Thesis Abstract and Dissertation Abstracts
Data Warehousing and Data Mining Applications for Upper Atmospheric Studies
Abstract Category: Science
Course / Degree: PhD
Institution / University: Sri Krishna Devaraya University, India
Published in: 2011
Meteorology is an important area of practice and research about the Atmosphere from the earth to the highest levels in the space. However the approach is vary to finding the things and improve the scope of forecasting and effects of weather changes. We have different techniques to implement on the Meteorological datasets and measure the related applications. In this research we are focusing on radiosonde datasets usage in upper atmospheric studies to find out the hidden facts and trends. In general Radiosonde observations are the primary source of upper-air data and will remain so into the predictable future. Radiosonde observations are used over a broad spectrum of efforts including: Input for computer-based weather prediction models, Local severe storm, aviation, and marine forecasts, Weather and climate change research, Input for air pollution research, and Ground truth for satellite data.
The global atmospheric radiosonde data has millions of soundings from worldwide upper air stations over the period 1997 – present. Each sounding is composed of records of pressure, geopotential height, temperature, dew-point temperature, wind-speed and wind-direction at standard and significant pressure levels proposed by World meteorological Organization (WMO) from the surface to approximately 20-30 km. This data was extracted from the daily occurrences of atmospheric soundings compiled by British Atmospheric Data Centre (BDAC) and is made available for individuals and organizations for atmospheric studies. Most of the individuals or organizations are choosing the specific to their regional or country level datasets with the limited frequencies and pickup the conventional statistical approaches for their studies. Hence few studies have assessed atmospheric practice in general and critical success factors in particular and we have many data level limitations for traditional scientific study.
We have plenty of guidelines for implementation while only a few have been subjected to investigational testing; Facilities currently do not exist to manipulate the large volume of radiosonde data in a traditional scientific approach. Further, there needs to be a better understanding of implementation factors and their effects on atmospheric studies. The atmospheric radiosonde data is typically varying in size which leads to high maintenance cost. Additionally it is expensive to process and we usually get a partial result which is difficult to comprehend. Hence we propose to implement a simplified data warehousing and data mining solution. These techniques are easily process the large volume of datasets using various cleansing, integration, modelling analysis and knowledge management and information discovery by using the OLAP and Data Mining applications for effective scientific decision supporting and weather forecasting.
As part of the exercise in this endeavor we are processing 10 years of standard resolution radiosonde data from worldwide upper-air observation stations over the period 1997 to 2007 and implementing the end to end data warehousing and data mining solutions to measure the weather parameters. As a case study comparing the radiosonde temperature variations in the time periods of 2 years Low-Term, 5 years Mid -Term and 10 years High-Term of analytics for effective atmospheric forecasting and predict the future temperature trends in the Equatorial Regions, Northern Hemispheric and Southern hemispheric zones of the Earth’s Upper Atmosphere. This process will help us to overcome the potential impacts in conventional approaches and increase the productivity of the scientific decision supporting.
Thesis Keywords/Search Tags:
Data Warehousing, Data Mining, Atmospheric Sciences, Radiosonde Data
This Thesis Abstract may be cited as follows:
No user preference. Please use the standard reference methodology.
Submission Details: Thesis Abstract submitted by Venkata Sheshanna Kongara from Singapore on 13-Jan-2011 20:50.
Abstract has been viewed 4559 times (since 7 Mar 2010).
Venkata Sheshanna Kongara Contact Details: Email: kv.sheshu@gmail.com
Disclaimer
Great care has been taken to ensure that this information is correct, however ThesisAbstracts.com cannot accept responsibility for the contents of this Thesis abstract titled "Data Warehousing and Data Mining Applications for Upper Atmospheric Studies". This abstract has been submitted by Venkata Sheshanna Kongara on 13-Jan-2011 20:50. You may report a problem using the contact form.
© Copyright 2003 - 2024 of ThesisAbstracts.com and respective owners.