Preface to the special issue of Dense Array Seismology

Both earthquake seismology and structural seismology rely on observations, which in turn contribute critically to the development of seismology, especially in recent years. In order to understand specific geologic structures and interior processes of the Earth, seismic arrays are widely deployed in regions of interests, which produce more coherent waveforms for further data analysis and interpretation. Due to the high cost, broadband seismic arrays (linear or 2-D) typically consist of seismic stations with a spacing interval ranging from about 10–100 km. This limits the spatial resolution and lateral extent of tomographic or imaging results.

Recently, dense arrays, typically consisting of hundreds of stations, have been more and more widely used in seismological studies and greatly facilitated our understanding of Earth’s structures of various scales, earthquake activities, earthquake rupture processes, etc. These dense arrays either have very large spatial coverage (e.g., the Japanese Hi-Net array, USArray, and ChinArray), or have very close station spacing in a local region (e.g., the Long Beach array in southern California; Lin et al., 2013). In particular, for these local dense arrays, instead of using expensive broadband stations, integrated geophones or short-period instruments with build-in battery and digitizer have been popularly used to obtain high-resolution images of regional crust structures, volcanic regions, fault zones, urban areas, oil and gas as well as mineral deposit fields, etc. These stand-alone instruments are much cheaper and easier to deploy, which make ultra-dense arrays possible, sometimes with a large number (so called “Large N”) of integrated geophones deployed in a very small region with average station spacing reaching about several hundred meters or less.

A very successful example of this dense array is the Long Beach array in southern California, which consists of about 5, 300 high-frequency single-component geophones with 10-Hz corner frequency, deployed between January and June 2011 in a 10 km×6 km region. The seismic data of this array have been successfully used in analysis of teleseismic body-waves for crust and uppermost mantle structure, high-frequency ambient noise (surface-wave) tomography (Lin et al., 2013), body wave retrieval from ambient noise interferometry and body wave traveltime tomography (Nakata et al., 2015), analysis of site amplification across the Long Beach region from ambient noise correlation (Bowden et al., 2015), high-resolution detection of local seismicity, etc. The success of the Long Beach array has encouraged popular dense geophone array studies worldwide. Various studies based on dense seismic arrays have been conducted in both methodology development and applications, termed “dense array seismology”.

This special issue contains 9 papers, which focus on recent advances in dense array seismology in China. Among the 9 papers, one paper deals with fast computation of dense array ambient noise cross-correlation functions (NCFs) using public cloud computing from ALIYUN (Wang et al., 2018), which offers a very efficient way to compute NCFs for millions of station pairs using continuous waveforms of thousands of stations. This new approach can reduce yearly computational time of NCFs to less than 12 hours.

Four papers focus on applications of ambient noise tomography of different scales. Hui et al. (2018) obtained phase velocity maps from ambient noise tomography in the Ordos region using data from 257 ChinArray and permanent broadband stations, which reveals regional crustal structures and tectonic features. Wang et al. (2018), Zhou et al. (2018) and Liang et al. (2018) applied dense array ambient noise tomography using short-period instruments to reveal high-resolution shallow crustal structures in regions of horizontal scales about 10 km. The study region of Wang et al. (2018) is the Shangrao section of the Xinjiang basin in southeastern China, occupied by 203 three-component stations with inter-station distance of ~400 m. Zhou et al. (2018) investigated the shallow crustal structure of the Luoyang basin in the western He'nan uplift with 107 three-component short period seismometers with an average station spacing of about 2 km. Liang et al., (2018) used 47 stations with an average spacing of 0.5–2 km to study the shallow crustal structure of the Ji'nan urban area in Shandong Province.

Using the same stations as in the study by Zhou et al. (2018), Tan et al. (2018) estimated the site effects of the Luoyang basin using the horizontal-to-vertical spectral ratio (HVSR) method from ambient noise records, which is helpful to estimate ground motion for earthquake hazard mitigation. Xu et al. (2018) used earthquake waveforms, ambient noise data, and the active source data from the Fixed Airgun Seismic Transmitting Station in Binchuan, which were recorded by the very dense array (510 seismometers) in the Binchuan basin, Yunnan, to investigate the basin effects. This new dense array covers a 30 km × 40 km region with about 2 km station spacing, which aims to reveal the geological features of the Binchuan basin and to provide seismic hazard estimates of this area.

For local scale 3-D ambient noise (surface-wave) tomography, most studies ignore the topographic effect on tomographic inversion. However, for dense arrays with station spacing about several hundred meters, the surface topography in mountain regions may have a significant effect on ambient noise tomography. Wang and Sun (2018) investigated this issue using synthetic waveforms from a 3-D model with topography and found topography does affect subsurface velocity inversion. And depth distribution of the structure can be distorted without considering the topography in the inversion. They further applied the tomographic algorithm considering surface topography to a dense array, which consists of 54 stations with station spacing about 250–500 m, in a polymetallic ore deposit region.

With dense array data and ambient noise interferometry technique, more and more studies have successfully recovered body waves between station pairs ( Nakata et al., 2015; Feng et al., 2017). Feng et al. (2018). used the dense regional broadband seismic array (438 stations) in southwest China to successfully recover the body wave reflections (P₄₁₀P, P₆₆₀P) from the mantle transition zone interfaces using ambient noise interferometry. Their results reveal apparent lateral variations of the 410-km and 660-km discontinuities in southwest China, which provides a complementary way to investigate deep interface structures.

Although this special issue only covers limited topics in dense array seismology, these papers open a window of dense array studies in the Chinese seismological community and may promote more innovative studies in this field in the future. Finally, we sincerely appreciate all the authors, reviewers, and editorial members for their great contributions to this special issue.