A Modified Method to Correct the Measurement Error of TSI3563 Integrating Nephelometer

Nan Ma,Xiuji Zhou,Peng Yan,Chunsheng Zhao

DOI: https://doi.org/10.11898/1001-7313.20150102

2015-01-01

Abstract:TSI3563 integrating nephelometer is designed for high-quality in-situ aerosol scattering measurement, which is widely used all over the world.However,the scattering coefficient measured by TSI3563 nephe-lometer contain two systematic errors:The truncation error (i.e.,the geometrical blockage of near-for-ward/backward-scattered light)and the non-Lambertian error (i.e.,the slightly non-cosine weighted in-tensity distribution of illumination light provided by the opal glass diffusor).These errors need to be cor-rected since they can typically cause a bias of about 10% in the measured scattering coefficient.Based on the aerosol properties measured in North China Plain during Hachi (Haze in China)Project,the correction factor is calculated with a traditional method and the Mie model (taken as reference)which requires aerosol number size distribution and refractive index as input.The traditional correction method is widely used all over the world since it requires only data from nephelometer itself.However,results show the traditional method cannot provide a good estimation of the correction factor.Due to the high concentration of submi-cron aerosol in PM10 ,aerosol number size distributions measured in North China Plain are different from those assumed in the traditional method.The traditional correction method is therefore inadequate for high-aerosol pollution region like North China Plain.It is found that the correction factor is sensitive on the volume fraction of supermicron aerosol in PM10 .Higher volume fractions would lead to higher correc-tion factors. <br> A modified correction method is proposed.The volume fraction of supermicron aerosol which can be obtained from PM1 and PM10 measurement is used in the new method.For different volume fractions,dif-ferent parameters are chosen for the calculation of correction factors.Testing with aerosol properties measured in North China Plain,the modified method provided a good estimation of the correction factors. 80% of correction factors calculated with the modified method are with a bias less than 1% and 100% are with a bias less than 3%.Compared with the traditional method,a distinct improvement is found in cor-rection results.It suggests that to estimate the correction factor for TSI3563 nephelometer measurement, the Mie model should be the first choice if a real-time measurement of aerosol number size distribution is a-vailable.Otherwise,the modified method proposed should be used if a real-time PM1 and PM10 measure-ment is available.Without those parallel measurements,the traditional method can be the last choice to es-timate the correction factor.