Digital Object Identifier (DOI) 10.1007/s00193-003-0212-3 Shock Waves (2003) 13: 237–252 Characterisation of underground blast-induced ground motions from large-scale field tests C. Wu 1 , Y. Lu 2 , H. Hao 1 , W.K. Lim 2 , Y. Zhou 3 , C.C. Seah 3 1 School of Civil and Resource Engineering, the University of Western Australia, 35 Stirling Highway, Crawley WA 6009, Australia 2 Protective Technology Research Center, Nanyang Technological University, c/o School of CEE, Singapore 639798 3 Defense Science and Technology Agency, Ministry of Defense, 1 Depot Road, Singapore 109679 Received 18 November 2002 / Accepted 8 July 2003 Published online 8 October 2003 – c  Springer-Verlag 2003 Communicated by H. Gr¨onig Abstract. An in-situ large-scale underground blast test program was conducted to investigate stress wave characteristics inside the medium, at the soil-rock interface and on the soil ground surface. The stress wave measurements were made through a vertical borehole and a horizontal borehole inside the rock mass, and along several horizontal lines arranged at the rock-soil interface and on the soil surface. Three different charge weights of 0.5, 2.5 and 10t with loading densities of 0.5, 2.5 and 10kg/m 3 , respectively, were used in the detonation. The measured stress wave time histories and their characteristic parameters, such as peak particle velocity (PPV) and peak particle acceleration (PPA), as well as the principal frequency (PF) at different locations are presented and analyzed in this paper. The results are also compared with those recorded from small-scale tests in a previous study and those from some other researchers, and the pertinent scale effects are discussed by observing differences of the stress wave intensities between the small-scale and the large-scale tests. Key words: Characterization, Underground blast, Ground motions 1 Introduction Accidental detonation in an underground ammunition storage chamber in a rock mass might cause serious dam- age to the rock mass around the chamber, to adjacent tunnels and chambers, to the ground surface; and in the worst case it could result in the collapse of above or below ground level structures. Therefore, when designing such underground ammunition storage magazines, the proper- ties of the blast-induced vibrations are always a concern to engineers to prevent damage to structures from acciden- tal blasts. Different criteria exist in different design reg- ulations (NATO 1993; DDESB 1996). Usually the peak particle velocity (PPV) is used as a criterion for limiting damage to structures due to blast-induced ground excita- tions, whereas the diversity of the frequency contents has not been particularly emphasized. In fact, the frequency contents of shock motions are expected to have significant influences on the damage of structures. Frequency-based criteria for the control of vibrations and structural dam- age, however, require methods to predict both the peak particle velocity and the principal frequency (PF). Unfor- Correspondence to: C. Wu (email: wu@civil.uwa.edu.au) tunately, very few publications are available in the unclas- sified literature mentioning about the PF prediction. Many empirical formulae have been proposed for the prediction of the ground motion amplitudes (PPV) due to underground blasting (Hendron 1977; Langefors and Kihlstrom 1978; Wiss 1981; McMahon 1994; Dowding 1985, 1996; Murrell and Joachim 1996; Zhao et al. 1997). Because the stress wave properties are affected by many factors, such as the charge weight, the type of explo- sives, the coupling of explosives, the chamber shape, the rock mass properties and, of course, the distance from the charge chamber, it was observed that the difference of PPV at the same scale distance from different blast tests could be more than 100 times (Dowding 1985). Moreover, most of the available empirical formulae did not differen- tiate among stress waves in the rock free field, on the rock surface, at the rock-soil interface and on the soil surface. The scale of the blasts, and the possible scale effects were never mentioned in these empirical formulae. Recently, a series of large-scale field blast tests for un- derground ammunition storage were conducted at a Swe- den site as a joint work program (Hanson 2001). During the tests, measurements were arranged to record the stress wave inside the rock mass, at the rock-soil interface and