On the relationship between vegetation and climate in tropical and northern Africa

Vegetation cover is a crucial component of the Earth’s climate system but, still, our understanding of the mechanisms governing the reciprocal influence between atmosphere and vegetation is limited. In this study, we investigate the unilateral atmospheric impact on vegetation cover in tropical and northern Africa, differentiated into regions with different circulation regimes and into detailed land-cover classes. In contrast to former studies, climate predictors from a regional climate model are used as input for a multiple regression model. Climate models provide consistent data without gaps at high spatial resolution, a considerably larger set of available climate variables and the perspective to transfer the statistical relationships to future projections, e.g., in the context of anthropogenic climate change. Indeed, robust climate predictors which drive up to 70 % of observed interannual vegetation variability could be extracted from the climate model. Besides precipitation and temperature, global radiation, and relative humidity play an important role. The statistical transfer functions are plausible in terms of the affected regions and land-cover classes and draw a rather complex picture of the atmosphere–vegetation relation in Africa.     

Variation de la stabilité atmosphérique urbaine en fonction de la direction du vent

Abstract

Hourly observations of temperature difference between the 61 ‐ and 6.1‐m levels of the Montréal Botanical Gardens meteorological tower from December 1967 to November 1972 have been stratifïed according to wind direction and season, and have been analysed to find any variation of the urban stability with respect to wind direction. A maximum in stability was found to be associated with easterly and southeasterly winds, and a maximum in instability with westerly and northwesterly winds. The amplitude of this variation is such that the instability for the three hours around noon moves, according to Pasquill nomenclature, from class A (very unstable) to class D (neutral), thus generating notable differences between the coefficients of dispersion used in the Gaussian dispersion and hence between the pollutant concentrations calculated with these various coefficients.