Скачать 479.15 Kb.
Aircraft used for atmospheric measurements typically fly at speeds ranging from 40 to 200ÊmÊs 1 while the flow speed through filter samplers or aerosol counting/sizing instrumentation is typically less than 3ÊmÊs 1 (Jonsson, Wilson et al. 1995). Furthermore, flight maneuvers lead to variations of ±10 in pitch and ±5 in roll and yaw (Baumgardner and Huebert 1993). These sampling conditions lead to significant uncertainty (positive and negative???) in aircraft aerosol measurements (Baumgardner and Huebert 1993; Foltescu, Selin et al. 1995). Uncertainties arise from the effect of the flow around the airframe on particle size distributions near the probe inlet, unknown efficiencies with which particles of various sizes enter the sampling probe, unknown transport efficiencies through the probe to the particle measurement device, and the increase in the aerosol temperature caused by the rapid deceleration. What about pressure differences within and outside the probe or is this covered within the first item in the list - flow around the airframe???)
Work by Huebert and coworkers (Huebert, Lee et al. 1990) demonstrated significant particulate species losses for marine sampling with isokinetic aircraft probes (i.e., probes for which the air flow speed through the inlet equals the speed of the aircraft). For an aircraft traveling at 100ÊmÊs 1, they found that 50% to 90% of the sampled sodium deposited within the probe immediately downstream of the inlet where the flow was highly turbulent due to the small probe diameter and the high sampling speed. The fractional loss of sulfate within the probe was less than for sodium but was still significant. Because losses by turbulent inertial deposition increase with size (Friedlander and Johnstone 1957; Liu and Agarwal 1974), they speculated that the higher losses likely resulted from larger mean sizes of sodium. Although size distribution data were not available, they felt it likely, based on previous work in a similar environment, that a significant fraction of the sulfate was submicron, which led them to believe that losses of such ÒaccumulationÓ mode particles may have been significant. They concluded that Òwe must consider the possibility that many of the existing literature values for aerosol concentrations above the surface are underestimates of the actual ambient values by factors of 2-10.Ó
Daum and coworkers obtained data which led them to believe that isokinetic probes collect and transport submicron particles with high efficiency, as would be expected based on available theories for aerosol sampling and transport (Belyaev and Levin 1974; Pui, Romay-Novas et al. 1987; Rader and Marple 1988; Tsai and Pui 1990). They found that concentrations of sulfates measured with an aircraft traveling at 50ÊmÊs 1 agreed to within experimental uncertainties with concentrations obtained at a fixed site sampler located on the ground (Peter Daum, personal communication, July 1997).
Fahey and coworkers (Fahey, Kelly et al. 1989) took advantage of the non-idealnonideal sampling characteristics of a sub-isokineticsubisokinetic inlet for NASAÕs ER-2 aircraft to obtain important information on heterogeneous chemical processes in the stratosphere. The air speed into a sub-isokineticsubisokinetic inlet is less than the speed of the aircraft. Therefore, air streamlines rapidly diverge as they approach the inlet. Particles with sufficient inertia cross streamlines and enter the inlet, leading to an enhancement in the concentration of large particles. They were able to confirm in this way that NOy is a major constituent of polar stratospheric cloud particles.
Another factor that can play an important role in aerosol sampling is the temperature increase associated with Òram heating,Ó which occurs when the flow speed is rapidly reduced in the probe. These temperature increases are in the range of 5 to 20C, depending on the aircraft speed, and lead to evaporation of water and other volatile species. Losses of such species can affect size distributions and chemical composition of the sampled aerosol. Wilson and coworkers (Wilson, Stolzenburg et al. 1992) analyzed the effect of such heating on stratospheric sulfate particles as they traveled through the particle sampling inlet of NASAÕs ER-2 aircraft, which travels at 200ÊmÊs 1, and concluded that sizes decrease by as much as 20%. More problematic than the loss of water, which can be estimated with reasonable confidence, is the loss of other labile species such as nitric acid, that are present in unknown quantities. Difficulties in measuring such species have made it difficult to quantify the composition of polar stratospheric cloud particles (Marti and Mauersberger 1993; Molina, Zhang et al. 1993; Worsnop, Fox et al. 1993). Is ram-heating a significant problem at near surface temperatures???
Experts who participated in the 1991 Airborne Aerosol Inlet Workshop in Boulder, CO (Baumgardner and Huebert 1993) concluded that more work on aircraft particle sampling inlets is required. They recommended a three-part program that includes modeling, aircraft studies, and wind tunnel studies. There has since been significant progress on this topic. Seebaugh and Lafleur (Seebaugh and Lafleur 1996) have shown that it is possible to draw a portion of the flow through porous walls of conical inlets, thereby drastically reducing the development of turbulent flow within the probe. Theory suggests that particle deposition should be negligible if turbulence is negligible. Other investigators have used multiple diffusers to reduce flow speeds and shrouded inlets to both reduce speeds and align particles with the sampling inlet so as to permit isokinetic sampling (Leifer, Albert et al. 1997).
В. С. Гребенникова " my world ") Introductory comment Ю. Н. Чередниченко, ст н с. Laboratories of biophysics нии of a General(common)...