Surface Runoff Dynamics in Irish Raised Peat Bogs: Synthesis and Modelling

Peatlands represent the largest terrestrial natural carbon storage providing important ecosystem services. However, they have experienced global land-use pressures. In Ireland, they cover over 20% of the land area but were historically drained for agriculture, forestry, and peat extraction, producing negative impacts on their long-term ecohydrological functioning. The primary goal of this research is to investigate surface runoff dynamics in peatlands and their responses to restoration and rewetting measures. The thesis aims to: (i) provide a systematic review of the scientific knowledge on the effects of peatland management on flood regulation; (ii) review and assess the most widely used models for simulating peatlands' dynamics; (iii) monitor the rainfall-runoff dynamics from a drained Irish peat bog site; and (iv) demonstrate the applicability of a flexible modular modeling framework to improve model performances through multi-model combinations. The thesis is structured into six chapters. Chapter 2 addresses the first objective, developing a new conceptual model elucidating the interactions between climate, vegetation, soils, land-use, and hydrology that influence flood regulation in peatlands. The systematic review shows that intact peatlands often attenuate flows and reduce flood peaks. However, drainage consistently disrupts soil structure and water storage capacity, leading to flashier flows with higher peaks and increase downstream flood risks. The review highlights that restoration can help recover water tables but is not always reversible due to the drainage legacy and alterations in soil structure and hydrologic flow paths. However, processes that act in the longer run (such as biogeochemical processes and landscape position) may mediate the outcomes and complicate the flood regulation functions. The review highlights potential gaps in the scientific knowledge base for predicting the impacts of land management measures on flood regulation through empirical evidence from controlled before-after restoration studies. Chapter 3 addresses the second objective and provides an exhaustive review of process-based models for simulating peatlands' dynamics and evaluates them based on their applicability to different peatland types and climate zones, which meets the FAIR (Findable, Accessible, Interoperable, Reusable) criteria. This chapter identifies current limitations in peatlands and identifies the need for a 'peatland community modeling platform'. Chapter 4 develops a monitoring protocol and presents the methodology for quality control procedures applied to the high-resolution rainfall-runoff monitoring program implemented in an industrial cutaway bog in Ireland, Corralanna. Chapter 5 employs a flexible modular modeling framework, MARRMoT, to demonstrate the first successful application of this toolbox to peatlands in ten different model structures, both individually and combined through multi-model approaches (14 different ensemble techniques) for the first time. The analysis indicates that the rewetted peatland exhibits distinct surface runoff dynamics from the drained peatland. The combined models produced new synergistic behaviors compared to single models and improved their performances in some techniques, but these were not consistent for certain techniques for all peatlands. These results point to the value of the ensemble modeling techniques, which could potentially compensate for individual models' deficiencies by producing combined and more robust behaviors for hydrological modeling studies. The findings from Chapters 3 to 5 stress the importance of sharing high-quality data for field studies and improved model benchmarking studies. Finally, Chapter 6 summarizes the key findings and recommendations for future work for a better understanding of peatland management impacts on their flood regulation functions and provides a roadmap for further research to support evidence-based and informed management decisions.