Particulate matter (PM) emissions from agricultural operations including cotton harvesting have come under increased regulatory pressure in the US. Regulators in California calculate annual emissions from cotton production operations based on emission factors developed from protocols using ambient PM10 samplers and dispersion modeling. The samplers used in these protocols most likely reported PM10 concentrations that were in error due to the interaction of the sampler performance characteristics and the particle size distribution of the dust emitted from the operation. Agricultural dusts have particle size distributions characterized by mass median diameters in excess of 15 micrometers and geometric standard deviations in the range of 1.6 to 2.3. It has been shown that an ambient PM10 sampler sampling dust with these characteristics could report PM10 concentrations in excess of true PM10 concentrations by as much as 300 – 400%. The dispersion models used to develop the emission factors were state of the art at the time they were used, but later science has provided better methods to use in modeling the downwind impacts of emissions.

This paper will focus on the development of a new system to measure emissions from cotton harvesting operations on a source testing basis. Preliminary design concepts will be discussed and evaluated to determine the feasibility of implementing the design in a new emission factor development protocol.