Previous Article in event
Previous Article in session
Next Article in event
Next Article in session
Evaluating extreme precipitation events on the Iberian Peninsula using TRMM satellite data
Published:
10 November 2017
by MDPI
in First International Electronic Conference on the Hydrological Cycle
session Extreme Events
Abstract:
An assessment of extreme precipitation events (EPEs) is performed using the high-resolution (0.2°) gridded daily precipitation database available for the IP, the accumulated precipitation from ERA-Interim reanalysis by EMCWF at 6-hour intervals, and the Tropical Rainfall Measuring Mission (TRMM) Multisatellite Precipitation Analysis (TMPA) dataset, based on multisatellite estimates of precipitation and gauges measurements, for the common period since 1998. This study presents an analysis and validation of the extreme precipitation characteristics over IP, using both satellite and ground observations. Results show that there is a good general agreement between total precipitation analysis from observational gridded and TRMM datasets, both temporal and spatially, although TRMM TMPA results are underestimated when compared to observations and ERA Interim data.
Keywords: extreme precipitation events, natural hazards, Iberian Peninsula, TRMM satellite, gridded precipitation dataset
Comments on this paper
Rosana Nieto Ferreira
15 November 2017
RNF comments
This is a very timely and relevant study of the comparison of TRMM and raingauges in the Iberian Peninsula (IP). It is also a well written and easy to follow study. I have the few minor questions/comments below:
-Section 2.1: what dataset is used to determine extreme events? The gauge data? TRMM? IB02?
-Would the same extreme events have been captured if you used the 0.25 degree resolution TRMM dataset?
-In table 1, what is the 'domain area' and how large is it? Is it only the IP or the whole domain in Figure 1? You explain this near the bottom of page 7 but it is needed when you introduce Table 1.
-A figure showing a typical distribution of the gauges that are used to produce the IB02 and PT02 datasets would have been useful.
-Are any of the rain gauges in the IB02 and PT02 datasets ingested into the TRMM 3B42V7?
-Since there are only 4 events, a brief discussion of the synoptic scale phenomena that produced the rainfall maxima would be easy to include and would be very informative. Are they all cold fronts?
-Page 7: What do you mean by 'Figures (b)'?
-It is not clear what data goes into the scatterplots in Figure 6. Is it the accumulated precipitation in each pixel at the resolution of each dataset?
-What does T stand for in 'T#7', etc?
Obrigada!
-Section 2.1: what dataset is used to determine extreme events? The gauge data? TRMM? IB02?
-Would the same extreme events have been captured if you used the 0.25 degree resolution TRMM dataset?
-In table 1, what is the 'domain area' and how large is it? Is it only the IP or the whole domain in Figure 1? You explain this near the bottom of page 7 but it is needed when you introduce Table 1.
-A figure showing a typical distribution of the gauges that are used to produce the IB02 and PT02 datasets would have been useful.
-Are any of the rain gauges in the IB02 and PT02 datasets ingested into the TRMM 3B42V7?
-Since there are only 4 events, a brief discussion of the synoptic scale phenomena that produced the rainfall maxima would be easy to include and would be very informative. Are they all cold fronts?
-Page 7: What do you mean by 'Figures (b)'?
-It is not clear what data goes into the scatterplots in Figure 6. Is it the accumulated precipitation in each pixel at the resolution of each dataset?
-What does T stand for in 'T#7', etc?
Obrigada!
Margarida L. R. Liberato
16 November 2017
Thank you for your interest in our work and for the careful revision and comments. Some details were not comprehensively described in this proceeding to prevent it from being too extended. We will take them into consideration on the future paper.
We will reply to each point below (R:Reply):
R1:The extreme events were previously determined as explained in Ramos et al. (2014). The dataset available from Ramos et al. (2014) was built using the high resolution daily gridded precipitation dataset for the Iberian Peninsula spanning from 1950 to 2008 (IB02). The datasets and additional information are available at (https://www.ipma.pt/en/produtoseservicos/index.jsp?page=dataset.pt02.xml; http://www.meteo.unican.es/es/datasets/spain02).
R2: It is not possible to reply to this question without extensive research. We would need to apply all the methodology explained in Ramos et al. (2014) to the TRMM 3B42V7 dataset. This work is a preliminary approach to try to reply to this question in the future.
R3: As it is referred in the Table 1 caption, the EPEs occur between 1998 and 2008 for the Iberian Peninsula (IP). Since the EPEs dataset is built using the IB02, the 'domain area' is the continental IP, as defined in Ramos et al. (2014).
R4: The IB02 dataset is built from the combination of two national datasets, “PT02” for mainland Portugal [Belo-Pereira et al., 2011] and “Spain02” for peninsular Spain and Balearic islands [Herrera et al., 2012]. All the details including available information on the gauges used are described comprehensively in these papers. We are users of a validated final dataset and all the information on gauges and data validation is out of scope of this paper.
R5: The Tropical Rainfall Measuring Mission (TRMM) Multisatellite Precipitation Analysis (TMPA) provides precipitation estimates from multiple satellites. TMPA is available both in real time and post-real-time products. Only the post-real-time product (which we are using in this work) incorporates gauge analyses in the final step of the calibration-based sequential scheme where feasible (Huffman et al., 2007). Huffman et al. (2007) state that it is highly advantageous to include rain gauge data in combination datasets. They refer that however, experience shows that on any time scale shorter than a month the gauge data are neither reported with sufficient density nor reported with consistent observational intervals to warrant direct inclusion in a global algorithm. Therefore all available 3-hourly merged estimates are summed over a calendar month to create a monthly multisatellite (MS) product. The MS and gauge are combined as in Huffman et al. (1997) to create a post-real-time monthly satellite-gauge (SG) combination, which is a TRMM research-grade product in its own right. Then the field of SG/MS ratios is computed on the 0.25° x 0.25° grid (with controls) and applied to scale each 3-hourly field in the month. The authors indicate that the result is to provide the high resolution typical of satellite data and the typically small bias of gauge analyses over land. We are aware of this procedure which uses in Version 7 the Global Precipitation Climatology Project (GPCP) Level 3 Full and Monitoring Precipitation Gauge Analyses.
R6: The goal of this proceeding is to present a first assessment of the ability of TRMM TMPA products to study EPEs over the IP, a continental region in the extratropics and to discuss its caveats for the analysis of precipitation extremes. The synoptic analysis of the EPEs is not dependent on the TRMM satellites’ datasets and is out of the scope of this proceeding. This is still work in progress and will soon be submitted.
R7: On this context, Figures (b) refer to all panels (b) from the Figures under consideration. These panels correspond to the evaluation of EPEs using reanalysis. In the first paragraph of Page 7 “Figures (b)” correspond to “panel (b) of Figures 2 to 5”. Similarly, “Figures (c) and (d)” refer to all panels (c) and (d) of Figures 2 to 5.
R8: Yes. Accumulated daily precipitation (mm.day-1) for: TRMM 3B42V7 versus IB02 at 0.2 degree resolution for panels (a) to (d); and TRMM 3B42V7 versus ERA-Interim at 1 degree for panels (e) to (h).
R9: As indicated in the text, “T#9, number 9 in the ranking; Table 1”. The “T#” stands for “Top number”.
Gracias! Un saludo!
The authors
We will reply to each point below (R:Reply):
R1:The extreme events were previously determined as explained in Ramos et al. (2014). The dataset available from Ramos et al. (2014) was built using the high resolution daily gridded precipitation dataset for the Iberian Peninsula spanning from 1950 to 2008 (IB02). The datasets and additional information are available at (https://www.ipma.pt/en/produtoseservicos/index.jsp?page=dataset.pt02.xml; http://www.meteo.unican.es/es/datasets/spain02).
R2: It is not possible to reply to this question without extensive research. We would need to apply all the methodology explained in Ramos et al. (2014) to the TRMM 3B42V7 dataset. This work is a preliminary approach to try to reply to this question in the future.
R3: As it is referred in the Table 1 caption, the EPEs occur between 1998 and 2008 for the Iberian Peninsula (IP). Since the EPEs dataset is built using the IB02, the 'domain area' is the continental IP, as defined in Ramos et al. (2014).
R4: The IB02 dataset is built from the combination of two national datasets, “PT02” for mainland Portugal [Belo-Pereira et al., 2011] and “Spain02” for peninsular Spain and Balearic islands [Herrera et al., 2012]. All the details including available information on the gauges used are described comprehensively in these papers. We are users of a validated final dataset and all the information on gauges and data validation is out of scope of this paper.
R5: The Tropical Rainfall Measuring Mission (TRMM) Multisatellite Precipitation Analysis (TMPA) provides precipitation estimates from multiple satellites. TMPA is available both in real time and post-real-time products. Only the post-real-time product (which we are using in this work) incorporates gauge analyses in the final step of the calibration-based sequential scheme where feasible (Huffman et al., 2007). Huffman et al. (2007) state that it is highly advantageous to include rain gauge data in combination datasets. They refer that however, experience shows that on any time scale shorter than a month the gauge data are neither reported with sufficient density nor reported with consistent observational intervals to warrant direct inclusion in a global algorithm. Therefore all available 3-hourly merged estimates are summed over a calendar month to create a monthly multisatellite (MS) product. The MS and gauge are combined as in Huffman et al. (1997) to create a post-real-time monthly satellite-gauge (SG) combination, which is a TRMM research-grade product in its own right. Then the field of SG/MS ratios is computed on the 0.25° x 0.25° grid (with controls) and applied to scale each 3-hourly field in the month. The authors indicate that the result is to provide the high resolution typical of satellite data and the typically small bias of gauge analyses over land. We are aware of this procedure which uses in Version 7 the Global Precipitation Climatology Project (GPCP) Level 3 Full and Monitoring Precipitation Gauge Analyses.
R6: The goal of this proceeding is to present a first assessment of the ability of TRMM TMPA products to study EPEs over the IP, a continental region in the extratropics and to discuss its caveats for the analysis of precipitation extremes. The synoptic analysis of the EPEs is not dependent on the TRMM satellites’ datasets and is out of the scope of this proceeding. This is still work in progress and will soon be submitted.
R7: On this context, Figures (b) refer to all panels (b) from the Figures under consideration. These panels correspond to the evaluation of EPEs using reanalysis. In the first paragraph of Page 7 “Figures (b)” correspond to “panel (b) of Figures 2 to 5”. Similarly, “Figures (c) and (d)” refer to all panels (c) and (d) of Figures 2 to 5.
R8: Yes. Accumulated daily precipitation (mm.day-1) for: TRMM 3B42V7 versus IB02 at 0.2 degree resolution for panels (a) to (d); and TRMM 3B42V7 versus ERA-Interim at 1 degree for panels (e) to (h).
R9: As indicated in the text, “T#9, number 9 in the ranking; Table 1”. The “T#” stands for “Top number”.
Gracias! Un saludo!
The authors
