1. UKM Learning and Research Repository
2. Master Thesis / Tesis Master
3. Institute of Climate Change / Institut Perubahan Iklim

Title:	The development of leo-neqo trapped particles forecasting system
Authors:	Gusrizal (P63982)
Supervisor:	Wayan Suparta, Assoc. Prof. Dr.
Keywords:	Low Earth Orbit (LEO) Trapped particles Forecasting system Dissertations, Academic -- Malaysia
Issue Date:	26-Feb-2015
Description:	The development of Low Earth Orbit (LEO) – near Equatorial orbit (NEqO) satellite is important for Equatorial countries like Malaysia because it is able to provide better coverage for the country than the polar satellites. However, LEO-NEqO satellites are exposed to greater radiation hazards from trapped particles due to frequent passing through the South Atlantic Anomaly (SAA) than polar satellites. Moreover, trapped particles are a major source for anomalies in LEO satellites. Hence, the main objective of this study was to build a trapped particle forecasting system for LEO-NEqO. This system applied Gaussian process (GP), a hierarchical Bayesian spatio-temporal (HBST) model to perform particles prediction process on a daily basis. Kriging interpolation technique was then employed to draw the forecasted distribution of trapped particles in the modeled area. The system was built by embedding R packages in Matlab platform and results were displayed in a web system. Data source was obtained from the Medium Energy Proton Electron Detector (MEPED) which was part of the Space Environment Monitor (SEM) 2 module installed at The National Oceanic and Atmospheric Administration (NOAA) satellites. Four omni-direction proton particles, i.e.: mepomp6 (> 16 MeV), mepomp7 (> 36 MeV), mepomp8 (> 70 MeV) and mepomp9 (> 140 MeV) were used in this study. The validation of this method was done over two areas, i.e.: the SAA and the NEqO regions. Results showed that the mean absolute percentage error (MAPE) for all particles was around 10 – 20% at the SAA and around 15 – 25% at the NEqO, which indicated that the accuracy of this method was 80 – 90% and 75 – 85 % for the SAA and the NEqO, respectively. This method has also produced a dynamic map that follow the trend of trapped particles distribution in accordance to solar activity. Finally, the HBST-Kriging combination has successfully displayed the results on a web-based. However, the HBST-Kriging system is valid only for low to mid altitudes of up to 800 km. On the contrary, a comparison to the National Aeronautics and Space Administration (NASA) AP-8 model implemented in Space Environment Information System (SPENVIS) showed that the NASA model produced a static map and its flux values vary by the order of two in logarithmic scale with real data. This helped to add a favour to the HBST-Kriging system. The forecasting system is named the UKMtrapcast and has the potential to contribute towards LEO-NEqO as its daily trapped particles forecasting system,Master
Pages:	119
Call Number:	CB158 .G847 2015
Publisher:	UKM, Bangi
Appears in Collections:	Institute of Climate Change / Institut Perubahan Iklim

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.