Research

The scientific work within SAMIRA is being carried out within six work packages, in addition to WP 1000, which is setup to deal with the Scientific coordination and Project management.

WP 2000 AOD retrieval from SEVIRI

The objective of this work is to develop techniques and algorithms for NRT retrieval of the aerosol optical depth (AOD) from the SEVIRI data over the territory of Poland, Romania, Czech Republic and, if feasible, southern Norway and Sweden (up to 63 -64oN). The work will build on the optimal interpolation approach to retrieve AOD from SEVIRI [see Zawadzka and Markowicz, 2014]. The SEVIRI detected radiances (temporal resolution 15 min and spatial resolution 3 km in nadir) give a good opportunity to expand the aerosol data set for Europe. AOD will be derived from a synergistic use of satellite and ground-based observations and/or aerosol transport model output, including MODIS data and/or NAAPS or MACC aerosol transport model output.  The most suitable algorithm (the one-channel versus the two-channel method, or a modification of both) will implemented for the now-casting in the SAMIRA target areas.

WP 3000 AOD-to-PM2.5 conversion

The aim is to retrieve column and near-surface hourly PM2.5 maps for the study area from SEVIRI-AOD and modelled data. WRF-Chem aerosol species will be grouped to reconstruct the aerosol components defined in GADS, for which microphysical parameters are available. T-Matrix computation will be applied to calculate the mass-to-extinction conversion factors for a wide range of aerosol classes in various humidity conditions and mass proportions. The mass-to-extinction conversion factor LUT’s will then be used to calculate the column PM2.5 from the SEVIRI-AOD. A parameterization to extract the near-surface PM2.5 will be introduced, by using WRF-Chem cross-section of PM2.5 concentration, and considering a single aerosol type in the atmospheric column. CALIPSO overpasses (and/or ground-based LIDAR measurements) in combination with collocated sun-photometer observations will be used to estimate the uncertainty due to the parameterization scheme, and to check the validity of the assumption referring to the homogeneity of the aerosol type in the atmospheric column.

WP 4000 Using satellite data and data fusion techniques for air quality mapping

The objective of this work package is to combine the in situ measurement data, chemistry transport model data and Earth Observation data using data fusion (and namely, residual kriging) methodology. Based on this, the air quality maps in different time steps (hourly, daily, and annual) will be produced. In the first stage of the project, the historical data will be examined. Then, the use of the up-to-date data will be tested. The data fusion will be demonstrated for two domains – for the Czech Republic (using CAMx model in) and for the major part of Europe (using WRF-Chem model). Within the work package, the data flow for the near real time data will be established, as well as the web presentation of the maps created by the data fusion methodology.

WP 5000 Downscaling of satellite-based air quality products using a high-resolution CTM

The purpose of this work package is to develop, test and demonstrate techniques for downscaling air quality-related EO products to a spatial resolution that is more in line with what is generally required for studying urban and regional scale air quality. This will be achieved using geo-statistical techniques by making use of the additional information provided by a chemical transport model, e.g. WRF-Chem, operated at high spatial resolution. We aim at demonstrating the methodology for NO2, SO2 or AOD/PM, if possible. We will focus on a set of study sites that include the capitals of the four countries (Bucharest, Warsaw, Prague, and Oslo) and regions of special interest that are traditionally severely affected by air pollution. The latter ones include the region of Silesia/Ostrava along the border of Poland and the Czech Republic, and Gorj and Dolj county in Romania

WP 6000 User interaction and Product Validation

The scope of this work package is to provide a robust and independent validation of all products obtained in SAMIRA. The products will be validated against specific measurement datasets that were not included in the development of the product. This will be done for SEVIRI AOD with sites not included in the retrieval, e.g. non-AERONET sun-photometer sites, and through satellite-satellite inter-comparison with AOD data, for example from the ESA Aerosol_cci project. Retrieved and downscaled AOD/PM, NO2 and SO2 concentrations will be compared with standardized air quality measurements that were not used in the retrieval and/or downscaling process (i.e. MAX-DOAS) and also with European inventories such as EMEP and AirBase. Furthermore, user requirements and feedback will be gathered, centralized and updated in order to ensure that the products developed will met the user’s needs. A User Requirements Document will be compiled and updated regularly from meeting minutes between partners and national stakeholders. User feedback on the products developed will be similarly compiled in a Service Utility Report based on Service Level Agreements signed within SAMIRA.

WP 7000 Pre-operational PM (PM2.5 and PM10) forecast system based on in situ data assimilation (PM10 only)

The purpose of this work package is to further develop the SiAiR prototype into a pre-operational PM10 air quality forecast system capable to produce more accurate PM2.5 and PM10 forecasts over the Czech Republic, Norway, Poland and Romania. The system will assimilate in-situ data (PM10 only) and its open architecture will allow a straightforward extension with satellite data by adding processor(s) for MODIS, Sentinel 3, Sentinel 5P, METEOSAT. Starting from the prototype assimilation system developed in the ESA SiAiR project, the pre-operational PM air quality system will be implemented around the Weather Research and Forecasting (WRF) coupled with Chemistry (WRF-Chem) model. The outcome of WP 7000 will be a pre-operational, tested PM Air Quality Forecasting System using assimilation (PM10 only) of observational data (in-situ), – a topic of high interest for environmental agencies, public health authorities, and local / regional / national authorities.