Implications of different spatial (and temporal) resolutions for integrated assessment modelling on the regional to local scale – nesting, coupling, or model integration?

摘要

Integrated assessment modelling (IAM) in general is currently applied to a range of environmental problems addressing aspects of air pollution and climate change, water pollution and many more. While different branches have emerged from applications within different disciplines, they share a similar view of the core features of IAM, i.e. multi-disciplinary approaches, integration across environmental compartments, and the application of models with the aim to provide decision support for complex problems. Examples of IAMs on a regional scale are the RAINS/GAINS model suite (International Institute for Applied Systems Analysis, IIASA), with versions for Europe and Asia. On a national scale, several European countries are currently developing and applying IAMs for policy development, in some cases using special adaptations of the IIASA RAINS/GAINS model (e.g. Italy), or own models (UK, Germany).udIAMs have been extensively used in the preparation of the Multi-Effect Protocol (United Nations Convention on Long-Range Transboundary Air Pollution, CLRTAP) and the European Clean Air For Europe (CAFE) strategy. In these applications, target setting included a mixture of health and ecosystem related indicators. State-of-the-art IAMs are typically operating on rigid spatial scales, and in most cases do not take into account the temporal patterns of emissions and effects in their assessment approaches. IAM results are typically provided on national or regional level (e.g. control measures, costs, benefits due to reduced environmental and health impacts) and for annual indicators (e.g. critical load exceedances or morbidity/mortality effects. However, scientific evidence is today capable of providing a better foundation to identify major aspects for uncertainties in these larger scale assessments, for instance investigating the distinct temporal patterns of air quality throughout the year and the detailed modelling and mapping of human exposure to air pollutants beyond statistical average exposures on total population level. This requires a more advanced and flexible design of IAMs to better model the temporal and spatial domains which are of relevance for the key issues to be assessed.udFirst steps towards bridging the gap between regional and national, respectively national and local scale models for integrated assessments have taken the route to derive parameters for e.g. the urban differential in ambient air quality outside of the models regular domain and integrate these parametric values into the IAMs assessments. While this approach is moderately labour intensive, the major flaw is the integration of static values into an intrinsically dynamic model. In other words, if input datasets and external drivers (e.g. meteorology, atmospheric composition and chemistry) change, all other parameters have to be recalculated and re-integrated. This paper will discuss emerging trends for IAMs with a specific focus on spatial and temporal aspects and aims to elaborate on the policy context which is a key driver for the development of IAMs. The growing understanding of how complex interactions e.g. between/within the nitrogen and carbon cycles, where both management options and effects arise/occur on different spatial scales and with different time scales, both feeds into and requires the development of next generation IAMs, which are capable of tackling these problems.ud