Ebook: Scientific Support for the Decision Making in the Security Sector

The main objective of the present conference was to bring together specialists from diverse areas that would make them a part of the think tank of the future security elite of Europe. The main purpose of the organizers was to invite people with mathematical, computer and information sciences specialization who would have in the future the chance to contribute actively to the security topics. The conference was thought to play the role of a round table where specialists with technical background were invited to meet specialists in security so that the stimulating atmosphere could make them think in the perspective of security issues, and eventually attract them to new security projects in the future. Thus the present proceedings reflect the initial state of this dialogue between specialists in security and specialists in mathematics, computer and information sciences. Respectively, the papers included in this volume are naturally subdivided into four parts showing the wide future perspective for synthesis between science and security:

1. Planning for Security

2. Mathematical, Computer and Information Sciences Methods for Security

3. Environmental Securit

4. Dynamic Optimization for Security.

* * *

Today's security environment is characterized by deep uncertainty. Threats are being posed not only by adversary (political) forces but may also come from natural challenges (be it energy, water, ecology or whatever). The types of operations that our civil security and military forces find themselves in today comprise a wide variety of tasks. The success criteria for these operations are a safe/secure environment for local population and stable conditions for state building rather than hit-kill ratio's against adversaries – the criteria are soft and the many actors involved may have divergent if not opposing objectives. And where actors intentionally share common objectives, they come from different cultural and organizational backgrounds, and their systems and modus operandi (doctrine) have loose or no connectivity.

Under these complex and uncertain conditions decision making is a challenging process. It is complex both for the long term planning process (concerned with the capabilities required from the own “security forces”) and for the operational planning process (concerned with how to best operate with the “security forces” available today). How to model today and future worlds and how to bring together the various organizations that have to make contributions to the security now and then is the subject of the papers that are presented in Part 1 of the volume.

Key element in the support to the Decision Making in the security sector is the understanding of models and modeling. K. Niemeyer lays the basis for a Theory of Models, where he conceptualizes the understanding of the environment (perception), the ambitions as of how to (re-)direct that environment (motivation) and then how to manipulate the environment into that desired direction (anticipation). He claims that the systematic formulation of a model theory and further work in this area will provide a considerable improvement of understanding the intelligent behavior of humans and the decision making processes of higher level human organizations including advanced constructs of information technology like simulation models and decision support tools.

From this theory onwards S. Malerud provides a framework to cope with uncertainty and complexity. This starts with how to deduce good and relevant criteria to judge whether operations have been successful, and whether they are on track according to the mission objectives. Decision making related to this kind of operations usually involves more than one decision criterion, and explicit approaches to cope with uncertainty are required. His framework combines elements of Problem structuring methods (PSM) such as Soft Systems Methodology (SSM) – to discuss and agree on which problems to address, elements of scenario planning – to cope with uncertainties, value modeling and multi-criteria analysis – to evaluate and prioritize decision alternatives.

In his paper R. Willems expands on the importance of scenario development and analysis in support of long term defense planning. Not only defense force capabilities must be much more adaptive and flexible than was required before, but also the planning and the decision making about which systems and doctrine to be implemented, requires a much more flexible approach then ever before. He presents a systematic approach and a discussion on how the development of scenarios (to create plausible future worlds) and scenario-analyses (to guide today's ambitions and requirements setting to cope flexibly with such futures) may support this planning process.

One of the key elements to consider in the planning process is technology. M. Rademaker elaborates in his paper on the role of technology and the potential (operational) impact that technology developments may have. With a special interest in the so-called disruptive technologies (technology developments that may change the game in just one or two generations and consequently may have a high impact on current and future capabilities) he proposes a gaming method to assess possible technology effectiveness. Bringing together the strategic planners' world with that of operational planners and of the technology and research & development community he seeks an assessment of the doctrinal potential of technology developments, hence guidance as of where and how to seek priorities in technology investments and to anticipate for new doctrinal concepts and changes.

While the previous papers are in support of the long(er) term planning processes, V. Shalamanov addresses computer assisted exercises to bring together national and international stakeholders. He concentrates on [preparing for] emergency planning where civil and military organizations are co-operating, but his approach is equally valid for intervention, stabilization and reconstruction operations abroad. A prerequisite for success is the understanding of cultural and organizational differences that the various players will bring with them including their respective supporting models and analysis facilities. Shalamanov recommends a proxy organization to provide the basis for an integrated computer assisted exercise environment. In such environment both the development of new concepts, the experimentation with new systems and the implementation in standing organizations can be tested in a close cooperation between administrations, academia and industry. Basis for his recommendations are recent experiences in emergency management exercises that were held in Bulgaria in 2006.

The paper of T. Tagarev and B. Mednikarov presents a framework for capability-based planning and capability development in the security sector, and examines a particular application in the area of maritime sovereignty. The approach is based on centralized planning of the capabilities for protection of maritime sovereignty and agency-based development of these capabilities. The authors propose a process that links objectives, ambitions, planning scenarios, tasks, required capabilities, and planning risks and examine major decision support requirements to capability planning for maritime sovereignty.

In the paper presented by Z. Minchev, scenario development for Computer Assisted eXercises (CAX) is analyzed. The focus is on the greatest challenge – the consideration of terrorist attacks representation, modeling and simulation, where the information uncertainty is too high. An ad hoc created methodology and tool for scenario development is presented and based on the application of Expert Systems, Intuitionistic Fuzzy Sets and random numbers, implementation in a real CAX system. A software, named “I-SCIP” has been already successfully experimented in a real CAX – EU TACOM SEE 2006 (conducted in July 2006 in Bulgaria) as a part of the Joint Training Simulation & Analysis Center in Civil-Military Emergency Planning/Response (JTSAC – CMEP) – Analytical Center. Additionally, this program is planned to be a part of the contribution of Center of Excellence in Operational Analyses (CoE-OA) in NATO RTO MSG-049 project.

The paper of V. Shemaiev and O. Velychko is addressing the issues of scenario simulation for the state military security maintenance on a systematic basis. The method of processes simulation is offered in the framework of the methodology of cognitive simulation. The verification of the offered method is done on a training exercise. The important result of a cognitive simulation is definition of scripts of actions which approach (with other things being equal) a situation to a desirable target condition. In this aspect the solution of a management problem has a practical sense. The further decomposition of the received script of the situation development allows to receive quantitative solutions. The further direction of researches in the given area is simulation interaction of subjects of military security.

* * *

It has become nowadays an obvious truth that any further essential progress in security is impossible without attracting the most recent advances in the area of natural sciences, and especially the newest developments in both pure and applied mathematics, computer and information sciences. One has to mention some of the areas which were represented at the Workshop, as pattern recognition theory, signal and image processing, compression and recognition, computer/geometric design and representation of images, communications and networks, etc.

Correct decision making in the security sector mainly depends on information, received from multiple sources. The paper by K. Alexiev and I. Nikolova contains an analysis of the fusion theory literature in the last years. The main objective is to provide an overview of the latest state-of-the art techniques for data and information fusion and to reveal the topics, on which the scientific society's efforts are nowadays concentrated. The authors also attempt to outline the most important and interesting topics for research in the field in the next few years.

The paper by N. Atreas and C. Karanikas emphasizes the role of approximation theory for fast pattern recognition. The authors build a new fingerprint function on the set of all words of length N written in an alphabet, which then they approximate by an appropriate hash function in order to reduce its computational complexity. In the paper of C. Karanikas and coauthors N. Atreas, A. Bakalakos, P. Polychronidou another aspects of recognition are addressed; namely, they propose a number of methods for fast exact and non-exact string matching, pattern recognition, and grammar detection, which may be applied to strings of symbolic information, e.g., biological data, biometric data, intelligence information, or any other form of information.

The paper Database Structure for Radiation Incidents and for Treatment of Affected People by N. Kirov, J. Djounova and K. Kirov describes an effort to create a national database of radiation incidents in Bulgaria. It is designed for the National Center for Radiobiology and Radiation Protection. The purpose of the database is to store specific description of radiation incidents and to trace the effects on the health of the people affected by the incident.

In the paper of O. Kounchev and H. Render some aspects of representation of geometric data is considered which are important from the point of view of efficiency and design of the data; in particular they consider a generalization of the famous Bernstein-Bezier representations and curves in the case of exponential models. Their methods are efficient especially in the case of data which arise from processes with an exponential background. The results presented form a basis for developing new tools for efficient multidimensional signal representation, manipulation, and control.

O. Kounchev (with H. Render and K. Gumenerov) consider some new methods for forecasting of time series are the framework of the Global systems dynamics. The idea is to attract tools that have found wide use in the financial industry, and may facilitate the forecasting of future critical states of the global system of human-societal dynamics.

The paper presented by S. Nikolov (with coauthors T.D. Dixon, J.J. Lewis, C.N. Canagarajah, D.R. Bull, T. Troscianko and J. Noyes) tackles decision making from the perspective of image processing. The authors study how different multi-modality fused image or video displays affect visual information perception, interpretation, and decision making. The presented results include findings about the way these displays affect rapid decision making with very short display times, target tracking in multi-sensor visible and IR surveillance videos, and multi-sensor image segmentation.

A. Tsankov focused on analysis of computer networks for crises management as complex systems which consist of stationary and mobile management centers, exchanging information via different communication channels. Crises management networks should have ability for sharing its resources differentially between various users and applications according to prior defined criteria or for supporting with high degree the Quality of Services (QoS) of network traffic from different users and applications. It is shown that the proposed method for QoS management of network traffic is an effective solution of the formulated problems. A further goal of the paper is the thorough research of interactions between two stages of network management and formulation of dependencies which will allow effective reconfiguration of QoS parameters in cases when changes in network environment are needed.

The paper by H. Ugail and E. Elyan addresses the issue of the 3D data representations for biometric data; this is an important aspect of security, in particular for public spaces where it is important to have face identification/authentication in a timely fashion. Such data require to be represented using a handful of key facial parameters which can be identified for efficient storage and verification.

* * *

Environmental Security has become a very important issue in nowadays security sector. The relation between the environment and the security of humans has been the object of much research and the subject of many publications in recent decades but it is only recently becoming an important focus of international environmental policy. A recent comprehensive overview of the environmental security field observes that the environment is the most transnational of transnational issues, and its security is an important dimension of peace, national security, and human rights that is just now being understood. Environmental security is central to national security, comprising the dynamics and interconnections among the natural resource base, the social fabric of the state, and the economic engine for local and regional stability.

The papers of R. San José (with coauthors), Z. Zlatev and of K. Georgiev are devoted to air pollution problems. R. San José (J.L. Pérez, J.L. Morant and R.M. González) presented a paper considering an Operational Air Quality Forecasting System for industrial plants, urban and regional areas. The analyzed system is called OPANA (OPerational Atmospheric Numerical model for urban and regional Areas), that is an evolution and result of more than 10 years of research on sophisticated state-of-the-art numerical air quality modeling systems and the implementation on state-of-the-art computer platforms. The OPANA tool is operating on different industrial plants in the south of Madrid area and also over urban areas such as Madrid (Spain), Leicester City (U.K.), Las Palmas de Gran Canaria (Canary Islands, Spain). A complex emission model (EMIMO) developed by the authors' team in UPM (Spain) to provide accurate hourly and high spatial resolution (1 km) pollution emission data at global level is also described. The results show that these tools can provide reliable and robust information to authorities and industrial managers to control – in forecasting mode – the air quality in the surrounding areas of the industrial plant or to forecast the air quality in whole city or region.

The paper of Z. Zlatev presents large-scale air pollution models which can successfully be used to design reliable strategies to control the pollution levels. The author points out that the decision about what kind of measures are to be taken is made after many runs of the available mathematical models with different scenarios. The questions of the reliability of the model results and the choice of a sufficiently large set of scenarios are discussed. The conclusions are illustrated by results obtained in several comprehensive environmental studies.

K. Georgiev presents model studies of sulfate and sulfur dioxide in the atmosphere and comparison with real life data. The model used in this study is a three dimensional global chemistry Transport Model. Comparisons of the model's output results with measurements for sulphate and sulphur dioxide reported by the EMEP (scientifically based and policy driven program for international co-operation to solve transboundary air pollution problems) stations over the territory of Europe are presented.

The papers presented by T. Vardanian, Ch. Hakopian and B. Mnatsakanyan (with K. Aghababyan) deal with the issue of forecasting natural hazardous phenomena – earthquakes, floods, mudslides caused by mountainous rivers' overflows, etc. In the first two papers it is pointed out that today's science can not forecast some of these phenomena well enough, let alone preventing them. Since the causes for some of the considered phenomena are man-made, they argue that record keeping and detailed classification of the various natural disasters can improve our ability to resist them and to better manage the consequences from them. In the third paper several methods for calculating maximal river flows are compared.

* * *

Real life processes develop within their own timeframe, and the ability to react to a dynamically changing situation is important for the decision makers responsible to confront a developing crisis. Dynamical optimization techniques, optimal control among them, were present in some of the talks at the Workshop. In some of the presented models, assuming that a precise forecast for the dynamics of the crisis exists, it is possible to find the solution a priori. In other models the response to a developing crisis is regularly updated on the basis of the regularly incoming information about its dynamics.

In the paper of R. Gabasov and F. Kirillova the authors develop a novel approach to the optimal synthesis problem. Their approach is based on a new parameterization of optimal controls and fast algorithms of correcting optimal open-loop solutions in real time. They investigate the basic problems of classical regulation theory and provide up-to-date optimization methods producing the solution in real time, which employs the advances of computer technologies. The results represent the achievements of the Minsk group on optimal control and applications. In the paper several important examples on control (regulation) problems, stabilization by bounded controls, damping oscillations and on pendulum control are presented.

A. Kalinin and J. Grudo consider the time-optimal problem for a nonlinear singularly perturbed system with a bounded multidimensional control. The authors study an algorithm for the construction of asymptotic approximation to the solution. The algorithm employs solutions to two unperturbed optimal control problems of lesser dimension than the original problem.

Methods of computational graph theory for studying and modelling of the critical infrastructure were presented by E. Kelevedjiev. The theoretical model proposed and an experimental computer interactive implementation are designed for predicting the critical states of a large flow network system. Linear programming technique is used to find solutions in multi-stage in time and multi-criteria optimization of the involved graph flow problem. Due to the ability of interactive re-computing with different sets of input and control data, an expert using the proposed implementation can perform adequate decision making. Real numerical experiments are made for two main cases: modeling of an existing water supplying system for the upper part of the Iskar river basin near Sofia, Bulgaria (with two main scenarios- normal operation and situation of a water shortage) and calculations for the high-voltage transmission network in Bulgaria for which operation behavior is modeled to minimize shortages at some type of critical accidents.

The paper of Ts. Tsachev presents deterministic models of evacuation activities. The models differ in the assumptions on how the authorities control the transport flows. The continuous time models are in the form of optimal control problems. The discrete time model reduces to a linear programming problem. It is pointed out that further research is needed to determine the specific form of some of the functions, assumed known in two of the models. These functions describe how individual decisions, e.g. whether to leave the evacuated area or stay at home, are affected by other people's behavior or by the media.

* * *

The reader may find additional information about the program, the lectures delivered, and the participants, at the website of the Workshop http://www.math.bas.bg/or/NATO_ARW/.

The editors express their special thanks to Ms. Volya Alexandrova and her team from “Mathematica Balkanica” for the excellent job in preparing this volume.

April 5, 2007

The Editors, The Hague and Sofia

During recent years the security environment has become more unpredictable and complex. The armed forces face a broader spectrum of tasks, with an increased focus on low intensity conflicts. The inherent uncertainty and complexity of these operations are indeed a challenge for long-term planning as well as for direct OR support to military operations. The first problem is to define the problems. Problem structuring methods (PSM) such as Soft Systems Methodology (SSM) are a useful approach for discussing and agreeing on which problems to address. The next challenge is how to include uncertainty in the analysis models and evaluations. Use of scenarios is one common approach to explore uncertainties in the future security environment. Uncertainties and inaccuracy related to model parameters and human decision making may be represented by, for instance, probability distributions and fuzzy sets. Decision making related to complex problems usually includes more than one decision criterion. Various methods for multi-criteria decision analysis offer different approaches for modelling preferences and to rank alternatives. We believe that combining different OR methods – soft and hard – are a fruitful approach to deal with complexity and uncertainty. Thus, we have developed a framework of methods that combines elements of SSM, elements of scenario planning, value modelling and a multi-criteria model for evaluation and prioritizing of decision alternatives. This framework is currently applied in a case study addressing the effectiveness of the Norwegian oil-spill preparedness system.

Scenario development for Computer Assisted eXercises (CAX) is a very ambitious area in the Crisis management field. Generally, the reason for this is that CAX allows real world crises and objects translation in the more flexible digital world. The greatest challenge here is the consideration of terrorist attacks representation, modelling and simulation, where the information uncertainty is too high. This paper presents an ad hoc created methodology and tool for scenario development, based on the application of brainstorming, Intuitionistic Fuzzy Sets and random numbers, implementation in a real CAX system.

The model and its application, simulation, have got an ever increasing weight as instrument in many disciplines since computers became available. At the same time new scientific theoretical approaches have been developed in order to classify and systematize this method. In this paper an approach is discussed, which describes the modelling method on the basis of the theory of science. The hypothetical statements as formulated in the paper perhaps serve as stimulus for further work in this area and could contribute to the more common understanding of the phenomena of modelling in the context of intelligent and rational behaviour. A model theory has been presented by Stachowiak /1973/ as a fundamental contribution to the theory of science, which is seen as the basis and starting-point for the contribution in this paper. Most important attributes of models are:

• purpose

• model-original-relation

• contraction.

The intellectual feedback cycle of models is the principal element of the paper which can serve as an extension to the theory of modelling. The distinction between perception models and anticipation models is introduced as well as the hierarchical or fractal approach to describe systems of model applications in the context of intellectual behaviour. Examples of real life systems are discussed and presented.

Technology is for modern defense and security organizations an important force multiplier. In their strategic planning they therefore have to take good notice of short and long-term technology developments. To have an assessment of premature technologies is not often made by military staff. A methodology is presented that is currently under development and tested by NATO RTO Task Group-062 to assess the potential disruptiveness of these premature technologies. The methodology supports both technology push and demand pull and has the potential to help technology experts and military planners to assess the disruptiveness of new technological developments for defense and security.

Paper presents experience in integrating real Command and Control systems with federation of modeling and simulation tools for performing Computer Assisted Exercises (CAX) in Crisis Management Area. Proposed Service Oriented Architecture for CAX support is based on experience of EU TACOM SEE-2006 exercise. Described approach is used to prepare Joint Training Simulation and Analysis Center for interagency concept development and experimentation in the area of civil security. Key element is development of federation of models management mechanism and connectivity with real systems transparently for the participants in the exercise.

Issues of scenario simulation for the military safety maintenance of the state on a systematic basis are considered. The method of processes simulation for maintenance of military safety is offered. The verification of the offered method is done on the training exercise.

Capability-based planning (CBP) is proving its efficiency in managing armed forces. This approach has a considerable potential for implementation in managing the development of the security sector. This paper briefly presents a planning framework and examines a particular application in the area of maritime sovereignty. The approach is based on centralized planning of the capabilities for protection of maritime sovereignty and agency-based development of these capabilities. We propose a process that links objectives, ambitions, planning scenarios, tasks, required capabilities, and planning risks. The distribution of capabilities among security sector organizations accounts for their traditions, experience, and current roles, but focuses on cost effectiveness. In the final section of the paper we examine major decision support requirements to capability planning for maritime sovereignty.

This paper discusses the importance of scenario development and analysis in support of long term defense planning in today's world. Today, the security environment is characterized by deep uncertainty. [Defense] forces may be engaged in a broad range of roles and missions; hence, force capabilities must be much more adaptive and flexible then was required during the cold war.

Consequently, planning what kind of capabilities nations desire/wish to have to support their defense ambitions and the decision making on what systems and doctrine to be fielded, requires a much more flexible approach then ever before.

A systematic approach to guide decision makers and a discussion on how the development of scenarios and scenario-analyses may support this process will be presented as a mechanism as of how to cope with today's challenges.

Correct decision making in the security sector mainly depends on information, received from multiple sources. Often, this information is insufficient, unreliable and contradictive. Multisensor data fusion systems seek to combine information from multiple sources and sensors in order to achieve inferences that cannot be achieved with a single sensor or source or in some sense better than single source information. This paper contains an analysis of the fusion theory literature in the last years. The main objective is to provide an overview of the latest state-of-the art techniques for data and information fusion and to reveal the topics, on which the scientific society efforts are nowadays concentrated. But, the outcome of this analysis would be insufficient and scanty if it concerns only the past years. That is why the authors turn on a risky deal – to forecast the future research in this field. The authors outlined the most important and interesting topics of research in the next few years. The authors hope that the paper could inspire a discussion about the future of data and information fusion systems and their effective application.

Living in a rapidly changing world, where new threads emerge very often, security is at the epicenter of the international dialogue. Security is a broad term covering many activities like forecasting, prevention, building new communication and detection methodologies and many others, therefore many branches of science including mathematics has a lot to contribute to this issue. In this text we wish to highlight the role of approximation theory for fast pattern recognition. In particular, we build a new fingerprint function on the set of all words of length N written in an alphabet. Since a hash function is a way for fast text processing, we approximate our fingerprint function by an appropriate hash function in order to reduce its computational complexity.

Inasmuch as security is dependent on most recent scientific and technological innovations, some of the relevant fundamental challenges are: detecting, recognizing, interpreting, and ultimately assigning meaning to strings of symbolic information, whether it is biological data, biometric data, intelligence information, or any other form of information. The research reviewed in this paper addresses the above problem from a mathematical/information-theoretic point of view, proposing a number of methods for fast exact and non-exact string matching, pattern recognition, and grammar detection. The proposed methods are: the Stern-Brocot Transform, the Cyclic Transform, the Unbalanced Haar Transform and the Haar-Riesz Product.

This article describes an effort to create a national database of radiation incidents in Bulgaria. It is intended for use by National Center for Radiobiology and Radiation Protection (NCRRP). The purpose of the database is to store specific description of the radiation incident and to trace the health effect on people who participated in the incident. We show also data from the recent radiation incident in Sliven stored in our database.

The main purpose of the present paper is to outline a new approach towards an efficient approximation and modelling of data having a dynamical background, in particular processes which are governed by ordinary (stochastic) differential equations.

The first-named author has been supported by the Greek-Bulgarian bilateral project B-Gr17, 2005-2008. The second-named author is partially supported by Grant MTM2006-13000-C03-03 of the D.G.I. of Spain. Both authors acknowledge support within the project “Institutes Partnership” with the Alexander von Humboldt Foundation, Bonn.

We introduce some new methods of Approximation theory to forecasting and we outline some possible application to Global Dynamics security management.

The first-named author has been partially supported by the Greek-Bulgarian bilateral project B-Gr17, 2005-2008, and by a DFG grant at IZKS-University of Bonn. The second-named author is partially supported by Grant MTM2006-13000-C03-03 of the D.G.I. of Spain. Both authors acknowledge support within the project “Institutes Partnership” with the Alexander von Humboldt Foundation, Bonn.

In this paper we study how different multi-modality fused image or video displays affect visual information perception, interpretation, and decision making. The human decision process can be aided significantly by reducing the cognitive load, and more importantly, by displaying task- and decision-relevant information. Results from three studies undertaken at the University of Bristol are presented. These include findings about the way different multi-modality displays affect rapid decision making with very short display times, target tracking in multi-sensor visible and IR surveillance videos, and multi-sensor image segmentation.

Computer networks for crisis management are complex systems which consist of stationary and mobile management centers, exchanging information via different communication channels. Due to the hard prediction, decentralization and broad range of potential crises, ensuring high-speed communications in all circumstances and between all management centers is a crucial task.

Therefore, to ensure efficient work of crisis management staffs and rescue teams, crisis management networks should have ability to differentially share its resources between various users and applications according to prior defined criteria. That means to provide utmost support to the Quality of Services (QoS) of network traffic of users and applications in process of crisis management.

An important issue in many of today's biometric applications is the development of efficient and accurate techniques for representing related 3D data. Such data is often available through the process of digitization of complex geometric objects which are of importance to biometric applications. For example, in the area of 3D face recognition a digital point cloud of data corresponding to a given face is usually provided by a 3D digital scanner. For efficient data storage and for identification/authentication in a timely fashion such data requires to be represented using a few parameters or variables which are meaningful. Here we show how mathematical techniques based on Partial Differential Equations (PDEs) can be utilized to represent complex 3D data where the data can be parameterized in an efficient way. For example, in the case of a 3D face we show how it can be represented using PDEs whereby a handful of key facial parameters can be identified for efficient storage and verification.

The aerosols in the atmosphere have a big and increasing importance for the assessment of air quality and climate forcing. They play a crucial role for global temperature modifications. The study of aerosols in the surface layer is mainly motivated of their impact on human health and possible ecological effects. Unfortunately, there are not enough measurements of most of the dangerous aerosols available for support of the decision makers. Therefore, the model studies in this field are of big importance for our community. Some model studies of three aerosols (sulfate, ammonium and nitrate) will be reported together with the comparisons with some measurements done. The TM5 model, which is a three dimensional global chemistry transport model, were chosen for our study. It allows two-way nested zooming which leads to possibility to run the model on a relatively fine space grid.

As is well known, the primitive man, who went hunting and collected fruit, and, later on, developed farming and livestock breeding, had the only aim - to obtain subsistence means. He could not and did not have sufficient knowledge, means, equipment, etc., to forecast and, moreover, prevent natural disasters like earthquakes, volcanoes, drought, fire, flood, and others. Today, in the years of the culmination of science development and modern, advanced equipment, man would be able to forecast disasters; however, he either cannot forecast them timely, in order to avoid material and human casualties, or he manages to forecast, but is unable to prevent them. Thus, even at its present level, science is not ‘strong’ enough to ensure security. That is why it is necessary that natural disasters be classified, measures of forecasting them and ensuring security be carried out, and a complex approach to the in-depth study and solution of issues retaining in present day science be used.

In mountainous countries with complex physical geographical conditions, such as the territory of the Republic of Armenia, the calculation of the maximal outlets of small rivers has a very important practical meaning, since these rivers bring mudflows and cause serious damage to many branches of economy. The methods used to evaluate the maximal outlets are often not accurate enough. A comparative research of the case study of the Marmarik river with application of a number of methods used in Soviet times and methods used abroad have been carried out. The research shows that in calculating of small river maximal outlets it is more efficient to use regional formulae or maps, since the values calculated with their use are close to the maximal outlets values and determined data for the rivers with long series of measurements.

In this contribution we show the description, implementation and operation of an Operational Air Quality Forecasting System for industrial plants, urban and regional areas. The system is called OPANA (OPerational Atmospheric Numerical model for urban and regional Areas) and is the evolution and result of more than 10 years of research on sophisticated state-of-the-art numerical air quality modeling systems and the implementation on state-of-the-art computer platforms. The system is in operation since 2005 although older versions started to operate at the end of the 90's. The OPANA tool is operating on different industrial plants in the south of Madrid area and also over urban areas such as Madrid (Spain), Leicester City (U.K.), Las Palmas de Gran Canaria (Canary Islands, Spain). OPANA modeling tool is actually operating with MM5 mesoscale meteorological model (PSU/NCAR, US) and the CMAQ (Community Multiscale Air Quality Modeling System) model developed by EPA (US). Additionally, a complex emission model (EMIMO) developed by our group in UPM (Spain) is also described and needed to provide accurate hourly and high spatial resolution (1 km) pollution emission data at global level. Results show that these tools can provide reliable and robust information to authorities and industrial managers to control – in forecasting mode – the air quality in the surrounding areas of the industrial plant or to forecast the air quality in whole city or region.