods (i.e., scaling, eQM, pQM, and gpQM) should be provided and discussed, especially for their advantages and limitations. Due to their respective advantages, the adaptive calibration approach, which could consider their advantages, is necessary. Meanwhile, the adaptive calibration approach is not the best in all cases, which is worth analyzing.
- Weather typing is important in this study. So, which five types? The description lacks detail.
- Lines 159-161: could the authors provide some literature to support this point?
- Lines 161-162: can the authors provide some study results (e.g., specific metric values) to support this point?
- Lines 166-167: what results?
- Why did the adaptive calibration method not improve the accuracy at the Rarotonga and Nu’uuli stations?
Citation: https://doi.org/10.5194/egusphere-2023-1402-RC2 -
AC2: 'Reply on RC2', Oscar Mirones, 14 Feb 2024
We are pleased to provide detailed responses to all the comments raised. Please find attached a document containing a point-by-point response to your feedback, including a "diff" file with the description of the changes made in the manuscript to address your concerns.
-
AC3: 'Reply on RC2', Oscar Mirones, 19 Feb 2024
We are pleased to provide detailed responses to all the comments raised. Please find attached a document containing a point-by-point response to your feedback, including a "diff" file with the description of the changes made in the manuscript to address your concerns.
Viewed
HTML | XML | Total | Supplement | BibTeX | EndNote | |
---|---|---|---|---|---|---|
365 | 137 | 45 | 547 | 38 | 39 | 28 |
- HTML: 365
- PDF: 137
- XML: 45
- Total: 547
- Supplement: 38
- BibTeX: 39
- EndNote: 28
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1