Search

We have combined chemical and micrometeorological measurements to investigate the formation and distribution of HONO throughout a forest canopy. HONO was measured simultaneously at two heights, close to the forest floor and just above canopy. The turbulent exchange between the forest and the atmosphere above was studied using vertical profiles of eddy covariance measurements of wind velocity, sonic temperature, water vapour and CO2. HONO mixing ratios at both heights showed typical diel cycles with low daytime values (~80 ppt) and high nighttime values (up to 500 ppt), but were influenced by various sources and sinks leading to mixing ratio differences (above canopy minus below) of up to +240 ppt at nighttime. In the late afternoon and early night mixing ratios increased at higher rates near the forest floor, indicating a possible ground source. Due to the simultaneous decoupling of the forest from the air layer above the canopy, mixing ratio differences reached about −170 ppt. From the late night until the early morning mixing ratios above the forest were typically higher than close to the forest floor. For some cases, this could be attributed to advection above the forest, which only partly penetrated the canopy. Measured photolysis frequencies above and below the forest canopy differed by a factor of 10–25 resulting in HONO lifetimes of about 10 min above and 100–250 min below the canopy at noontime. However, these differences of the main daytime HONO sink were not evident in the mixing ratio differences, which were close to zero during the morning hours. Effective turbulent exchange due to a complete coupling of the forest to the air layer above the canopy in the morning has offset the differences caused by the daytime photolytic sink and added to the interplay between different HONO production and loss processes.

The input of nitrogen (N) to ecosystems has increased dramatically over the past decades. While total (wet+dry) N deposition has been extensively determined in temperate regions, only very few data sets of N wet deposition exist for tropical ecosystems, and moreover, reliable experimental information about N dry deposition in tropical environments is lacking. In this study we estimate dry and wet deposition of inorganic N for a remote pasture site in the Amazon Basin based on in-situ measurements. The measurements covered the late dry (biomass burning) season, a transition period and the onset of the wet season (clean conditions) (12 September to 14 November 2002) and were a part of the LBA-SMOCC (Large-Scale Biosphere-Atmosphere Experiment in Amazonia – Smoke, Aerosols, Clouds, Rainfall, and Climate) 2002 campaign. Ammonia (NH3), nitric acid (HNO3), nitrous acid (HONO), nitrogen dioxide (NO2), nitric oxide (NO), ozone (O3), aerosol ammonium (NH4+) and aerosol nitrate (NO3-) were measured in real-time, accompanied by simultaneous meteorological measurements. Dry deposition fluxes of NO2 and HNO3 are inferred using the ''big leaf multiple resistance approach'' and particle deposition fluxes are derived using an established empirical parameterization. Bi-directional surface-atmosphere exchange fluxes of NH3 and HONO are estimated by applying a ''canopy compensation point model''. N dry and wet deposition is dominated by NH3 and NH4+, which is largely the consequence of biomass burning during the dry season. The grass surface appeared to have a strong potential for daytime NH3 emission, owing to high canopy compensation points, which are related to high surface temperatures and to direct NH3 emissions from cattle excreta. NO2 also significantly accounted for N dry deposition, whereas HNO3, HONO and N-containing aerosol species were only minor contributors. Ignoring NH3 emission from the vegetation surface, the annual net N deposition rate is estimated to be about −11 kgN ha-1 yr-1. If on the other hand, surface-atmosphere exchange of NH3 is considered to be bi-directional, the annual net N budget at the pasture site is estimated to range from −2.15 to −4.25 kgN ha-1 yr-1.

A detailed description of the WAD/SJAC and the analytical procedures is given elsewhere (Slanina et al., 2001; Wyers et al., 1993). A simplified sketch of the sampling system is shown in Fig. 2. The air flow through the instrument (∼17 l min−1, STP) was generated by a scroll pump outside the wooden house and could be adjusted with a needle valve. The flow was measured continuously (1 min time resolution) with a mass flow meter (Bronkhorst, F-112ACHA-55-V). After the sample air passed the steel elbow and/or pre-impactor and the PFA Teflon tubing, it entered a horizontally aligned WAD that scavenges soluble gaseous species. Trace gases (such as NH3, HNO3, HNO2, HCl and SO2) were collected in a 10−4 M NaHCO3 absorption solution. The liquid input was controlled automatically by an infrared sensor and a switching valve and the liquid was continuously pumped out of the denuder at a flow rate of 0.5– 0.6 ml min−1 by a peristaltic pump. The liquid effluent was collected in a sample reservoir (“gas sample”, see Fig. 2). Artifacts due to evaporation of aerosol phase species in the WAD can be excluded because the characteristic time for formation/evaporation of NH4NO3 is >10 s (Dlugi, 1993), while the mean residence time of the sample air in the WAD is ∼0.002 s (annulus volume: 0.0018 l, flow rate: ∼17 l min−1). After the WAD, the air entered a reservoir where it was mixed with steam of highly purified water. The supersaturation causes aerosol particles to grow rapidly (within 0.1 s) into droplets of at least 2μm diameter. These droplets, containing the dissolved aerosol species were then collected in a cyclone (Khlystov et al., 1995). The cyclone effluent (“aerosol sample”) was transferred into the sample reservoir by a peristaltic pump at a flow rate of 0.5–0.6 ml min−1.