Motivation fluid injections often show pure shear fractures hydraulic fractures are often non parallel to s1 some authors explain by combined sheartensile fracture this presentation shows why tensile fractures are so rare Tensile earthquakes in focal mechanisms tensile events – opening => normal stress must have been negative, σneff<0 => tensile events occupy limited area in Mohr diagrams σneff<0 σneff= σn− p < 0 p > σn (Vavryčuk, 2002) Mohr circle, 2D σ1 τ 45° 60° 30° 2 σ3 σ1 σ3 1 2 2 s n s 2 s 1 s 3 4 => sn and acting at any plane lie on the Mohr circle σ Material strength Coulomb (1773) – linear strength f = c + m sn shear failure strength cohesion friction Mohr (1880) – strength envelope is non-linear ! τ shear fracture c 2 tensile (Hydraulic) fracture σ s Empirical strength envelopes granites tensile strength << compression strength 5..10 MPa 50..300 MPa limestones (Parry, 1995) Intact rock failure (high σ1 - σ3) τ Big radius = σ1 - σ3 c 2 σ3 p σ1 σ1 σ-p => Mohr circle touches the strength envelope at σ>0 ISO = 0 σ3 DC = 100% high 45° < < ~50° Intact rock failure (small σ1 - σ3) τ Small radius = σ1 - σ3 c p => Mohr circle touches the strength envelope at σ<0 σ1 ISO 0 σ-p pure tension = HF DC > 0 σ3 small ~50° < 90° not well constrained, weak dependance of σn on the angle Depth dependance of σ1-σ3 KTB, Soultz: SH ~ 2 Sh SH – Sh = 2/3 Sv depth SH-Sh 100m 1.8 MPa 1 km 18 MPa 10 km 180 MPa tensile fractures possible at < 1-2 km depths (Brudy et al., 1997) How to achieve s < 0 high pore pressure – fluid injections stimulations in gas fields, < 3 km geothermal reservoir stimulations, < 5 km earthquake swarms, < 10 km very small depths (small lithostatic pressure) land/rock slides – extensional regime => monitoring of ISO component helpful Conclusions strength envelope is non-linear !! tensile earthquakes occur only if deviatoric stress σ1-σ3 is small enough preexisting fractures strike close to σ1 any tensile fracture shows some shear component tensile frac occurrence decreases with depth