WP12: The surface exchange and vertical transport of aerosols

WP12 Objectives

The aim is to standardize vertical exchange measurements of aerosol particles in the planetary boundary layer using a combination of state-of-the-art in-situ and remote-sensing techniques. There are three main objectives:
•    To standardise the retrieval and analysis of in-situ measurements of aerosol particle surface fluxes,
•    To develop, evaluate, and implement suitable methodologies for obtaining the vertical profile of aerosol particle fluxes from ground-based active remote sensing,
•    To combine both in-situ and remote-sensing observations of aerosol particle fluxes, to understand the horizontal scales over which each observation type is representative, and provide recommendations for measuring aerosol fluxes at an ecosystem level within a European infrastructure network.

Description of work and role of partners

The exchange of aerosol particles between the atmosphere and the underlying surface plays a significant role in determining the mass and number distribution of atmospheric aerosol particles. However, it is still unclear how different ecosystems emit particles of different sizes to the atmosphere, and more important, how different surfaces remove particles from the atmosphere via dry deposition. Measurements of particle surface-atmosphere exchange have been made in various ecosystems with different methods and different variables (Pryor et al. 2008), but to date there is no systematization of instrumentation, data analysis and representativeness of the measurements. Only recently, laser spectrometers (HR-ToF-AMS and TD-PTR-MS)to measure the aerosol composition with temporal resolution suitable for flux measurements have become available. These new and innovative methods will further improve our understanding and quantification of aerosol emission sources as well as dry deposition from the atmosphere.

Vertical aerosol transport within the planetary boundary layer (PBL) must also be monitored to fully understand the local and background dispersion mechanisms. The importance of such measurements is twofold: 1) They allow the quantification of the surface emission of aerosol particles relevant for haze/smog formation and atmospheric composition; 2) the vertical transport and mixing of aerosol is integral to understanding the relative impact of local sources when attempting to quantify the link between aerosol and cloud droplet formation. Up until now, aerosol fluxes have only been possible to measure in expensive aircraft campaigns, but recent advances in the capabilities of ground-based remote-sensing techniques such as lidar allow the routine measurements and yield vertical profiles of many atmospheric parameters. Not only can lidar be used to retrieve aerosol size parameters from backscatter profiles (Raman, UV, VIS), the techniques to measure wind speeds with Doppler lidar have recently reached maturity. The combination of backscatter and Doppler lidar can thus be applied to measure vertical aerosol fluxes throughout the PBL (Engelmann et al. 2008). These techniques can be complemented with radar profilers to extend investigations towards aerosol-cloud interaction.

Task 12.1: Surface flux measurements of particles and in-situ method standardisation

Measurements of tower-based particle fluxes will be made at six sites above different land surface types. Building on previous long-term and campaign-based measurements at these sites, ACTRIS-2 will support long-term routine particle flux measurements at Hyytiälä, (UHEL, Finland), Cabauw (KNMI, the Netherlands) and AUC (NERC, UK). In addition, measurements in Kosetice (CHMI, Czech Rebublic), AGORA (UGR, Spain) and Pallas (FMI, Finland) will be started during the ACTRIS-2 project. Since these sites are all also ICOS sites, there is a mutual benefit between the two infrastructures. Different instruments and methods to calculate the particle fluxes are used at each site. These differences cover measurement techniques from direct eddy covariance to disjunct eddy and relaxed eddy accumulation techniques, differences in instrumentation measuring either total particle (by mass or by number) or size-segregated number concentrations, and differences in particle flux post-processing tools including instrument-dependent applied corrections. At present, many corrections routinely applied to other scalar fluxes, such as those monitored in the context of ICOS, are estimated or even ignored in particle flux studies due to limited understanding.

The following will be conducted to tackle the disparity in the different methods and to obtain uncertainty estimates for tower-based particle flux measurements (UHEL, NERC, KNMI, CHMI, FMI, UGR):
a)    Inter-comparison campaign in Hyytiälä, where the instruments from the six sites are run side-by-side, and particle fluxes calculated and compared,
b)    Assessment of the post-processing tools used to calculate the final flux values from comparing multi-month periods of particle fluxes from different measurements,
c)    Improve the poor theoretical understanding of particle fluxes, examine the impact of potential corrections, their correct utility, including those for temperature and water vapour fluctuations.

The particle flux in-situ measurements will be complemented with chemically resolved flux measurements of submicron non-refractory aerosol components. Such observations, using the eddy covariance and disjunct eddy covariance techniques, will be made during campaigns at Hyytiälä (UHEL), AUC (NERC) and Cabauw (KNMI) to provide more detailed information on deposition/emission rates by compounds, and to help provide information on the effect of thermodynamic partitioning/aerosol evaporation as a crucial element in the interpretation of the particle number fluxes.
The tower-based measurements are representative for local scale, but in Hyytiälä (UHEL) and in Cabauw (KNMI) the regional scale fluxes will be calculated based on radon observations in the boundary layer combined with atmospheric transport models such as FLEXPART.

Task 12.2: Boundary-layer measurements of aerosol fluxes by ground-based remote sensing

Vertical profiles of PBL particle fluxes can be determined from co-located Doppler and aerosol lidars by combining the turbulent vertical-wind component derived from the first instrument with aerosol variance and microphysical properties obtained from the second (Engelmann et al. 2008). There have only been a few studies concerning this method. To ascertain whether we can provide vertical profiles of aerosol fluxes routinely, with sufficient accuracy, we will evaluate the following challenges of the technique (FMI, KNMI, TROPOS, UGR):
a)    Are the measurement volumes of the two lidars sufficiently matched?
b)    Is similarity theory (i.e. isentropic assumption) applicable under different atmospheric conditions?
Other issues to be evaluated are:
c)    Retrieval of uncertainties in turbulent and aerosol properties, their mutual correlations and uncertainties in retrieving the impact of aerosol modification during transport within the PBL (hygroscopic growth, thermodynamic gas-aerosol partitioning),
d)    Instrument sensitivity,
e)    Appropriate and relevant measurement scales (temporal and spatial).
Turbulent properties and their associated uncertainties can be derived from measurements of the vertical air motion with commercially available Doppler lidars (O’Connor et al., 2010). Obtaining the relevant aerosol properties from the same atmospheric volume and timescale is more difficult. High-quality aerosol volume and mass concentrations are usually determined by inversion from long (1 hour) time averages of multi-wavelength Raman lidar data. But, the turbulent fluctuations of the microphysical properties should be determined from lidar data with high temporal resolution (< 1minute).To achieve this, existing aerosol lidars (such as PollyXT from TROPOS and FMI) must first be modified by installing fast acquisition channels and deploying to zenith (current pointing angles are typically 5° off zenith).
A series of co-located measurements with aerosol and Doppler lidar will be performed in terms of field campaigns at Pallas (FMI) and in KOS (CHMI) and extended to other sites with appropriate instrumentation where the strong requirement of horizontal homogeneity for flux measurements is met and comparisons with in-situ methods are possible (AGORA, Cabauw, Hyytiälä).
The remote-sensing measurements will cover a vast range of turbulent scales with the vertical profile. What volumes in time and space are appropriate for deriving the fluxes, how does this vary in the vertical and with atmospheric conditions, and how do they relate to measurements at the surface? These questions will be answered using large-eddy simulations (KNMI) to estimate the footprint and spatio-temporal scales of fluxes relevant to each measurement event. The potential for implementing standardized flux measurements at sites operating profiling instruments will be assessed at combined Cloudnet-EARLINET stations. Many sites operate a co-located Doppler lidar and a ceilometer. Can scaling parameters for the turbulent flux within the boundary layer be derived so that only aerosol gradients and the dissipation rate of turbulent kinetic energy are required (gradient method)? How well does this agree with the more sophisticated eddy-covariance techniques? This knowledge will address whether such methods can be implemented in an operational sense, with an obvious benefit for ICOS in enabling the capability of generating mixing-layer heights and other turbulent characteristics.

Task 12.3: Integration of the in-situ and remote-sensing particle fluxes

The aerosol particle fluxes measured both with in-situ and remote-sensing methods will be compared and integrated to obtain ecosystem-scale aerosol particle fluxes. The in-situ measurements are more widely used, in contrast to the remote-sensing technique. Here the objectives and sub-tasks are:
a)    Comparison of the various retrievals from measurement systems providing different moments (number, surface area, mass) of the same size distribution,
b)    Comparison of the in-situ and ground-based remote-sensing techniques at overlapping altitudes,
c)    Calculation of dry deposition using both tower-based and remote-sensing methods,
d)    Defining and providing recommendations for appropriate combinations of the different techniques suitable for understanding the vertical transport and surface-atmosphere exchange of aerosol particles.
Within ACTRIS-2 the applicability and performance of the two techniques will be evaluated in measurement campaigns at five sites, which are Hyytiälä (UHEL), Pallas (FMI), Cabauw (KNMI), AGORA (UGR) and KOS (CHMI). These sites cover different canopies allowing us to examine also the impact of different surfaces to the performance of the methods and their mutual behavior.