Search

International audience

Monthly mean satellite data from the Global Precipitation Climatology Project (GPCPv2) are used to examine the year-by-year variability of precipitation over the Mediterranean Basin and its changes over the period 1979-2010. The results show that the mean annual precipitation averaged over the study area has slightly increased from 1979 to 2010 by 1.28mm or by 0.2% (trend not statistically significant at the 95% confidence level). Nevertheless, examining the trends at a local scale, spatial and temporal patterns are revealed, with opposite trends in adjacent areas and increasing precipitation in summer and autumn against almost unchanged or decreasing precipitation in winter and spring, respectively. Inter-decadal changes of precipitation are detected, with precipitation decreasing in the 1980s, then increasing through the late 1990s and finally declining in the 2000s before levelling off since 2007. These changes are significantly anti-correlated (R=-0.57, up to -0.66 in winter) with the North Atlantic Oscillation (NAO) index, thus confirming the critical role of this large-scale teleconnection for the regional precipitation over the basin.

The precipitation regime over the Mediterranean basin is investigated for the period 1979-2010 using monthly mean satellite data from the Global Precipitation Climatology Project (GPCPv2). The results show that a clear contrast exists between the more rainy northern part of the study region (Southern Europe) and the drier southern area (North Africa, Iberian Peninsula) and between the western sides (rainsides) of the Iberian, Italian and Balkan peninsulas and their eastern sides (rainshadows). The mean annual precipitation averaged over the study area is P = 593 ± 203 mm year-1, but it has a strong spatial variability ranging from 20 mm year-1 (North Africa) to 1500 mm year-1 (Alps). A significant seasonal variability exists, with the early winter and late autumn months (November and December) being the wettest with precipitation amounts larger than 60 mm month-1. The GPCPv2 satellite precipitation data are satisfactorily correlated with rain gauge measurements from 433 stations within the study area (correlation coefficient R = 0.78 for all stations on a yearly basis, with values ranging between 0.72 and 0.82, depending on the season) with a slight overestimation. They also compare well with the higher spatial and temporal resolution Tropical Rainfall Measuring Mission (TRMM) data, which supports the validity of the present study.

A prototype system for the analysis and investigation of extreme rainfall events

Of the three primary sources of spatially contiguous precipitation observations (surface networks, ground-based radar, and satellite-based radar/radiometers), only the last is a viable source over ocean and much of the Earth's land. As recently as 15 years ago, users needing quantitative detail of precipitation on anything under a monthly time scale relied upon products derived from geostationary satellite thermal infrared (IR) indices. The Special Sensor Microwave Imager (SSMI) passive microwave (PMW) imagers originated in 1987 and continue today with the SSMI sounder (SSMIS) sensor. The fortunate longevity of the joint National Aeronautics and Space Administration (NASA) and Japan Aerospace Exploration Agency (JAXA) Tropical Rainfall Measuring Mission (TRMM) is providing the environmental science community a nearly unbroken data record (as of April 2012, over 14 years) of tropical and sub-tropical precipitation processes. TRMM was originally conceived in the mid-1980s as a climate mission with relatively modest goals, including monthly averaged precipitation. TRMM data were quickly exploited for model data assimilation and, beginning in 1999 with the availability of near real time data, for tropical cyclone warnings. To overcome the intermittently spaced revisit from these and other low Earth-orbiting satellites, many methods to merge PMW-based precipitation data and geostationary satellite observations have been developed, such as the TRMM Multisatellite Precipitation Product and the Climate Prediction Center (CPC) morphing method (CMORPH. The purpose of this article is not to provide a survey or assessment of these and other satellite-based precipitation datasets, which are well summarized in several recent articles. Rather, the intent is to demonstrate how the availability and continuity of satellite-based precipitation data records is transforming the ways that scientific and societal issues related to precipitation are addressed, in ways that would not be otherwise possible. These developments have taken place in parallel with the growth of an increasingly interconnected scientific environment. Scientists from different disciplines can easily interact with each other via information and materials they encounter online, and collaborate remotely without ever meeting each other in person. Likewise, these precipitation datasets are quickly and easily available via various data portals and are widely used. Within the framework of the NASA/JAXA Global Precipitation Measurement (GPM mission, these applications will become increasingly interconnected. We emphasize that precipitation observations by themselves provide an incomplete picture of the state of the atmosphere. For example, it is unlikely that a richer understanding of the global water cycle will be possible by standalone missions and algorithms, but must also involve some component of data, where model analyses of the physical state are constrained alongside multiple observations (e.g., precipitation, evaporation, radiation). The next section provides examples extracted from the many applications that use various high-resolution precipitation products. The final section summarizes the future system for global precipitation processing.

The intense rainfall systems typically forming on the Mediterranean basin are often characterized by trajectories which are directed to the Southern part of the Italian Peninsula. With exception of the so called Tropical Like-Cyclone (TLC), that is a well-organized system rotating around a central eye with a strong surface wind, the majority of the flash-flood events are generated by deep convective rainfall characterized by a small scale domain and a very short life time. As demonstrated by CHARAVALLOTI & GABRIELE (2009) these systems can be extremely localized up to the basin scale and the intense precipitation associated to them are responsible for strong runoffs. For this reason, these systems are very difficult to model and appear substantially unpredictable. Furthermore, the observations with meteorological satellites are often not representative of the system characteristics and the use of rain gauges is generally inadequate to fully reconstruct the ground rainfall field.

Two heavy rain events over the Central Mediterranean basin, which are markedly different by genesis, dimensions, duration, and intensity, are analyzed. Given the relative low frequency of this type of severe storms in the area, a synoptic analysis describing their development is included. A multispectral analysis based on geostationary multifrequency satellite images is applied to identify cloud type, hydrometeor phase, and cloud vertical extension. Precipitation intensity is retrieved from (i) surface rain gauges, (ii) satellite data, and (iii) numerical model simulations. The satellite precipitation retrieval algorithm 183-Water vapor Strong Lines (183-WSL) is used to retrieve rain rates and cloud hydrometeor type, classify stratiform and convective rainfall, and identify liquid water clouds and snow cover from the Advanced Microwave Sounding Unit-B (AMSU-B) sensor data. Rainfall intensity is also simulated with theWeatherResearch and Forecasting (WRF) numerical model over two nested domains with horizontal resolutions of 16 km (comparable to that of the satellite sensor AMSU-B) and 4 km. The statistical analysis of the comparison between satellite retrievals and model simulations demonstrates the skills of both methods for the identification of the main characteristics of the cloud systems with a suggested overall bias of the model toward very low rain intensities. WRF (in the version used for the experiment) seems to classify as low rain intensity regions those areas where the 183-WSL retrieves no precipitation while sensing amixture of freshly nucleated cloud droplets and a large amount ofwater vapor; in these areas, especially adjacent to the rain clouds, large amounts of cloud liquid water are detected. The satellite method performs reasonably well in reproducing the wide range of gauge-detected precipitation intensities. A comparison of the 183-WSL retrievalswith gauge measurements demonstrates the skills of the algorithm in discriminating between convective and stratiform precipitation using the scattering and absorption of radiation by the hydrometeors.