Three thermally activated mechanisms are considered to be of particular relevance during slip in thermally unstable rocks such as carbonates: 1) flash heating at highly stressed frictional micro-contacts (asperities); 2) thermal pressurization of heated fluids released and/or trapped in the slip zone; and 3) the lubrication effects of nanoparticles produced by thermally-induced chemical decomposition reactions (decarbonation). In order to investigate whether such chemical and physical reactions in carbonate fault zones can make faults extremely weak and favour the continued propagation of earthquake ruptures, we propose here a multidisciplinary research program where mechanical, mineralogical, microstructural, fluid flow properties and modelling data, obtained from both field and laboratory studies are integrated. Fieldwork studies will be carried out in carbonate rocks from the Italian Apennines to reconstruct the natural fault zone geometries, identify the different structural domains and their associated fault rock assemblages. The integration of field observations and microstructural/mineralogical analyses will provide important geological constraints in the analysis and interpretation of the dominant deformation processes observed in experimentally deformed samples. These results will be used to produce a new classification scheme for seismic fault rocks in carbonates, based on the identification and description of diagnostic associations of fault rocks and microstructures which are indicators of earthquake slip events. This classification will aid in the recognition of fossil earthquakes along exposed fault segments and, therefore, can be used to interpret records of palaeo-seismic faulting in other parts of the world, aiding in risk/hazard assessment. High velocity friction experiments will be performed on solid and granular carbonate rocks, sheared at speeds similar to that seen in large earthquakes (1.3m/s), in order to assess the likely dynamic frictional strength Tf of fault rock materials collected from active fault zones in the study areas. Synthetic nano-powders obtained by thermal decomposition of carbonates in a furnace will also be tested. The integration of laboratory friction test results and microstructural studies from both experimental and natural faults should allow the identification of the dominant weakening mechanisms and constrain their operational conditions in natural environments. The permeability of granular materials is controlled by grain size distribution, grain shape, solid volume fraction and pore connectivity. All of these geometric parameters vary across a fault zone. Permeability laboratory measurements will be performed on field samples collected along transects oriented orthogonal and parallel to principal slip surfaces in the fault zones. These data can provide useful 'snapshot' information on the evolution of permeability of slip zones under known/controlled conditions (friction, displacement, timing of fluid emissions), which can be used to calibrate/test fluid flow modelling results. We will use a state-of-the-art numerical model (name) to determine the permeability tensor for the range of geometric parameters obtained from quantitative microstructaral analyses. This numerical approach will allow us to explore systematic permeability variations in a way that cannot be achieved through laboratory experiments alone. The rate of dissipation of the fluids generated by thermal decomposition of the carbonate during slip will be quantified, allowing the role they play in fault zone lubrication to be determined.