| Summary: | The University Archives has determined that this item is of continuing value to OSU's history. It is determined that much of the variation in the kinds, concentration, and size distribution of microparticles (larger than 0.6 micron diameter) in snow accumulation on the Antarctic and Greenland ice sheets is the result of seasonal and regional differences in temperature and precipitation rate. The observed variations in microparticle deposition over intervals of several years and several centuries may be due to small- and large-scale climatic changes, respectively. Microparticle deposition on polar ice sheets is almost entirely accomplished through the process of precipitation. Some of the deposited particles are active as nuclei in the formation of snow, and others are scavenged (swept) from the atmosphere by falling snow. The sizes and kinds of particles able to form precipitation by sublimation nucleation depend on the temperature. Smaller particles and particles with less suitable nucleating surface characteristics serve as nuclei at lower temperatures. Intervals of high precipitation rate deplete the supply of atmospheric dust, and snow that forms during a time when dust is depleted forms on smaller and less efficient nuclei and scavenges fewer dust particles than snow that forms at times of higher atmospheric dust concentrations. The effects of temperature on nucleation and of atmospheric dust concentration (controlled by precipitation rate) on scavenging lead to seasonal patterns of microparticle deposition. Spring accumulation contains relatively high dust concentrations and greater relative abundances of particles of about 1 to 2 microns diameter compared to the particle distributions in winter accumulation. This is because the lower limit of sublimation nucleus size increases and moves into our range of measurement (0.6 to 3 microns diameter) with rising temperature in the spring. The population of particles deposited in the spring is composed of a "background" population of modal diameter less than 0.6 micron and a "critical" population of modal diameter within our range of measurement. The "background" population, made up of scavenged particles and perhaps condensation-freezing nuclei, has a size distribution very nearly the same as the distribution of dust in the troposphere in the temperate regions. The "critical" population is distributed similarly and is composed of sublimation nuclei. The mixture of the critical and background population is a nearly lognormal distribution (a convex-upward curve on a log-log plot of cumulative number versus size) when the critical population is small diameters. When the critical population moves to larger diameters, the composite distribution departs from log-normality. Summer accumulation contains lower concentrations of particles as a result of depletion of atmospheric dust in the spring, and the sublimation nucleus population is very large because temperatures are high. The composite population of deposited particles plots as a convex-downward curve on log-log scales as a result of the large size of the sub-sublimation nuclei. The background population has a modal size smaller than 0.6 micron. Fall accumulation is characterized by higher measured concentrations of particles as a result of the recharging of the atmospheric dust reservoir and a decrease in modal diameter of sublimation nuclei within our range of measurement. In the winter, the lower limit of sublimation nucleus size is smaller than 0.6 micron, and the measured size distribution of deposited dust is very similar to the distribution of dust in the atmosphere. The effect of mean annual temperature on mean sublimation nucleus size is reflected in the size distribution of particles from different sites at different elevations in Antarctica. The size distribution becomes more log-normal (convex-upward) at higher elevations as the average modal size of sublimation nuclei decreases with decreasing annual temperature. Thus, the size distribution of dust in snow is an index to mean annual temperature. The relationship between mean annual temperature and size distribution has been used in interpreting particle distributions in deep ice samples. The systematic decrease in log-normality of the distributions in progressively younger samples indicates a warming rate of about 1.7°C century over the past 14 centuries in Antarctica. National Science Foundation Grant GA 530.
|
|---|