Climate Rainfall Products
A large number of climate rainfall products from a variety of sources are currently available to the end user. All of them have strengths and weaknesses which may or may not be important depending on the application. While several merged satellite and rain gauge datasets have been developed specifically for climate applications, significant climate variations exist in the component datasets used in these products. The method used to combine the various components as well as the weighting used for each can and does impact variability in the climate record. Because of this we have broken down the various climate rainfall products into those from research satellites, operational satellites, merged products, and ground-based products. To aid the user in understanding how the merged products differ from the component products and from state-of-the-art rain estimates from research satellites such as TRMM we have developed an interactive product intercomparisons page [Plot Intercomparison].
Research satellites often involve new technology being flown on a single satellite. As a result, such missions rarely provide long enough time series for most climate applications. As in the case of TRMM, however, which flew the first precipitation radar, the hardware (sensor) and software (algorithm) are both designed to provide state-of-the-art rainfall information. For this reason, comparing long-term climate rainfall datasets with those from research satellites can provide valuable insight into limitations and/or issues with the climate record.
Because all of the various rain products currently available have advantages and disadvantages, a number of climate rainfall products attempt to merge these products to take advantage of the best attributes of each while minimizing the negatives. Satellite IR rain estimates from geostationary satellites provide excellent sampling over both space and time, however, the observed cloud-top temperatures are only indirectly related to rainfall. Passive microwave techniques provide a more physically direct measure of rainfall, but poorer spatial resolution and temporal sampling leads to biases in climate products. Rain gauge measurements are the most direct measure of surface rainfall, however, they are not available over oceans and high latitude regions, and have limited coverage over many land regions.
Operational satellites generally consist of a series of satellites with the same or similar sensor systems. Datasets produced from operational satellites/sensors provide continuity over longer time periods, which is valuable for creating climate records. A number of rainfall datasets created from operational sensors go back to January of 1979, however, the IR window channel and microwave sounding instruments used for these long-term satellite rainfall records were not designed for rainfall estimatation. Given the indirect relationship between the satellite observed brightness temperatures and rainfall for many of these sensors, large errors and climate regime biases are evident from these datasets. They can still be useful for climate studies, but the user should be careful regarding how they are used.
Ground-based rainfall measurements from gauges/radars often provide higher spatial and temporal sampling as well as the most direct measure of rainfall over many regions. Often such data are used for purposes of calibration, however, over regions with sufficient coverage they can provide better large-scale climate estimates than any satellite remote sensing technique. For this reason, accumulations of global gridded rain gauge data have been incorporated into several of the merged products.