PhD Research Project

Despite all the advances in Numerical Weather Prediction in the last decades, accurate precipitation forecasts are still a challenging task. One of the most important reasons is the poor specification of the initial state of the atmosphere. In that sense, the assimilation of radar observations is a promising tool to obtain better initial conditions because they provide detailed information of the structure and temporal evolution of convective systems.
Nonetheless, despite all the efforts, the positive impacts of radar Data Assimilation (DA) on precipitation forecasts typically do not persist in time. Since the factors that control this persistence are not fully understood, in this PhD research we will focus in answering the following questions:
  1. Why and when radar DA improvements are short-lived?
  2. What factors in the DA techniques control the level and persistence of the positive impacts on precipitation forecast?

Software projects


The SkewTplus package provides tools to easily read atmospheric sounding data from different formats (University of Wyoming and ARM) and create SkewT sounding plots along with parcel diagnostics (CAPE,CIN,etc.).
This package is an extended version of the SkewT Python package developed by Thomas Chubb. The original SkewT package was extended in a way that the vertical soundings plots are handled by a new class skewT.figure. The new SkewT class extends the base matplolib’s Figure class with an interface similar to matplolib’s pyplot. It also allows to create Skew-T type plots in a simple way. This new class allows a complete control over the Figure properties like multiple plots (normal axis and Skew-T axis). In addition, the thermodynamics module was improved. All the intensive computations were migrated to Cython and parallelized.

GitHub page Documentation

pyVET -- Variational Echo Tracking Algorithm


The pyVET provides a python implementation of the Variational Echo Tracking presented by Laroche and Zawazdki (1995). This algorithm is used by the McGill Algorithm for Prediction by Lagrangian Extrapolation (MAPLE) described in Germann and Zawadzki (2002).