Carlos Gomez

This thesis investigates the role of advanced atmospheric inversion techniques and ∆¹⁴CO₂ observations in improving the accuracy of regional CO₂ flux estimates across Europe. Accurate quantification of fossil fuel emissions and biospheric fluxes is essential for understanding regional carbon budgets and supporting climate policy. However, significant uncertainties persist due to limited observational coverage and discrepancies in bottom-up inventories. The research integrates dual-tracer, regional isotope budget, and CO₂ inversion methods with CO₂ and ∆¹⁴CO₂ measurements to enhance fossil fuel emission and Net Ecosystem Exchange (NEE) estimates. The results demonstrate that ∆¹⁴CO₂ serves as a critical tracer for distinguishing fossil fuel emissions from biospheric CO₂ fluxes, significantly reducing biases in NEE estimates caused by prior assumptions about fossil fuel emissions. These methods produced the most reliable results in regions with dense observational networks, such as Western and Central Europe, while revealing larger uncertainties in under-sampled areas like Southern and Eastern Europe. Using Italy as a case study, a southern European country with sparse observational coverage, the thesis evaluates the optimization of the Integrated Carbon Observation System (ICOS) network. Strategic placement of monitoring stations, particularly in Chieti and Lecce, significantly improved CO₂ flux estimates, while additional stations provided diminishing returns. These findings underscore the importance of targeted, cost-effective network expansion guided by inverse modeling frameworks. The research also examines sampling strategies to maximize the value of ∆¹⁴CO₂ observations, highlighting the importance of prioritizing high fossil CO₂ signal and minimizing nuclear radiocarbon contamination. A well-designed sampling strategy is critical for enhancing the signal-to-noise ratio in ∆¹⁴CO₂ measurements and improving flux estimates, especially in regions with complex emission dynamics. This thesis demonstrates the potential of integrating ∆¹⁴CO₂ data into inversion systems to provide independent and robust constraints on carbon fluxes. It also highlights the utility of inverse modeling as a decision-making tool for stakeholders involved in network design and sampling strategies. Additionally, the work explores the intercomparison of inversion approaches as complementary tools for validation, offering valuable insights for improving greenhouse gas monitoring systems, supporting climate policy, and advancing the understanding of regional carbon dynamics.