الفهرس | Only 14 pages are availabe for public view |
Abstract This thesis is a study of the spatial and temporal variations of Sinai seasonal rainfall and an investigation of its teleconnection with El-Niño Southern Oscillations (ENSO) by employing Wavelet Principal Component Analysis (WPCA) and Wavelet Coherence during two main rainy seasons in the period 1981-2015. Rainfall spatial characteristics divide Sinai into two main zones: the northern zone and the central-southern zone. Moreover, the dominant frequency period of Sinai seasonal rainfall ranged between 2 to 8 years. The study focuses on El-Arish and Wadi Watir in Sinai and found a strong teleconnection between their seasonal rainfall and ENSO with a 3-month lead time. Consequently in this study, Artificial Neural Networks is applied to forecast seasonal rainfall over El-Arish and Wadi Watir using ENSO data. After that, forecasted rainfall was compared with the observed ones and provided a good fit where the coefficient of determination (R2) ranged from 0.71 to 0.97 and the root mean square error (RMSE) is between 0.09 and 0.52 during calibration and validation stages within the two studied seasons. Rainfall forecasting helps in taking proper precautions before flash floods to eliminate their impacts. Furthermore, extreme winds can cause flash floods in coastal regions such as El-Arish city. Therefore, this thesis calculated the extreme values of wind speeds over El-Arish. This study found a poor teleconnection between El-Arish seasonal winds and ENSO with a 3-month lead time. Using Gumbel distribution (R2=0.98 & RMSE=0.18), the extreme winds ranged from 16.4 to 22.6 m/sec within return periods from 1 to 100 years. |