A mixed application of an objective synoptic classification and spatial regression models for deriving winter precipitation regimes in the Eastern Pyrenees

Abstract

Management of hydric resources in alpine mountains requires spatial knowledge of precipitation as a variable of noteworthy importance in the study of hydrological hazards (avalanches, landslides, floods, etc.), especially during the winter season. We therefore study the spatial distribution of mean daily precipitation (MDP) and daily precipitation probability (DPP) in the Eastern Pyrenees, based on a previous objective synoptic classification defining the most frequent atmospheric patterns during the winter season (November-May) between 1990 and 2015. The synoptic classification provided 12 circulation weather types. For each of these, we obtained MDP maps and DPP maps possessing a threshold equal to, or greater than, 2, 5, 10, 20 and 50 mm. The best fit of the models was obtained in the DPP ≥ 2 mm, with an adjusted R 2 of around 0.8, followed by the models showing MDP and DPP ≥ 5 mm, with an adjusted R 2 generally between 0.7 and 0.8. Finally, we performed an unsupervised classification of the 12 MDP models in order to obtain a categorical and simplified cartography explaining the winter precipitation regimes in the Eastern Pyrenees.

Spatio‐temporal patterns of snow in the Catalan Pyrenees (NE Iberia)

“…Using a S‐mode matrix and with varimax rotation, the components retained, 4 in this case (71% of variance), derive in 8 potential CT. Later, the real cases were classified thanks to the higher and lower values of the scores and the multidimensional distances by K‐means clustering method, obtaining the catalogue of CT. This methodology has been successfully applied in Andorra to analyse heavy snowfalls (Esteban et al ., 2005), and in the study area to characterize lightning activity (Pineda et al ., 2010) or precipitation (Lemus‐Canovas et al ., 2018). We used NCEP/NCAR Reanalysis 2 database (Kalnay et al ., 1996) for obtain the CT.…”

“…From November to May, the average precipitation above 2,000 m is ca. 700 mm at the eastern Catalan Pyrenees and Pre‐Pyrenees and around 900 mm at the axial zone of the Pyrenees (Lemus‐Canovas et al ., 2018). The northern slopes record higher amounts of precipitation than the southern ones, receiving also less solar radiation.…”

Precipitation in peninsular Spain: Influence of teleconnection indices and spatial regionalisation

“…The components selected are rotated by means of the Varimax rotation method (Esteban et al ., 2006; Lemus‐Canovas et al ., 2019b). This technique enables the variance to be redistributed in order to prevent most of the series from being grouped in the first region (Lemus‐Canovas et al ., 2018). This resulted in a series of new variables called scores, which show the degree of representativeness associated with the mode of variation of each principal component.…”

Characterisation of Extreme Precipitation Events in the Pyrenees: From the Local to the Synoptic Scale

“…For example, Esteban et al [17] modelled daily mean precipitation and temperature in Andorra for different circulation types, the latter obtained through principal component analysis (PCA) applied to an Smode matrix [7] and subsequent clustering with k-means; similarly, Lemus-Canovas [18] employed statistical regression techniques to perform a synoptic classification focused on SW Europe aimed at mapping the daily mean precipitation of each weather type over the whole Pyrenees. Other examples of this type of approach can be found in different parts of the Iberian Peninsula [19][20][21][22][23], frequently using the objective Lamb approach [19,24]. However, another approach for characterising atmospheric patterns related to torrential events is known as "Environment-to-Circulation".…”