Nutrient input from the atmosphere into Brazilian savanna oxisols under corn

Above-canopy precipitation (ACP) reaches the soil under corn (Zea may's L.) as throughfall (TF) and stemflow (SF). We hypothesized that this results i n heterogeneous nutrient inputs to soil and that the corn canopy modifies t he chemical composition of ACP because of dry deposition and crop leaching. Therefore, we examined the quantity and quality of ACP, TF, and SF in corn fields on Oxisols of the Brazilian savanna during two vegetation periods a nd estimated the dry deposition rates with a Na tracer technique. Throughfa ll accounted for 70 to 75% of the ACP. Evaporation losses of intercepted wa ter were close to zero. In TF and SF, volume-weighted mean (VWM) concentrat ions in mug L-1 of Ca (TF: 581/SF: 565), Cl- (826/1330), Cu (2.4/1.9), K (2 110/5710), Mg (199/403), Mn (1.3/1.2), N (1550/770), Na (493/586), S (376/1 84), and Zn (9.3/24) were generally higher than in ACP (Ca: 117, Cl-: 265, Cu: 1.0, K: 152, Mg: 44, Mn: 0.48, N: 199, Na: 246, S: 62, and Zn: 5.6 mug L-1). The average contributions of SF to the total nutrient input were simi lar to its contribution to the total water input (27%) for Ca (27%), Mn (28 %), and Na (30%), higher for Cl- (37%), Mg (43%), Zn (50%), and K (51%), an d lower for Cu (23%), N (16%), and S (15%). This indicated leaching of Cl-, Mg, Zn, and K from the stems and uptake of Cu, N, and S by the plants. Dry deposition contributed 50 to 55% to the average total deposition during th e rainy season (October-April) of 3520 g Ca (Ca), 8050 g Cl-, 35 g Cu, 4690 g K, 1370 g Mg, 15 g Mn, 7460 g N, 7590 g Na, 2090 g S, and 199 g Zn ha(-1 ). Leaching from the corn canopy was most pronounced for K (41 kg ha(-1)) > Ca (5.1) > S (2.6) > Mg (2.4). On average, 76% of the 12 kg N ha(-1) that reached the soil was organic, indicating that N was also leached. In contra st, there was almost no leaching of Mn (3.5 g ha(-1)) and, on average, even net uptake of 1.3 g Cu and 0.1 g Zn ha(-1). The total deposition, on avera ge, accounted for 3.9% (K) to 46% (Zn) of the nutrient demand of the corn p lants. However it represented <6.5% of the nutrient input with fertilizers. Our results demonstrate that the spatial distribution of the nutrient inpu t to soils under corn is element-specific and that: deposition contributes to the micronutrient nutrition of corn grown on nutrient-poor Oxisols.