Contributors: H. Konrad (DWD), M. Schröder (DWD), A. C. Mikalsen (DWD), R. Hollmann (DWD), T. Sikorski (DWD)
History of modifications
List of datasets covered by this document
Related documents
Acronyms
General definitions
In the scope of the Copernicus Climate Change Service (C3S), a Climate Data Record (CDR) always has a fixed end point in time, whereas an Interim Climate Data Record (ICDR) is extended continuously, often serving as an extension of a respective completed CDR.
In contrast to this, the record by the Global Precipitation Climatology Project (GPCP) is continuously extended in time, with a latency of two to three months due to the latency of the rain gauge product (see below). In order to avoid confusion, we state here that we will label the GPCP record as CDR in the scope of the brokering to C3S until December 2017, which is the end point for the first delivery to the CDS. Later extensions of the record will be labelled as ICDR in the scope of brokering to C3S. In this sense, the term ‘ICDR’ is used here according to C3S terminology.
The GPCP provides an interim product for the monthly solutions, too. It is based on the same satellite data and respective algorithms, but relies on the input of only a preliminary version of the rain-gauge dataset by the Global Precipitation Climatology Centre (‘First Guess’ in contrast to the ‘Monitoring Product’ or the ‘Full Data Monthly’ product). Consequently, it is available much earlier than the fully processed data. When the rain-gauge dataset becomes available and the processing can be completed, the files containing the interim dataset are replaced by the fully processed ones. In the scope of brokering the GPCP data to C3S, we choose not to broker the interim monthly data in order to avoid confusion and frequent updates. Users interested in these more up-to-date files are referred to the original data repositories.
Please note: The version numbering within C3S (4th column in List of datasets on page 3) is for internal use and should not be used when citing the dataset.
Scope of the document
This Product Quality Assurance Document provides a description of the product validation methodology for the satellite- and rain gauge-based precipitation estimates by the Global Precipitation Climatology Project (GPCP) at the University of Maryland (UMD). These are brokered to the Climate Data Store (CDS) by the Copernicus Climate Change Service (C3S). The validation described in this document is carried out within the scope of C3S, whereas the intellectual property rights of the products themselves remain with the GPCP. In this sense, this document is not part of the official GPCP documentation, but produced only in the scope of brokering the data to the CDS.
Executive summary
Estimates of precipitation by the Global Precipitation Climatology Project (GPCP) from the University of Maryland (UMD) are brokered to the Climate Data Store (CDS) by the Copernicus Climate Change Service (C3S). The GPCP is part of the international World Climate Research Programme (WCRP) and its Global Energy and Water Exchanges (GEWEX) project. All intellectual property rights remain with the GPCP.
The precipitation products provided as monthly (v2.3) and daily (v1.3) means are assessed in the context of their inclusion in the CDS. The validation consists of initial quality (sanity) checks of missing data points and range of the available data points, a comparison to other gridded datasets (ERA-5 global atmospheric reanalysis, TMPA lower latitude satellite-based rainfall estimates, radar-based NIMROD rainfall estimates over Europe) and in-situ observations (PACRAIN for the Pacific Ocean; OceanRAIN for ship-borne sensors). The comparison features biases between GPCP and the respective datasets and their evolution over time, including trends, i.e. stability. Biases and their trends are also related to the Key Performance Indicators which are reported to C3S continuously. Other validation procedures include the comparison of climatologies, i.e. temporally averaged spatial fields, and a data point-by-data point comparison, e.g. in the form of a histogram of differences. A brief literature review is appended if appropriate.
1. Validated products
The validation procedures explained in this document are designed to validate the surface precipitation estimates of the Global Precipitation Climatology Project (GPCP), brokered to the Climate Data Store (CDS) by the Copernicus Climate Change Service (C3S) from the University of Maryland (UMD). GPCP offers these estimates as daily and monthly means, see below. The products are extended regularly; within the scope of C3S they will be formally described as Climate Data Record (CDR) until 12/2017 and as Interim Climate Data Record (ICDR) in the subsequent regular updates (see above, General definitions). We will not distinguish between CDR and ICDR during validation – when the ICDR finally becomes available – with the exception of this section where the validated products are listed and where we include this distinction for the sake of completeness.
1.1 GPCP monthly means of precipitation v2.3 (CDR)
The CDR for monthly mean estimates covers the period from 01/1979 to 12/2017. It is a level-3 gridded dataset, available on a regular latitude-longitude grid with a 2.5 degree spacing. Uncertainty information is provided together with the monthly means.
1.2 GPCP daily means of precipitation v1.3 (CDR)
The CDR for daily mean estimates covers the period from 01/10/1996 to 31/12/2017. It is a level-3 gridded dataset, available on a regular latitude-longitude grid with a 1.0 degree spacing. No uncertainty information is provided.
1.3 GPCP monthly means of precipitation v2.3 (ICDR)
The ICDR for monthly mean estimates extends the CDR in time, starting in 01/2018. All characteristics listed in Section 1.1 apply.
1.4 GPCP daily means of precipitation v1.3 (ICDR)
The ICDR for daily mean estimates extends the CDR in time, starting on 01/01/2018. All characteristics listed in Section 1.2 apply.
2. Description of validating datasets
If not stated otherwise, we expect all validating datasets to be extended continuously.
2.1 Gridded datasets
2.1.1 TMPA 3B42/3B43 v7 satellite-based estimates
Based on satellite observations by the Tropical Rainfall Measuring Mission (TRMM), the TRMM Multi-satellite Precipitation Analysis (TMPA) products are available from 01/1998 to 12/2019 between 50°S and 50°N on a 0.25° equidistant grid. TMPA 3B43 (reference TRMM, 2011) represents monthly means; TMPA 3B42 (reference Goddard Earth Sciences Data and Information Services Center, 2016) represents daily means. Note that since the TRMM satellite mission ended in 2015, the record has been extended based on other satellite input data. The data can be accessed through the Data and Information Services Center at the Goddard Space Flight Center (https://disc.gsfc.nasa.gov/). Known issues with the data are summarized in the documentation at this repository.
The temporal aggregations of the mentioned TMPA products equal those of the GPCP products. However, the spatial information in TMPA needs to be aggregated in order to match the spatial resolution of GPCP products. For this, averages of all TMPA grid cells that fall into one GPCP grid cells are computed.
The TMPA products have been decommissioned and are only served until 12/2019.
2.1.2 ERA-5 reanalysis
The ERA5 (reference C3S, 2017) global climate reanalysis assimilates observations of different meteorological quantities across the globe into a numerical model of atmospheric dynamics. It produces a large variety of output quantities including precipitation. ERA5 output is available in the CDS by C3S (https://cds.climate.copernicus.eu/cdsapp#!/dataset/reanalysis-era5-single-levels?tab=overview) in hourly intervals and as monthly means at 0.25° spatial resolution. Known issues are listed in the documentation; none of these relates directly to precipitation (as of 21/12/2018), but links between precipitation-related processes and respective issues may well exist.
For collocation of level 3 gridded values, the respective hourly ERA5 precipitation is temporally accumulated to match the daily GPCP resolution. For comparing spatial patterns, the temporally averaged daily as well as the directly available monthly ERA5 fields are then re-gridded to the 1.0°/2.5° geographic grid by averaging values in grid cells that are contained in the respective GPCP grid box.
2.1.3 Met Office Rain Radar Data from the NIMROD System
The United Kingdom’s Met Office provides precipitation estimates across Europe based on observations at rain radar stations on a 5 km grid (reference Met Office, 2003). These data can be accessed at http://catalogue.ceda.ac.uk/uuid/82adec1f896af6169112d09cc1174499. As with the previous gridded datasets, the 5 km spatial grid and the temporal aggregation in 15 minute-slices are harmonized with the respective grids of the GPCP products via accumulation in the temporal dimension and averaging in the spatial dimensions.
2.2 In-situ (instantaneous) datasets
2.2.1 PACRAIN
The Pacific Rainfall Database (PACRAIN, Greene et al., 2008) comprises daily and monthly in-situ observations of precipitation from the tropical Pacific basin. For the purpose of this document, we will use observations between 1980 and 2015. The environment of the various stations is logged so that specific comparisons, such as for ocean-like conditions only, can be carried out. The data can be accessed at http://pacrain.ou.edu/.
2.2.2 OceanRAIN
The Ocean Rainfall And Ice-phase precipitation measurement Network (OceanRAIN, Klepp et al., 2017, Klepp, 2015) provides ship-borne, temporally highly resolved observations of precipitation. Observations are available from eight ships between 2010 and 2017. The data can be accessed at https://cera-www.dkrz.de/WDCC/ui/cerasearch/entry?acronym=OceanRAIN-W.
3. Description of product validation methodology
3.1 Initial quality checks
For each given available month/day, the number of missing values in the two-dimensional precipitation fields for GPCP monthly v2.3 and GPCP daily v1.3 is extracted and displayed, so that users can assess at which times in the past the record is (in-)complete. Likewise, simple statistics are extracted at each given time, namely minimum, mean, and maximum values. These simple statistics show whether the estimates are in a reasonable range.
3.2 Comparison with gridded datasets
3.2.1 Temporal evolution of spatial averages
The GPCP product and the validating datasets are spatially averaged over the mutual spatial extent for each comparison, i.e. globally for the GPCP-ERA5 comparison, between 50°S and 50°N for the GPCP-TMPA comparison, and over Europe for the GPCP-NIMROD comparison. Respective differences (biases) are discussed on the basis of figures showing the time series of these respective mean values.
3.2.1.1 Key Performance Indicators
A routine quality check on the data within C3S is the evaluation of the accuracy and stability of the product in the scope of the Key Performance Indicators (KPI). The respective procedures are based on the bias over the chosen spatial region. Consequently, a reference dataset needs to be chosen. Here, we opt for the TMPA products and their respective spatial coverage (between 50°S and 50°N) as the region over which the KPIs will be evaluated until December 2019 (when TMPA products end). From January 2020 onwards ERA5 (with global coverage) will be used as the reference dataset (note: this is only relevant for ICDR deliveries).
For the KPI accuracy of the TCDR (until 12/2017), absolute differences of spatial averages of the GPCP products and of a reference dataset are compared to a threshold accuracy of 0.3 mm/d. The accuracy is expected to remain below this threshold at all times in the case of the GPCP monthly product v2.3 and at 90% of all dates in the case of the GPCP daily product v1.3. The KPI accuracy for the ICDR (from 01/2018) is evaluated with respect to targets derived from the TCDR accuracy time series instead of ad hoc thresholds. Switching from TMPA to ERA5 as the reference product for the ICDR KPI accuracy implies that these targets also change, due to the differences between TMPA and ERA5 relative to GPCP mentioned in the PQAR [D3].
For the KPI stability (applies only to the TCDR, i.e., until 12/2017), a polynomial of degree 1 is fitted to the spatial averages of the GPCP products and the respective reference in a least squares sense. The absolute differences of the slopes, i.e. the absolute of the slope of the bias, are expected to remain below 0.034 mm/d/dec.
3.2.2 Climatology
For each of the three re-gridded datasets (ERA5, TMPA, NIMROD), the long-term temporal mean in each GPCP grid cell is computed and contrasted with the respective GPCP long-term mean. The respective time period will be defined by the temporal overlap between each two compared datasets. The results are visualized as spatial differences between each pair of datasets.
3.2.3 Single collocated grid cells
A brief point-by-point comparison between each pair of datasets in the form of a histogram of differences or a scatterplot and respective statistics is carried out for all GPCP grid cells and all GPCP time step. Unlike the comparisons in Sections 3.2.1 and 3.2.2, this does not include spatial or temporal averaging or accumulation except for bringing the validating datasets onto the same spatiotemporal grid as the GPCP products.
3.3 Comparison with in-situ data
The comparison of the GPCP products with PACRAIN and OceanRAIN is carried out similarly to the procedures in Section 3.2. The results of the respective comparisons are expected to mainly reflect local variability, as one grid cell of the GPCP products represents a larger area whereas the locations of the rain gauges in the PACRAIN and OceanRAIN datasets do not reflect the conditions in a larger environment around the respective instruments.
3.4 Literature review
The GPCP products have a longstanding history of usage by many researchers worldwide. We therefore rely on the many dataset intercomparison exercises published in scientific journals if possible. However, the relatively recent upgrade of the GPCP products to the current versions in 2017 will likely prohibit the inclusion of previously published literature about GPCP products.
4. Summary of validation results
In the following, we list the main points of the assessment carried out within C3S. The details can be found in the respective Product Quality Assessment Report [D3].
The GPCP monthly dataset is spatially and temporally complete and the global minimum, maximum, and average values are stable and plausible. The GPCP daily dataset features some smaller data gaps and outlying maximum values.
The comparison of spatially averaged values between GPCP and the respective monthly TMPA satellite product shows that the GPCP monthly product meets the initially formulated accuracy target of 0.3 mm/d and the stability target of 0.034 mm/d/dec. The GPCP daily product also satisfies the same stability target, but violates the accuracy target of 0.3 mm/d in ~7.5% of all days, as it features a higher spatial and temporal resolution and thus variability. The comparison with ERA5 reanalysis data confirms an earlier reported overestimation of global mean precipitation in ERA5 with respect to GPCP or TMPA products.
Spatial evaluation of GPCP, ERA5, and TMPA products shows that ERA5 and to a lesser extent TMPA contain higher rates of precipitation in the tropics compared to GPCP. TMPA sees smaller rates over mid-latitudinal oceans, where GPCP and ERA5 agree better.
The uncertainty budget provided with the GPCP monthly dataset has been evaluated against the mismatch of the precipitation rate in GPCP vs. ERA5 and TMPA. The results indicate that the provided uncertainties represent one sigma or more.
GPCP overestimates precipitation over the North Sea and North Atlantic with respect to the ground-based precipitation radar database NIMROD for Northwest and Central Europe. Over most of NIMROD-covered continental Europe, this situation is reverse. A seasonal cycle can be detected in these differences as well.
The comparison of GPCP with in situ observations at PACRAIN stations across the Pacific Ocean and on research vessels (OceanRAIN) remains difficult due to the very distinct sampling characteristics of satellite and in situ data.
References
Copernicus Climate Change Service (C3S) (2017): ERA5: Fifth generation of ECMWF atmospheric reanalyses of the global climate. Copernicus Climate Change Service Climate Data Store (CDS), 21/12/2018. https://cds.climate.copernicus.eu/cdsapp#!/home
Goddard Earth Sciences Data and Information Services Center (2016), TRMM (TMPA) Precipitation L3 1 day 0.25 degree x 0.25 degree V7, Edited by Andrey Savtchenko, , Goddard Earth Sciences Data and Information Services Center (GES DISC), doi: 10.5067/TRMM/TMPA/DAY/7
Greene, J.S., M. Klatt, M. Morrissey and S. Postawko, 2008: The Comprehensive Pacific Rainfall Database. J. Atmos. Oceanic Technol. 25(1), 71-82.
Klepp, C., 2015: The oceanic shipboard precipitation measurement network for surface validation – OceanRAIN, Atmospheric Res. 163, 74-90, doi:10.1016/j.atmosres.2014.12.014.
Klepp, Christian; Michel, Simon; Protat, Alain; Burdanowitz, Jörg; Albern, Nicole; Louf, Valentin; Bakan, Stephan; Dahl, Andrea; Thiele, Tanja (2017). Ocean Rainfall And Ice-phase precipitation measurement Network - OceanRAIN-W. World Data Center for Climate (WDCC) at DKRZ. https://doi.org/10.1594/WDCC/OceanRAIN-W
Met Office (2003). Met Office Rain Radar Data from the NIMROD System. NCAS British Atmospheric Data Centre, http://catalogue.ceda.ac.uk/uuid/82adec1f896af6169112d09cc1174499
Tropical Rainfall Measuring Mission (TRMM) (2011), TRMM (TMPA/3B43) Rainfall Estimate L3 1 month 0.25 degree x 0.25 degree V7, Greenbelt, MD, Goddard Earth Sciences Data and Information Services Center (GES DISC), doi: 10.5067/TRMM/TMPA/MONTH/7
This document has been produced in the context of the Copernicus Climate Change Service (C3S).
The activities leading to these results have been contracted by the European Centre for Medium-Range Weather Forecasts, operator of C3S on behalf of the European Union (Delegation agreement signed on 11/11/2014). All information in this document is provided "as is" and no guarantee or warranty is given that the information is fit for any particular purpose.
The users thereof use the information at their sole risk and liability. For the avoidance of all doubt , the European Commission and the European Centre for Medium - Range Weather Forecasts have no liability in respect of this document, which is merely representing the author's view.
