Introduction to NZJGG special issue in honour of John Beavan's scientific contributions

John Beavan made profound contributions to geodetic, crustal deformation and active tectonics research both in New Zealand and around the world. His research involved investigations of tectonic processes in many far-flung and diverse locations including Alaska, California, New York, the Philippines, the Marianas Islands, Micronesia, Tonga, Samoa and, most notably, in New Zealand. He arrived in New Zealand in 1994 to begin working for GNS Science, New Zealand's national geoscience research organisation. Prior to his time in New Zealand, he was a Senior Research Scientist at Columbia University's Lamont-Doherty Earth Observatory for nearly two decades, where he made major contributions to the field of crustal deformation measurements using precise levelling, tiltmeters, strainmeters, tide-gauges and eventually GPS. John's arrival at GNS forever changed the face of geodetic and crustal deformation research in New Zealand, where he immediately dedicated himself to building up world-class geodetic datasets throughout the New Zealand plate boundary zone. He tirelessly continued this work until his death in November 2012. As a consequence of John's efforts, New Zealand is the site of one of the most comprehensive crustal deformation networks of any major plate boundary on Earth. These data have enabled John and his very fortunate colleagues to gain unprecedented insights into active plate boundary processes in New Zealand.

The first manuscript in this issue (Beavan et al. 2016) presents the New Zealand national velocity field resulting from campaign GPS measurements undertaken largely with John's leadership. We expect that this velocity field release will be a vital reference for future research. Throughout his time in New Zealand, John was instrumental in the integration of the New Zealand GPS velocity field into a dynamic surveying datum to account for the impact of the ongoing tectonic contortions on the New Zealand national surveying datum. The paper by Blick & Donnelly (2016) represents an overview of 150 years of geodetic datums in New Zealand, marked by major developments including NZGD2000 and the transition from triangulation-based control surveys to the extensive use of GNSS surveys and a national CORS network (PositioNZ network). John Beavan played a leading role in the development of NZGD2000, in terms of both the secular deformation model and earthquake dislocation models that are necessary to define NZGD2000 coordinates. The paper by Crook et al. (2016) details the continued development of the New Zealand deformation model which forms an integral part of the NZGD2000 coordinate system. Since the early 2000s, New Zealand has been rocked by a number of large earthquakes causing large surface displacements, triggering the need to implement an update to the datum. John was heavily involved in the development of the GeoNet and PositioNZ continuous GPS networks, which have also contributed to maintaining the New Zealand survey datum. The paper by Gentle et al. (2016) overviews the history of development of GeoNet, and its role in surveying/datum applications as well as crustal deformation studies.

John contributed greatly to the understanding of tsunami generation and subduction zone processes, particularly in the western Pacific. One of the most significant discoveries of John's career was his recognition from GPS measurements in northern Tonga that the 2009 M_w c. 8.0 outer rise earthquake (and devastating tsunami) in Samoa was accompanied by a near-simultaneous M_w c. 8.0 rupture of the subduction interface at the Tonga Trench (Beavan et al. 2010a). He has also contributed greatly to work using GPS to characterise the slip behaviour of the North Island's Hikurangi subduction zone, including interseismic locking (Darby & Beavan 2001; Wallace et al. 2004) and slow-slip events (Douglas et al. 2005; Wallace & Beavan 2006, 2010; Wallace et al. 2012). Power et al. (2016) present a comprehensive overview of current research on earthquake and tsunami potential of the southern Hikurangi subduction interface that is strongly influenced by this work. They present the most updated information from geodesy, paleoseismology, paleotsunami and passive and active source seismology, which feed into new tsunami wave modelling. A key result is the high sensitivity of the tsunami impact in Cook Strait to how far ruptures extend across the strait. A contribution from Jacobs et al. (2016) outlines the relationships between seismicity and slow-slip events (SSEs) in the Hikurangi margin. The authors calculate seismicity rates during SSEs using the raw GeoNet earthquake catalogue and a derived sequence catalogue from 2002 through 2011. They show an increase in seismicity rates for most regions during times of slow slip. Dimitrova et al. (2016) use the latest campaign GPS velocity field from the North Island to gain a broader-scale picture of contemporary deformation adjacent to the Hikurangi subduction margin. Using newly developed methods for deriving high-resolution strain rates, the authors identify previously unrecognised features of the strain field such as: large-scale contraction within the Taupo Volcanic Zone; and extension in the Raukumara Ranges and southern Hawke's Bay that might be related to gravitational collapse of the forearc. The paper by Dimitrova et al. represents an update on earlier work by John Beavan and John Haines to develop the first GPS-based national strain-rate model for New Zealand (Beavan & Haines 2001).

The field of active faulting and seismic hazards in New Zealand has benefited greatly from John's efforts. For example, results from New Zealand's continuous and campaign GPS networks have underpinned some aspects of the latest version of the New Zealand National Seismic Hazard model (Stirling et al. 2012). Results provided in the paper by Nicol et al. (2016) will further help to improve the use of active fault data when developing robust models of seismic hazard. Nicol et al. use a combination of geological data and simulated earthquakes on over 100 active New Zealand faults to quantify the variability of recurrence interval and single-event displacement. New Zealand's Active Faults Database constitutes one of the largest data sources to underpin the NZ National Seismic Hazard model. Langridge et al. (2016) outline the latest release of the new NZ Active Faults Database. The new release includes more accurate active fault locations and characteristics (including slip rate, recurrence interval, single-event displacement). In future, it will be revealing to undertake more detailed comparisons between long-term deformation (e.g. >10,000 years) informed by the NZ Active Faults Database and that observed in the contemporary deformation field from GPS.

In recent years, advances in geodetic monitoring have enabled the detection and analysis of ever smaller deformation signals, and we expect that the groundwork laid by John will lead to better understanding in the future of the hazard posed by low-slip-rate faults to New Zealand's major population centres such as Auckland. The Kerepehi Fault in the Hauraki rift is one such fault, and the paper by Persaud et al. (2016) presents the latest paleoseismological investigations of the earthquake potential of the Kerepehi Fault. By incorporating high-resolution LIDAR and traditional paleoseismic trenching methods, the authors provide new constraints on fault geometry and rupture history in a region where the current geodetic network is too sparse to derive the contemporary slip rate. Another crustal deformation monitoring tool that has seen advances in the last decade is Interferometric Synthetic Aperture Radar (InSAR). John was involved in many of the early studies using InSAR data for measuring crustal deformation across New Zealand, most notably during the Canterbury earthquakes. The paper by Haghshenas Haghighi & Motagh (2016) documents the use of InSAR for monitoring extremely slow deformation associated with the Taihape landslide in the central North Island. Using satellite data acquired between 2003 and 2011, the authors estimate surface displacements of up to 10 mm a^–1 in the satellite line-of-sight and identify the swelling and shrinkage of the ground surface in response to changes in the groundwater level.

John made fundamental contributions to active tectonics and seismological topics in the South Island throughout his nearly 20 years in New Zealand. These include the first-ever measurements of the vertical uplift rate of the southern Alps (Beavan et al. 2010b), the first GPS estimates of slip rates and interseismic coupling across the Alpine Fault system (Beavan et al. 1999) and detailed investigations of the Dusky Sound earthquake sequence (Beavan et al. 2011a) and other broader South Island problems (Beavan et al. 2007; Wallace et al. 2007). Further contributions to the understanding of crustal deformation in the northwestern South Island have been made by Barnes & Ghisetti (2016). Seismic reflection data have been used to characterise six major active reverse faults of the North Westland deformation front that extends for 320 km offshore between Cape Farewell and Hokitika. They define nine new potential earthquake sources with M_w ranging from 6.7 to 7.8 and recurrence intervals from 7600 to 30,400 years. Measurements of absolute gravity at existing continuous GPS points in the Southern Alps is presented by Bilham et al. (2016), and were inspired by discussions with John in 1999. Confronting difficulties for measuring absolute gravity in rough alpine terrain (as well as the effects of surface and subsurface water changes), Bilham et al. quantify changes in gravity that have occurred over the past 15 years across the Southern Alps and in the epicentral region of the 2010 Canterbury earthquakes. They use Beavan et al.’s (2010a) calculations of vertical velocities from GPS co-located with the gravity sites to extrapolate elevation changes to 2015 levels. They observe a Bouguer surface gradient of approximately 1.9 μGal cm^–1, consistent with the observed uplift. A contribution from Denys et al. (2016) builds on earlier geodetic work in the South Island and presents results from a newly densified GPS network in the Otago region (the Central Otago Deformation or COD network), which reveals a much clearer picture of active tectonic deformation in the Otago region than was previously possible.

In the last few years of his life, John dedicated extraordinary time and energy to using GPS and InSAR to understand the unusually complex source mechanisms and postseismic deformation in the devastating sequence of earthquakes that have struck the Canterbury region of New Zealand since 2010 (Beavan et al. 2010c, 2011b, 2012). In addition to new insights into the evolution of the complex earthquake sequence, John's work underpins ongoing efforts to forecast what might be expected to occur in future earthquakes in that area as the Canterbury earthquake sequence evolves. Holden & Kaiser (2016) undertake stochastic ground motion modelling of many of the largest aftershocks in the Canterbury sequence. To help constrain these models, they use detailed source information underpinned by John's geodetic work on the Canterbury sequence. Holden and Kaiser find that appropriate stress drops and site-specific amplification functions are critical to reproduce the observed peak ground acceleration and response spectra. Ellis et al. (2016) draw on John's work to better constrain the regional stress field before and after the 2010 Darfield earthquake. Making use of seismic tomography, fault geometries from Beavan et al. (2012), GPS deformation rates and 3D numerical modelling, the authors investigate static stress changes from the earthquake superimposed on the regional stress field. Their models suggest that stress rotations resulting from the earthquake sequence cannot explain those observed from aftershock focal mechanisms, concluding that observed changes in maximum horizontal stress directions near the Greendale fault might not be due to stress changes resulting from the earthquake sequence itself.

John's colleagues will remember him as an incredibly rigorous, generous, humble and insightful collaborator. His passing leaves a huge gap in the field of geodesy both in New Zealand and globally. However, John has left an important legacy in New Zealand geodesy and active tectonics that will continue for decades to come.