Last modified on Dec 11, 2024 11:17

Contributors: M. Pondrom (Deutscher Wetterdienst)

Issued by: Deutscher Wetterdienst (DWD) / Marc Pondrom

Date: 22 Jan 2024 

Ref: C3S2_D312a_Lot1.1.4.1_202311_PQAD_ECV_Precipitation_GIRAFE_v1.1

Official reference number service contract: 2021/C3S2_312a_Lot1_DWD/SC1

History of modification

List of datasets covered by this document

Related documents

Acronyms

List of figures

List of Tables

General definitions

Scope of the document

The Product Quality Assurance Document (PQAD) provides a description of the validation methodology for the Global Interpolated RAinFall Estimation (GIRAFE) version 1 (v1) Climate Data Record (CDR) for the Essential Climate Variable (ECV) Precipitation. GIRAFE v1 is a product brokered from the EUMETSAT Satellite Application Facility on Climate Monitoring (CM SAF). It merges precipitation data from microwave (MW) imagers and sounders over land and ocean with infrared (IR) observations from several geostationary platforms along the equator (Geo-Ring).

This document describing the validation methodology of the GIRAFE precipitation product for the period from January 2002 to December 2022 refers extensively to the methodology presented in the CM SAF Precipitation Product Validation Report [D1], used in the validation of the TCDR product. The validation results are to be found in [D1].

Executive summary

The CM SAF GIRAFE Precipitation Thematic Climate Data Record (TCDR) version 1 (v1) has been brokered to the Copernicus Climate Change Service (C3S) to upgrade the Copernicus micrOwave-based gloBal pRecipitAtion (COBRA) product of the preceding phase of C3S. This new product has been designed to meet the requirements specified by users of global Precipitation data that have first been formulated in a series of dedicated workshops hosted by DWD and CM SAF. It is a global 1° x 1° latitude-longitude data record that is produced at a daily temporal resolution, as well as on a monthly mean basis, including a unique uncertainty sampling estimate. It utilises Level 2 data from passive MW imagers and sounders onboard polar-orbiting satellites over land and ocean, combined with Geo-Ring IR imagery. The CDR covers the time period January 2002 to December 2022.

This report outlines the methodology used for the validation of CM SAF GIRAFE (v1) Precipitation product. A wide range of reference (e.g. rain gauges) and intercomparison (e.g. satellite, reanalysis) datasets are used to characterise the TCDR in terms of accuracy, precision, and stability. The requirements for accuracy, precision, and stability formulated during CM SAF procedures are presented here.

The evaluation of the CM SAF GIRAFE daily and monthly products are fully described in the Validation Report [D1]. Therefore, in the present document, the 1^st section briefly presents the CM SAF Precipitation dataset, while the following two sections provide information on the validating datasets and the validation approach, respectively. Validation results are discussed in details in [D1].

1. Validated products

The GIRAFE product consists of a global 1°x1° daily accumulated precipitation amount together with uncertainty estimates, and a global 1°x1° monthly mean of daily accumulation. Unlike other CM SAF products, GIRAFE is based on a large variety of sensors and platforms whose contributions to the final product are outlined in Figure 1-1, Figure 1-2 and Figure 1-3.

All details on the Precipitation product, including the GIRAFE retrieval algorithm and other processing elements to generate the dataset, have been given in the Algorithm Theoretical Basis Document (ATBD)[D2]. The main elements of the processing chain for the generation of the GIRAFE Precipitation TCDR are briefly summarised below.

Based on the Tropical Amount of Precipitation with an Estimate of Errors (TAPEER) approach developed at the French Centre National de Recherche Scientifique (CNRS), GIRAFE merges MW sounder- and imager-based estimations of instantaneous surface precipitation over the 55°S-55°N region with IR imagery from several geostationary platforms along the equator (Geo-Ring). Poleward of 55°, precipitation retrievals are derived from MW observations only. To the maximum feasible extent either Fundamental Climate Data Records (FCDRs) are utilised or in-house quality control is applied to ensure high-quality MW and IR input to the precipitation product. The retrieval of precipitation from MW imagers over ocean relies on the HOAPS method [Andersson et al., 2010] in its most recent, temporally extended version [Andersson et al., 2017]. MW sounder-based retrievals PNPR-CLIM and PRPS utilise methods described in Bagaglini et al. [2021] and Kidd et al. [2021]. The MW-based precipitation estimates are harmonized in a dedicated pre-processing step and then used to compute conditional precipitation rates and to train GIRAFE locally to detect the fraction of precipitating clouds in the quality-controlled IR brightness temperatures. The merging process is described in detail in Roca et al. [2018, 2020]. A dedicated error model is used to estimate the sampling uncertainty in the final amount, based on regional analyses of (de-)correlation of the IR-derived precipitation fields.

The CDR covers the time period January 2002 to December 2022.

Figure 1-1: Overview of the geostationary platforms used for the GIRAFE CDR. Spatial resolution and wavelength (channel) are specified in parentheses. In the vertical are indicated the central longitudes of the different satellites and in the horizontal their respective temporal coverage. Same colour indicates same sensor. (Figure from the CM SAF GIRAFE ATBD [D2])

Figure 1‑2: Temporal coverage of the satellite MW imagers used to derive precipitation estimates (over ice-free ocean) for GIRAFE CDR. (Figure from the CM SAF GIRAFE ATBD [D2])

Figure 1-3: Temporal coverage of the satellite MW sounders used to derive precipitation estimates for GIRAFE CDR. (Figure from the CM SAF GIRAFE ATBD [D2])

2. Description of validating datasets

The assessment of the performance of GIRAFE Precipitation TCDR v1 has been carried out based on reference and intercomparison datasets as recommended in the CM SAF during CM SAF requirements elicitation. Some of them consist in mean precipitation rates [mm/d], while others (like GIRAFE CDR v1) in daily accumulations [mm]. Therefore, all the data are converted to 1° x 1° gridded mean precipitation rates [mm/d] before the evaluation of GIRAFE.

2.1 Reference datasets

The reference datasets are considered the “ground truth” in the sense that a discrepancy between GIRAFE CDR v1 (or any other intercomparison dataset) and these would be the consequence of a deficiency in GIRAFE CDR v1. There are several types of reference datasets used for precipitation products validation, such as the rain gauge networks and radar-derived instantaneous precipitation rate estimates used for the validation of GIRAFE.

The “African Monsoon Multidisciplinary Analysis – Couplage de l'Atmosphère Tropicale et Cycle Hydrologique” (AMMA-CATCH) dataset consists of high-resolution rain gauge networks in West Africa [Lebel et al., 2009], curated since the 1990s and widely used for the validation of satellite-based precipitation estimates [Gosset et al., 2018]. The network in Niamey, Niger, that represents a dry Sahelian climate constituting the longest and most stable record, is used to evaluate the quality of GIRAFE Precipitation TCDR v1 during the rainy season (June-September, JJAS) for the time period 2002-2019.

The European climatological high-resolution gauge-adjusted radar precipitation (EURADCLIM) dataset [Overeem et al., 2023] is based on radar-derived instantaneous precipitation rate estimates at 15 minutes resolution across Europe on a 2 km x 2 km grid. After post-processing and matching to GIRAFE’s default settings, the 24h accumulated precipitation rate is used to assess the quality of the Precipitation TCDR.

2.2 Intercomparison datasets

Table 2-1 presents the general characteristics of the datasets used for the validation of GIRAFE.

Table 2-1: List of all the (near-)global datasets used for the validation of GIRAFE Precipitation TCDR v1.

The intercomparison – global or near-global – datasets consist of global rain gauges (over land only), reanalysis or satellite-based precipitation records. To evaluate the quality of GIRAFE Precipitation TCDR v1, they have been compared not only directly to the daily and monthly products, but also to the reference datasets, for an easier interpretation of the analysis. They have been used in daily aggregations and taken from the Frequent Rainfall Observations on GridS (FROGS) database curated by [Roca et al., 2019], which also bring the datasets from sub-daily and sub-1° resolution to the desired 1° x 1° x 1d resolution.

The ERA5 global climate reanalysis [Hersbach et al., 2020] assimilates observations across the globe into a numerical model of the atmosphere, coupled to a land-surface and to an ocean-wave model, hence producing a consistent and complete record of the Earth System over a wide range of atmospheric and surface variables. ERA5 has been available globally with an hourly time step from 1940 onwards and at spatial resolution of 0.25°/0.5° (HRES/EDA). FROGS utilises the “total precipitation” variable of the hourly ERA5 reanalysis product [Hersbach et al., 2023] for building the 1° x 1° x 1d dataset by accumulating over 24 hours and interpolating to the coarser grid. For the evaluation of the GIRAFE monthly TCDR, ERA5 monthly product is used and aggregated to the coarser 1° x 1° grid.

The Global Precipitation Climatology Project version 3.2 (GPCP v3.2) provides global 0.5° x 0.5° estimates of precipitation, at monthly resolution from 1983 until 2020, and, since 2000 at daily resolution [Huffman et al., 2022a, b]. The monthly product combines MW-based precipitation observations from SSM/I or SSMIS with IR-based precipitation observations. Over land, the satellite-based data are then merged with GPCC data. The daily GPCP data are derived from temporally highly resolved IR observations.

The TRMM Multi-satellite Precipitation Analysis (TMPA) products are available between 50°S and 50°N on a 0.25° x 0.25° grid from January 1998 until December 2019. TRMM stands for “Tropical Rainfall Measuring Mission”. Monthly means are provided in TMPA 3B43 [TRMM, 2011]. Daily data are provided in TMPA 3B42 [Huffman et al., 2016].

The CMORPH – Satellite-based precipitation dataset based on the Climate Prediction Center (CPC) Morphing Technique (MORPH) – product uses the morphing technique. MW-based instantaneous precipitation rate estimates are interpolated in space and time to a regular grid by assuming advection along IR-derived cloud motion vectors [Xie et al., 2017] inside 60°N/S. The daily product is available at 0.25° resolution, both in a rain gauge-corrected and uncorrected version.

The Integrated Multi-satellitE Retrievals for GPM (IMERG) v6 product [Huffman et al., 2019] merges MW-based observations of precipitation (with a focus on the GPM constellation) with IR observations by using – among others – procedures developed for CMOPRH. The “final (un)calibrated” products used for the comparison against GIRAFE TCDR have been regridded from the native 0.1° to the usual 1° spatial resolution and are available through FROGS from 2001 to 2020.

The Tropical Amount of Precipitation with an Estimate of ERrors (TAPEER) dataset [Chambon et al., 2012; Roca et al., 2010] uses the same method as GIRAFE for the merging of MW and IR data, but it differs in the underlying MW database [Viltard et al., 2006] and in the restriction to the 30°N/S latitude band. TAPEER has been used for the verification of GIRAFE implementation and is currently available in the FROGS database from end of 2011 until 2021.

The Copernicus micrOwave-based gloBal pRecipitAtion (COBRA) dataset [Konrad et al.; 2021, 2022] relies only on MW observations that are mostly the same as GIRAFE (HOAPS, PNPR-CLIM). COBRA is available for the time period 2000-2017. The comparison between COBRA and GIRAFE is helpful in exploring the impact of IR observations in GIRAFE.

3. Description of product validation methodology

This chapter provides an overview of the validation methodology to assess the performance of the global 1° x 1° daily and monthly GIRAFE Precipitation products. A detailed explanation can be found in the CM SAF GIRAFE validation report [D1].

3.1 Methodology

This section describes the approaches, methods and metrics that are used to validate GIRAFE and the other intercomparison datasets with the reference datasets.

3.1.1 Quality indicators

In the following, \( p_{t,i} \) and \( p_{r,i} \) indicate a 1° x 1° daily or monthly average precipitation rate [mm/d] of the validated and validating datasets respectively.

3.1.1.1 Bias

The mean difference \( µ \) (also frequently reported in the literature as ‘mean error’ or ‘bias’) results from the arithmetic mean of the difference over the members of the data records. This measure should indicate how close the parameter estimation is on average to a reference observation (representing the truth, or more correctly the best estimation of the truth). It indicates whether the data record on average over- or underestimates the reference data record and is defined as:

where the weights \( w_i \) , the cosine of the latitude, is used as a proxy to account for the varying area of 1° x 1° grid cells when averaging in space. The intercomparison datasets are often adjusted towards other datasets which have the largest influence on the bias (locally and globally) and adds a certain extent of inter-dependence in the ensemble of intercomparison datasets, so that this measure does not necessarily point to the weak spots of GIRAFE as that dataset which is ultimately assessed in these activities.

3.1.1.2 Bias-corrected root-mean-square difference

The sample standard deviation \( σ \) (also frequently reported in literature as ‘precision’ or ‘bias-corrected RMS error’) is a measure of the spread around the mean value of the distribution formed by the differences between the test and the reference data records. This measure should tell how individual parameter estimations are distributed relative to the mean difference and is defined as: \[ σ = \sqrt{\frac{\sum_iw_i(p_{t,i} - p_{r,i} - µ)^2}{\sum_iw_i}} \ \ (Eq. 3.2) \]

3.1.1.3 Stabiliby

The stability \( s \) indicates whether one or more accuracy metrics are stable or if they are changing over time. The CM SAF has chosen to monitor only the first metric here (the mean difference) where criteria have been defined for the maximum changes being acceptable per decade for each product. The slope of the time series of mean differences is calculated using a least-squares fit of a 1-degree polynomial to the respective bias timeseries.

3.1.1.4 Correlation coefficient

The Pearson correlation coefficient ( \( CC \) ) is computed as: \[ CC = \frac{\sum_i(p_{t,i} - \overline{p_t})(p_{r,i} - \overline{p_r})}{\sqrt{\sum_i(p_{t,i} - \overline{p_t})^2)}\sqrt{\sum_i(p_{r,i} - \overline{p_r})^2}} \ \ (Eq. 3.3) \] where \( \overline{p_t} \) (respectively \( \overline{p_r} \) ) is the arithmetic mean in the \( p_{t,i} \)  (resp. \( p_{r,i} \) ) sample.

The Pearson correlation coefficient ( \( CC \) ) and bias-corrected root-mean-square difference ( \( σ \) or \( bc-RMSD \) ) indicate how well the validated dataset (“t”) reproduces the validating dataset (“r”) quantitatively.

3.1.1.4 Detection scores

The detection scores are based on the contingency table for the binary classification “precipitation”/”no precipitation” in the two datasets “t” and “r”. A value of 1 mm/d is used for distinguishing between these two outcomes for a given 1° x 1° x 1d grid cell. The contingency table for a given sample in space, time or both is of the form:

Table 3-1: Contingency table used for binary classification “precipitation”/”no precipitation”.

Where a is the number of instances in which both datasets see precipitation above 1 mm/d, b the number of falsely detected precipitation events, c the number of missed precipitation events and d the number of cases for which none of the datasets detected precipitation.

The hit rate ( \( HR \) ) is the fraction of equal classification in the two datasets (the closer to 1 the better):

The probability of detection (POD) is the fraction of correctly detected precipitation in dataset “t” among all precipitation events in dataset “r”, i.e. including those classified as not precipitating in dataset “t” (the closer to 1 the better):

The false alarm rate ( \( FAR \) ) is the fraction of falsely detected precipitation events in dataset “t” (i.e. where dataset “r” sees no precipitation) among all events classified as precipitating in dataset “t” (the closer to 0 the better):

The Heidke Skill Score ( \( HSS \) ) measures all correct and false classifications in dataset “t” at once (the closer to 1 the better); it is defined as:

3.1.1.4 Frequency of error bar overlap

Finally, the validity of the GIRAFE uncertainty budget has been evaluated as a 1-sigma uncertainty by computing the frequency of error bar overlap ( \( FEBO \) ). With \( σ_{t,i} \) and  \( σ_{r,i} \) being the uncertainties of \( p_{t,i} \) and  \( p_{r,i} \) respectively, it is computed as:

where \( n \) is the number of non-overlapping 1° x 1° x 1d instances (i.e. in which \( p_{t,i} + σ_{t,i} < p_{r,i} - σ_{r,i} \)  or \( p_{t,i} - σ_{t,i} > p_{r,i} + σ_{r,i} \) ) and  \( N \)  is the total number of analysed instances.

3.1.2 Homogeneity analysis

An important quality aspect in climate analysis is the degree of homogeneity that indicates how strongly the data records are affected by breakpoints. Taking the anomaly differences of GIRAFE Precipitation TCDR as input, the breakpoint analysis detects abrupt changes in the time series of Precipitation and produces the time and the strength associated with the breakpoint as output. Table 3-2 provides an overview of considered regions and references.

Table 3-2: Cases considered for the analysis of homogenity.

3.2 Compliance with product requirements

The requirements for the performance of GIRAFE Precipitation TCDR with respect to the global or near-global datasets chosen for this kind of analysis are listed in Table 3‑3. Accuracy is defined to be the theoretical degree of conformity of the measurement to the unknown ‘true’ value. The CM SAF uses the mean difference (or ‘mean error’ or ’bias’) with respect to a reference dataset to assess the accuracy for a given dataset [D3]. Precision is defined to be the closeness of agreement between independent measurements of a quantity under the same conditions. The quantity used by CM SAF to express the precision of a dataset is the standard deviation of the error, or standard deviation of the differences between the dataset and a reference dataset [D3]. Finally, stability is defined as the consistency of measurements for a given parameter over time. The CM SAF has chosen to monitor the mean error (bias) where the decadal trend is compared to a reference data record [D3].

Table 3-3: Requirements for the uncertainty characteristics of GIRAFE v1 as per PRD [D3].

3.3 Gaps in the data record

The spatial and temporal completeness of the GIRAFE v1 CDR has been assessed in [D1], first in details for the daily resolved dataset, and then briefly for the monthly dataset.

Inside the Geo-Ring, missing grid cells are due to the temporal non-availability of a geostationary platform or the absence of MW-IR collocations. The latter issue generally occurs more often over land where GIRAFE uses MW observations from fewer platforms, and is amplified by the presence of varying scan modes of the IR platforms.

More generally, the absence of MW observations is an issue in the high-latitude regions where fewer platforms are able to provide observations and the grid cell geometry prevents a complete coverage near the poles.

4. Summary of validation results

The key results from the validation of the GIRAFE Precipitation TCDR v1, performed against the above-mentioned ground-based observations and intercomparison (near-)global reference datasets, are thoroughly analysed in the upcoming Validation Report of CM SAF [D1].

References

Andersson, A., Fennig, K., Klepp, C., Bakan, S., Graßl, H., and Schulz, J. (2010): The Hamburg Ocean Atmosphere Parameters and Fluxes from Satellite Data – HOAPS-3, Earth Syst. Sci. Data, 2, 215–234, https://doi.org/10.5194/essd-2-215-2010.

Andersson, A., Graw, K., Schröder, M., Fennig, K., Liman, J., Bakan, S., Hollmann, R., Klepp, C. (2017): Hamburg Ocean Atmosphere Parameters and Fluxes from Satellite Data - HOAPS 4.0, Satellite Application Facility on Climate Monitoring, DOI:10.5676/EUM_SAF_CM/HOAPS/V002, https://doi.org/10.5676/EUM_SAF_CM/HOAPS/V002.

Bagaglini, L., Sanò, P., Casella, D., Cattani, E., Panegrossi, G. (2021) The Passive Microwave Neural Network Precipitation Retrieval Algorithm for Climate Applications (PNPR-CLIM): Design and Verification. Remote Sens., 13, 1701. https://doi.org/10.3390/rs13091701

Chambon, P., Jobard, I., Roca, R., and Viltard, N. (2012): An investigation of the error budget of tropical rainfall accumulation derived from merged passive microwave and infrared satellite measurements, Q. J. Roy. Meteor. Soc., 139, 879–893, https://doi.org/10.1002/qj.1907.

Hans, I., Burgdorf, M., Buehler, S.A., Prange, M., Lang, T., and John, V.O (2019): An Uncertainty Quantified Fundamental Climate Data Record for Microwave Humidity Sounders, Remote Sensing, 11, 548, https://doi.org/10.3390/rs11050548.

Hanschmann, T., John, V., Roebeling, R., Grant, M., and Schulz, J.: Algorithm Theoretical Basis Document (ATBD) Fundamental Climate Data Record of Microwave Humidity Sounder, Copernicus Climate Change Service (C3S), C3S_311b Task 1.1 D1.1, D1.2, D1.3, 2019.  Available from https://www-cdn.eumetsat.int/files/2021-03/C3S_311b_T1_1_Microwave_FCDR_ATBD_D1_1_D1_2_D1_3.pdf.

Gosset, M, Alcoba, M., Roca, R., Cloché, S., Urbani, G. (2018): Evaluation of TAPEER daily estimates and other GPM-era products against dense gauge networks in West Africa, analysing ground reference uncertainty. QJR Meteorol Soc ; 144 (Suppl. 1): 255–269. https://doi.org/10.1002/qj.3335.

Hersbach, H., Bell, B., Berrisford, P., Hirahara, S., Horányi, A., Muñoz-Sabater, J., Nicolas, J., Peubey, C., Radu, R., Schepers, D., Simmons, A., Soci, C., Abdalla, S., Abellan, X., Balsamo, G., Bechtold, P., Biavati, G., Bidlot, J., Bonavita, M., De Chiara, G., Dahlgren, P., Dee, D., Diamantakis, M., Dragani, R., Flemming, J., Forbes, R., Fuentes, M., Geer, A., Haimberger, L., Healy, S., Hogan ,R.J., Hólm, E., Janisková, M., Keeley, S., Laloyaux, P., Lopez, P., Lupu, C., Radnoti, G., de Rosnay, P., Rozum, I., Vamborg, F., Villaume, S., Thépaut J.-N. (2020): "The ERA5 global reanalysis." Quarterly Journal of the Royal Meteorological Society 146 (730), 1999-2049. https://doi.org/10.1002/qj.3803.

Hersbach, H., Bell, B., Berrisford, P., Biavati, G., Horányi, A., Muñoz Sabater, J., Nicolas, J., Peubey, C., Radu, R., Rozum, I., Schepers, D., Simmons, A., Soci, C., Dee, D., Thépaut, J-N. (2023): ERA5 hourly data on single levels from 1940 to present. Copernicus Climate Change Service (C3S) Climate Data Store (CDS), https://doi.org/10.24381/cds.adbb2d47.

Huffman, G.J., D.T. Bolvin, E.J. Nelkin, and R.F. Adler (2016): TRMM (TMPA) Precipitation L3 1 day 0.25 degree x 0.25 degree V7, Goddard Earth Sciences Data and Information Services Center (GES DISC), https://doi.org/10.5067/TRMM/TMPA/DAY/7.

Huffman, G.J., E.F. Stocker, D.T. Bolvin, E.J. Nelkin, Jackson Tan (2019): GPM IMERG Final Precipitation L3 1 day 0.1 degree x 0.1 degree V06, Goddard Earth Sciences Data and Information Services Center (GES DISC), https://doi.org/10.5067/GPM/IMERGDF/DAY/06.

Huffman, G.J., A. Behrangi, D.T. Bolvin, E.J. Nelkin (2022a): GPCP Version 3.2 Satellite-Gauge (SG) Combined Precipitation Data Set, Goddard Earth Sciences Data and Information Services Center (GES DISC), https://doi.org/10.5067/MEASURES/GPCP/DATA304.

Huffman, G.J., A. Behrangi, D.T. Bolvin, E.J. Nelkin (2022b): GPCP Version 3.2 Daily Precipitation Data Set, Goddard Earth Sciences Data and Information Services Center (GES DISC), https://doi.org/10.5067/MEASURES/GPCP/DATA305.

Kidd, C., T. Matsui and S. Ringerud (2021): Precipitation Retrievals from Passive Microwave Cross-Track Sensors: The Precipitation Retrieval and Profiling Scheme. Remote Sens., 13(5), 947; https://doi.org/10.3390/rs13050947.

Konrad, H., Sikorski, T., Schröder, M., Bobryshev, O., Bagaglini, L., Sanò, P., Panegrossi, G., Cattani, E. (2021): Product Quality Assessment Report, Copernicus Microwave-based Global Precipitation (COBRA). Reference: C3S_D312b_Lot1.2.1.6-v1.0_202106_PQAR_COBRA v1.1. URL.

Konrad, H., Panegrossi, G., Bagaglini, L., Sanò, P., Sikorski, T., Cattani, E., Schröder, M., Mikalsen, A., Hollmann R.  (2022): Precipitation monthly and daily gridded data from 2000 to 2017 derived from satellite microwave observations. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). DOI: https://doi.org/10.24381/cds.ada9c583.

Merchant, C. J., Paul, F., Popp, T., Ablain, M., Bontemps, S., Defourny, P., Hollmann, R., Lavergne, T., Laeng, A., de Leeuw, G., Mittaz, J., Poulsen, C., Povey, A. C., Reuter, M., Sathyendranath, S., Sandven, S., Sofieva, V. F., and Wagner, W. (2017): Uncertainty information in climate data records from Earth observation, Earth Syst. Sci. Data 9, 511-527, https://doi.org/10.5194/essd-9-511-2017.

Overeem, A., van den Besselaar, E., van der Schrier, G., Meirink, J. F., van der Plas, E., and Leijnse, H.: EURADCLIM: the European climatological high-resolution gauge-adjusted radar precipitation dataset, Earth Syst. Sci. Data, 15, 1441–1464, https://doi.org/10.5194/essd-15-1441-2023, 2023.

Roca, R., Chambon, P., Jobard, I., Kirstetter, P. E., Gosset, M., and Bergés, J. C. (2010): Comparing satellite and surface rainfall products over West Africa at meteorologically relevant scales during the AMMA campaign using error estimates, J. Appl. Meteorol. Clim., 49, 715–731, https://doi.org/10.1175/2009JAMC2318.1, 2010.

Roca, R., Taburet, N., Lorant, E., Chambon, P., Alcoba, M., Brogniez, H., Cloché, S., Dufour, C., Gosset, M., Guilloteau, C. (2018): Quantifying the contribution of the Megha-Tropiques mission to the estimation of daily accumulated rainfall in the Tropics. Quart. J. Roy. Met. Soc., 63 (144), 49–682 63, https://doi.org/10.1002/qj.3327.

Roca, R., Alexander, L. V., Potter, G., Bador, M., Jucá, R., Contractor, S., Bosilovich, M. G., and Cloché, S. (2019): FROGS: a daily 1° x 1° gridded precipitation database of rain gauge, satellite and reanalysis products, Earth Syst. Sci. Data, 11, 1017-1035, https://doi.org/10.5194/essd-11-1017-2019.

Roca, R., Guérou, A., Jucá Oliveira, R. A., Chambon, P., Gosset, M., Cloché, S., and Schröder, M. (2020): Merging the Infrared Fleet and the Microwave Constellation for Tropical Hydrometeorology (TAPEER) and Global Climate Monitoring (GIRAFE) Applications. Advances in Global Change Research, 67, 429–450.

Sen, P. K. (1968): Estimates of the regression coefficient based on Kendall's tau, J. Am. Stat. Assoc., 63, 1379–1389.

Schneider, U., Hänsel S., Finger P., Rustemeier, E., Ziese, M. (2022): GPCC Full Data Monthly Product Version 2022 at 1.0°: Monthly Land-Surface Precipitation from Rain-Gauges built on GTS-based and Historical Data. https://doi.org/10.5676/DWD_GPCC/FD_M_V2022_100.

TRMM (2011): TRMM (TMPA/3B43) Rainfall Estimate L3 1 month 0.25 degree x 0.25 degree V7, Goddard Earth Sciences Data and Information Services Center (GES DISC), https://doi.org/10.5067/TRMM/TMPA/MONTH/7.

Viltard, N., Burlaud, C., and Kummerow, C. D. (2006): Rain Retrieval from TMI Brightness Temperature Measurements Using a TRMM PR–Based Database, J. Appl. Meteorol. Clim., 45, 455–466, https://doi.org/10.1175/jam2346.1.

Xie, P., Joyce, R., Wu, S., Yoo, S.-H., Yarosh, Y., Sun, F., and Lin, R. (2017): Reprocessed, Bias-Corrected CMORPH Global High-Resolution Precipitation Estimates from 1998, J. Hydrometeorol., 18, 1617–1641, https://doi.org/10.1175/JHM-D-16-0168.1.

Related articles

• Page:
  CP: Climate Change Initiative (CCI)
• Page:
  SRB CCI-ICDR: System Quality Assurance Document (SQAD)
• Page:
  SRB CCI-ICDR: Product Quality Assessment Report (PQAR)
• Page:
  SRB CCI-ICDR: Product Quality Assurance Document (PQAD)
• Page:
  SRB CCI-ICDR: Product User Guide and Specification (PUGS)