iEMSs 2002 Sessions

• ACCOUNTING FOR LANDUSE AND LANDSCAPE CONTROLS IN HYDROLOGICAL MODELLING
• AIR POLLUTION ANALYSIS
• APPLICATIONS OF AGENT-BASED SIMULATION IN INTEGRATED MODELLING AND ASSESSMENT
• BAYESIAN NETWORKS
• ENVIRONMENTAL FLUID DYNAMICS
• ENVIRONMENTAL APPLICATIONS OF ARTIFICIAL NEURAL NETWORKS
• INTEGRATED CATCHMENT MODELLING
• INTEGRATING ECOLOGICAL MODELS
• INTEGRATING MANAGEMENT AND DECISION SUPPORT
• SOCIO-ECONOMIC AND ENVIRONMENTAL MODELLING
• METHODS OF UNCERTAINTY ANALYSIS
• MODELLING CLIMATE AND GLOBAL CHANGE
• PARSIMONIOUS CATCHMENT-SCALE RAINFALL-STREAMFLOW MODELLING: METHODS, APPLICATIONS AND UNCERTAINTIES
• STRUCTURAL CHANGE, MODEL EVALUATION (VALIDATION), AND ENVIRONMENTAL FORESIGHT
• ENVIRONMENTAL MODELLING OF PHYSICAL, BIOPHYSICAL AND CHEMICAL PROCESSES AS A COMPONENT OF SOIL-PLANT-ATMOSPHERE INTERACTION
• INTRODUCING A MASTERS PROGRAMME IN INTEGRATED ASSESSMENT (a one-day workshop)

Estimates of hydrological variables are required as a prerequisite for solving a number of engineering and environmental problems. Whenever possible, models applied for delivering those estimates should be calibrated against existing records. However, hydrological expertise is often searched particularly in such cases where relevant data suitable for model calibration do not exist. One might wish to know how a proposed treatement, say a clearfelling or installing artificial drainage, affects the hydrological behaviour of a catchment. Or streamflow estimates could be required for a site which has not been gauged for streamflow. Either of the above cases calls for methods which link the hydrological response of a catchment to its physical attributes. Methods for establishing such a link can be based on statistical analysis of catchment behaviour as observed in sites with relevant hydrological records, or on physically consistent process descriptions. A lot of caution must be taken in applying any of these two approaches. However, due to the great labour involved in collecting time-series of hydrological data there is a clear need for regionalisation and impact assessment methods. This session welcomes submissions dealing with such issues including:

• identifying effects of landuse changes in hydrological behaviour
• regionalisation of hydrological models
• incorporating landuse and landscape controls in hydrological models

The session deals with modelling and simulation of air pollution in different scales.Topics : Meso scale and micro scale models, Eulerian models, Lagrangian models, GIS,environmental data bases, remote sensing data,traffic induced pollution,ozone forecast,model coupling,regional and urban areas models,air pollution management, networking and distributed computing. This session is part of the 9th ERCIM meeting within the conference.

Agent-based simulation (ABS) is becoming increasingly popular for use in integrated assessment, especially with regards to developing models which integrate social and environmental systems. This session calls for papers which describe such applications of ABS for use in environmental/resource management. General topics of interest include, but are not restricted to: the use of ABS to explore the link between observed system behaviour at different scale levels; the use of ABS to explore the influence of social behaviour on the success of environmental policy; approaches to the validation of ABS; and critical assessments of the advantages and disadvantages of adopting ABS techniques, based on current or completed case studies. Due to the important role that stakeholders have in integrated assessment, descriptions of methods for communicating the results of ABS to stakeholders and, more generally, the use of ABS in participatory settings are also of interest.

This session welcomes papers addressing the use of Bayesian Networks in the analysis of environmental management problems, in stochastic simulation of environmental systems (including feedback), and as Decision Support Systems (or as part of them) for environmental management.

Many environmental problems, ranging from global climate change to  turbulent diffusion, involve fluid dynamic processes. Thus, environmental fluid dynamics  is a fundamental  component of environmental modelling and simulation. This session is dedicated to new theories, modelling techniques, observational evidence and applications in the vast field of research. Topics related to

• atmospheric and oceanic flows;
• boundary layer flows;
• large eddy simulation;
• direct numerical simulation;
• flow over complex terrains, e.g., hills, city canyon, wind breaks, etc;
• two and multiphase flows;
• flow in porous media; and
• flow in continental water bodies, e.g., lakes, rivers, reservoirs, etc
• other related subjects

will be discussed.

The integrated planning and management of water resources involves several aspects, among which the joint regard of ground and surface water resources and the joint regard of water quantity and quality are the most relevant to this session.  The main intention of the session is to discuss the methods
of river catchment modelling currently available, the related methodological problems and deficits, and the issues related to data assimilation.

Ecological systems are complex. Models become complex too. How do we deal with this complexity? What can be a building block for an ecological model? How do we integrate them? How do we integrate ecological models with models from other disciplines?

Neural networks are proving more and more to be an interesting and practical paradigm to solve complex problems, such as those arising in the simulation and management of environment. However, there is not yet a final answer on how they should be structured, trained and used in different environmental fields. The session aims at bringing together researchers and users in different areas such as flow and pollution forecasting or water manegement to share their experience and characterize future research and application perspectives

Models provide for the encapsulation and transfer of knowledge about the processes and relationships, which underpin natural resource management. To be used by and useful to managers, they need to be augmented by many other tools, which allow for scenario description, data exploration, explanation and assessment. Traditionally this has been achieved by developing Decision Support Systems (DSSs) which contain and enhance a suite of models, and are tailored to the clients' needs and expertise. We are seeking papers that describe practical examples of decision support frameworks (including but not limited to DSS) which move beyond mere model integration and/or linking. We are particularly interested in the non-modelling components, their role in delivery, support and enhancement of the underlying models and their contribution to improved decision making

Models of environmental (and most other) systems can only be applied with confidence if the uncertainty in their parameters, inputs and structure, and the sensitivity of their conclusions to it, can be properly assessed. Conventional approaches to analysis of uncertainty, such as those based on analytical probabilistic descriptions, on Monte Carlo trials or on deterministic sensitivities defined as normalised derivatives, are often inadequate in such applications. The assumptions required to make probabilistic analysis feasible are frequently restrictive and hard to justify; sufficient coverage at sufficient resolution by Monte Carlo trials may be difficult or impossible; derivatives may give misleading sensitivities outside the range over which they are computed. The advent of bound-based methods of uncertainty analysis has mitigated some of these difficulties, but these methods also have severe limitations, particularly when non-linear or time-varying dynamics are present, the number of uncertain quantities is large or the system is distributed. The aim of this session is to present some of the more recent uncertainty-analysis techniques suitable for environmental models, and to expose their strengths and shortcomings.

Sophisticated global and regional climate models are the most powerful tools we currently have with which to investigate key problems of climatic and global change. Examples of the problems that these models can be used for include evaluating the effects of warmings due to increases in greenhouse gases, and coolings due to increases in atmospheric sulfate and smoke from biomass burning. The models can also be used to provide input for assessment of the impacts of climatic change on key societal concerns such as agriculture. These climate models can be linked up with ecological and biogeochemical  models to form complex models of the entire 'earth system'. Knowledge of climatic variability is also essential to proper interpretation of model predicitons of future change. Papers on any topic pertaining to the use of global and regional climate models in evaluating climatic and global change are welcomed.

Within the overall context of the Conference (Integrated Assessment and Decision Support), the session on Parsimonious catchment-scale rainfall-streamflow modelling: methods, applications and uncertainties will address the following practical issues:

• balance between sufficient (or justifiable) conceptual complexity in catchment-scale hydrological models and their ‘fitness for intended purpose’
• advantages of parsimonious models over physically-based, deterministic, models (better identifiability of fewer parameters, relatively modest data needs, ease and speed of application, suitability for application by stakeholders, etc.)
• methods and benefits of multi-objective parameter identification (with case-study examples)
• developments and applications of unit hydrograph-based models for runoff event and continuous river flow simulation
• evaluation and presentation of model output uncertainties
• catchment characterisation versus curve fitting: spuriously good calibration model-fits; validation of models in simulation mode
• research and project case-study applications of parsimonious models (low flow hydrology, flood frequency analysis, regionalisation/ungauged catchments)
• examples of parsimonious catchment-scale hydrological models as a component of decision support systems (DSS)
• delivery of parsimonious modelling methodologies to non-researchers, teachers, stakeholders, etc.

Socio-economics is interpreted broadly to include analytical and policy interactions among the social, economic and natural environments. The portability of modelling techniques, specifically, identification, estimation, testing, evaluation, selection, and forecasting, across the various disciplines in Socio-economics is strongly emphasized. Theoretical and applied papers in economics, econometrics, finance, risk and uncertainty, mathematics, statistics, time series analysis, environmental modelling, marketing, management, information systems, technology policy, R&D investment, registration of patents, travel and transportation, and tourism research, are suitable for presentation. Papers dealing with recent mathematical, statistical and econometric advances in modelling time-varying and stochastic volatility, as well as analysing outliers and extreme observations, for financial, economic, environmental, technological and patents data, are strongly encouraged.

Some of the most interesting challenges in developing and using models of environmental systems concern exploration of future patterns of behavior that may differ significantly from what has been observed in the past. Essentially, the problem is that the number of state variables in the model, how they interact, and the form of their interactions, may be evolving over time. What may have appeared to have been an insignificant mode of behavior in the past - buried within the uncertainty of the model and the historical data - may come to dominate behavior in the future. Technically, we may call this a change of structure. What methods of analysis are available for detecting changes of structure in the past, or for exploring the future under structural change? How, in the absence of appropriate historical data, should we assess the quality of a model to be used for such exploratory purposes? What might be the role of scientifically lay stakeholders in this process, both in terms of evaluation (validation) of model performance and in generating environmental foresight more generally? Are special types of models required to simulate structural change in the behavior of an environmental system? Could we design our models with the goal of discovering our ignorance of the system's behavior, and at the earliest possible moment - could we use our models as a kind of "radar" for detecting unexpected threats to the environment lying just over the "horizon"? The session seeks to encourage the presentation of papers and discussion that will help us to begin to answer these and related questions.
Note: This session will be coordinate with the session on "Methods of uncertainty analysis".

An experience with numerical modelling of biophysical and chemical processes at the land-atmosphere interface has progressed over the decades. The environmental modelling community has come to recognize that various aspects of atmosphere-ecosystem-ocean system that once were thought play a relatively minor role are actually very important in the biophysical and chemical dynamics and circulations of the lower part atmosphere environment. Ecosystem and soil-plant-atmosphere processes and their effect on the atmosphere are certainly in this category. In providing a reliable sophisticated information, either for scientific or operational purposes, the environmental models dealing with the foregoing processes still meet with methodological, mathematical as well as software problems. In that sense this session welcome submissions dealing with such issues including:

• Modelling of flux exchanges between non-homogeneous surfaces and atmosphere
• Modelling of transport of water and chemicals in the soil
• Closure problem inside and above the tall vegetation
• Modelling of planetary boundary layer land surface processes
• Modellling in population dynamics
• Uncertainties in environmental modelling and occurrence of deterministic chaos
• Predicting the occurrence of plant diseases
• Modelling of environmental fate of pesticides
• Modelling of planetary boundary climatic change and future state in agriculture and forestry