84th AMS Annual Meeting

Monday, 12 January 2004
Climate analysis and prediction over the Arno river basin, Italy
Hall 4AB
Francesco Meneguzzo, Institute of Biometeorology /National Research Council, Firenze, Italy; and G. Menduni, G. Maracchi, M. Baldi, G. Brandani, A. Crisci, F. Marrese, M. Pasqui, and F. Piani
Poster PDF (1.2 MB)
1. Introduction. The long term management and planning of a catchment require the accurate knowledge of heavy rainfall and drought regimes for flood protection and water resource conservation. The project “Climate reanalysis and prediction over the Arno river basin”, funded by the Arno River Basin Authority (Italy), aims at providing quantitative information concerning the past, current and future variability and trends of heavy rainfall and drought occurring over the Arno River Basin, Italy (about 9200 km2). The broad scope is the support to the periodic update of the distributed constraints system around the basin, to the design and management of flood protection works, to the targeting of the local weather forecasting system, and to the water quality and conservation policies.

2. Data and methods. The ongoing project is based on the historical series of in-situ rainfall data, the CRU, GPCC and GPCP gridded precipitation products, the NCEP/NCAR global atmospheric and surface reanalyses, and the climate scenarios produced by the Canadian Centre for Climate Modelling and Analysis (CCCma). The global atmospheric reanalyses were processed to identify decadal changes and trends in the average seasonal circulation, represented in terms of the storm tracks over the North Atlantic, Europe and the Mediterranean. The gridded precipitation data and the global atmospheric reanalyses were processed to identify mechanisms driving the summer climate over Mediterranean and the larger area. The gridded and the in-situ precipitation data were analyzed during at least 50 years in the past to feature the seasonal inter-annual variability and trends of the frequency of rainy days, precipitation totals and average daily intensity, the frequency of local and basin-scale rainfall events above few thresholds. The significance of any trend has been evaluated by means of the Mann-Whitney test. The same analyses concerning the historical in-situ data were performed on the climate scenarios provided by the CCCma, on the (coarse) grid cell covering the Arno River basin.

3. Results. At the large North Atlantic, European and Mediterranean scale, the storm tracks have undergone significant decadal changes at least in winter, summer and fall, suggesting global mechanisms forcing such changes. The precipitation patterns over the Mediterranean land areas partly follow the changes of the storm tracks, especially in winter, and partially in fall, while other background forcings could play a major role in the other seasons. The regional air masses around the Arno river basin appear to have changed their own features, revealed by means of the moisture load (precipitable water and relative humidity), the drying being the most impressive result (even if precipitable water sometimes grows, likely due to increasing evaporation). The regional warming is also significant and continuous, strongest in summer and at low altitudes, so that the vertical lapse rate in the low to mid-troposphere is generally increasing. Remote links of the summer Mediterranean climate were investigated, motivated by the occurrence of both extreme rainstorms and prolonged droughts in that season. The Asian monsoon and above all the West African monsoon appear to be the most likely candidates to drive the inter-annual variability: stronger West African monsoons coincide with increased mid-tropospheric geopotential heights and reduced precipitation over central and western Mediterranean. Among the relevant results concerning the precipitation variability around the Arno river basin, the following are worth to be mentioned. 3.1 The total annual precipitation has not changed significantly since 1950, but the frequency of rainy days has decreased until early 1980s’. 3.2 The year-round average daily rainfall intensity increases significantly. 3.3 In winter, the total precipitation and the frequency of rainy days have decreased significantly until mid-1990s’. 3.4 In spring, the total precipitation has increased as a result of the increasing average daily rainfall intensity. 3.5 In summer, the total precipitation is decreasing on the upper mountainous portion of the basin, while the average daily rainfall intensity is significantly increasing elsewhere, with a pronounced inter-annual variability. 3.6 In fall, statistically robust 30-years increasing trends are shown for total precipitation, frequency of rainy days and precipitation intensity, superimposed to a relevant inter-annual variability. 3.7. The daily precipitation events exceeding given thresholds, relevant for local flash-floods and floods, are growing in frequency both locally and averaged over the sub-basins, so that they are today more frequent than ever, at least in the last 150 years: such frequency has increased from 20% in the upper portion of the basin to about 150% in the lower portion. 3.8 Extreme annual precipitations have increased, but only at very short duration (1 hour, 3 hours). 3.9 The CCCma climate scenarios suggest for the next decades, beyond insignificant changes of the total precipitation and of the frequency of rainy days, a further increase of the frequency of excessive daily rainfalls until at least 2015, henceforth it should remain constant, about 30% higher than in the current climate. 3.10 The CCCma scenarios suggest higher and higher summer daily rainfall intensities in the next very few decades, along with an abrupt increase of the chance of extreme summer droughts in about 20 years. 3.11 The winter precipitations should gradually increase, so to completely offset the future summer drying on an yearly basis. 3.12 The regional surface warming is predicted in significant acceleration, in summer more than in winter, so that future summer precipitation deficits could result in extreme droughts.

4. Conclusions. Both flood and drought risks appear to undergo significant changes over the Arno River basin, both in the recent past and in the near future. While the flood risk is already significantly increasing, and is projected to increase rapidly in the next several years or few decades, the drought risk is just starting to show, and is predicted to become a potentially dramatic challenge for water quality and availability in next decades. The climate of late spring and summer 2003 has been a very large outlier with regard to precipitation (in the lower 5th percentile of the last 150 years) and temperature (May-August anomaly about 3°C over the 1961-1990 climatological average), much above any past record and largely exceeding the variability simulated by the CCCma climate scenarios.

Supplementary URL: http://www.ibimet.cnr.it/clima