Abstract:
Dust storms over North Africa represent a critical component of the regional and global climate
system, influencing radiation balance, precipitation processes, ecosystem functioning, and human
health. Despite their importance, recent studies suggest a multidecadal decline in dust activity
across the region, though the driving mechanisms remain debated. This study investigates the
spatiotemporal trends in North African dust storm frequency over the past four decades and
explores the meteorological and climate drivers underlying these changes using a combination of
observational datasets, reanalysis products, and statistical and machine learning methods. Seasonal
and annual dust frequency trends derived from surface visibility records reveal a pronounced
decrease in dust activity, particularly across the Sahel and central Sahara. Using Theil-Sen trend
estimation and Mann-Kendall significance testing, we detect statistically significant declines in
surface wind speeds and increases in vegetation cover (leaf area index) and precipitation,
especially between 10°N and 15°N. Concurrently, the Saharan Heat Low (SHL) shows signs of
intensification and expansion, suggesting possible suppression of dust uplift due to modifications
in regional circulation and thermodynamic stability. Correlation analyses further highlight strong
seasonal associations between dust storm frequency and drivers such as 10-m wind speed,
precipitation, SHL strength, and climate indices (Atlantic Multidecadal Oscillation (AMO), North
Atlantic Oscillation (NAO)). A Self-Organizing Map (SOM) classification of sea level pressure
and 925 hPa wind patterns during dust storm days reveals dominant atmospheric configurations
associated with dust generation, with a clear seasonality and regional preference in SOM node
activation. Additionally, Long Short-Term Memory (LSTM) models using climatic and
environmental predictors demonstrate skill in reconstructing the historical evolution of dust
storms, reinforcing the predictability of dust activity based on climate variability. Overall, the results support the hypothesis that recent declines in North African dust storm activity are linked
to a combination of decreased surface wind stress, increased vegetation and rainfall, and changes
in the SHL and large-scale climate drivers influence. These findings provide an updated
understanding of dust-climate interactions and underscore the importance of land-atmosphere
coupling and climate teleconnections in shaping dust variability.
Description:
A Thesis submitted to the West African Science Service Centre on Climate Change and Adapted Land Use and the Federal University of Technology, Akure, Nigeria, in partial fulfillment of the requirements for the degree of Doctor of Philosophy Degree in West African Climate Systems