ECOSIM
An Urban Environmental Management Information System
Dr.Kurt Fedra
ECOSIM EN1006
Environmental Software & Services GmbH
PO Box 100 Gumpoldskirchen A-2352 Austria
E-mail: kurt@ess.co.at, http://www.ess.co.at/ECOSIM
ABSTRACT
ECOSIM is an environmental management information system that integrates monitoring and simulation modeling for environmental decision support in urban areas. The project develops and demonstrates an environmental decision support system based on a modular and distributed client-server architecture using wide-area network technology and the Internet to connect clients, monitoring networks, and high-performance model servers. Traffic generated air pollution including photochemical smog, coastal water quality, and groundwater are the initial application domains, analysed by a set of state-of-the-art simulation models with a multi-media user interface. ECOSIM involves participants from Austria, Germany, Greece, Italy, Poland, and the UK, and works with the cities of Berlin, Athens, and Gdansk as validation sites and initial end users.
The successful use of ECOSIM will result in more effective environmental planning within urban areas which, for the citizen, should mean a cleaner environment. ECOSIM will also provide more information on pollution levels of general interest to the public, but can also be used as the basis for a more general environemntal information system, distributing information over the Internet, to the general public.
ECOSIM is designed to support the implementation of European environmental policy, guidelines and standards such as 96/62/EC; the Directive on Air Quality Assessment and Management, (the Air Quality Framework Directive). The general regulatory framework is described in the Guide to the Approximation of European Environmental Legislation.
INTRODUCTION
The ECOSIM project builds a model-based information and decision support system for urban environmental management. It integrates data acquisition and monitoring systems, GIS, and dynamic simulation models in a flexible client-server architecture based on standard protocols.
The main objectives of ECOSIM is to develop a tool that:
- Is easy to use for administrators, planners, engineers, and decisionmakers;
- Supports environmental planning and decisionmaking by providing scientifically sound information from state-of-the-art tools in a format that is of directly relevance to the decisionmaking process;
- Integrates numerous information sources, from online monitoring to databases, geographic information systems, and complex simulation models, in one common framework and presentation format;
- Is cost-efficient by using a flexible architecture that minimises investment requirements by allowing more expensive components, such as high-performance computing equipment but also specific expertise, to be shared over the Internet.
THE TECHNICAL APPROACH
The main objectives are to demonstrate the integration of a range of distributed information resources including monitoring systems, and simulation models including HPC (high performance computing) models implemented on parallel machines, in an easy to use multi-media framework that provides useful information in support of environmental planning and decisionmaking processes based on state-of-the art scientific methods and information technology. Integration of information resources, a range of forecasting and simulation tools, GIS and expert systems, and decision support components as well as their validation in three parallel case studies are the focus of the second project phase. The main product is the generic ECOSIM Server and the underlying generic client-server architecture together with a set of generic and partly specific environmental simulation models, implemented and tested in three case studies covering a range of climatological, environmental, and technological conditions. ECOSIM integrates, through a distributed client/server architecture:
- Data acquisition from environmental monitoring networks;
- Data analysis and visualization tools;
- A range of numerical environmental simulation models (ranging from very simple, but fully interactive, scoping models to multi-layer forecasting models and ultimately fully 3D dynamic simulation models);
- A hybrid geographical information system;
- An embedded expert system;
- Strong visualization components.
These components are integrated into a common interactive, graphical user interface and implemented through a distributed architecture based on TCP/IP and http (hypertext transfer protocol).
ECOSIM supports integrated quantitative and qualitative analysis of the environment in urban and industrial areas across different environmental domains in a common framework. The tools range from strategic planning models to daily operational forecasting tools. By using a broad range of distributed information resources and existing databases and models, the ECOSIM solution offers a new and efficient tool for urban environmental management. Three public authorities at Berlin, Athens and Gdansk are providing day-to-day feedback to the ECOSIM developers and use the system to manage current environmental problems during the project validation phase.
Rapid prototyping, object oriented design, and distributed client-server implementation are the main technological features. ECOSIM makes use of high performance computing technology where necessary to provide sufficient power to conduct its complex simulations. The wide range of data and results which ECOSIM can include are managed by a multi-media tool-kit and a Geographic Information System.
THE BASIC ARCHITECTURE
ECOSIM is based on distributed a client-server concept. This simply means, that different functions can, where feasible and desirable, be implemented on different computers connected through the TCP/IP protocol, the core of the Internet. However, all the available functionality can also be made operational, in principle, on a single workstation.
The ECOSIM client-server architecture takes advantage of the http protocol. The main server provides the basic user interface and controls the user dialogue, displays information, and connects to external information resources (monitoring data, databases, simulation models) as required.
This communication is based on public http protocol, and can be based on the Internet, or dedicated connections (such as ISDN phone lines) for the physical communication layer. This protocol also forms the basis of World Wide Web browsers like Netscape or Internet Explorer. The following diagram summarizes this architecture:
The same method that can connect various computers supporting different tasks such as models, can link online monitoring systems into the ECOSIM framework. While monitoring provides valuable, and often mandatory information on the environment, it is also an essential element of modeling used for strategic analysis or operational forecasting. One class of conceptual clients in the ECOSIM framework provides a user interface which is fully interactive and menu-driven, and offers graphical interaction supported by an embedded rule-based expert system that helps less experienced users, and extensive graphical visualisation of data and model results.
ECOSIM ARCHITECTURE DIAGRAM. Also see: http://www.ess.co.at/ECOSIM/architecture.html
BASIC FUNCTIONALITY
The generic ECOSIM information system framework consists of:
- Graphical User Interface, menu system and interaction control;
- Geographic Information System (GIS);
- Object database including object editing and display functions;
- Embedded rule-based expert system;
- Hypertext system and embedded WWW browser.
Within this framework, using a generic client-server architecture based on the TCP/IP and the http protocol, individual information resources such as databases and a set of environmental simulations models as the analytical core of the system can be linked and integrated.
The Graphical User Interface
The information system framework relies on a fully menu driven, largely icon-based, graphical user interface. Almost all user interaction consists of point, drag, and click operations. Text input, wherever possible, uses multiple-choice selection rather than text input (typing) to minimize the potential for operator error. Menus, selections, and a range of graphical analogue data entry tools (sliders, compass, etc.) are all context sensitive and constrained to feasible ranges or choice sets to ensure consistent and plausible user input. The various icon menus and option selectors are coordinated by event loops and callbacks that translate the interactive user specifications into the appropriate actions. Consistent layout across pages, use of colours and fonts aims to simplify interpretation and learning within the system. A status bar and an embedded help and explain hypertext system (see below) provides guidance for the novice or infrequent user.
Geographic Information System (GIS)
Almost all objects in the environmental information system are spatially referenced. As a major integrating tool, an embedded GIS manages this spatial data. The GIS provides for the smooth integration of different formats, including vectors, graphs, rasters, cell-grids, digital elevation models (DEMs), triangulated irregular networks (TINs), finite elements (FE), and finite difference (FD) model grids, and dynamic model output. Its main function is to provide a spatially organized interface to the system's data objects and associated functions, as well as the display and analysis of (spatially distributed) model results.
Object Database
The information system is based on an object oriented design. This approach supports the representation and structuring of the problem domain in terms of natural, real-world objects and related information requests in a format and style that is directly understandable and useful for the end-user. It supports a problem (and therefore user) oriented rather than a data or model-oriented view of the problem domain.
Objects or object classes in this framework include:
- Individual functional components such as emission sources, power plants, treatment;
- Plants, observation stations, etc;
- Related data such as observation time series or emission inventories;
- The individual variables or parameters that define the attributes of compound objects (descriptors in the expert systems knowledge base);
- Structural and geographic elements such as road segments, city blocks or districts, hydrographical catchments, parks, etc.;
- Tasks or decision scenarios;
- Individual analytical tools such as simulation models.
Diagram 1: Display of monitoring data: Air quality observations, Athens.
Embedded Rule-based Expert System
As an important generic function of the framework, an embedded rule-based expert system is used to assist the user in the definition of decision and input variables, as well as in the interpretation of results. In addition, the expert system can be used to serve as a classification tool, for example, for Environmental Impact Assessment tasks. The system uses near-natural language 'Rules' (first order logic) and as its variables, 'Descriptors'. Through an interactive dialogue, the system compiles information from the user through an inference process is used to deduce answers (i.e., updated attributes of objects in a changed context).
Hypertext and WWW Integration
The hypertext system is embedded within the overall user interface as a help and explain function associated with information icons throughout the user interface. It also provides meta-data and background information on individual objects such as the descriptors of the expert systems, a model, or a decision-related scenario. In line with the distributed network architecture of the system, the hypertext browser can use both an internal hypertext format as well as a (parallel) embedded web browser HTML documents. This supports direct access to local HTML pages as well as to the entire Internet.
THE BASIC SIMULATION MODELS
What distinguishes ECOSIM from existing systems is its capacity to perform integrated quantitative and qualitative analysis of the environment, in urban and industrial areas across different environmental domains and sub-domains. In addition, the interrelations between the domains and their dynamic behaviour is analysed and the available multi-media data sources and modeling results are cross-calibrated (see Diagrams 2 and 3).
Diagram 2: Ozone forecasting and comparison with observation data: REGOZON, Berlin.
The system at each local site includes the following numerical modeling tools, which are connected on-line to the users' individual monitoring networks:
- Atmospheric wind-field model;
- Air pollution dispersion and air chemistry model;
- Ground and surface water level and pollution model;
- Coastal water pollution model;
- Traffic and traffic emission model.
The flexible modular architecture, enables the simple addition of new components into the same integrated framework for other environmental domains and problems such as waste management, noise abatement, surface water pollution, etc.
The ECOSIM Demonstrator includes the following models:
- MEMO - mesoscale atmospheric model;
- DYMOS - air pollution dispersion and air chemistry model system;
- POM - Princeton Ocean Model;
- MODFLOW/MT3D - groundwater modeling system.
In addition to these detailed models, ECOSIM also includes a number of forecasting and screening level models that combine fast response with a solid scientific basis (see Diagram 3 below). The technical details on the model can be found under: http://www.ess.co.at/ECOSIM/models.html.
Diagram 3: Coastal water quality modeling: POM, Athens.
SUMMARY
ECOSIM provides a powerful, but simple tool for environmental planning, management, and decisionmaking that conforms to European environmental policy, guidelines, and regulations. The main concepts are the integration of distributed and diverse information resources through wide area networking (Internet based) methods, but with an easy-to-use interface that makes the technical complexity completely hidden from the user. Menu driven, graphical, and supported by an embedded expert system, the interface makes interaction with complex models easy. Being able to run these demanding codes remotely, through a client-server connection, makes it possible to use them as an external service, on demand, without a need to invest in powerful computers, expensive training and code maintenance. The modular structure simplifies the configuration of the system for any urban or regional domain and its corresponding environmental problems, and the integration of the available data sets and tools within a new, powerful information and decision support system framework.
REC * PROGRAMS * TELEMATICS * DETERMINE
