Fine particulate matter generation under controlled laboratory and wind tunnel conditions

Evaluation of the propensity of soils for PMio (particulate matter < 10 //m in diameter) emission and identification of the causes for variations in PMio production are critical to improve the estimation of airborne PMio emission by wind erosion. Development of equipment and techniques to satisfactorily generate and analyze dust aerosols in the laboratory would greatly facilitate PM10 studies.

Preliminary studies were conducted using the Lubbock dust generation, analysis and sampling system (LDGASS) to determine the appropriate operation and set of conditions which would generate reproducible and reliable results of dust analysis. Results indicated that the effect of the soil sample mass used in the LDGASS on PM10 emission was significant (P < 0.05) and the emission of PMio was proportional to the sample mass. Airflow rate through the system significantly affected PMio production. Particle size distributions of generated dust plumes were also affected by the airflow rate.

Similar experiments were then conducted separately using the LDGASS and a wind tunnel to evaluate the comparability of PMio emissions produced by the kinetic energy applied by the LDGASS with those obtained by reproducing the wind erosion process in a wind tunnel. The effect of soil clay and calcium carbonate (CaCOa) content on PMio emission from eight agricultural soils near Lubbock, Texas, was evaluated. Aerodynamic roughness of a highly erodible agricultural field near Lubbock, Texas, was simulated in the wind tunnel. Sand abrader was fed into the wind tunnel to abrade soil aggregates and to generate PM10 aerosol.

Wind tunnel results showed that PMio emissions significantly (P < 0.05) increased as soil CaCOa content increased and as clay content decreased. Average PM10 concentrations of 69.1 and 127.1 /ug/m^ were observed for low and high CaCOs levels, respectively. Average PMio concentrations for low and high day levels were 118.9 and 91.2 lug/m^, respectively. Accordingly, the soil with high CaCOs and the lowest clay content among soils high in CaCOa produced more PMio (174 fig/m3) than the remainder soils, and the soil with low CaCOs and the highest day content generated less PMio (42.6 /dg/m^) than the other soils. From 96 to 99 % of the total variation in PMio concentrations was accounted for selected soil properties.

Analysis of LDGASS and wind tunnel results and estimates of kinetic energy suggested that the LDGASS imparted much more kinetic energy to a dust source than the wind tunnel. The LDGASS was modified and the new results were in agreement with the results obtained for the wind tunnel. The correlation coefficients (r) between PMio measured in the wind tunnel by a DataRAM and PMio measured by the modified LDGASS by a DataRAM and by a MiniVOL were .62 (P = 0.002) and 0.66 (P = 0.001), respectively. Functional equations to estimate PMio emissions in the wind tunnel from PMio generated by the modified LDGASS produced determination coefficients (r^) of 0.60 and 0.61 for DataRAM and MiniVOL data, respectively.