.[1] Nitric acid (HNO 3 ) is the dominant end product of NO x (= NO + NO 2 ) oxidation in the troposphere, and its dry deposition is considered to be a major removal pathway for the atmospheric reactive nitrogen. Here we present both field and laboratory results to demonstrate that HNO 3 deposited on ground and vegetation surfaces may undergo effective photolysis to form HONO and NO x , 1 -2 orders of magnitude faster than in the gas phase and aqueous phase. With this enhanced rate, HNO 3 photolysis on surfaces may significantly impact the chemistry of the overlying atmospheric boundary layer in remote low-NO x regions via the emission of HONO as a radical precursor and the recycling of HNO 3 deposited on ground surfaces back to NO x .

[1] Ambient measurements of HONO and HNO 3 , using a highly sensitive coil scrubbing/ HPLC/visible detection technique, were made at a rural site in southwestern New York State from 26 June to 14 July 1998, along with concurrent measurements of NO x , NO y , O 3 , and various meteorological parameters. The mean (and median) half-hour concentrations of HONO and HNO 3 during this period were 63 (and 56) pptv and 418 (and 339) pptv, respectively. On average, there were two HONO concentration peaks, the first around 0200-0300 LT and the second around 0700-0800 LT, and a minimum at about 2000 LT. The sum of NO x , HONO, and HNO 3 (AENO yi ) was highly correlated with the measured NO y concentration (r 2 = 0.64). The average HONO/NO x ratio was 0.07, while the average AENO yi /NO y ratio was 0.66. During the early morning hours, the photolysis of HONO appeared to be a dominant source of HO x radicals in boundary layer near the ground surface. The average daily radical production from HONO photolysis was 2.3 ppbv, accounting for 24% of the total production from photolyses of HONO, O 3 , and HCHO at the measurement height of 4 m above the ground. Diurnal patterns of HONO and relative humidity suggest that the ground and vegetation surfaces were sinks for HONO in the boundary layer when dew droplets were formed at night and that the subsequent release of the trapped nitrous acid/nitrite from the surfaces acted as a strong HONO source in the morning as the dew droplets evaporated. Our data also suggest that, in order to maintain the observed daytime HONO concentration of $60 pptv, there should be a strong daytime source of 220 pptv hr À1 , which was much greater than the nighttime source of 13 pptv hr À1and the estimated production of $ 40 pptv hr À1 from the gas-phase NO-OH reaction. Photolysis of HNO 3 , which deposits and accumulates on the ground and vegetation surfaces, may contribute significantly to the ''missing'' daytime HONO sources.

[1] Significant production of HONO was observed on glass sample manifold wall surface when exposed to sunlight during the PROPHET 2000 summer measurement intensive. It is hypothesized that the artifact HONO was produced by photolysis of adsorbed nitric acid/nitrate on the manifold wall surfaces followed by the subsequent reaction of produced NO 2 and adsorbed H 2 O on surface. This observation suggests against the use of an unshielded glass manifold as a sampling inlet for the measurement of atmospheric HONO. It may also have some implications in interpreting field NO x data measured using similar glass inlet manifolds, especially from the clean remote environments where NO x is low and HNO 3 is a major fraction of NO y .