Saharan dust is shown to enter the Central Amazon Basin (CAB) in bursts which accompany major wet season rain systems. Low‐level horizontal convergence feeding these rain systems draws dust from plumes which have crossed the tropical Atlantic under the large‐scale circulation fields. Mass exchange of air between the surface and 4 km over the eastern Amazon basin is calculated using rawinsonde data collected during storm events. Mean concentrations of dust observed by aircraft over the western tropical Atlantic are used to calculate the amount of dust injected into the Basin. Individual storm events inject some 480,000 tons of dust into the north‐eastern Amazon Basin. Storm and dust climatology suggest that the annual importation of dust is in the order of 13 Mtons. In the north‐eastern basin, this may amount to as much as 190 kg ha−1 yr−1. Deposition of trace species, such as phosphate, associated with this dust ranges from 1‐4 kg ha−1 yr−1. Uncertainties in these estimates are not believed to be greater than ± 50% and may be as low as ± 20%. The deposition fluxes from Saharan dust are essentially identical to the CAB wet deposition fluxes from precipitation in the wet season; a result that implies that the major ionic composition of rain water in the CAB wet season may be strongly influenced by inputs of material originating on the African continent nearly 5000 km away. The total amount of Saharan dust calculated to enter the Amazon basin is 1/2 to 1/3 of that estimated to cross 60°W longitude between 10° and 25°N latitude. We conclude that part of the productivity of the Amazon rain forest is dependent upon critical trace elements contained in the soil dust originating in the Sahara/Sahel. This dependence should be reflected by expansions and contractions of the Amazon rain forest in direct relationship to expansions and contractions of the Sahara/Sahel. Turnover rates for nutrient species deposited with Saharan dust in the Amazon Basin suggest a time scale of 500 to 20,000 years. We believe the dependence of one large ecosystem upon another separated by an ocean and coupled by the atmosphere to be fundamentally important to any view of how the global system functions. Any strategy designed to preserve the Amazonian rain forest or any part thereof should equally concern itself with the inter‐relationship between the rain forest, global climate and arid zones well removed from Amazonia.

[1] Measurements of the column-integrated aerosol optical properties in the southern African region were made by Aerosol Robotic Network (AERONET) Sun-sky radiometers at several sites in August-September 2000 as a part of the Southern African Regional Science Initiative (SAFARI) 2000 dry season field campaign. Fine mode biomass burning aerosols dominated in the northern part of the study region (Zambia), which is an active burning region, and other aerosols including fossil fuel burning, industrial, and aeolian coarse mode types also contributed to the aerosol mixture in other regions (South Africa and Mozambique), which were not as strongly dominated by local burning. The large amount of smoke produced in the north lead to a north-south gradient in aerosol optical depth (t a ) in September, with biomass burning aerosol concentrations reduced by dispersion and deposition during transport. Large average diurnal variations of t a (typical diurnal range of 25%) were observed at all sites in Zambia as a result of large diurnal trends in fire counts in that region that peak in midafternoon. However, for all sites located downwind to the south, there was relatively little ($5-10%) average diurnal trend observed as the aerosol transport is not strongly influenced by diurnal cycles. AERONET radiometer retrievals of aerosol single scattering albedo (w 0 ) in Zambia showed relatively constant values as a function of t a for t a440 ranging from 0.4 to $2.5. The wavelength dependence of w 0 varied significantly over the region, with greater decreases for increasing wavelength at smoke-dominated sites than for sites influenced by a significant coarse mode aerosol component. Retrievals of midvisible w 0 based on the fitting of Photosynthetically Active Radiation (PAR; 400-700 nm) flux measurements to modeled fluxes for smoke in Mongu, Zambia yielded an average value of 0.84. This is in close agreement with the estimated average of 0.85 derived from interpolation of the AERONET retrievals made at 440 and 675 nm for August-September 2000. The spectral dependence of w 0 independently retrieved with the AERONET measurements and with diffuse fraction measurements in Mongu, Zambia was similar for both techniques, as a result of both methods retrieving the imaginary index of refraction ($0.030-0.035 on one day) with very little wavelength dependence.

Nitrogen (N) cycling was analyzed in the Kalahari region of southern Africa, where a strong precipitation gradient (from 978 to 230 mm mean annual precipitation) is the main variable affecting vegetation. The region is underlain by a homogeneous soil substrate, the Kalahari sands, and provides the opportunity to analyze climate effects on nutrient cycling. Soil and plant N pools, 15N natural abundance (δ15N), and soil NO emissions were measured to indicate patterns of N cycling along a precipitation gradient. The importance of biogenic N2 fixation associated with vascular plants was estimated with foliar δ15N and the basal area of leguminous plants. Soil and plant N was more 15N enriched in arid than in humid areas, and the relation was steeper in samples collected during wet than during dry years. This indicates a strong effect of annual precipitation variability on N cycling. Soil organic carbon and C/N decreased with aridity, and soil N was influenced by plant functional types. Biogenic N2 fixation associated with vascular plants was more important in humid areas. Nitrogen fixation associated with trees and shrubs was almost absent in arid areas, even though Mimosoideae species dominate. Soil NO emissions increased with temperature and moisture and were therefore estimated to be lower in drier areas. The isotopic pattern observed in the Kalahari (15N enrichment with aridity) agrees with the lower soil organic matter, soil C/N, and N2 fixation found in arid areas. However, the estimated NO emissions would cause an opposite pattern in δ15N, suggesting that other processes, such as internal recycling and ammonia volatilization, may also affect isotopic signatures. This study indicates that spatial, and mainly temporal, variability of precipitation play a key role on N cycling and isotopic signatures in the soil–plant system.

The mean annual rainfall in southern Africa is found to explain over half of the observed variance in the stable nitrogen (N) isotopic signatures of C3 vegetation in southern Africa (r2=0.54, P<0.01). The inverse relationship between the stable N isotopic signatures of foliar samples from C3 vegetation and long‐term southern African rainfall is found on a scale larger than previously observed. A modest relationship is found between stable carbon (C) isotopic signatures of C3 vegetation and rainfall across the region (r2=0.20, P<0.01). No such relationship is found between stable C and N isotopic signatures of C4 vegetation and rainfall. The explanation of the relationship between 15N in C3 vegetation and the mean annual rainfall presented here is that nutrient availability varies inversely with water availability. This suggests that water‐limited systems in southern Africa are more open in terms of nutrient cycling and therefore the resulting natural abundance of foliar 15N in these systems is enriched. The use of this relationship may be of value to those researchers modeling both the dynamics of vegetation and biogeochemistry across this region. The use of the isotopic enrichment in C3 vegetation as a function of rainfall may provide an insight into nutrient cycling across the semi‐arid and arid regions of southern Africa. This finding has implications for the study of global change, especially as it relates to vegetation responses to changing regional rainfall regimes over time.