Please use this identifier to cite or link to this item:
|Title:||Extended Worksheets for Case-Study Areas and Variables, Including Sampling Frequency, Number of Sites and Analyzed Variables: A General Protocol Including Methodological Aspects Related to Variations and Uncertainties in Empirical Data Needed to Run and Validate Predictive Models in Coastal Management|
|Authors:||HAKANSON Lars; ANDERSEN Tom; CARSTENSEN J.; ZALDIVAR COMENGES JOSE'; HERNANDEZ GARCIA Emilio|
|Abstract:||The member states of the European Union spend considerable amounts of money on environmental monitoring programs designed to detect the ongoing status of, e.g., aquatic systems, so that critical thresholds and regime shift may be identified in good time and the proper remedial actions applied. Such data-bases are essential in management and science, e.g., to critically test models to optimize remedial actions and to obtain a reliable information about trends. Several European and International data-bases related to marine coastal areas have been studied and one aim has been to highlight factors and information of importance in future work to harmonize the structure of worksheets and the information and data in coastal monitoring programs. We have also given examples on the use of these data-bases to detect critical thresholds. A key message in this report is that we have used much more information than anticipated when the project started and the title of the deliverable was written. So, this report is considerably more ambitious than initially planned, which was to make a compilation of the data from the case-study areas in the project (e.g., Ringkobing Fjord, the North Sea, the Varna Bay, as described in the DOW). Since we have used data not from just a few but from literally hundreds of coastal areas, this report gives a compilation of the actual data-bases used and highlights how the information from such very comprehensive sets of data may be utilized to address important and basic questions in threshold science. The aim of this report is not, however, to discuss statistical or modelling tools in detail, since this is being done in other reports from the project. The aim here has been to provide examples to illustrate that even if measurement error could be completely eliminated, coastal ecosystems are inherently variable, and this sets limits on the predictability of many critical forcing functions, bioindicatirs and operational variables of ecosystem effects. How do we incorporate the sources of uncertainty in our ecosystem models and determine the confidence in model output? Using data from the available comprehensive data-bases, this report addresses these questions by means of fundamental concepts in predictive ecosystem modelling, including coefficients of variation within and among coastal areas, the sampling formula, methods to rank how x-variables influence the variability in a target y-variable, the use of data to reveal new thresholds and critically test proposed regime shifts, sensitivity and uncertainty analyses of models (using Monte Carlo techniques).|
|JRC Institute:||Institute for Environment and Sustainability|
Files in This Item:
There are no files associated with this item.
Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.