Ph.D. Research Profile: Mikhail Titov
The application of mesoscale atmospheric numerical models to understanding local atmospheric structures and circulations
Research Overview
The fundamental aim of the PhD research is to use limited-area atmospheric modelling tools to improve understanding of local and regional atmospheric circulations and their application to air pollution dispersion. This includes implementation of the most appropriate coupled atmospheric and air pollution models driven by existing and proposed scenarios of chemical composition of air pollutants, to improve understanding of the accumulation and dispersal of air pollution, especially in regions of complex terrain.
The main practical aim of the project is to apply the numerical modelling system to extend knowledge of issues in air pollution, including improved understanding of meteorological controls on spatial and temporal variations in air pollution in complex terrain, knowledge of spatial variability in human exposure to air pollution, improvements in predicting the impact of proposed new sources of pollution, assessment of the effectiveness of air quality management strategies, optimal location of air pollution monitoring sites, and the use of statistical techniques to assess the accuracy of emissions inventories.
A SUMMARY OF PROGRESS SINCE THE START (1.1.05)
- The final version of PhD proposal was completed and submitted in June 2005;
- Literature research dedicated to numerical studies of mesoscale air circulation over complex terrain, analysis of observed PM 10 – PM 2.5 distribution for different parts of urbanized areas (kerb - tunnel, industrial - residential, residential - suburban sites), and chemical analysis (with application of different methods) of particulate material composition has been completed during this year 2005;
- Initial experiments with the state-of-the-art WRF (Weather Research and Forecasting) were completed on a Sun computer workstation for 8 ideal and 2 real cases with 1-2 nested grids. Total time used by WRF to make 12-hour forecast provided evidence for the need to purchase a multi-parallel computer (PC cluster) to properly realise the advantages of using this advanced model;
- Active work with global datasets from the National Centre for Environmental Prediction (NCEP), USA allowed the creation of essential global and regional meteorological, orographic and soil distribution databases for the time period 2004 to 2005. These databases are used to extract input-nudging meteorological and environmental information during model applications;
- Near-surface meteorological and air pollution data from the Coles Place site (St.Albans, Christchurch, winter 2005) was utilized and analysed with application of the statistical and graphical interface. Basic statistical analysis helped to assimilate the data and identify the most interesting pollution episodes for the Christchurch area during winter 2005;
- The above-mentioned data assimilation allowed the start of very complex experiments using the MM5–CAMx (meteorological–chemical) modelling system to investigate the spatial and temporal distribution of particulate material (aerosol pollution) in the form of coarse (PM 10) and fine (PM2.5) size fractions over the Christchurch area for winter time in the emission source groups “Total”, “Domestic” and “Transport” for different time periods during an ordinary winter day;
- The MM5–CAMx numerical model system is used, together with application of statistical regression methods to the modelling system output, for analysis of chemical composition of fine and coarse particulate material. The aim is to identify the basic components of aerosol pollution by applying the “most appropriate scenario” technique. This new experiment will investigate the most appropriate ratios between 6 basic chemical compounds of the aerosol (that comprise more than 90% of the total airborne particle in ambient air) for use in applying the CAMx model to obtain more accurate results of winter pollution time-space distributions. Additional data received from the Environment Canterbury observation site (Coles Place, St. Albans) about gaseous (CO, NO, NO 2, NO x and SO 2) precursors of the primary and secondary aerosol will be used to increase the level of sophistication of the chemical scenario applied to the Christchurch region (during winter time);
- As a side project, evaluation of the state-of-the-art Meteorological Model generation 5 (MM5) over the 2-year modelled time interval (2004-2005) has also helped to assess the potential wind energy resource for a site in the North Island and for New Zealand as a whole. Principal components analysis and cluster analysis were applied to MM5 output data not only to identify the basic factors that influence long- and short-time spatial variability of the near-surface wind (10-100 metres above the surface), but also to ascertain correlations between existing wind farm sites (like the Tararua wind farm, near Wellington) and potential new wind farm sites. The cluster analysis technique provide the possibility to extract basic regional wind patterns for New Zealand;
- Several experiments have also been run for the sites in the Australian-New Zealand region of Antarctica to compare the results of MM5 modelling with observation data obtained at Scott Base for spring and summer 2003;
- Several initial modelling experiments have also been run (as part of joint research with Iranian meteorologists) to reproduce the processes of dust storm development resulting from passage of a warm front over major Iranian cities (like Yazd). This work also aims to develop more accurate physical and dynamical parameterization used in MM5 by using observed data to be collected from Iran over the next 2 years;
- An inter-comparison of the MM5–CAMx and TAPM model systems was undertaken to provide a better understanding of the basic physical processes driving two air circulation-air pollution models.
The results produced in the research conducted to date have been reported in several international conferences:
- Zawar-Reza, P., Titov, M. and Sturman, A. 2005 Dispersion modelling of PM 10 for Christchurch, New Zealand: an intercomparison of performance between Mesoscale Model (MM5) and The Air Pollution Model (TAPM). 17th International Clean Air & Environment Conference, 3 to 6 May, Hobart.
- Titov, M., Zawar-Reza, P. and Sturman, A. 2005 Application of MM5 and CAMx4 to local scale dispersion of particulate matter for the city of Christchurch, New Zealand. 98th Annual Conference and Exhibition of the Air and Waste Management Association, 21–24 June, Minneapolis.
- Titov, M., Sturman, A.P., and Zawar-Reza P. 2005 Application of MM5 to study of air quality in Christchurch, New Zealand – some problems of using MM5 with global analysis data, Joint MM5 & WRF Workshop, Boulder, Colorado, USA, July-August 2005.
- Sturman, A.P., Zawar-Reza, P., Green, M. and Titov, M. 2005 Developments in wind mapping and forecasting of wind energy in New Zealand. Annual Conference of the New Zealand Wind Energy Association, 30-31 August 2005, Wellington.
- Sturman A.P., Zawar-Reza P., Green M., Titov M., Developments in the analysis and forecasting of the wind energy resource in New Zealand, 4 th Annual Wind Energy Association Conference,3-4 November 2005, Melbourne, Australia.
- Zawar-Reza, P., Dehghanpour, A. and Titov, M. (2005) Numerical simulation of a dust storm event using mesoscale models MM5 and TAPM. The first Korea-Iran Joint Conference on Climate, 16-17 November, Mashad, Iran.
- Titov M., Sturman A.P., Numerical prediction of spatial-temporal variability of suspended particulates during winter air pollution episodes over Christchurch, 26 th Annual Conference of the Meteorological Society of New Zealand, 23-25 November 2005, Wellington.
OUTLINE OF THE PROPOSED PROGRAMME FOR THE NEXT SIX MONTHS
- Literature research of numerical studies of air circulation over complex terrain, global model–global analysis system, numerical modelling of PM 10–PM 2.5 ratios for urban sites, chemical analysis and test modelling of participation of carbons, inorganic soluble ions and crustal material in aerosol. Research of contemporary literature concerning environmental numerical modelling;
- To continue the evaluation experiments with WRF using the Sun workstation, installation of graphic user interface for WRF to make the model more accessible for ordinary users. Installation of WRF on a cluster of multi-parallel processors (if we have one) with adjustment of the model to the meso-climate of New Zealand, and especially the South Island, as the next compulsory step after installation. Development of the input–output and plotting–transforming software interface for WRF;
- Global datasets from NCEP used to extend meteorological, orographic and soil distribution databases to include the 2006 year. Purchase, download, utilize and store NCEP Final Analysis (FNL) global database DS090.00 for the time period 1948 – 2003 (55 years) for the purpose of long time numerical modelling (given adequate funding);
- To continue the complex experiments using the MM5–CAMx (meteorological–chemical) modelling system to research spatial and temporal distribution of airborne particulate material (aerosol pollution – both coarse (PM 10) and fine (PM 2.5) size fractions) over the Christchurch area and other urban areas of New Zealand, mostly for the winter season and to assess contributions of “Domestic” and “Vehicle” groups to “Total” pollution concentration levels;
- The MM5 (WRF)–CAMx numerical system will be very actively used for precise analysis (with application of statistical procedures to the modelling system output) of fine and coarse particulate material chemical composition for the basic components of aerosol pollution. The “most appropriate scenario” technique will be found for the Christchurch area for a typical winter day split over 4 time periods. Environment Canterbury gaseous pollution data measured at Coles Place, St. Albans for precursors (CO, NO, NO 2, NO x and SO 2) of the primary and secondary aerosol will be used for development of appropriate chemical scenarios for the city of Christchurch;
- Evaluation of MM5 over a 55-year modelled time interval will be of prime importance to assess the potential long-term wind energy resource for both North and South Islands of New Zealand, as well as for potential and existing wind farm sites. Principal components analysis and cluster analysis will be applied to 55 years of MM5 output meteorology to identify basic time trends and the main factors that influence long- and short-term variability of the wind over New Zealand (and in specific regions), and to assess correlations between different wind farm sites;
- Another secondary modelling topic would be to run a new series of experiments to reproduce more accurately the process of dust storm formation in central Iran and the influence of dust transport on the level of aerosol pollution in Iranian main cities (such as Yazd);
- Inter-comparison of the MM5–CAMx and TAPM models will be continued in relation to wind energy investigations and dust storm events, with the aim of providing a better understanding of the basic physical-dynamical processes driving the two air circulation-air pollution models.
The results obtained in these experiments will be presented at several international conferences (as presentations and in proceedings), possibly including:
- “Mesoscale Processes in Atmosphere, Ocean and Environmental Systems”, New Delhi, India, 14–17 February 2006;
- “Track on Modelling and Simulation Methodologies”, the 20 th European Conference of Modelling and Simulation, Bonn, Germany, 28–31 May 2006;
- Air and Waste Management Association 99 th Annual Conference, New Orleans, USA, 20–23 June 2006;
- International Geographical Union 2006 Conference, Brisbane, Australia, 3–7 July 2006;
and in few journal papers (e.g. Journal of Applied Meteorology, Atmospheric Environment)
OUTLINE OF DIFFICULTIES EXPERIENCED IN RESPECT OF RESOURCES
Two key areas of concern are:
- The absence of a powerful multi-parallel computer cluster very seriously increases the time required to complete modelling tasks and reduces the academic level of the numerical experiments that can be attempted. Sometimes there is just no possibility at all for me to undertake the atmospheric modelling tasks that are very important not only for my thesis research, but also for environmental investigations in New Zealand;
- Conference money restrictions do not allow me to present my results at some very important international conferences, symposiums and workshops, which limit two-way exchange of scientific information in this rapidly developing and important field of research.