A novel approach for the characterization of transport and optical properties of aerosol particles near sources – Part I: Measurement of particle backscatter coefficient maps with a scanning UV lidar

Abstract

The physical and chemical properties of aerosols emitted from a livestock farm were determined by a novel approach which combines high-resolution lidar measurements (0.33 s, 30 m) with simulations of a microphysics–chemistry–transport model. This first of two companion papers describes the scanning lidar measurements of optical particle properties. The lidar system employed laser radiation at a wavelength of 355 nm with a power of 9 W and a pulse repetition rate of 30 Hz. The laser beam was expanded before transmission to the atmosphere so that it became eye-safe at distances >270 m to the lidar. The elastic backscatter signal was detected with a resolution of 0.033 s and 3 m. A receiving telescope with a primary-mirror diameter of 40 cm was used. For this system, we developed a novel method for two-dimensional retrievals of the particle backscatter coefficient. With this set up and approach, the lidar was able to identify the aerosol plume up to a range of ∼2.5 km from the source, a farm in northern Germany, in daytime. The measurements confirm that the optical particle properties of the emission plume vary largely with distance from the source and that the maximum particle backscatter coefficient is found away from the source. Within a close-to-horizontal scan (elevation angle of 2.3°), we found a mean particle backscatter coefficient of 1.5·10⁻⁵ m⁻¹ sr⁻¹ inside the plume between 1.5 and 2.0 km distance from the source. Subtraction of the mean particle backscatter coefficient of the background aerosol present in this case (4.1·10⁻⁶ m⁻¹ sr⁻¹) yields a particle backscatter coefficient of the livestock aerosols of 1.1·10⁻⁵ m⁻¹ sr⁻¹. The limited extend of the plume is revealed with the scanning lidar: Scans with a slightly higher elevation angle of 4.8° did not pick up the plume.

Introduction

The impact of livestock facilities on the environment, especially in areas with dense animal populations, can become significant (e.g., Denby et al., 2008). Livestock farm emissions are partly particulate matter and partly gaseous. The emitted gases, like, e.g., ammonia, are partly aerosol precursors, so that the farm emissions contribute directly and indirectly to the aerosol load in the environment (Lammel et al., 2004, Lammel et al., 2005). The emissions pose a hazard to the health of the inhabitants surrounding the sources and to the farmers working at the facilities (e.g., Cambra-López et al., 2010). Furthermore, aerosol particles from livestock farm emissions contribute to the effect of aerosols on anthropogenic climate change. The uncertainty of this contribution is still large (IPCC, 2007). The experimental quantification of the aerosol emission from livestock facilities is a prerequisite to characterise the potential impact of these particles. Little is known, about their optical properties which, however, is an important aspect for assessing their climate impact. A complicating fact in this context is that livestock aerosol properties are undergoing rapid changes close to their sources that affect the composition of particles far away from the source (Lammel et al., 2005).

So far only a few studies on aerosols originating from livestock buildings exist. Some research efforts have been performed to investigate the aerosol emissions of livestock farming using point measurements around the facilities. Takai et al. (1998) made field surveys of indoor dust concentrations and dust emissions from various farms with cattle and poultry. Lammel et al. (2004) investigated the constituents of the aerosols emitted from a livestock facility in southern Germany with ground measurements collected downwind and upwind of the farm. Some studies exist on the yearly or monthly amount of emission (e.g. Costa and Guarino, 2009). But no study allowed yet for volume integrated mass budgeting because of the limitations the experimental studies based on point measurements have: (1) they cannot adequately provide the spatial and temporal distribution of plumes, (2) they cannot resolve the transport processes under different environmental conditions, (3) emissions from the farms are not continuous and thus measurements at certain times give erroneous results in case of varying emission rates. As a result, our knowledge is still insufficient to describe the spatial extent and the transformation during the transport of livestock aerosols in order to assess their significance.

This problem calls for the application of remote sensing techniques: Lidar can determine the range-resolved distributions of optical properties of atmospheric aerosols with high temporal and spatial resolution up to a range of several kilometers (see, e.g., Weitkamp, 2005 for an overview of the lidar technique). In order to investigate the spatial distribution of aerosol particles, scanning lidars can be applied (e.g., Piironen and Eloranta, 1995, Mayor and Spuler, 2004).

While lidar has been used for a large number of other atmospheric studies, only a few lidar studies on livestock aerosols have been reported to date. Hartung et al. (1998) characterised the aerosol emission from a farm by tilling and reported an increase of the backscatter signal intensity in one profile. The transformations of aerosol properties during transport could not be addressed, partly because of the restrictions for operation of their non-eye-safe lidar. Prueger et al. (2008) investigated the dispersion of particulate matter outside an animal husbandry by means of eddy covariance on towers and supported their findings with scanning elastic lidar measurements that detected the transport of the plume. Both studies did not quantify optical properties of the aerosols.

In this study, we employ the scanning aerosol lidar system of the University of Hohenheim (UHOH) that emits laser pulses with a wavelength of 355 nm with 30 Hz and 9 W average power. The use of the frequency-tripled ND:YAG laser radiation compared to the frequency-doubled or fundamental Nd:YAG laser radiation has two main advantages: (1) eye-safety for higher laser power and (2) the possibility to calibrate the lidar measurements with the strong molecular backscatter signal in the UV and to derive therefore the particle backscatter coefficient at the laser wavelength with the method based on Fernald (1984). Larger laser light intensity is also uncritical for the eye at wavelengths around 1.5 μm. To reach this wavelength range, however, special lasers have to be developed as transmitters for aerosol lidar (Spuler and Mayor, 2007, Petrova-Mayor et al., 2008). Furthermore, the Rayleigh backscatter cross section is a factor of (1500/355)⁴ = 319 weaker at 1.5 μm compared to 355 nm which is a disadvantage for calibrating the data. At the time of this experiment, the UHOH lidar detected only the elastic backscatter signal. Later, rotational Raman channels were added (Radlach et al., 2008) to the system that allow the independent measurement of the particle backscatter coefficient and the particle extinction coefficient (Behrendt et al., 2002) as well as temperature measurements (Behrendt, 2005, Groenemeijer et al., 2009, Radlach, 2009, Behrendt et al., 2010).

To allow for volume integrated mass budgeting, our lidar measurements were combined with in-situ measurements and large-eddy-simulation modelling within a collaborative project of UHOH and Max Planck Institute of Meteorology. The field campaign (hereafter named PLUS1 campaign) was conducted in the framework of the BW-PLUS (Baden-Württemberg Programm Lebensgrundlage Umwelt and ihre Sicherung) programme. This paper, the first of the two parts, provides a description of the UHOH scanning lidar system and discusses the technique to derive the particle backscatter coefficient. In a companion paper (Valdebenito et al., 2011), results obtained with the UHOH scanning lidar and with a high-resolution atmosphere–microphysics–chemistry model are combined.

This paper is organised as follows: Technical details of the UHOH scanning lidar system are presented in Section 2. A description of the lidar data analysis technique is given in Section 3. An overview of the lidar results of the PLUS1 campaign is given in Section 4. The combination of the lidar measurements with results of a large-eddy-simulation aerosol-optical-property model are outlined in Section 5. A summary and an outlook are given in Section 6.