Trend Analysis of Radiosonde Atmospheric Temperature Products for Assessing Climate (RATPAC)

Author Information

David M. Armas, ERAUFollow

Authors' Class Standing

David Armas, Senior

Lead Presenter's Name

David Armas

Faculty Mentor Name

Chris Herbster

Abstract

In this paper a statistical trend analysis of upper atmospheric data will be outlined. The data set was created by Free, et al. (2005) to remove temporal inhomogeneities caused by changes in instrumentation for the purpose of studying long term climate variation. This analysis will be performed by investigating annual global, hemispheric and latitudinal mean temperature anomalies at various pressure levels over the latter half of the twentieth century for statistically significant trends. In order to accomplish this, the author will conduct an in depth exploratory data analysis (EDA) for each region and level over which the anomalies were averaged. Trends that appear to be evident in the EDA will be studied by inferential analysis to determine their statistical significance. While making no assertion as to why they are occurring the author will seek to verify the presence and direction of any existent trends.

Location

Flight Deck

Start Date

9-4-2014 10:00 AM

Share

COinS
 
Apr 9th, 10:00 AM

Trend Analysis of Radiosonde Atmospheric Temperature Products for Assessing Climate (RATPAC)

Flight Deck

In this paper a statistical trend analysis of upper atmospheric data will be outlined. The data set was created by Free, et al. (2005) to remove temporal inhomogeneities caused by changes in instrumentation for the purpose of studying long term climate variation. This analysis will be performed by investigating annual global, hemispheric and latitudinal mean temperature anomalies at various pressure levels over the latter half of the twentieth century for statistically significant trends. In order to accomplish this, the author will conduct an in depth exploratory data analysis (EDA) for each region and level over which the anomalies were averaged. Trends that appear to be evident in the EDA will be studied by inferential analysis to determine their statistical significance. While making no assertion as to why they are occurring the author will seek to verify the presence and direction of any existent trends.