Réf. Ciccarelli & al. 2008 - A

Référence bibliographique complète
Ciccarelli N., von Hardenberg J., Provenzale A., Ronchi C., Vargiu A., Pelosini R. Climate Variability in North-Western Italy during the Second Half of the 20th Century. Global and Planetary Change, 2008, in press, 25 p.

Abstract: A large set of daily temperature and precipitation time series measured by a dense observational network in north-western Italy was analysed, in the period from 1952 to 2002. Average temperatures display a significant increase of about 1°C over the period of observation. The increase is particularly evident for maximum daily temperatures in winter and summer months. By contrast, precipitation time series display no significant trend over the last fifty years. The statistical properties of interannual fluctuations in temperature and precipitation was also determined, and their correlation with large-scale atmospheric patterns and global indices such as the North Atlantic Oscillation (NAO), the Scandinavian pattern (SCAN) and European Blocking (EB) was quantified. The positive phase of the Scandinavian pattern and the presence of frequent blocking episodes are found to be significantly correlated with increased summer and fall precipitation and cold temperatures in the study area.

Mots-clés
Climate variability, time series analysis, impacts of climate change, regional studies.

Organismes / Contact
ISAC-CNR, Corso Fiume 4, I-10133 Torino, Italy. j.vonhardenberg@isac.cnr.it
ARPA Piemonte, Corso Unione Sovietica 216, 10134 Torino, Italy.

(1) - Paramètre(s) atmosphérique(s) modifié(s)
(2) - Elément(s) du milieu impacté(s)
(3) - Type(s) d'aléa impacté(s)
(3) - Sous-type(s) d'aléa
Temperature, Precipitation,
Circulation patterns
     

Pays / Zone
Massif / Secteur
Site(s) d'étude
Exposition
Altitude
Période(s) d'observation
North-western Italy Piedmont and Valle d'Aosta       1952-2002

(1) - Modifications des paramètres atmosphériques
Reconstitutions
 
Observations
Precipitation
Even if the ensemble averages of precipitation variables do not show any significant trend, these can instead be found for individual station data. Overall, stations showing a significant positive trend are offset by others showing significant negative trends. No elevation-dependent pattern emerges from these results, as the stations with significant positive or negative precipitation trends are evenly spread in elevation.

Even if no significant trend is associated with the longest duration of a dry spell in a year, this statistic shows extremely strong fluctuations, especially in recent years. Frequent episodes with very long dry spells are evident, particularly starting from 1989.

Extension of the analysis to other statistics, which characterize the intra-annual temporal distribution of observed precipitation and the nature of precipitation events, indicates the absence of significant trends in all the variables considered.

Temperature
Ensemble averages of annual means of minimum and maximum temperature anomalies show a significant and quite abrupt increase in temperatures starting in the mid 80s. A linear fit to the ensemble-averaged annual means of minimum and maximum temperature anomalies confirms significant and quite large trends of 0.023°C/year for maximum temperatures and of 0.012°C/year for minimum temperatures.

Summer and winter months show significant trends in maximum temperatures while minimum temperatures show a significant trend only in summer.

The plot confirms a positive trend for most stations and does not suggest a dependence of trend magnitude on station elevation. Further, temperature trends of individual stations show no dependence on station latitude (not shown).

The analysis of the ERA40 data provides results that are consistent with the analysis of the station dataset. ERA40 maximum temperatures over north-western Italy show a significant linear increase, of about the same magnitude, when considered at the annual scale or for winter and summer months. Minimum temperatures also show a similar overall trend for annual means, but, differently from the station data, they do not show a significant trend in summer months.

Large scale circulation patterns
While temperatures are positively correlated with the EA pattern all over the year, NAO plays a role, with a positive correlation, only in winter. In other seasons it is the SCAN index which affects (negatively) temperatures in the study area. Precipitation is (negatively) correlated with NAO only in winter, while a positive phase of SCAN leads to increased precipitation all over the year. In general these results are consistent with the role of SCAN for precipitation in Italy discussed in Wibig (1999).

The Euro-Atlantic blocking index has correlations that are similar to those of the negative phase of NAO: Low temperatures in winter and spring are associated with the positive phase of EAB. The index shows no significant correlation with precipitation, in contrast with the results reported in Quadrelli et al. (2001). The European blocking (EB) index shows, not surprisingly, a behaviour analogous to the SCAN index: It is associated with high precipitation in summer and fall and with low temperatures from winter to summer. The correlation with temperature is consistent with the impact of blocking episodes on winter temperatures in Europe discussed in Trigo et al. (2004).
Modélisations
 
Hypothèses
 

Informations complémentaires (données utilisées, méthode, scénarios, etc.)
Time series of daily cumulated precipitation and of daily minimum and maximum temperatures, measured by a dense and uniformly distributed observational network located in the Piedmont and in the Valle d'Aosta regions in north-western Italy, were analysed. The network includes 119 rain gauges and 40 thermometric stations. This study has focused on the period 1952-2002.

All available time series were checked for internal consistency and for the presence of outliers. Some stations provided valid data only for a fraction of the entire period of interest and some years are incomplete. Temporal interpolation techniques were used to fill these gaps and analyse only the measured data. Data in a year or in a season are used only if the number of missing days is not too large; annual and seasonal averages are computed only if no more than 31 days in a year and/or 7 days in a season are missing. “Ensemble averages” mean averages taken over the set of measurement stations. Annual day counts and annual averages start on December 1st: The data corresponding e.g. to the year 1987 go from December 1st, 1986 to November 30th, 1987.

Monte-Carlo methods were extensively used to estimate sampling errors and to assess the significance of the results. The results provided by this method were compared with a standard Mann-Kendall test at the same significance level, obtaining the same results (apart one single exception mentioned below).

The following statistics were defined from the precipitation time series, all averaged on a yearly or seasonal time scale: average precipitation, precipitation intensity (precipitation averaged only over rainy days), the average duration of dry periods and of rainy events in each year/season, the average cumulated precipitation in rainy events and the percentage of non-rainy days. Extremes were analysed by computing also the longest duration of dry spells in a period, the longest duration of rainy episodes and maxima of cumulated precipitation per event in each year. For each station, these statistics were standardized to unit variance and zero mean before further analysis. A standard definition of winter (DJF), spring (MMA), summer (JJA) and autumn (SON) was used. A standard threshold of 1 mm/day was used to define a rainy day.

Results on temperatures from station data were compared with those obtained from the ERA40 reanalysis project from the European Centre for Medium-Range Weather Forecasts (Uppala et al., 2005) spanning the period 1958-2002. The reanalysis data are statistically independent of the station data used here as the latter were not part of the ERA40 assimilation cycle. Daily minimum and maximum surface temperature anomalies from the ERA40 dataset were extracted, using a grid resolution of 1.25°, to obtain a single timeseries by averaging data from the 9 grid points covering the region of interest.

The authors computed the correlation of the seasonally-averaged precipitation and temperature time series with the NAO (North Atlantic Oscillation), SCAN (Scandinavia), EA (East Atlantic) and EAWR (East Atlantic/West Russia) teleconnection indices. The teleconnection patterns were all computed from NCEP reanalysis data in the period 1950-2000. Significance of correlations was determined using both a standard Student's t-test and a shuffling Monte-Carlo method, obtaining the same results in all cases.

(2) - Effets du changement climatique sur le milieu naturel
Reconstitutions
 
Observations
The lack of trends in total precipitation and the observed increase in temperatures have led to a significant decrease of snow cover in the western Alps in the last twenty years.
The presence of a significant warming trend in summer months and the absence of any precipitation trend implies increased aridity, associated with larger evapo-transpiration due to the higher temperatures.
Modélisations
 
Hypothèses
 

Sensibilité du milieu à des paramètres climatiques
Informations complémentaires (données utilisées, méthode, scénarios, etc.)
   

(3) - Effets du changement climatique sur l'aléa
Reconstitutions
 
Observations
 
Modélisations
 
Hypothèses
 

Paramètre de l'aléa
Sensibilité du paramètres de l'aléa à des paramètres climatiques
Informations complémentaires (données utilisées, méthode, scénarios, etc.)
 
 

(4) - Remarques générales
 

(5) - Syntèses et préconisations
 


Références citées :

Quadrelli, R., Lazzeri, M., Cacciamani, C., Tibaldi, S., 2001. Observed winter Alpine precipitation variability and links with large-scale circulation patterns. Climate Research 17, 275–284.

Trigo, R. M., Trigo, I. F., DaCamara, C. C., Osborn, T. J., 2004. Climate impact of the European winter blocking episodes from the NCEP/NCAR reanalyses. Climate Dynamics 23, 17–28.

Uppala, S. M., Kallberg, P. W., Simmons, A. J., et al., 2005. The ERA-40 re-analysis. Quarterly Journal of the Royal Meteorological Society 131, 2961–3012.

Wibig, J., 1999. Precipitation in Europe in relation to circulation patterns at the 500 hPa level. International Journal of Climatology 19, 253–269.