Understanding the subsurface better is on many grand challenges lists.  The interest  is not new, and much progress has been made over the decades to better see into the earth.  Nevertheless the perception continues to be that only through actual drilling or excavation can one for sure know the subsurface in detail.  Such drilling and excavation leads to exceptional expense, disruptions, and often environmental impacts.  This author believes that understanding the heterogeneity of the rock mass physical and chemical complexity including discontinuities such as layer interfaces, fractures, and inclusions is the highest priority for better understanding of the subsurface.  The need for better understanding of the subsurface is driven by economic, environmental, and public perception considerations.  Better subsurface characterization is important.

            Introduction:  Dr. Charles Fairhurst made an analogy of space exploration to underground exploration.  Unfortunately, to the lay person space may be quite glamorous while the underground is--well it's underground.  For those that get past what's glamorous and what's not, and think of what is important to mankind, the comparison of underground exploration and space exploration becomes meaningful.  We could not exist without harvesting underground resources, maybe the most visible are water and today fossil energy sources, not to mention harvesting minerals required to build bridges, cars and airplanes, minerals to generate and carry electricity, minerals to make windmills and batteries to store electric energy, harvesting underground space for storage and to dispose of certain wastes including toxic substances, and more.  And not to mention the impact of the shifting subsurface on inhabited areas including landslides and earthquakes, and the National security applications both defensive and offensive.  And the list goes on.

            Glamorous or not to the lay person, better subsurface characterization seems to be on any well thought out great challenges list that scientists and engineers would generate.  For example, the NAE Fourteen Goals, Grand Challenges contains six areas where the earth’s subsurface plays a significant role.  More thought provoking, however, is what is needed for better subsurface understanding and how this would be achieved. 

            Programs and Previous Recommendations:  The current Department of Energy ("DOE") SubTER project (Subsurface Technology and Engineering Research, Development, and Demonstration) is a good example of moving forward toward better subsurface characterization.  This is a "crosscutting" interagency program aimed at US energy common technical challenges.  It focuses on major technical and socio-political issues regarding energy production and storage and the management of waste in the subsurface.  There are other "field laboratory" type subsurface programs underway.  The DOE Frontier Observatory for Research in Geothermal Energy ("FORGE") program is aimed at a field demonstration.  The National Energy Technology Laboratory ("NETL") test wells are field, subsurface sites.  The Department of Defense ("DOD") has used many field sites for large experiments over the decades.  And there are others.

            A 2000 National Research Council ("NRC") report, "Seeing into the Earth", addresses noninvasive methods to investigate the subsurface.  Its first recommendation, "Scientists and engineers should improve their ability to integrate multidisciplinary data for modeling, visualizing, and understanding the subsurface", is indeed profound.  The power of multidiscipline approaches is huge.  A follow-up 2001 NRC report is, "Conceptual Models of Flow and Transport in the Fractured Vadose Zone".  The conclusions speak to further research needed [to better characterize the subsurface] including development and testing of conceptual models specifically related to fluid flow and transport. 

            An earlier 1978 NRC report, "Limitations of Rock Mechanics in Energy-Resource Recovery and Development" addressed the role of the "rock" for problems involving the underground.  One might note "that's not very profound" as the rock is the underground, but recently individuals pursuing oil/gas recovery from the shales have restated "rocks matter" as thought this is something new.  And a follow-on 1981 NRC report, "Rock-Mechanics Research Requirements for Resource Recovery, Construction, and Earthquake-Hazard Reduction", is a follow-on to rock mechanics research needs.  The conclusions speak to "insitu" modeling of the "rock mass"--i.e. subsurface characterization stated in a different way.

            Several American Rock Mechanics Associations ("ARMA") meetings have addressed subsurface characterization directly or indirectly.  The 1998 ARMA "Asilomar" forum, "New Directions in Rock Mechanics" considered fractured rock mass--weak rock formations--strength and fluid flow conditions.  The conclusions emphasis was on better modeling requiring better characterization--i.e. better subsurface understanding. 

            Several National Science Foundation ("NSF") sponsored ARMA workshops/studies addressed the underground research requirements.  "Workshop on Industry-Government-University Partnership in Rock Mechanics and Rock Engineering: Challenges and Opportunities", 1999; "Deep Underground Science and Engineering Laboratories (DUSELs) in Conjunction with the 10th Congress of the International Society for Rock Mechanics in Johannesburg, South Africa in 2003; and "Engineering Research Opportunities in the Subsurface:  Geo-hydrology and Geo-mechanics" 2003 by Derek Elsworth and Charles Fairhurst.  All emphasize characterizing the rock mass with laboratory and field experiments, and overall to develop a better subsurface understanding.

            The American Rock Mechanics Association has the vision of the need for better subsurface understanding.  ARMA has created a Fellows member category of experts who have extensive experience in the mechanical behavior of the earth’s subsurface and understanding of engineering applications of rock mechanics.  This group of ARMA Fellows can offer technical advice to the public, and much more, regarding understanding the subsurface.

             Geology, Geomechanics, Geophysics, Geochemistry, Rock Mechanics:  It is worth dwelling on technologies a bit, as often there is a mixed understanding of geology, geomechanics, geophysics, and rock mechanics--all are important.  Geology is the rock, the rock matrix and the rock mass.  Geomechanics is the application for solving a problem--such as borehole stability analysis or slip along a natural fracture or fault.  Geomechanics requires as input the rock properties and the insitu stresses (as noted above).  To some extent this may be viewed as "an end product", the "answer".  Geophysics most often may be associated with exploration of the earth, including seismic, borehole logging, electro, magnetic, or gravity measurements, and certain insitu stress tests and analysis. This is looking into the earth and such measurements are absolutely critical.  But, "voltages" obtained in the measurement instruments alone aren't so helpful unless they can be calibrated to definite understandable properties like stiffness, strength, density and porosity, pore fluid saturations, and the like. That requires geology, geomechanics (for example to estimate insitu stresses)--and, rock mechanics.  Geochemistry involves the very important rock-fluid interactions.  Rock mechanics most often may be associated with physical measurements of the rock matrix--most often laboratory measurements. 

            Even having noted the above, it is realized that to some extent such differentiation is like art--in the eye of the beholder.  Characterizing the subsurface--seeing into the earth--involves geology, geomechanics, geophysics, geochemistry, and rock mechanics.  In fact, one alone will not characterize the subsurface.  The 2000 NRC report, "Seeing into the Earth", recommendation to integrate multi-disciplinary data for understanding the subsurface is indeed profound. 

             What is Missing for Understanding the Subsurface:  When one asks what are the number one issues missing from subsurface understanding, what has emerged to this author is understanding the heterogeneity of the underground.  Some would say, characterization of the rock mass, in addition to characterization of the rock matrix.  That is, defining the physical makeup of the rock mass including the ever present pore fluids that always exist.  Second is identifying the insitu stresses that exist down to the local scale, which must be complicated and variable for highly heterogeneous formations. To model the earth one must know the material properties and the stress loadings that exists (including the pore pressure).

            Characterizing the rock mass as well as the rock matrix is not new to the rock mechanics "real world" practitioners.   However, the relatively new recovery on a very large scale of oil and gas from unconventional formations has brought to the forefront the importance of understanding heterogeneities and insitu stresses.  The inventions that have made oil and gas recovery from shales possible is horizontal well drilling and hydraulic fracture completions of the horizontal well. Industry has become incredible efficient at the operations of drilling and fracturing.  But industry has accomplished this at the great expense of drilling, drilling, drilling which has allowed the details of the shale rock and of the insitu stresses to small scale to be effectively "calibrated out".  The same can be noted for mining applications, where "drilling" is always a large part of the budget as an ore body is being identified.  Drilling, drilling, drilling is always a large part of any underground characterization.

            Digging Deeper:  As one 'digs deeper' into what is missing--not deeper into the earth, but deeper into what are the issues--three details emerge.  These issues have to do with 1) calibrating the geophysics and geochemistry measurements to quantify usable and understandable quantities, 2) measuring to the small and large scale of things, and 3) determination of the insitu stresses.   And all of these become more complicated for highly heterogeneous formations.

            First, is the calibration matter.  Rock mechanics (including core analysis and cuttings analysis) tends to be the vehicle most often quoted for calibration of geophysics and geochemistry measurements.  There is a big problem here however, in that rock mechanics tests tend to be performed on extracted rock samples, rocks that have been extracted from the earth--most often either drill cuttings, core, or outcrop samples.  Extracting the rock sample--like coring--will at least to some extent damage the sample mechanically, alter the pore fluid, change the temperature, and will relieve the insitu stresses.  The latter means that the laboratory samples will be stress relieved, expanded, at the minimum.  Layer interfaces and fractures will be changed.  Stress conditions that exist from one layer to another will be changed and interface shear stress and/or displacement correspondingly will be changed.

            To gain insitu properties, standard rock mechanics practice is to put the samples back to "insitu conditions", something the author and others worked fifty years ago.  And for many formations this is done by simply reapplying loads to simulate the insitu stresses, adjustment of the pore pressure, and heat or cool the sample to the insitu temperature.  Standard rock mechanics practice usually would assume homogeneous conditions within the sample--whatever the size, assuming a statically determinant test.  It doesn't take much imagination to see the problems for heterogeneous geologically complex formations.

            Second, is the matter of the scale.  Critical features may be at a very small scale as for the shales mentioned above, or at a large scale such as defining an ore body or an oil/gas play.  Geophysical measurements tend to be at a large scale but average over a significant volume of rock; for example seismic may average over meters and logs may average over many centimeters.  Unfortunately a layer interface or a filled natural fracture is at a scale too small to be discretely detected.  [High resolution logs may measure to the millimeter scale, but sample very close to the wellbore wall, where drilling induced damage and stress change occur.]  At the pore scale, features may be in the nano-meter scale as for the shales.  So we see the scale is from the very small nano-meter scale to the very large kilometer scale.  If we look at the small scale our field-of-view is small, maybe 50 microns or so; if we look at the large scale our field-of-view is large but our resolution is low, meters for seismic to tens of centimeters for well logs.

            The "scale of things" also comes into play in rock mechanics testing, in that practical laboratory size tests are limited as are the size of sample that can be extracted for laboratory tests.  And, even a large block laboratory test (using outcrop samples) of about a meter performed by a few laboratories may not be large enough to represent the rock mass in bulk nor the phenomena being observed.

            Thirdly, the insitu stresses to the local scale tend to be a big unknown.  Insitu stresses can't be "measured"; they must be inferred by stress relief strain measurements, rock failure observations calibrated with hydraulic fracturing and strength/stiffness understanding, or measurements that are calibrated by laboratory tests like sonic velocities. 

            Indeed "digging deeper" uncovers major issues with respect to characterizing the subsurface.

            Conclusions:  Characterizing the subsurface is not a new issue.  However, what is new is that the need for better characterization of the subsurface is today driven more than ever by strong economic considerations, by important environmental considerations, and by public perception considerations.  The economic considerations need not be repeated here; they are indeed very obvious and they are very large.  The environmental considerations have become stronger drivers with increased population, with increased awareness, and with increased resource exploitations.  The environmental issues where the underground is a 'player' are obvious and they are extensive--oil/gas recovery, induced seismicity, water availability and delivery, nuclear waste and other toxic waste storage, carbon sequestration, geothermal energy recovery, national security applications, and the list goes on.  Public perceptions considerations are driven by uncertainty by the unknown.  Uncertainty in knowing the subsurface leads to public perception problems.  As the author was once told, "It's not the facts that count, but what is perceived to be the facts that count".  And what is perceived to be the subsurface is not necessarily the subsurface.

            The largest subsurface unknown has to do with the heterogeneity of the underground; the variability, the discontinuities that exist, the pore fluids that exist, and the insitu stresses.  As was once said, "We know everything about the ground [subsurface] except for understanding the rock and understanding the insitu stresses--but we know everything else". 

The focus needs to be on how to ‘know’ the rock and how to infer the insitu stresses better.  At present industry tends to "calibrate out" the rock and the insitu stresses by drilling, drilling, drilling (for oil/gas, for mining, for civil constructions).  We need an easier less invasive way to see into the earth with more detail.