Observational uncertainties in the estimation of precipitation extremes
Abstract
Observational evidence of the precipitation amounts falling during extreme events is vital to better understand how this might change in a future warmer climate. Nowadays, a number observational datasets are available for precipitation, and this number has largely increased since the beginning of the satellite era. However, for a focus at the global scale, there is little consensus on the reliability of satellite products for precipitation extremes and more generally on which dataset to use. Over the terrestrial regions of a quasi-global domain (50°S:50°N; 130°W:180°E), we inter-compare the representation of extreme precipitation in more than 20 observational products from a variety of sources: in situ-based, satellite observations with or without a correction to rain gauges and reanalyses. Our aim is to evaluate the consistency of precipitation extremes across these different observations, and also to give guidance on the use of the observational products for precipitation extremes at the global scale. Our results reveal larger observational uncertainties for annual extreme than annual total precipitation. We show that reanalyses generally present a pronounced and widespread bias, either wet or a dry, in the tropics in particular. We find that extreme precipitation intensity is relatively similar between in situ-based and satellites estimates and that the satellite version with a correction to rain gauges is preferred to the uncorrected version as it compares generally better to the ensemble. This inter-comparison allows the selection of a subset of datasets, which reduces the observational uncertainties for the estimation of precipitation extremes over global land. Extreme precipitation intensity as well as the timing of extreme precipitation throughout the year compare better within that selection compared to all datasets, giving further confidence on the use of this selected products. Given the level of uncertainties associated with the estimation of extreme precipitation in the observations, none of the datasets can be thought as the best estimate. We recommend avoiding reanalyses but considering in situ-based and satellite data to gather an ensemble of products for a better estimation of observational uncertainties.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFM.A41E..05B
- Keywords:
-
- 3305 Climate change and variability;
- ATMOSPHERIC PROCESSES;
- 3354 Precipitation;
- ATMOSPHERIC PROCESSES;
- 1854 Precipitation;
- HYDROLOGY;
- 4313 Extreme events;
- NATURAL HAZARDS