Geotechnology: Paradigm Shifts in the Information Age J. Carlos - PowerPoint PPT Presentation

GeoCongress 2006 - Geotechnical Engineering in the IT Age Geotechnology: Paradigm Shifts in the Information Age J. Carlos Santamarina Georgia Institute of Technology

Information Technology - Synergism: microelectronics computers data storage and display sensors digital data analysis inverse problem solving numerical methods communications (cell phones - internet)

Interwoven History 1910's Fredholm: generalized inverse 1920's Consumer electronics (radios, electronic phonographs) 1930's Car radios and portable radios Digital computer 1940's Transistor at Bell Labs Digital signal processing starts Sony pocket-size transistor radio 1950's Integrated circuits at Texas Instruments Feynman: nano-technology Computers emerge 1960's Growth of digital signal processing: FFT algorithm Microprocessors: computers = chip 1970's Consumer electronics begin transition to digital Computerized tomography Personal computers & CD players, commercial cellular phones 1980's Texas Instrument: single-chip digital signal processor Micromachining Digital memory and storage 1990's IBM Deep Blue defeats G. Kasparov (1997) World wide web Submicron electronic devices 2000's More than 30 nano-technology research centers in the US.

Microelectronics – Moore's Law doubles: 24 months 1.E+09 Transistors per Chip Pentium IV Pentium III 1.E+07 Pentium & 80786 80486 80386 1.E+05 80286 8086 8080 4004 1.E+03 1970 1980 1990 2000 2010 Year data from Birnbaum and Akinwande

Storage doubles: 14 months 100000 1000 Kilobytes per dollar 10 0.1 0.001 2006: < $10/GB 0.00001 1950 1960 1970 1980 1990 2000 2010 Year data from Kurzweil

The brain - Storage each neuron stores 1 bit brain ~1 TB 10,000 $ each synapses stores 1 bit brain ~100 TB 1 million$ 2006 Computer Capabilities brain ~10 7 TB each molecule stores 1 bit 100 billion$

Calculations per second doubles: 19 months (Calculations/second) / $1000 100000000 1000000 10000 100 1 2006: 10 4 MIPS computers 0.01 Brain: 10 8 MIPS 0.0001 0.000001 1900 1920 1940 1960 1980 2000 Year data from Kurzweil; Moravec

Communications doubles: 10 months 100000000 10 MBytes per second per $ 1000000 MBytes per second 0.1 doubles: 10000 7 months 100 0.001 1 0.00001 0.01 wireless 0.0001 0.0000001 1940 1950 1960 1970 1980 1990 2000 2010 Year data from Kurzweil

Lenses: Paradigm Shifts Galileo geocentric heliocentric (pre-Copernicus) Telescope Leeuwenhoek sterile biotic Microscope

Observations Underlying technology: doubles every 7-to-24 months At present rate: computers ~ brain in 10-to-20 years How is our field changing? What are possible paradigm shifts?

Building Blocks Sensors Signals Inversion Content Databases

Nano and Micro Technology Sensors - MEMS

Nano-Control Nano-manipulation (Eigler 1990) Montmorillonite (MDL) 9.6 Å H H H H Surface control NaPAA C C C C H H C C O O O O Na Na

Micro-electrical mechanical systems MEMS Cantilever displacement sensor Yaralioglu et al

Micro-electrical mechanical systems MEMS Micro-mirror array Bell Labs

Fiber optic based pressure transducer 0-to-70 kPa to 0-to-7 MPa www.fiso.com

Distributed Optical TDR Sensors laser Signal Processing Strain (Dowding) Pore fluid chemical properties Moisture content (Brillouin - Pamukcu) Temperature (Raman - SENSA) 30 km … every 1 m … 1 o C resolution

Soil = innate sensing system 0 0 1 1 2 2 3 3 4 4 5 5 6 6 7 7 al al Meausred Sign Meausred Sign 8 8 9 9 10 10 11 11 12 12 (N. Skipper – UCL 2002) 13 13 14 14 15 15 16 16 0 0 100 100 200 200 300 300 400 400 500 500 600 600 700 700 Time [microsec] Time [microsec]

Signal Processing Data Fusion

Signals → Information Katrina (8/29) Dennis (7/4) Emily (7/9) Rita (9/23) 2 Water Level [m] 1 0 -1 0 10 20 30 40 50 60 70 80 90 100 7/1/05 Days 9/30/05 Pilots Station, Louisiana – NOAA

Before Katrina

After Katrina NSF - D. Fratta

NSF - D. Fratta

Massive data → Display → Information Pile 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Biloxi I-110 Bridge D’Iverville Bathymetry: 200 kHz Sub bottom profiling: 20 kHz NSF - D. Fratta

Data Fusion: Same Mode Fuse multi-sensor data to gain new information http://www.pc.rhul.ac.uk/zanker/teach/PS1061/L6/braille.JPG

Data Fusion: Multi Mode Navigational Homing in sunsite.tus.ac.jp/multimed/pics/animals/bat.jpg www.moorhen.demon.co.uk

Cementation - Elastic waves 700 sand-cement 600 500 Vs (m/sec) 400 300 σ ’= 70 kPa 200 100 0 10 100 1000 10000 Time (min)

Cementation - Electromagnetic waves 1.2 1 0.8 o k'/k' 0.6 0.4 0.2 0 1.5 1 k" eff /k" effo 0.5 bentonite-cement 0 0.0001 0.001 0.01 0.1 1 10 time [days]

Observations Signal processing = information extraction noise control similarities between signals simple algorithms may be sufficient Data fusion = new information from: multiple same-mode sensors multi-modal sensors spatially distributed sensors concurrent or time-shifted data streams

Inversion Sensing at boundaries … learning about the body

From CAUSE to EFFECT q d forward z t

From EFFECT back to CAUSE ? q d inverse z t

Tomography ? Unknown internal conditions

S 3 S 4 ∗ ∗ invert R 1 S 1 ∗    h h 0 0    t 1 / V 1 1 , 1 1 , 2 1       t 0 0 h h 1 / V 1 2       2 , 3 2 , 4 2 = ⋅ 2       h 0 h 0 t 1 / V 3 , 1 3 , 3 3 3       0 h 0 h t 1 / V       4 4 , 2 4 , 4 4 S 2 ∗ R 2 4 3 R 3 R 4

Micro Computed Tomography Alshibli - www.eng.lsu.edu

Inversion: Ubiquitous in Geotechnology Measured Values Inverted Values triaxial F- δ constitutive model parameters oedometer u(t) C v k pollutant c(z,t) location and timing of leak V Rayleigh ( ω ) V s (z) from SASW settlement f(t) C v C s δ h (z) along a pile k h (z) along the pile ground vibration evolution of G during event Conceive all e xperiments within inverse problem solving framework

Distributed Content Development many + internet = collective intelligence

Great Backyard Bird Count Northern Cardinal (2/17/06 - 2/20/06) Responses: 31,515 www.birdsource.org/gbbc

Community Internet Intensity Map Nisqually 2/28/2001 did you feel it? www.usgs.gov

Wiki-Geo-Pedia? "Thousands of people, all over the world, from all cultures, working together in harmony to freely share clear, factual, unbiased information… [with the] simple and pure desire to make the world a better place." Wikipedia Founder Jimmy Wales

Observations Distributed sensing Many not necessarily "sophisticated sensors" Specific task / protocol Proper data gathering / transfer Distributed content development Unprecedented opportunities Development of large databases New information… new understanding… new questions…

Databases From data to knew understanding

To identify the critical parameters Risk of heart complications (Database: 10,682 patients - 7 hospitals) Q-waves in electrocardiograms low systolic blood pressure abnormal respiratory sound with fine crackles exacerbation of known reduced blood flow to the heart Better practice/diagnosis Lower cost Enhanced understanding Guide to further research

To identify the n th control variable 1.4 maximum e max 1.2 1.0 0.8 0.6 0.8 minimum e min 0.6 0.4 0.2 1 2 3 4 5 6 10 coefficient of uniformity, C u Youd

To explore causal relations rotational frustration (e ↑ ) ? vs. chain collapse (e ↓ ) 50 CS friction angle cv 40 30 20 φ = − ⋅ 42 17 R cv 10 0 0.2 0.4 0.6 0.8 1 Roundness R

Spatial Systematic Organization Mendeleev (1860's) Bronowski

Spatial organization + analyses: GIS Paris www.brgm.fr

Paradigm Shifts The future ain’t what it used to be … Yogi Berra

"inert soils" → "self-sensing media" 1000 1000 β β σ σ + + σ σ     ' ' ' ' x x y y     α α V V = =     S S 2 2 P P     a a V s [m/s] V s [m/s] 100 100 10 10 100 100 1000 1000 σ ’ v [kPa] σ ’ v [kPa]

A1 C1 A2 C2 A8 C8 B1 B2 B6

V s (m/s) 35 50 65 80 95 110 >125 Pixel Parametric (RLSS) (L-norms) Fernandez, Lee

"n-simple tests" → "one information-rich test" q q p' p' q q p' p' See also A. Rechenmacher: spatial variability