Integrated System for Intelligent Regional
Environmental Monitoring & Management:
ISIREMM Project description
5.2 Architecture and Components
ISIREMM is designed as a flexible, modular system with a
set of interchangeable components. The
basic elements of the system are:
the ISIREMM core (or main server) which co-ordinates all
information resources and the display clients;
the core include the data base management for monitoring data and model
scenarios and results, the emission inventories, the embedded
expert system for parameter estimation, and the GIS tools;
the monitoring system, linked as a logical server to the core;
the simulation models components, which are either
embedded with the core for simple screening-level models are
implemented as computer servers;
the DSS/optimisation components related to the simulation models;
the embedded web server and cgi model components for the
support of Internet access and related HTML/Java components for
the browser based user interface.
These components are integrated through the standard protocols TCP/IP and http.
5.3.1 Systems architecture
ISIREMM is based on a client-server architecture that uses the
public domain Internet protocol (TCP/IP and http) to link the
various components and information resources of the system with the
central analytical engine and a range of clients.
Exploiting public domain tools under the GNU software licensing as
much as possible, the implementation will be based on a Linux server
implementation with an attached Apache web server.
This will support networked clients under either the X11
protocol and display functions for high-bandwidth local clients,
or an http based protocol for low-bandwidth PC based clients running
standard web browsers for XML/HTML and Java based user interfaces.
5.3.2 Modeling and monitoring
ISIREMM involves three sets of complementary model components:
- The original set of models of the ECOSIM/AirWare system,
including simple Gaussian short- and long-term screening models,
and a dynamic photochemical box model;
- The 3D dynamic meteorological and photochemical models used for
- The advanced set of locally developed models with the
full integration and utilisation of monitoring data integrated
into the final ISIREMM system.
The screening models
The ECOSIM/AirWare system has a number of
embedded and fully interactive screening level
regulatory models, primarily based on USEPA Gaussian guideline
models such as ISC-3/AERMOD, and PBM, a dynamic photochemical
box model, as well as a dynamic multi-layer
Eulerian code using the diagnostic wind model DWM as a pre-processor.
These simple but efficient models are coupled to both the emission
inventories, meteorological data bases, and monitoring data, and can
be run interactively and with a fully graphical user
interface with embedded GIS functionality in both short-term
(episode) mode or as long-term models to generate seasonal or
annual average results.
The results of the long-term model, in the form of a
source-receptor matrix, are the main input to
the optimisation (emission control) module.
The baseline models
The air pollution model system to be used in the frame
of ISIREMM was developed by Aristotle University of
Thessaloniki, and represents one of the most widely
utilised European model systems developed for the description
of local-to-regional dispersion and chemical transformation processes.
In its present version, the model system takes fully into account
the manifold interactions between the various scales
influencing the pollution patterns in the airshed considered
and comprises of the models MEMO and MUSE.
MEMO is a prognostic mesoscale model, which allows describing the
air motion over complex terrain. Within MEMO, the conservation
equations for mass, momentum, and scalar quantities as
potential temperature, turbulent kinetic energy and specific
humidity are solved in terrain-influenced co-ordinates.
Non-equidistant grid spacing is allowed in all directions.
The numerical solution is based on second-order discretisation applied
on a staggered grid. Conservative properties are fully
preserved within the discrete model equations.
The discrete pressure equations are solved with a
fast elliptic solver in conjunction with a generalised conjugate
gradient method. Advective terms are treated with the TVD scheme.
Turbulent diffusion can be described with either a zero-, one- or two-
equation turbulence model. At roughness height similarity theory is applied.
The radiative heating/cooling rate in the atmosphere is calculated
with an implicit multilayer method for short-wave radiation.
The surface layer over land is computed from the surface heat budget equation.
The soil temperature is calculated by solving a one dimensional heat
conduction equation for the soil.
MEMO allows performing "telescopic" simulations, via nesting
MEMO-in-MEMO. In this sense, meteorological information may be
input to MEMO at rather large distances from the area of
interest. As one of the core models of the European Zooming Model,
MEMO has been successfully applied for various European airsheds including
the Upper Rhine Valley and the areas of Heilbronn, Basel, Graz,
Barcelona, Lisbon, Madrid, Milano, London, Cologne, Lyon, The Hague and Athens.
MUSE is a multilayer dispersion model for reactive species in the
local-to-regional scale. The atmospheric boundary layer is divided
into individual layers (at least three) the thickness of which is
allowed to vary in the course of the day. This variation reflects
adequately the dynamics of the atmospheric boundary layer.
The upper layer serves as a reservoir layer located just above the
boundary layer. A shallow layer adjacent to the ground is used for
simulating dry deposition (with the resistance model concept) and
other sub grid phenomena. Thanks to the modular structure of
MUSE, chemical transformations can be treated using any suitable
chemical reaction mechanism.