How To Download Precipitation Data [2021]
Elcira Acfalle
NCEI, the National Centers for Environmental Prediction (NCEP), and the GeophysicalFluid Dynamics Laboratory provide remote access to high-volume numerical weatherprediction and global climate models and data.
The Precipitation Frequency Data Server (PFDS) is a point-and-click interface developed to deliver NOAA Atlas 14 precipitation frequency estimates and associated information. Upon clicking a state on the map above or selecting a state name from the drop-down menu, an interactive map of that state will be displayed. From there, a user can identify a location for which precipitation frequency estimates are needed.
Estimates and their confidence intervals can be displayed directly as tables or graphs via separate tabs. Links to supplementary information (such as ASCII grids of estimates, associated temporal distributions of heavy rainfall, time series data at observation sites, cartographic maps, etc.) can also be found.
The main difference between the IMERG Early and Late Run is that Early only has forward propagation (which basically amounts to extrapolation), while the Late has both forward and backward propagation (allowing interpolation). The additional 10 hours of latency allows lagging data transmissions to make it into the Late run, even if they were not available for the Early.
We always advise people to use the Final Run for research unless their application will require the use of Early or Late data due to latency. In such a case, the application should be developed using the long-record of the Early or Late, as appropriate. The vast majority of grid boxes have fairly similar Late and Final values over ocean, and to a lesser extent over land. Extreme value statistics are more sensitive to these details; medians, means, and root-mean square difference are less sensitive.
3GPROF, 'GPROF Profiling', produces global 0.25 degree x 0.25 degree gridded means using Level 2 Gprof data. Vertical hydrometeor profiles and surface rainfall means are computed. Various pixel counts are also reported. The PI is Joyce Chou. The product can be monthly or daily.
As of the GPM Version 6 reprocessing cycle, the radars on both the TRMM and GPM satellites have their data products written in the HDF5 file format. Also as of Version 6 the research products are stored in the same FTP archive for both satellites, The FTP archive is organized into directories whose names are "yyyy/mm/dd/radar/" where yyyy, mm, and dd are the four-digit year and the two-digit month and day of month, respectively. In prior reprocessing cycles, TRMM and GPM data products were stored in different FTP archives. As of May 2020, PPS distributes near-realtime GPM data via FTPS and HTTPS rather than FTP. A similar switch is expected to occur with research data products later in 2020.
GPM Dual-frequency Precipitation Radar (DPR) and TRMM Precipitation Radar (PR) single-orbit rainfall estimates. The objective of these radar algorithms is to generate f radar-only derived meteorological quantities on an instantaneous FOV (field of view) basis working from the level 1 radar products. A subset of these estimates serves as input data to the level 2 combined radar-radiometer algorithm and the level 3 combined and radar-only products. The general idea behind these level 2 algorithms is to determine general characteristics of the precipitation, correct for attenuation, and estimate profiles of the precipitation water content, rainfall rate. When dual-wavelength data are available, the algorithm also estimates the particle size distributions in the phase of the precipitation (i.e., liquid or frozen). GPM's dual-wavelength data will provide better estimates of rainfall and snowfall rates than those of the TRMM PR data.
For instruments currently in orbit, there are near-realtime (NRT) products and standard-research products. All instruments have climate products that are designated 2A-CLIM GPROF or 2A-CLIM PRPS. The difference between climate products and standard-research products is that climate products use ancillary data that are not produced until approximately 3 months after the satellite observations are made.
Level 1A: Reconstructed, unprocessed instrument data at full resolution, time referenced, and annotated with ancillary information, including radiometric and geometric calibration coefficients and georeferencing parameters (i.e., platform ephemeris), computed and appended, but not applied, to Level 0 data.
The Level 1C algorithms calculate calibrated brightness temperature values so that the brightness temperature form an individual sensor is consistent with the brightness temperature from other passive-microwave sensors. Currently, the GPM Microwave Imager (GMI) is the reference instrument for calibrating the other sensors. The 1C data product is the lowest level passive-microwave data product that most researchers are interested in examining, although some instruments have more primitive data products available from the GPM project.
1AGMI contains unpacked packet data from GMI science data from the GMI passive microwave instrument flown on the GPM satellite. Swath S1 has 9 channels which are similar to TRMM TMI (10V 10H 19V 19H 23V 37V 37H 89V 89H). Swath S2 has 4 channels similar to AMSU-B (166V 166H 183+/-3V 183+/-8V). Data for both swaths is observed in the same revolution of the instrument. Swath S3 has ScienceDataHeader. Swath S4 has full rotation for low freq channels (S1). Swath S5 has full rotation for high freq channels (S2)
This file is the percent water surface coverage file used for the IMERG products in binary format. It is on a 0.1x0.1 grid as REAL*4 (3600,1800). The orientation is the same as the IMERG data. The file displays percent water surface coverage with 100% = all water and 0% = all land.
This file is the percent water surface coverage file used for the IMERG products. It is on a 0.1x0.1 grid as REAL*4 (3600,1800). The orientation is the same as the IMERG data. The file displays percent water surface coverage with 100% = all water and 0% = all land.
This file is the percent water surface coverage file used for the TRMM 3B42/43 products, in NetCDF format. It is on a 0.25x0.25 grid as REAL*4 (1440,720). The orientation is the same as the 3B42/3B43 data. The file displays percent water surface coverage with 100% = all water and 0% = all land.
The TRMM Visible and Infrared Scanner (VIRS) Level 1B Calibrated Radiance Product (1B01) contains calibrated radiances and auxiliary geolocation information from the five channels of the VIRS instrument, for each pixel of each scan. The data are stored in the Hierarchical Data Format (HDF), which includes both core and product specific metadata applicable to the VIRS measurements. A file contains a single orbit of data with a file size of about 95 MB. The EOSDIS "swath" structure is used to accommodate the actual geophysical data arrays. There are 16 files of VIRS 1B01 data produced per day.
Geolocation and channel data are written out for each pixel along the scan, whereas the time stamp, scan status (containing scan quality information), navigation, calibration coefficients, and solar/satellite geometry are specified on a per-scan basis. There are in general 18,026 scans along the orbit pre-boost and 18,223 post-boost, with each scan consisting of 261 pixels. The scan width is about 720 km pre-boost and 833 km post-boost.
The Level-1A product is a simple concatenation of Level-0 data with a Header, which could easily be reversed back to Level-0. Level-1A remains in a binary format and is not in HDF. The Level-1B format groups like data together.
Precipitation data from the GPM and TRMM missions are made available free to the public in a variety of formats from several sources at NASA Goddard Space Flight Center. This section outlines the different types of data available, the levels of processing, the sources to download the data, and some helpful tips for utilizing precipitation data in your research.
There are several sources for downloading and viewing data which allow you to subset the data to only include specific parameters and/or geographic locations. These include the GES DISC, Giovanni and STORM. In Giovanni you can obtain data for a specific country, U.S. state, or watershed by using the "Show Shapes" option in the "Select Region" pane.
The TRMM satellilte has been decommissioned and stopped collecting data in April 2015. The transition from the Tropical Rainfall Measuring Mission (TRMM) data products to the Global Precipitation Measurement (GPM) mission products has completed as of August 2019. The GPM IMERG dataset now includes TRMM-era data from June 2000 to the present, and other TRMM-era data has been reprocessed with GPM-era algorithms and is now available on the GPM FTP servers. TMPA data production ended as of December 31st, 2019 and the TRMMOpen FTP server has been shut down. Historical TMPA data is still available to download from the NASA GES DISC at: =TMPA&page=1
Browse our directory of GPM & TRMM data products to locate your desired algorithm, then click on the links in the algorithm description under "Documentation". All documentation is also available at the Precipitation Processing System website.
The resolution of Level 0, 1, and 2 data is determined by the footprint size and observation interval of the sensors involved. Level 3 products are given a grid spacing that is driven by the typical footprint size of the input data sets.
GPM data products can be divided into two groups (near real-time and production) depending on how soon they are created after the satellite collects the observations. For applications such as weather, flood, and crop forecasting that need precipitation estimates as soon as possible, near real-time data products are most appropriate. GPM near real-time (GMI & DPR) products are generally available within a few hours of observation. For all other applications, production data products are generally the best data sets to use because additional or improved inputs are used to increase accuracy. These other inputs are only made available several days, or in some cases, several months, after the satellite observations are taken, and the production data sets are computed after all data have arrived, making possible a more careful analysis.
For the GPM IMERG dataset, IMERG Early and Late Runs are the near real-time products, while IMERG Final Run is the research / production product. Click here to learn more about the differences between IMERG Early, Late and Final.