The lack of paleoecological records from the montane Atlantic Rainforest of coastal Brazil, a hotspot of biological diversity, has been a major obstacle to our understanding of the vegetational changes since the last glacial cycle. We present carbon isotope and pollen records to assess the impact of the glaciation on the native vegetation of the Serra do Mar rainforest in São Paulo, Brazil. From ca. 28,000 to ∼ 22,000 14C yr BP, a subtropical forest with conifer trees is indicative of cool and humid conditions. In agreement carbon isotopic data on soil organic matter suggest the presence of C3 plants and perhaps C4 plants from ∼ 28,000 to ∼ 19,000 14C yr BP. The significant increase in the sedimentation rate and algal spores from ∼ 19,450 to ∼ 19,000 14C yr BP indicates increasing humidity, associated to an erosion process between ∼ 19,000 and ∼ 15,600 14C yr BP. From ∼ 15,600 14C yr BP to present there is a substantial increase in arboreal elements and herbs, indicating more humid and warmer climate. From ∼ 19,000 to ∼ 1000 14C yr BP, δ13C values indicated the predominance of C3 plants. These results are in agreement with studies in speleothems of caves, which suggest humid conditions during the last glacial maximum.
The study place is in the Barreirinhas region, Maranhão State, northeastern Brazil. A vegetation transect of 78 km was studied among four vegetation types: Restinga (coastal vegetation), Cerrado (woody savanna), Cerrad"o (dense woody savanna), and Forest, as well as three forested sites around Lagoa do Caçó, located approximately 10 km of the transect. Soil profiles in this transect were sampled for δ13C analysis, as well as buried charcoal fragments were used for 14C dating. The data interpretation indicated that approximately between 15,000 and ∽9000 14C yr B.P., arboreal vegetation prevailed in the whole transect, probably due to the presence of a humid climate. Approximately between ∽9000 and 4000–3000 14C yr B.P., there was the expansion of the savanna, probably related to the presence of drier climate. From ∽4000–3000 14C yr B.P. to the present, the results indicated an increase in the arboreal density in the area, due to the return to a more humid and probably similar climate to the present. The presence of buried charcoal fragments in several soil depths suggested the occurrence of palaeofires during the Holocene. The vegetation dynamic inferred in this study for northeastern Brazil is in agreement with the results obtained in areas of Amazon region, based on pollen analysis of lake sediments and carbon isotope analysis of soil organic matter (SOM), implying than similar climatic conditions have affected these areas during the late Pleistocene until the present.
Field quantification of ammonia (NH3) volatilization from applied nitrogen (N) is limited because there are no simple inexpensive methods which measure losses without interference from the measuring equipment. Five greenhouse experiments were conducted at the Center for Nuclear Energy in Agriculture, Piracicaba, Brazil in order to estimate NH3 volatilization from applied N using semi‐open static collectors and the 15N balance method. Urea, ammonium sulfate [(NH4)2SO4] and aqua ammonia labeled with 15N were applied at rates from 5 to 100 mg N kg‐1 soil in pots containing 600 g of Oxisols and Ultisols (0–20 cm). Efficiency of the equipment in capturing NH3 was determined by the ratio of volatilized NH3‐N from the fertilizer labeled with 15N retained in the collector to the volatilized quantity obtained by the 15N balance method. Efficiencies varied from less than 1% to about 50% based on the mean rate of NH3‐N from the fertilizer and the mean rate of total NH3‐N (fertilizer‐N + soil‐N) retained in the collector. The functions were adjusted by correlating all collector efficiency values with the corresponding rates of fertilizer NH3‐N and total N retained in the collector regardless of soil and climatic conditions or of fertilizer application method. The models provided estimates of the efficiency of the collector for retaining NH3‐N derived from fertilizer and the total NH3‐N lost from the soil. The NH3‐N losses were calculated based on the efficiency estimates. Losses from fertilizers were not significantly different from losses determined by the 15N method when fertilizer N losses were greater than 0.11 mg day‐1. Total loss estimates did not differ from the losses determined by 15N balance method for collector measured losses greater than 0.26 mg day‐1. These results showed that, if the quantity of NH3‐N retained in the collector and the efficiency of the equipment are known, the collector is a simple and inexpensive device for estimating actual NH3‐N losses under varying soil, climatic, and N fertilizer management conditions, without using expensive 15N‐fertilizer.