Exploring W. Africa Mesoscale Convective Complexes (MCCs) Spatio-Temporal and Convective Characteristics in Satellite Data using Open Climate Workbench

Mesoscale convective complexes (MCCs) are a category of mesoscale convective systems that last less than 24hours and contribute large amounts to the daily and monthly rainfall totals. In West Africa, ~170 MCCs occur annually during the boreal summer with a contribution of ~75% of the annual wet season rainfall. The spatial expanse of MCCs and the spatio-temporal variability in their convective characteristics make them difficult to characterize even in dense networks of radars and/or surface gauges. As such there exist criteria for identifying MCCs with satellite images – mostly using infrared (IR) data, and methods for determining the amount of rainfall associated with these features from satellite data including TRMM. However, methods are limited to studying a particular MCC or at most a season of MCCs, and are at best semi-automatic in their implementation. The main objective of this research is to create and demonstrate a fully automated and transferable method for determining MCCs and their rainfall characteristics over West Africa between 2001- 2012, from various sources of hourly resolved satellite data, leveraging the existing Python-based software package the Open Climate Workbench (OCW, http://climate.incubator.apache.org). OCW is an Apache Software Foundation (an open-source software project community) that maintains the high-level functionality of the Regional Climate Model Evaluation System (RCMES) but allows the user the ability to extract, edit and add functionality as required for individual projects, and greater flexibility in creating a project workflow. The Regional Climate Model Evaluation System (RCMES, http://rcmes.jpl.nasa.gov), developed collaboratively by NASA's Jet Propulsion Laboratory and the University of California, Los Angeles, is a software tool designed to facilitate seamless regional climate model (RCM) evaluation through easy access to the database (RCMED) inclusive of NASA remote sensing data, e.g. TRMM, MERG, other observations and reanalysis data, and the toolkit (RCMET) that provides capabilities for regridding and calculating commonly used evaluation metrics, e.g. bias, correlation coefficients, and root-mean-square errors. Leveraging these existing packages not only promotes software sustainability and reuse, but also promotes model evaluation and verification with the MCC information.