Get Complete Project Material File(s) Now! »
Assimilation of ground versus lidar observations for PM10 forecasting
In this chapter, a tool for assimilating PM10 concentration measurements on the vertical profile is developed using the optimal interpolation (OI) method. This tool is meant to investigate the potential impact of future ground-based lidar networks on analysis and short-term forecasts of PM10. We make the hypothesis that there exists a relation between PM10 mass concentration and optical properties of aerosols. Such relation was determined for aerosol pollution over Greater Paris by Raut and Chazette .
Raut and Chazette  reported two methods, a theoretical method and an empirical method, to estimate PM10 from extinction coefficients.
The theoretical relationship between PM10 and aerosol extinction coefficients at 355 nm is given as a function of the density of particle ρ, the mean cubic radius r3 and the mean extinction cross-section σext,355:
where the extinction coefficient αext,355,dry can be determined by coupling lidar measurements retrieved from a Rayleigh-Mie lidar and photometer measurements, or by measurements re-trieved from a Raman-N2 lidar. The density of particle ρ is estimated as a constant which depends on the aerosol type. The mean cubic radius r3 can be estimated by the size distri-bution [Raut et al., 2009a]. The mean extinction cross-section σext,355 can be estimated from the model. This method was applied to assess the pollution level of PM10 from mobile lidar observations over Greater Paris by Raut and Chazette  and Royer et al. . As de-tailed in Raut and Chazette , Figure 2.1 shows the route followed by a mobile lidar on 25 May 2005 and the lidar-derived PM10 concentrations. Also, this method was used to determine ash concentrations in Paris, following the eruption of the Icelandic volcano Eyjafjallajökull on 14 April 2010 [Chazette et al., 2012]. Figure 2.2 shows the temporal evolutions of AOT and aerosol mass concentrations derived from a ground-based lidar at 355 nm.
The empirical relationship between dry PM10 and aerosol extinction coefficients at 355 nm is given as follows:
Figure 2.1: Left panel shows the route followed by a small vehicle embarking a lidar on 25 May 2005 on the Paris Peripherique before rush-hour traffic and the hourly averaged PM10 values measured at AIRPARIF network stations. The black line circling Paris is the geographic of Paris city. Right panel corresponds to the temporal evolution of lidar-derived PM10 concentra-tions along the vehicle route. The mean PM10 profile retrieved from lidar signals is shown in white. From Raut and Chazette .
where C0 is the slope of regression analysis between the nephelometer scattering coefficients at 700 nm and the TEOM PM10 measurements, ω0,355 is the albedo at 355 nm, a is the Angström exponent between 355 and 700 nm. Royer et al.  generalised those relationships to wet PM10 concentrations by take into account the effect of humidity.
Presently, several lidar networks could perform regular measurements at continental scales to establish a comprehensive dataset of the aerosol vertical distribution and also to assess vol-canic, dust, fires or pollution events (see Chapter 1). However, no lidar network performed operational measurements before December 2012. Thus, an Observing System Simulation Ex-periment (OSSE) is built over western Europe for one month from 15 July to 15 August 2001 in this chapter (see section 2.4). Although there exists the European Aerosol Research Lidar Network (EARLINET) comprising 28 lidar stations over Europe (see Figure 2.3), their instru-mentation is rather inhomogeneous. Therefore, in OSSE, we defined a network of 12 fictitious ground-based lidar stations covering western Europe, which is based on the lidar sites of exist-ing observation stations, i.e. a subset of EARLINET stations.
Representing the “true” atmosphere by a simulation called “nature run” in OSSE, the ef-ficiency of assimilating the lidar network (with 12 stations) measurements is compared to the efficiency of assimilating surface concentration measurements from the AirBase ground net-work, which includes about 500 stations in western Europe in 2001. It is found that assimi-lating the lidar observations decreases by about 54 % the root mean square error (RMSE) of PM10 concentrations after 12 h of assimilation and during the first forecast day, against 59 % for the assimilation of AirBase measurements. However, the assimilation of lidar observations leads to similar scores as AirBase’s during the second forecast day. The RMSE of the second forecast day is improved on average over the summer month by 57 % by the lidar DA, against 56 % by the AirBase DA. Moreover, the spatial and temporal influence of the assimilation of lidar observations is larger and longer.
A sensitivity study on the number and location of required lidars is also performed to help define an optimal lidar network for PM10 forecasts (section 2.7). Two lidar networks of 12 lidar stations, whose locations are very different, were compared. It is found that spreading out the lidars regularly over Europe can improve the PM10 forecast. Moreover, the efficiencies of assimilating measurements from the lidar network of 26, 76 and 488 (the same as the number of AirBase stations) lidar stations are presented. We find that increasing the number of lidar improves the forecast scores. For example, during the first forecast day, the assimilation of 76 lidar stations measurements leads to a better score (the RMSE decreased by about 65 %) than AirBase’s (the RMSE decreased by about 59 %).
Aerosols have an impact on regional and global climates [Ramanathan et al., 2001; Léon et al., 2002; Sheridan et al., 2002; Intergovernment Panel on Climate Control (IPCC), 2007] as well as on ecological equilibrium [Barker and Tingey, 1992] and human health by penetrating the res-piratory system and leading to respiratory and cardiovascular diseases [Lauwerys et al., 2007; Dockery and Pope, 1996]. Aerosols influence the photo-dissociation of gaseous molecules [Randriamiarisoa et al., 2004] and can thus have a significant impact on photochemical smog [Dickerson et al., 1997]. Thus the accurate prediction of aerosol concentration levels has signi-fication human and economic cost implications.
Various chemistry transport models are used to simulate or predict aerosol concentrations over Europe, e.g. EMEP (European Monitoring and Evaluation Programme) [Simpson et al., 2003], LOTOS (Long Term Ozone Simulation) – EUROS (European Operational Smog) [Schaap et al., 2004], CHIMERE [Hodzic et al., 2006], DEHM (Danish Eulerean Hemispheric Model) [Brandt et al., 2007] and POLYPHEMUS [Sartelet et al., 2007]. However, uncertainties in mod-elling atmospheric components, in particular aerosols are high [Roustan et al., 2010], which leads to significant differences between model simulations and observations [Sartelet et al., 2007]. Data assimilation (DA hereafter) can reduce the uncertainties in input data such as the initial conditions or the boundary conditions by coupling models to observations [Bouttier and Courtier, 2002]. In meteorology, DA has been traditionally applied to improve forecasts [Kalnay, 2003; Lahoz et al., 2010]. In air quality, Zhang et al.  review chemical DA techniques developed to improve regional real-time air quality forecasting model performance for ozone, PM10, and dust. However, applications of DA to PM10 forecasts are still sparse. They include Tombette et al.  and Denby et al.  over Europe and Pagowski et al.  over the United States of America. They demonstrated the feasibility and the usefulness of DA for aerosol forecasts.
As in Tombette et al. , in situ surface measurements are often assimilated, e.g. Air-Base, BDQA (Base de Données de la Qualité de l’Air) or EMEP. However, they do not pro-vide information on vertical profiles. Niu et al.  used both satellite retrieval data and surface observations to assimilate dust for sand and dust storm (SDS) forecasts. They found that information on the vertical profiles of the SDS was needed for the SDS forecasts. Al-though satellite passive remote sensing can provide vertical observations, it is very expensive and data are often limited to low horizontal (e.g. 10 × 10 km2 for the Moderate Resolution Imaging Spectroradiometers (MODIS) [Kaufman et al., 2002]) and temporal resolutions (e.g., approximately twice a day for polar orbiting satellites). Passive instruments can only retrieve column-integrated aerosol concentration [Kaufman et al., 2002]. Spaceborne lidar promises to improve the vertical resolution of aerosol measurements at the global scale [Winker et al., 2003; Berthier et al., 2006; Chazette et al., 2010]. Nevertheless, the spaceborne lidar measurements are only performed along the satellite ground track.
Thanks to the new generation of portable lidar systems developed in the past five years, accurate vertical profiles of aerosols can now be measured [Raut and Chazette, 2007; Chazette et al., 2007]. Such instruments document the mid and lower troposphere by means of aerosol optical properties. Lidar measurements were used in several campaigns, such as ESQUIF (Étude et Simulation de la Qualité de l’air en Île-de-France) [Chazette et al., 2005], MEGAPOLI (Megacities: Emissions, urban, regional and Global Atmospheric POLlution and climate ef-fects, and Integrated tools for assessment and mitigation) summer experiment in July 2009 [Royer et al., 2011] and during the eruption of the Icelandic volcano Eyjafjallajökull on 14 April 2010 [Chazette et al., 2012]. Raut and Chazette  established a reliable relation between the mass concentration and the optical properties of PM10. Because the surface-to-mass ratio for fine particles (PM2.5, particulate matter with a diameter smaller than 2.5 µm) is high, they largely contribute to the measured lidar signal. However, the contribution of coarse particles may not be negligible as shown by Randriamiarisoa et al.  who estimated it to be about 19 %. The relative contribution of PM2.5 may increase with altitude [Chazette et al., 2005], but it is difficult to quantify. Thereby, the PM10 concentrations above urban areas can be retrieved from a ground-based lidar system with an uncertainty of about 25 %.
Because a lidar network with continuous measurements does not yet exist, lidar observa-tions have not yet been used for DA. This work aims to investigate the usefulness of future ground-based lidar network on analyses and short-term forecasts of PM10 and to help future lidar network projects to design lidar networks, e.g. number and locations of lidar stations. Building and maintaining observing systems with new instruments is very costly, especially for ground-based lidars. Therefore, an Observing System Simulation Experiment (OSSE) can be used to effectively test proposed observing strategies before a field experiment takes place, and it can provide valuable information for the design of field experiments [Masutani et al., 2010].
An OSSE is constituted by a nature run, simulated observations, and DA experiments. The nature run is usually a simulation from a high-resolution state-of-the-art model forecast, and is used to create observations and validate DA experiments [Chen et al., 2011]. Many ap-plications use OSSEs, such as for investigating the accuracy of diagnostic heat and moisture budgets [Kuo and Anthes, 1984], studying carbon dioxide measurements from the Orbiting Carbon Observatory using a four-dimensional variational assimilation [Chevallier et al., 2007; Baker et al., 2010], demonstrating the data impact of Doppler wind lidar [Masutani et al., 2010; Tan et al., 2007], defining quantitative trace carbon monoxide measurement requirements for satellite missions [Edwards et al., 2009], comparing the relative capabilities of two geosta-tionary thermal infrared instruments to measure ozone and carbon monoxide [Claeyman et al., 2011], evaluating the contribution of column aerosol optical depth observations from a future imager on a geostationary satellite [Timmermans et al., 2009], and studying the impact of ob-servational strategies in field experiments on weather analysis and short-term forecasts [Chen et al., 2011].
This paper is organised as follows. Section 2.2 provides a description of the DA methodol-ogy used in this study. Section 2.3 describes the experiment setup, i.e. the chemistry transport model used and real observations. An OSSE is built in Sect. 2.4. Results of the OSSE are shown in Sects. 2.5 and 2.6. Sensitivity studies with respect to the number and locations of li-dar stations are conducted in Sect. 2.7. The findings are summarised and discussed in Sect. 2.8.
Choice of DA method
Data assimilation couples model with simulated observations in an OSSE. Different DA al-gorithms may be used, e.g. OI, reduced-rank square root Kalman filter, ensemble Kalman filter (EnKF) and four-dimensional variational assimilation (4D-Var). Wu et al.  have illustrated their limitations and potentials. They found that in the air quality context the OI provides overall strong performances and it is easy to implement. In terms of performance, the reduced-rank square root Kalman filter is quite similar to the EnKF. Denby et al.  compared two different DA techniques, the OI and EnKF, for assimilating PM10 concentration at the European scale. They showed OI can be more effective than the EnKF. Although aerosol assimilation could be performed with 4D-Var [Benedetti and Fisher, 2007], it may be limited to the use of a simplified aerosol model, as it is quite expensive for computation.
Table of contents :
1.1 Atmospheric particulate matter
1.1.1 Health effects
1.1.2 Visibility effects
1.1.3 Climate effects
1.2 Aerosol monitoring
1.2.1 Surface measurements
1.2.2 Satellite remote sensing
1.2.3 Ground-based lidar networks
1.3 Air quality modelling of aerosols
1.3.1 Historical model development
1.3.2 Important processes
1.3.3 Numerical approach
1.3.4 Model performance evaluation
1.4 Data assimilation for aerosol forecasting
1.4.4 Ensemble Kalman filter
1.4.5 Choice of DA method
1.5 Objectives and plan of thesis
2 Assimilation of ground versus lidar observations for PM10 forecasting
2.2 Choice of DA method
2.3 Experimental setup
2.3.2 Input data
2.3.3 Observational data
2.4 Observing system simulation experiment
2.4.1 Nature run
2.4.2 Simulated observations and error modelling
2.4.3 Control run
2.4.4 Parameters of the DA runs
2.5 Choice of the horizontal and vertical correlation lengths
2.6 Comparison between AirBase and 12 lidars network DA
2.7 Sensitivity to the number and position of lidars
3 Modelling and assimilation of lidar signals over Greater Paris
3.2 Experiment setup
3.2.1 POLAIR3D model
3.2.2 Modelling setup and observational data
3.3.1 Modelling of lidar signals
3.3.2 Estimation of zref
3.4 Model evaluation
3.4.1 Model evaluation with Airparif data
3.4.2 Model evaluation with AERONET data
3.5 Comparisons with lidar vertical profiles
3.6 Assimilation test of lidar observations
3.6.1 Basic formulation
3.6.2 Construction of error covariances
3.6.3 DA setup
3.6.4 Results and discussions
4 Assimilation of lidar signals: Application to the Mediterranean basin
4.2 Modelling system
4.3.1 Lidar observations
4.3.2 Observations for validation
4.3.3 Case study
4.4 Assimilation parameter tests
4.4.1 Assimilation period length
4.4.2 Assimilation correlation length
4.4.3 Assimilation altitude range
4.5 Results and discussions
4.5.1 Validation with the BDQA network
4.5.2 Validation with the Barcelona network
4.5.3 Validation with the EMEP-Spain/Portugal network
4.5.4 Validation with the AERONET network
5.2.1 Aerosol modelling
5.2.2 Data assimilation
5.2.3 Lidar observations