Welcome to SessionD3: How can statistics help making sense of data from multiple sources?

 

Date: Thursday 2016-05-19
Time: 14.00 - 17.00
Room: to be confirmed later
Session Coordinator: Dr. R. Benestad

 

This session welcomes contributions related to statistical analysis techniques that make use of such large datasets, innovative methods to combining different sources of regional downscaled climate information/data and approaches dealing with the cascade of uncertainty inherent in these datasets.

Oral: Invited speaker (1 hr, incl discussion) - Richard Chandeler

Group work: how to distill information from lots of data?

• Priming: Information before data, sources of information, and methods
• Background: traditional approaches
• Aim: explain how one can answer a certain question
• Task: lay out a strategy to answer a question

 

Submitted talks (1hr)

• Improving multimodel medium range forecasts over the Greater Horn of Africa using the FSU superensemble - O. Kipkogei, A. Bhardwaj, V. Kumar, L. A. Ogallo, F.J. Opijah, J.N. Mutemi and T.N. Krishnamurti
• Africa regional climate multi-data analysis and management - N. Jibo
• A statistical downscaling model with uncertainty quantification for engineering infrastructure design adaptation - E. Linder, M. Zhao, Y. Liu, J. Jacobs and A. Stoner
• Assessment of the performance of CORDEX-South Asia experiments for monsoonal precipitation over the Himalayan region during present climate: Part I  - S. Ghimire, A. Choudhary and A. P. Dimri

 

Poster introduction (10 min)

Questions/discussion/break-out-group (30-1.5 hr min)

Questions:

1. How can we make use of statistics and the vast volume of data to provide answers to our questions regarding climate change consequences?
2. Why do different projections vary?
   • methods for distilling information from multiple sources/dimensions (e.g. common PCA)
   • statistics to explore uncertainties (adding more information & regression)
   • statistics to gauge the model skill & evaluation
   • extremes
   • statistics to visualise information
   • various ways of using statistics in downscaling 

 

Posters

• Complex algorithms for multiple-station long-term data processing as a first step for optimal probability estimates mapping - I. Osetinskoys
• Inter-variable relations in regional climate model outputs - R. Wilcke
• An R-package Designed for Climate and Weather Data Analysis, Empirical-Statistical Downscaling, and Visualisation - A. Mezghani, R. E. Benestad, and K. Paring