Introducing a rainfall compound distribution model based on weather patterns sub-sampling

Abstract

Abstract. This paper presents a new probabilistic model for daily rainfall, using sub-sampling based on meteorological circulation. We classified eight typical but contrasted synoptic situations (weather patterns) for France and surrounding areas, using a "bottom-up" approach, i.e. from the shape of the rain field to the synoptic situations described by geopotential fields. These weather patterns (WP) provide a discriminating variable that is consistent with French climatology, and allows seasonal rainfall records to be split into more homogeneous sub-samples. An exponential POT model is used to fit the distribution of each sub-sample. The distribution of the multi-exponential weather patterns (MEWP) is then defined as the composition, for a given season, of all WP sub-sample marginal distributions, weighted by the relative frequency of occurrence of each WP. The MEWP distribution appears able to fit various shapes of distributions using a simple and robust approach for asymptotic behaviour. It is a new contribution to the ongoing debate on the probabilistic tools used to study the asymptotic behaviour of extreme rainfall from observed records. The paper is illustrated throughout with the example of the Lyon (France) rainfall record for the period 1953–2005.

Drop Size Distribution Climatology in Cévennes-Vivarais Region, France

“…A daily weather classification [46] was provided by EDF/DTG (Electricité de France, Direction Technique Générale) for the studied period (2012-2016). It is derived from geopotential height fields, at 700 and 1000 hPa fields, at 00 h and 24 h, for the area delimited by the grey dashed rectangle in Figure 2.…”

An Application of a Stochastic Semi-Continuous Simulation Method for Flood Frequency Analysis: A Case Study in Slovakia

“…The observed centred three-day rainfall events are replaced with synthetic values. The value for the central rainfall is randomly drawn from values between 1 mm and an extreme value using a Multi-Exponential Weather Pattern (MEWP) distribution introduced by Garavaglia et al (2010). The rainfall adjacent to the central rainfall was estimated using a ratio relative to the central value.…”

“…At the end, the global MEWP distribution is defined by combining all the individual exponential distributions (for further information on combining the distributions, see Garavaglia et al, 2010, or Brigode et al, 2014 using the relative probability of each WP within the season.…”

“…5). The central rainfall is generated using the Multi-Exponential Weather Pattern (MEWP) probabilistic model developed at EDF by Garavaglia et al (2010) and thoroughly tested on datasets from France in Garavaglia et al (2011), and from Norway in Blanchet et al (2015). The model is based on both seasonal and weather pattern (WP) subsampling of rainfall series.…”

Slice sampler algorithm for generalized Pareto distribution

“…In this study, we use MRL plot as a commonly used tool for threshold selection, see [5,16,17,18,19,20]. …”