\magnification=1200 \pageno=59 %\nopagenumbers \hsize 6.0 true in \hoffset=1.25 true in \voffset=1.0 true in \vsize=9.0 true in \baselineskip=0.2 true in \medskip\noindent {\bf 11. Conclusions, Recommendations \& Potential Pitfalls} \smallskip During the summer of 1994, IETS and the Geological Survey of Canada conducted an ambitious feasibility study into {\sl "Seismics for Mineral Exploration"}. Major results of the current IPP research program are: \smallskip \item{(1)} Lithoprobe-type high-frequency, 2-D seismic reflection profiling has matured to a point where this technique, when $\underline{properly \ integrated}$ with comprehensive physical rock studies, borehole geophysical logging and the available geological data base, can be used for regional reconnaissance studies across existing mining camps or other areas of interest. \smallskip \item{(2)} Physical rock property studies, borehole seismic logs, and vertical seismic profiling demonstrate that massive sulphides are unique. They are characterized by high seismic impedances, the product of compressional wave velocity and density. The seismic impedances for massive sulphides are high compared to those of most felsic and mafic host rocks. These physical rock property data provide an excellent basis for the $\underline{direct \ detection}$ of massive sulphides by seismic methods. \smallskip \item{(3)} High impedance bodies, such as massive sulphides, cause scattering of seismic energy. A 2-D seismic profile may detect the characteristic scattering response but cannot easily distinguish between off-line scatterers and ones located beneath the seismic profile. Thus 3-D seismic methods should be employed to both detect and delineate local high impedance bodies in the subsurface. Our 3-D seismic modeling studies suggest that high resolution seismic methods offer a $\underline{large \ detection \ radius}$ (i.e, in the order of hundreds to thousands of meters). \smallskip \item{(4)} Conventional processing of seismic reflection data may attenuate the scattering response. Because of the difficulty to image gross structure and scattering responses at the same time we propose a $\underline{hybrid \ data \ processing}$ approach: conventional (CMP) processing to define the overall geological setting, and scattering analysis to detect high impedance bodies. \bigskip\noindent {\bf 11.1 Potential Pitfalls} \smallskip Our recommendations wouldn't be complete if we didn't point out some potential pitfalls, data acquistion problems, and special requirements: \smallskip \item{(1)} Massive sulphides are characterized by high seismic impedance values and should generate strong seismic reflections when juxtaposed against most common host rocks. However, metagabbros in the footwall complex could provide comparable impedance contrasts, too. On the other hand, po-rich ores inbedded in thick high-density sublayermay not cause significant scattering. Thus the interpretation of seismic scatterers, like any other geophysical anomaly, must incorporate knowledge of the local geological setting. \smallskip \item{(2)} Cavities, mined out ore bodies, ramps, and underground exploration galleries are characterized by zero impedance values (!). These will tend to produce scattering anomalies at least as bright as massive sulphide ore bodies. Thus 2-D seismic profiling should not be conducted in the vicinity of such features and 3-D seismic profiling should avoid such areas (if possible). \smallskip \item{(3)} Interpretation of seismic data cannot be done without routine applications of borehole geophysical logs (velocity, density, VSP, etc.) and physical rock property studies. However, conventional seismic tools may fail in a highly conductive/inductive mining environment (see chapter 3). In addition, inclined boreholes may cause severe friction between cable and borehole wall, putting radioactive density logging tools at risk. \smallskip \item{(4)} Current high frequency acquisition parameters (30-140 Hz, 20m receiver group spacing) and processing schemes work best for depths $z \ge 600 \ m$. In the Canadian Shield, shallow depths $z \le 200 \ m$ are often characterized by deep weathering, fractures, etc. and are poorly suited for seismic imaging. \smallskip \item{(5)} Surface seismic methods cannot be applied to image steeply dipping structures ($ \ge 60^o$). However, boreholes seismic surveys could be used for efficient imaging of steep contacts. \medskip\noindent {\bf 12.2 The Scattering Potential of Pyrrhotite Ores} \smallskip \by \pageno=1 \medskip\noindent \centerline{\bf ANNEX A} \medskip\noindent {\bf Resolution and Detectability} \smallskip Can Ore Bodies be Reflectors? From the preceding discussion, it is clear in terms of their physical properties, common sulphide ore should make strong reflectors. Wether or not an ore body is a seismic reflector, however, depends on a number of factors, including its size, shape and geologiacl setting. Whether or not an ore body can be imaged seismically, depends critically on acquisition parameters such as seismic frequencies and suitable source-receiver configurations. While theoretical resolution is controlled by the diameter (d), width (w), thickness (t), depth of burial (z) of a deposit and dominant seismic frequency f used during data acquisition, there is not a single relationship that that controls the resolving power of reflection seismic profiling. Using a dominant seismic frequency (f) of 100 Hz, a target depth of 1 km and assuming a formation velocity (v) of 6.0 km/s, we discuss some of the contraints that control resolution. It is well known from modeling studies, that a minimum diameter $d_min$ of one wavelength is required to avoid severe attenuation of reflection seismic amplitudes (i.e., Berryhill, 1977): $$ d_min \ge v/f = 60 m.$$ In practise, we have to consider both vertical and lateral resolution (for more details see Yilmaz, 1987). For vertical resolution, the minimum thickness $t_min$ of a seismic reflector is guided by the quarter- wavelength criterion: $$ t_min \ge v/4f = 15 m.$$ For lateral resolution, the width of the smallest deposit $d_{min}$ that can be resolved at depth (z) is defined by the Fresnel zone: $$ w_min > 0.5 v (2z/vf)^{0.5} = 173 m.$$ Note that the width of the Fresnel zone increases with depth. Thus, the resolving power of the seismic method decreases with depth. Under ideal circumstances, in data processing, migration will collapse the Fresnel zone w to about the dominant wavelength d. Since large base metal deposits commonly meet the criteria listed above, we conclude that in pronciple, such deposits should be resolvable to practical exploration depth. For exploration purpose, the choice of the dominant frequency f defines a important threshold, a minimum size ore body that can be resolved by the seismic method. As critical as size, shape and depth of a massive sulphide body are for the theoretical resolution of the reflection seismic method, are directional characteristics, geological setting and survey geometry for detectability. For example, the greater the dip of the ore body, the larger the source-receiver offsets are required to record the reflected wavefield. In pratice, this may require to center seismic surveys in the down-plunge direction away from the target. The effect of geological setting is most difficult to quantify and we recommend accurate forward modeling studies to be applied prior to conducting a field experiment. In the absence of any priory information on the geological setting, long continuous profiles with large source-receiver offsets should be acquired in order to record the reflected wavefield from dipping ore bodies. Otherwise, the elected acquistion geometry will act as a powerful dip filter. Finally, data processing requirements impose additional constraints: (1) the use of high frequencies requires small separation of sensors to avoid spatial aliasing, (2) the low signal-to-noise ratio in the crystalline environment requires digital recording equipment with large dynamic range, and (3) the need for large source-receiver offsets demands simultaneous recording of hundreds of sensors. Only recently, state-of-the-art exploration equipment has become available and affordable that meets these minimum requirements (i.e., Milkereit et al., 1994). \bye