Mixed nonlinear regression for modelling historical temperatures in Central–Southern Italy

This paper has exploited, for Central and Southern Italy (Mediterranean Sub-regional Area), an unprecedented historical dataset as an attempt to model seasonal (winter and summer) air temperatures in pre-instrumental time (back to 1500). Combining information derived from proxy–documentary data and large-scale simulation, a statistical downscaling approach in the form of mixed regression model was developed to adapt larger-scale estimations (regional component) to the sub-regional temperature pattern (local component). It interprets local temperature anomalies by means of monthly based Temperature Anomaly Scaled Index in the range ?5 (very cold conditions in June) to 2 (very warm conditions). The modelled response agrees well with the independent data from the validation sample (Nash–Sutcliffe efficiency coefficient, >0.60). The advantage of the approach is not merely increased accuracy in estimation. Rather, it relies on the ability to extract (and exploit) the right information to replicate coherent temperature series in historical times.     

Multiscale modelling of rainstorm-induced historical intermediate floods (June–November) in the Rhone River Basin

This study reveals the changes and evolution of rainstorm-driven intermediate floods occurring and driving multiple damaging hydrological events in the Rhone River Basin (RRB), since 1500 until 2010. A parsimonious approach was developed to simulate the major hydroclimatological flood-producing forcing, the Multiscale Rainstorm Climate Model (STORMCLIMM). We collected the frequency of intermediate floods—a type of particularly hazardous floods commonly taking place between June and beginning of November—from the RCB to be compared to STORMCLIMM estimates. The latter, smoothed by a moving window of 21 years, results in a high-pass filter in the time domain, which magnifies the signal of forcing variations causing intermediate floods. The RRB showed large temporal variations in both extreme rainstorms and associated multidecadal intermediate-floods (MUDIF) frequency at different climatic periods and land-use systems through historical times. An important peak was observed in the Maunder Minimum (1645–1715 AD). The model allowed detecting MUDIF occurred in the historical times. The situation becomes interesting with respect to recent times, because the Rhone landscape looks more vulnerable in the last decades as a consequence of land-use changes and climate shift towards more erratic and intensive storms. This evidence suggests that the interactions of land-use and climatic changes may turn into considerable vulnerability to fluvial flooding and agro-ecosystem connected to them for upcoming years. The Rhone, for example, provides basis for use of hydrological indicators (such as the one represented by STORMCLIMM) for one site or region and which, through minor modifications, can be made relevant to specific needs.     

Assessing and modelling changes in rainfall erosivity at different climate scales

An understanding of the weather drivers of soil erosion necessitates an extended instrumental meteorological series and knowledge of the processes linking climate and hydrology. The nature of such linkages remains poorly understood for the Mediterranean region. This gap is addressed through a composite analysis of long‐term climatic controls on rain erosivity in the Calore River Basin (southern Italy) for the period 1869–2006. Based on a parsimonious interpretation of rainstorm processes, a model (comparable with the Revised Universal Soil Loss Equation) was adapted to generate erosivity values on different time‐aggregation scales (yearly and seasonal). The evolution of the generated series of cumulated and extreme erosivity events was assessed by two return period (T) quantiles via a 22‐year moving window analysis (low return period, T = 2 years; high return period, T = 50 years). Erosivity extremes are shown to be characterized by increasing yearly trends (at a 100‐year rate of ～150 MJ mm ha^–1 h^–1 for T = 2 years and ～800 MJ mm ha^–1 h^–1 for T = 50 years), especially during the spring and autumn seasons. Quantile patterns on the extremes are also shown to be decoupled from trends in the cumulated values. The Buishand test was applied to detect the presence of temporal change points, and a wavelet spectrum analysis used for time‐frequency localization of climate signals. A change‐point in the evolution of climate is revealed over the 1970s in the spring series, which correlates to a distinct rain erosivity increase. The results indicate that soil erosion risk tends to rise as a consequence of an escalation of the climate erosive hazard, predominantly between April and November (associated with cultivation and tillage practices). Copyright © 2009 John Wiley & Sons, Ltd.     

Mapping soil erodibility from composed data set in Sele River Basin, Italy

Evaluation of soil erodibility is an important task for Mediterranean lands, in which fertility and crop yield are significantly affected by soil erosion. The soil physicochemical parameters affecting soil erodibility are highly variable in space and, as for many other environmental variables, sample measurements are generally not enough for assessing its spatial variability with an acceptable level of uncertainty at the scales of practical interest. This study illustrates the procedure applied for estimating the pattern of soil erodibility across the Sele Basin (Southern Italy), where soil properties have been measured on a limited number of sparse samples. Sampled data were integrated with other sparse data estimated by local regression functions, which relate soil erodibility to auxiliary variables, such as terrain attributes and land system class memberships. Sampled and estimated data were merged in a composed data set to assess the spatial pattern of soil erodibility by ordinary kriging. The proposed approach offers effective spatial predictions, and it is exportable to regions where financial costs for soil sampling are not feasible.     

500-year temperature reconstruction in the Mediterranean Basin by means of documentary data and instrumental observations

The paper reports the main results of the EU project Millennium in the Mediterranean area over the last 500 years. It analyses a long series of temperature from Portugal, Spain, France, Italy and Greece. The series are obtained by combining indices from documentary sources from AD 1500 to the onset of regular instrumental observations. There is an ongoing discussion regarding the proper way of combining documentary and instrumental data and how to translate accurately the conventional indices from ??3 to +?3 into modern units, i.e. degree Celsius. This paper produces for the first time a number of early instrumental observations, in some cases (i.e. Italy and France) covering 350 years, including thereby the earliest regular observations after the invention of the thermometer. These Mediterranean data show that anomalous temperatures usually had only a locally limited effect, while only few extreme events had a widespread impact over the whole region, such as the summer of 2003. During the period from 1850 to the present day, the Mediterranean temperature anomaly was close to the Northern Hemisphere in spring and summer, while it was warmer in autumn and winter. Compared with the long-term instrumental records (i.e. 1655 onwards), the recent warming has not exceeded the natural past variability characterized by heating–cooling cycles with no significant long-term trends.     

Interpolation processes using multivariate geostatistics for mapping of climatological precipitation mean in the Sannio Mountains (southern Italy)

The spatial variability of precipitation has often been a topic of research, since accurate modelling of precipitation is a crucial condition for obtaining reliable results in hydrology and geomorphology. In mountainous areas, the sparsity of the measurement networks makes an accurate and reliable spatialization of rainfall amounts at the local scale difficult. The purpose of this paper is to show how the use of a digital elevation model can improve interpolation processes at the subregional scale for mapping the mean annual and monthly precipitation from rainfall observations (40 years) recorded in a region of 1400 km² in southern Italy. Besides linear regression of precipitation against elevation, two methods of interpolation are applied: inverse squared distance and ordinary cokriging. Cross‐validation indicates that the inverse distance interpolation, which ignores the information on elevation, yields the largest prediction errors. Smaller prediction errors are produced by linear regression and ordinary cokriging. However, the results seem to favour the multivariate geostatistical method including auxiliary information (related to elevation). We conclude that ordinary cokriging is a very flexible and robust interpolation method because it can take into account several properties of the landscape; it should therefore be applicable in other mountainous regions, especially where precipitation is an important geomorphological factor. Copyright © 2005 John Wiley & Sons, Ltd.