Overpressure and Fluid Diffusion Causing Non-hydrological Transient GNSS Displacements GIULIANA ROSSI, 1 PAOLO FABRIS, 2 and DAVID ZULIANI 2 Abstract—In this work, global navigation satellite system (GNSS) observations from the northern tip of the Adria microplate are analysed to differentiate non-periodic (transient) tectonic sig- nals from other deviations from the linear trends primarily due to hydrological loading effects. We tested a recently proposed hypothesis that a porosity wave generated by fault-valve mecha- nisms in a seismogenic fault in the Bovec basin (western Slovenia) propagated throughout the surrounding region. After excluding potential spatially correlated common-mode errors in the consid- ered time series, we investigated the relationship between the GNSS observations and periodic hydrological loading variations. The tests demonstrated that subtracting the hydrological term was effective at the global scale and that the frequency band of the transient signal (1.5 \ T \ 3.5 years) was not correlated with hydrological effects at the local scale (within a few kilometres of the station). Next, the results of previous works are used to cal- culate the permeability values and pore-pressure state at the source of the transient signal. The permeability values for the four main rock formations in the region are consistent with independent observations for similar lithotypes. The ratio between the effective stress and lithostatic load for different vertical profiles in the Bovec area indicated a state of overpressure, with pore-pressure close to the value of the lithostatic load. Thus, our results help define a scenario in which the porosity wave could have originated. Indeed, the formation of the domains of interconnected fractures, such as during the formation of a porosity wave, increases the permeability values, thereby relieving overpressure and restoring a state of equilibrium. Key words: Transient signal in cGNSS, hydrological load, porosity wave, rock permeability, effective stress. 1. Introduction The last 20 years have witnessed a rapid increase in the number of continuously operating global navigation satellite system (cGNSS) networks to monitor crustal deformations in tectonically active regions or provide a reference for surveying. Time series of more than decadal length enable us to detect non-periodic (transient) displacements resulting from volcanic, hydrological, or tectonic processes in vari- ous global regions (e.g., Feng and Newman 2009; Chamoli et al. 2014; Devoti et al. 2015; Silverii et al. 2016). Recently, Rossi et al. (2016, 2017) reported significant deviations from regional linear trends in cGNSS time series in a small area along the northern edge of the Adria microplate (across NE Italy and W Slovenia, see Fig. 1). In particular, these authors focused on a transient displacement with duration of roughly two and half years, as recorded by 13 cGNSS stations. The transient signal propagated through the region with an approximately circular/elliptical pat- tern, thereby spurring a movement that was initially oriented upward (with the exclusion of a site to the east of the area) and subsequently oriented downward. The horizontal component showed similar behaviour, with a smaller oscillation that was parallel to the main tectonic faults and fractures (Fig. 1b). A tomographic inversion of the arrival travel time of the transient signal supplied a laterally and vertically varying propagation velocity field and revealed the source location and origin time. Rossi et al. (2016, 2017) used the 3D tomographic algorithm described in detail by Bo ¨hm et al. (1999) and Vesnaver and Bo ¨hm (2000). The disturbance appeared to have originated approximately 3.5 months prior to the M w = 5.2 event that occurred near Bovec (Slovenia) in 2004 and was located 6.5 km NW of the main shock epicentre along the continuation of the recognized seismogenic Ravne fault at a depth of 9.2 km. The authors hypothesized that this transient displacement process could have been attributed to deep fluids that were mobilized by valve behaviours of the Ravne fault and 1 Istituto Nazionale di Oceanografia e di Geofisica Speri- mentale - OGS, Centro di Ricerche Sismologiche, Borgo Grotta Gigante 42/c, Sgonico (Trieste), Italy. E-mail: grossi@inogs.it 2 Istituto Nazionale di Oceanografia e di Geofisica Speri- mentale - OGS, Centro di Ricerche Sismologiche, via Treviso 55, Udine, Italy. Pure Appl. Geophys. Ó 2017 Springer International Publishing AG, part of Springer Nature DOI 10.1007/s00024-017-1712-x Pure and Applied Geophysics