This deliverable documents the progress on characterising the main building blocks for representing uncertainty in an integrated anthropogenic CO2 emissions monitoring system. All components of this CHE Prototype are considered here to provide a consolidated view on the characterization of uncertainties. The individual components themselves are discussed in their respective (progress) reports: Observations in D5.1, emissions and transport models in D5.3, inversion / data assimilation methodology in D5.5 and D5.9 provides a synthesis of these building blocks for an end-to-end prototype system. The focus of this report is first to list the uncertainty components of the individual building blocks for the prototype and how they contribute to the overall posterior uncertainty estimations, second to identify tools and metrics to evaluate and benchmark the posterior uncertainty estimates of the prototype, and third to identify the immediate development needs of the prototype with respect to the uncertainty representation as well as longer term research priorities together with an estimate of the required efforts. The overall posterior uncertainty estimate of the CHE prototype arises from uncertainties in the prior data, the model uncertainty, uncertainties in the observations as well as from the ability of the inversion / data assimilation to correctly represent the posterior uncertainty given the uncertainties of the before listed ingredients. An important aspect for the development of the CHE prototype is to identify the research needs for the near-term (next 3-4 year) development steps. This report lists these immediate development needs with an estimate of the required effort for the prototype with respect to the uncertainty representation in Table 3. It also provides a high-level overview of research priorities beyond the near-term development needs (Table 4). The report also emphasises the importance of benchmarking and evaluating the posterior uncertainty estimates of the prototype system. This aspect of the report will be extended for in the final version.
Abstract
Files