[1] As part of the Large-Scale Biosphere-Atmosphere Experiment in Amazonia (LBA)-Cooperative LBA Airborne Regional Experiment (CLAIRE) 2001 campaign, separate day and nighttime aerosol samples were collected in July 2001 at a ground-based site in Amazonia, Brazil, in order to examine the composition and temporal variability of the natural ''background'' aerosol. A combination of analytical techniques was used to characterize the elemental and ionic composition of the aerosol. Major particle types larger than $0.5 mm were identified by electron and light microscopy. Both the coarse and fine aerosol were found to consist primarily of organic matter ($70 and 80% by mass, respectively), with the coarse fraction containing small amounts of soil dust and sea-salt particles and the fine fraction containing some non-sea-salt sulfate. Coarse particulate mass concentrations (CPM % PM 10 À PM 2 ) were found to be highest at night (average = 3.9 ± 1.4 mg m À3 , mean night-to-day ratio = 1.9 ± 0.4), while fine particulate mass concentrations (FPM % PM 2 ) increased during the daytime (average = 2.6 ± 0.8 mg m À3 , mean night-to-day ratio = 0.7 ± 0.1). The nocturnal increase in CPM coincided with an increase in primary biological particles in this size range (predominantly yeasts and other fungal spores), resulting from the trapping of surface-derived forest aerosol under a shallow nocturnal boundary layer and a lake-land breeze effect at the site, although active nocturnal sporulation may have also contributed. Associated with this, we observed elevated nighttime concentrations of biogenic elements and ions (P, S, K, Cu, Zn, NH 4 + ) in the CPM fraction. For the FPM fraction a persistently higher daytime concentration of organic carbon was found, which indicates that photochemical production of secondary organic aerosol from biogenic volatile organic compounds may have made a significant contribution to the fine aerosol. Dust and sea-salt-associated elements/ions in the CPM fraction, and non-sea-salt sulfate in the FPM fraction, showed higher daytime concentrations, most likely due to enhanced convective downward mixing of long-range transported aerosol.

[1] Real-time measurements of ammonia, nitric acid, hydrochloric acid, sulfur dioxide and the water-soluble inorganic aerosol species, ammonium, nitrate, chloride, and sulfate were performed at a pasture site in the Amazon Basin (Rondônia, Brazil). The measurements were made during the late dry season (biomass burning), the transition period, and the onset of the wet season (clean conditions) using a wet-annular denuder (WAD) in combination with a Steam-Jet Aerosol Collector (SJAC). Measurements were conducted from 12 September to 14 November 2002 within the framework of LBA-SMOCC (Large-Scale Biosphere Atmosphere Experiment in Amazonia -Smoke Aerosols, Clouds, Rainfall, and Climate: Aerosols From Biomass Burning Perturb Global and Regional Climate). Real-time data were combined with measurements of sodium, potassium, calcium, magnesium, and low-molecular weight (LMW) polar organic acids determined on 12-, 24-, and 48-hours integrated filter samples. The contribution of inorganic species to the fine particulate mass (D p 2.5 mm) was frequently below 20% by mass, indicating the preponderance of organic matter. The measured concentration products of NH 3 Â HNO 3 and NH 3 Â HCl persistently remained below the theoretical equilibrium dissociation constants of the NH 3 /HNO 3 /NH 4 NO 3 and NH 3 /HCl/NH 4 Cl systems during daytime (RH < 90%). The application of four thermodynamic equilibrium models (EQMs) indicates that the fine mode aerosol anions NO 3 À , Cl À , and SO 4 2À were balanced predominantly by mineral cations (particularly pyrogenic K + ) during daytime. At nighttime (RH > 90%) fine-mode NH 4 NO 3 and NH 4 Cl are predicted to be formed in the aqueous aerosol phase. Probably, Cl À was driven out of the aerosol phase largely by reaction of pyrogenic KCl with HNO 3 and H 2 SO 4 . As shown by an updated version of the equilibrium simplified aerosol model (EQSAM2), which incorporates mineral aerosol species and lumped LMW polar organic acids, daytime aerosol NH 4 + was mainly balanced by organic compounds. -H 2 O aerosol system and its gas phase precursors at a pasture site in the Amazon Basin: How relevant are mineral cations and soluble organic acids?,

Abstract. Within the project EUropean Studies on Trace gases and Atmospheric CHemistry as a contribution toLarge-scale Biosphere-atmosphere experiment in Amazonia (LBA-EUSTACH), we performed tower-based eddy covariance measurements of O 3 flux above an Amazonian primary rain forest at the end of the wet and dry season. Ozone deposition revealed distinct seasonal differences in the magnitude and diel variation. In the wet season, the rain forest was an effective O 3 sink with a mean daytime (midday) maximum deposition velocity of 2.3 cm s −1 , and a corresponding O 3 flux of −11 nmol m −2 s −1 . At the end of the dry season, the ozone mixing ratio was about four times higher (up to maximum values of 80 ppb) than in the wet season, as a consequence of strong regional biomass burning activity. However, the typical maximum daytime deposition flux was very similar to the wet season. This results from a strong limitation of daytime O 3 deposition due to reduced plant stomatal aperture as a response to large values of the specific humidity deficit. As a result, the average midday deposition velocity in the dry burning season was only 0.5 cm s −1 . The large diel ozone variation caused large canopy storage effects that masked the true diel variation of ozone deposition mechanisms in the measured eddy covariance flux, and for which corrections had to be made. In general, stomatal aperture was sufficient to explain the largest part of daytime ozone deposition. However, during nighttime, chemical reaction with nitrogen monoxide (NO) was found to contribute substantially to the O 3 sink in the rain forest canopy. Further contributions Correspondence to: U. Rummel (udo.rummel@dwd.de) were from non-stomatal plant uptake and other processes that could not be clearly identified.Measurements, made simultaneously on a 22 years old cattle pasture enabled the spatially and temporally direct comparison of O 3 dry deposition values from this site with typical vegetation cover of deforested land in southwest Amazonia to the results from the primary rain forest. The mean ozone deposition to the pasture was found to be systematically lower than that to the forest by 30% in the wet and 18% in the dry season.

[1] Measurements of NO-NO 2 -O 3 trace gas exchange were performed for two transition season periods during the La Niña year 1999 (30 April to 17 May, ''wet-dry,'' and 24 September to 27 October, ''dry-wet'') over a cattle pasture in Rondônia. A dynamic chamber system (applied during the dry-wet season) was used to directly measure emission fluxes of nitric oxide (NO) and surface resistances for nitrogen dioxide (NO 2 ) and ozone (O 3 ) deposition. A companion study was simultaneously performed in an oldgrowth forest. In order to determine ecosystem-representative NO 2 and O 3 deposition fluxes for both measurement periods, an inferential method (multiresistance model) was applied to measure ambient NO 2 and O 3 concentrations using observed quantities of turbulent transport. Supplementary measurements included soil NO diffusivity and soil nutrient analysis. The observed NO soil emission fluxes were nine times lower than oldgrowth rain forest emissions under similar soil moisture and temperature conditions and were attributed to the combination of a reduced soil N cycle and lower effective soil NO diffusion at the pasture. Canopy resistances (R c ) of both gases controlled the deposition processes during the day for both measurement periods. Day and night NO 2 canopy resistances were significantly similar (a = 0.05) during the dry-wet period. Ozone canopy resistances revealed significantly higher daytime resistances of 106 s m À1 versus 65 s m À1at night because of plant, soil, and wet skin uptake processes, enhanced by stomatal activity at night and aqueous phase chemistry on vegetative and soil surfaces. The surface of the pasture was a net NO x sink during 1999, removing seven times more NO 2 from the atmosphere than was emitted as NO.

Abstract. Within the project EUropean Studies on Trace gases and Atmospheric CHemistry as a contribution to Large-scale Biosphere–atmosphere experiment in Amazonia (LBA-EUSTACH), we performed tower-based eddy covariance measurements of O3 flux above an Amazonian primary rain forest at the end of the wet and dry seasons. Ozone deposition revealed distinct seasonal differences in the magnitude and diel variation. In the wet season, the rain forest was an effective O3 sink with a mean daytime (midday) maximum deposition velocity of 2.3 cm s−1, and a corresponding O3 flux of –11 nmol m−2 s−1. At the end of the dry season, the ozone mixing ratio was about four times higher (up to maximum values of 80 ppb) than in the wet season, as a consequence of strong regional biomass burning activity. However, the typical maximum daytime deposition flux was very similar to the wet season. This results from a strong limitation of daytime O3 deposition due to reduced plant stomatal aperture as a response to large values of the specific humidity deficit. As a result, the average midday deposition velocity in the dry burning season was only 0.5 cm s−1. The large diel ozone variation caused large canopy storage effects that masked the true diel variation of ozone deposition mechanisms in the measured eddy covariance flux, and for which corrections had to be made. In general, stomatal aperture was sufficient to explain the largest part of daytime ozone deposition. However, during nighttime, chemical reaction with nitrogen monoxide (NO) was found to contribute substantially to the O3 sink in the rain forest canopy. Further contributions were from non-stomatal plant uptake and other processes that could not be clearly identified. Measurements, made simultaneously on a 22 years old cattle pasture enabled the spatially and temporally direct comparison of O3 dry deposition values from this site with typical vegetation cover of deforested land in southwest Amazonia to the results from the primary rain forest. The mean ozone deposition to the pasture was found to be systematically lower than that to the forest by 30% in the wet and 18% in the dry season.