Long sequence time series analysis of Moroccan ecosystem dynamics

مرکز و کتابخانه مطالعات اسلامی به زبان های اروپایی

منو

" Long sequence time series analysis of Moroccan ecosystem dynamics "
Abdelkrim Marzouk J. R. Eastman

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	52699
Doc. No	:	TL22653
Call number	:	‭3194220‬
Main Entry	:	Abdelkrim Marzouk
Title & Author	:	Long sequence time series analysis of Moroccan ecosystem dynamics\ Abdelkrim Marzouk
College	:	Clark University
Date	:	2005
Degree	:	Ph.D.
student score	:	2005
Page No	:	139
Abstract	:	The principal technique used in this investigation was a Principal Components-based time series analysis (TSA). Using a set of 240 monthly composite multi-band AVHRR images for Morocco, a 20-year time series of monthly NDVI images was analyzed to understand the environmental dynamics in arid and semi-arid regions using different vegetation indices. In addition, a second analysis was conducted using a vegetation index specifically designed for use in areas of sparse vegetation--the Modified Soil-Adjusted Vegetation Index (MSAVI). The results showed that the NDVI archive produced a series of readily-interpreted components, including variations in biomass relating to geographic context (Component 1) and seasonality (Component 2). Of particular interest was one that was found to relate to the North Atlantic Oscillation (NAO) and another that showed a linear trend of increasing biomass levels over the entire series. The NAO index has a great impact on Moroccan agriculture. Success to model its occurrence results in a prediction of good and bad years in yield. Such process will help to plan for agriculture planning. In this research I have mapped spatially the occurrence of the NAO index and have found the concordance between component five and the occurrence of NAO. One problematic element in the NDVI analysis was its sensitivity to orbital decay. As the time of equatorial crossing decays over time, the NDVI measure is affected. This was picked up in several components and led to uncertainty in the interpretation of several. As a result, an initial exploratory analysis was undertaken of the MODIS instrument aboard the TERRA and AQUA satellites. Although the archive is currently very short, this product has a higher spatial resolution and is very well calibrated. An analysis of a three year sequence for Morocco clearly demonstrated the superiority of the product. However, it also corroborated the main interpretations of the AVHRR NDVI sequence. Surprisingly, the MSAVI analysis proved to be quite inconsistent with that from the NDVI analyses using AVHRR and MODIS data. Its greater sensitivity in areas of sparse vegetation was substantiated. However, the components were unusual and very difficult to interpret.
Subject	:	Applied sciences; Earth sciences; Arid; Ecosystem; Moroccan; Remote sensing; Geography; 0799:Remote sensing; 0368:Geography
Added Entry	:	J. R. Eastman
Added Entry	:	Clark University