Hydrogeophysics: Expectations, limitations and challenges

Andrew Binley, Lancaster University, UK

EAGE Near Surface 2005 Workshop on Hydrogeophysics, Palermo 2005

The value of hydrogeophysics

Geophysics has been widely used to support groundwater investigations for many years*. However, much of the earlier approaches concentrated on using geophysics to define lithological boundaries and other subsurface structures.

* For example, Zohdy et al. (1974), Application of surface geophysics to ground-water investigations, USGS.

During the 1990s there was a rapid growth in the use of geophysics to provide quantitative information about hydrological properties and processes.

Much of this was driven by the need to gain information of direct value to hydrological models, particularly given the developments of ‘data hungry’ stochastic hydrology tools.

The value of hydrogeophysics

Perhaps more significant is that there is a clear demand by government regulators and agencies for tools and technologies to allow characterisation of groundwater systems, for example linked to the EU Water Framework Directive

Advantages

Geophysics offers advantages over conventional sampling to the hydrologist because of:

High data sampling density

Larger measurement volume – more consistent with modelling needs

Relative lower cost of measurements – may avoid use of boreholes and/or allow quicker sampling

Minimally invasive – may allow investigations without affecting the hydrology of the system

0

10

20z

(m)

20 30 40 50 60 70 80x (m)

structure(e.g. hydraulic conductivity)

process(e.g. solute transport)

Time-lapseimaging

Geophysicalimaging

hydrologicalmodel

petrophysicalmodel

petrophysicalmodel

Hydrogeophysical approach

After Kemna (2003)

0 5 10

0

5

10

15

20

25

30

35

40

Distance (m)

Dep

th (m

)

(µS/m)1 10 100

σ ′′1 10 100

0 5 10

0

5

10

15

20

25

30

35

40

Distance (m)

(mS/m)'σ

Gamma (c.p.s.)

200

400

600

Sand/Gravel

Sand/Silt

Silt/Clay

Sandstone

Gamma (c.p.s.)

200

400

600

real electricalconductivity

imaginary electricalconductivity

Kemna, Binley and Slater (2003)

Structural characterisation example

Complex resistivity at the Drigg nuclear site, UK

day

electrical conductivitychanges (%)

Kemna, Vanderborght, Kulessa and Vereecken (2002)

Tracer experiment at the Krauthausen test site

Process characterisation example

Expectations

The hydrologist may expect the following:

1. Coverage over a large area at high resolution

2. Significant depth penetration

4. That the geophysicist will use the most appropriate method available

3. To make use of existing infrastructure

5. An (error free) image of the hydraulic property that they are interested in

Expectations versus reality

The hydrologist may get:

1. Coverage of a small plot of the site of interest

2. Limited depth penetration due to surface cover and conditions

4. The geophysicist used the method that he/she is most familiar with

3. Gaps in coverage or anomalies due to steel cased boreholes, for example

5. An image of a geophysical property (with unquantified uncertainty) that is somehow related to a hydraulic property

Expectations – essential communication

It is important, therefore, to communicate in order to establish:

1. What exactly does the hydrologist want ?

2. Does geophysics offer any solution ? There may be a better solution.

4. How will the geophysical property be related to what the hydrologist wants ? Can this be quantified ?

3. What exactly is realistic given the site conditions (geology, access, cover, etc.)

5. Is it possible to determine some level of uncertainty in the results that are communicated ?

Limitations – hydrogeophysical relationships (1)

For static imaging there must be a contrast in a geophysical property that can be related to hydrological parameters

This may be site specific

Binley, Slater, Fukes and Cassiani (2005)

Spectral induced polarisation of Triassic Sandstone

1m

Pore size (µm)

Cumulativeinjectionvolume(ml/g)

EC7-5EC15-5EC16-1

1000 100 10 1 0.1 0.010

0.04

0.08

0.12

0.16

0.2

0.01 0.1 1 10 100 1000 100000

10

20

30

40

50

-10

-20

-30

Frequency (Hz)

ρ(Ωm)

φ(mrad)

VEC16-1 depth = 17.61 mVEC15-5depth = 16.07 mVEC7-5depth = 8.22 m

ρ

φ

Limitations – hydrogeophysical relationships (1)

Limitations – hydrogeophysical relationships (1)

Is there a universal relationship ? Should there be one ?

0.0001 0.001 0.01 0.1 1 10

10-8

10-7

10-6

10-5

10-4

10-3

10-2

10-1

100

Graph 1Binley et al sandstoneLesmes sandstone

Spectral IP relaxationtime τ (s)

Hydraulic conductivity (m/d)

Time-lapse measurements may be easier to interpret in order to study processes but appropriate petrophysical relationships are needed

Limitations – hydrogeophysical relationships (2)

Moisture content changes due to natural inputs

Binley, Winship, Middleton, Pokar and West (2001)

0.06

-0.06

0Change inmoisturecontent

0 1 2 3 4 5

23456789

1011

29-Jul-99

Dep

th (

m)

0 1 2 3 4 5

23456789

1011

15-Sep-99

0 1 2 3 4 5

23456789

1011

22-Apr-99

0 1 2 3 4 5

23456789

1011

12-Mar-99

Dep

th (

m)

Distance (m)

0 1 2 3 4 5

23456789

1011

14-Dec-99

0 1 2 3 4 5

23456789

1011

11-Jun-99

TransmitterReceiver

Limitations – hydrogeophysical relationships (2)

Petrophysical relationships may be site specific and also highly uncertain

West, Handley, Huang and Pokar (2002)

Volumetric moisture fraction

Relativepermittivity

Limitations – hydrogeophysical relationships (2)

Limitations – variability of sensitivity in an image

The sensitivity of the imaging varies within the image – in some areas uncertainty in the geophysical property is high – will lead to mass balance errors in tracer studies

High confidence

Low confidence

Dep

th (

m)

Distance (m)0 2 4 6 8

-14

-12

-10

-8

-6

-4

-2

0

0 2 4 6 8 10 12 14

-14

-12

-10

-8

-6

-4

-2

0

Distance (m)

Wide borehole spacingNarrow borehole spacing

Binley, Winship and Gomez (2005)

Vadose zone tracer experiment Hatfield, UK – ERT results21-Mar-03

Dep

th (

m)

Distance (m)

Dep

th (

m)

Distance (m)

02-Apr-03

15-Mar-03

Dep

th (

m)

Distance (m)

Dep

th (

m)

Distance (m)

24-Mar-03

Dep

th (

m)

Distance (m)

Distance (m)

27-Mar-03

16-Mar-03

Dep

th (

m)

Distance (m)

Distance (m)

Distance (m)

Distance (m)

Distance (m)

Distance (m)

H-E3

H-E4

Tracer array

H-E

2H

-R2

H-R1

H-E1

H-I2

Limitations – variability of sensitivity in an image

Limitations

The sensitivity of the imaging varies within the image and is dependent on the technique and may be a function of the geophysical parameter

Day-Lewis, Singha and Binley (2005)

ERT

Radar

Limitations – variability of sensitivity in an image

Limitations – imaging artefacts

Data inversions can be strongly affected by regularisation (needed to get stable solutions)

This may lead to hydrologicallymeaningless results or may give the false impression of something hydrologicallymeaningful

0 2 4 6 8

-14

-12

-10

-8

-6

-4

-2

0

0

20

40

60

80

100

120

140

160

180

200

Elev

atio

n (m

)

Distance (m)

Resistivity(Ωm)

Limitations – impact of noise on images

Data inversions are strongly affected by data and modelling noise levels and these must be characterised for accurate assessment of geophysical properties

We can ignore this for ‘anomaly hunting’ but not if we want toget quantitative information from theimages

0 2 4 6 8

-14

-12

-10

-8

-6

-4

-2

0

0

20

40

60

80

100

120

140

160

180

200

Elev

atio

n (m

)

Distance (m)

Resistivity(Ωm)

Inversion of synthetic datawith 5% error added but

2% error assumed

Limitations – measurement scale

We need to ensure that measurement and model scales are comparable

0.08 0.12 0.16 0.2

Moisture content (-)

16

14

12

10

8

6

4

2

0

Dep

th (

m)

0.08 0.12 0.16 0.2Moisture content (-)

16

14

12

10

8

6

4

2

0

Dep

th (

m)

(a) (b)

0.08 0.12 0.16 0.2Moisture content (-)

16

14

12

10

8

6

4

2

0

Dep

th (

m)

0.08 0.12 0.16 0.2Moisture content (-)

16

14

12

10

8

6

4

2

0

Dep

th (

m)

(a) (b)

0.08 0.12 0.16Moisture content (-)

16

14

12

10

8

6

4

2

0

Dep

th (

m)

(c)

Cassiani and Binley (2005)

No averagingof model values

2m averagedmodel values

5m averagedmodel values

model

Radardata

Challenges

This workshop will allow us to assess the challenges we face but here are a few to start with:

1. Better understanding of the processes that lead to some geophysical signals (e.g. IP, SP)

2. Development of multi-geophysics and hydro-geophysics data fusion techniques

3. Improved assessment of uncertainty in geophysical and hydrogeophysical results

4. Advances in techniques for large area coverage (in complex terrains) to allow better watershed characterisation