Recharge rates to deep aquifer layers estimated with 39 Ar, 85 Kr and - PowerPoint PPT Presentation

Recharge rates to deep aquifer layers estimated with 39 Ar, 85 Kr and 14 C data: A case study in Odense (Denmark) Roland Purtschert University of Bern, Switzerland Troels Bjerre, Johann Linderberg VCS Denmark Klaus Hinsby Geological Survey of Denmark and Greenland 1

Starting point, Objectives Graphics, USGS Increasing interest in pre-modern groundwater due to  Overexploitation F ast circulation, residence times of decades → vulnerabe  Contamination Slower circulation, RT of centuries -> less vulnerable?  Effects of climate change  Investigation of potential and limitations of tracers beyond the 50 years age limit.  Determination of recharge rates, Deep groundwater residence times and renewal rates  Combination of tracer methods and numerical modelling

Study site: Odense river catchment, Funen Denmark Photo Jan Kofoed Winther Total area 1046 km 2 Direction of groundwtaer flow  Odense Water Ltd is one of the largest water utilities in Denmark with a Well fields groundwater abstraction permission of ~14 mill. m³/year  The groundwater is mainly abstracted from rather shallow aquifers (quaternary sand deposits) on 7 wellfields around the city of Odense.  The catchments of the wellfields are dominated by conventional farming and urban areas  The groundwater table is close to the surface and the low lying areas are 0 5 10 generally drained by tile drains kilometers

Topography The catchment area is characterised by two high areas rising to 130 m above sea level divided by a wide and shallow depression stretching NE – SW from Hansen et al, 2009

Hydrogeology Sand/gravel Clayey till Sandy till Fractured Clay Chalk  Several ice advances and subsequent ice retreats during the Weichselian glaciation have formed the present landscape and geology (Houmark-Nielsen & Kjær 2003)  Main aquifers are found in semi-confined units of glaciofluvial sand and gravel deposits at varying depths overlain by glacial till.  Tertiary marls and clay forms the lower boundary of the Quaternary aquifer system  Geology of the quaternary deposits is rather complex and heterogeneous  Interconnected hydrostatigraphic units with a typical thickness of 10-15 met er Troldborg, 2004

Timescales of groundwater dating methods (Half-life) Methods (10.7 yr) (269 yr) years Dating range Sampling in GAB, Australia 6

85 Kr -39 Ar: Key data 85 Kr 39 Ar  Half-life: 10.7 269 b b  Decay mode:  Modern isotope ratio (2008) ~3 · 10 -11 8.1·10 -16  Atoms/Liter water 58 ‘ 000 8 ‘ 700  Activity (Bq/L water) 7.1 · 10 -7 → 22 decays/yr 1.2 10 -4  Input function Water sample volume: 1-2 tons! Note: Water residence times are based on a isotope ratio ( 39 Ar/Ar, 85 Kr/Kr etc) and are therefore (rather) insensitive to  Details of recharge conditions (the addition of excess air, recharge temperature)  Degree of degassing (both in nature and during sampling) (In contrast to 3 H/ 3 He, SF 6 etc that are based on absolute concentrations) 7

Sampling and detection method 39 Ar and 85 Kr activity measurement by Water degassing in the field Low Level Counting Active and passive shielding in underground lab Sampling for 85 Kr, 39 Ar and 14 C

39 Ar production Subsurface secular equilibrium 300 n Atmospheric Production 4n  + 300%mod  0  40 Ar(n,2n) 39 Ar 250  2  -    p 0.104 dpm/L Argon  n + (100% modern) e - 200  - e + %modern  + e + T 1/2 =269 years n e - 200%mod  0  -   - 150 p n n 100%mod  n 100 p p  - 50%mod n n  n t 50  p  - h  - 0%mod + n n e - 0 cosmic rays  - 0 200 400 600 800 1000 n th groundwater residence time (years)  n p  n n n p n Spontaneous fission Subsurface Pore- U (n) Production Water Th 39 K(n,p) 39 Ar Al, Mg..( α ,n)

39 Ar depth profile 0 20 screen depth (m) 40 0.15m/yr 60 80 100 25 50 75 100 125 150 175 200 39 Ar (% modern)

39 Ar spatial distribution 0 39 Ar (%modern) 181.5 Lunde 180.0 20 160.0 140.0 screen depth (m) 120.0 40 Odense Northing 100.0 80.00 0.15m/yr 60.00 60 40.00 Borreby 80 Soeby Holmehaven 100 25 50 75 100 125 150 175 200 Easting 39 Ar (% modern)

39 Ar- 85 Kr depth profile 0 0 20 20 screen depth (m) modern water 40 40 0.15m/yr 60 60 80 80 100 100 0 10 20 30 40 50 60 70 80 25 50 75 100 125 150 175 200 85 Kr (dpm/cc Kr) 39 Ar (% modern)

39 Ar- 85 Kr depth profile 0 0 DL 20 20 screen depth (m) 40 40 0.15m/yr 60 60 80 80 100 100 0 2 4 6 8 10 12 14 25 50 75 100 125 150 175 200 85 Kr (dpm/cc Kr) 39 Ar (% modern)

Single well age gradient 39 Ar (%modern) 50 55 60 65 70 75 80 85 90 95 50 55 Top screen V mean : 0.25 m/yr 60 65 depth (m) 70 Bottom screen 75 80 85 90 95

Expected age distribution and dispersion in heterogeneous alluvial aquifer system WATER RESOURCES RESEARCH, VOL. 38, 2002 A wide rather than a piston piston-flow-like age distribution can be expected because of  The heterogeneity of the system  The spatially distributed recharge  Mixing in the extended screen intervals of the extraction wells

Age distribution: 39 Ar and 85 Kr data 20 Assumed Age Distributions 120 (T m =100 yr) 18 Data 16 AD1 proportion AD3 AD3 85 Kr (dmp/cc Kr) 14 20 30 12 AD2 10 AD1 40 AD2 8 6 50 30 0 100 200 300 400 60 4 70 age (yrs) 40 90 2 Detection limit 85 Kr 120 50 120 80 70 0 20 40 60 80 100 120 39 Ar (%modern)  Measured 85 Kr and 39 Ar activities can consistently be interpreted if dispersive mixing is taken into account (AD2-AD3)  Modern 39 Ar values in samples low in 85 Kr are suspicious for produced underground

Age spectra 39 Ar- 14 C 39 Ar (35 samples) 16 14 12 10 8 # 6 350 yrs 4 2 Explanation? 0 0 500 1000 1500 2000 4 14 C (7 samples) <2000 yrs 2 # 0 0 500 1000 1500 2000 Correted 14 C ages (F&G Model)

Geochemical correction of 14 C activities 120 Geochemical Evolution 100 0 50 80 39 Ar (%modern) 100 2 Komponentmixing? 60 200 300 40 400 piston flow 500 ages Geochemical correction 600 20 800 1000 decay curve 0 5000 20 25 30 35 40 45 50 55 60 14 C (pmC)  Two component mixing is not consistent (or at least very unlikely) within the hydro-geologial context  Geochemical correction models are not sufficient to eliminate the discrepancy between 39 Ar and 14 C ages

Diffusive exchange with stagnant zones (Sudicky, 1981) Aquitard layers Aquifer layers 120 100 0 50 80 39 Ar (%modern) 100 60 200 300 b D  40 L L  400 piston flow a 500 ages 600 20 Aquifer(active flow zone) + diffusion 800 Aquitard (stagnant) 1000 decay curve 0 5000 20 25 30 35 40 45 50 55 60 Time Steady State 14 C (pmC) Averages estimates (System of parallel layers; Sanford, 1997) Porosity (Active & stagnant) : 0.3 Thickness flow zone: 20 m Thickness stagnant zone: 40 m Diffusioncoeff. ( 14 C): 3.15 10 -3 m 2 /yr Diffusioncoeff. ( 39 Ar): 2.6 10 -3 m 2 /yr

Mixing-Dispersion (T m =100 yr) 120 AD1 proportion AD3 100 0 50 80 39 Ar (%modern) 100 AD2 60 200 0 100 200 300 400 300 age (yrs) 40 400 piston flow 120 500 ages 600 20 + dispersion + diffusion 100 800 1000 decay curve 80 0 5000 Activity 20 25 30 35 40 45 50 55 60 60 14 C (pmC) 40 20 Dispersive mixing reduces the apparent decay rate of 39 Ar relative to 14 C 0 0 500 1000 1500 2000 Time

Subsurface Production of 39 Ar Assumed subsurface 120 secondary equilibrium: 100 0 30%modern Tracer model 50 ages 100 80 39 Ar (%modern) 100 190 280 60 200 370 460 + underground production 300 550 40 400 2000 1000 piston flow 500 ages 600 20 + dispersion + diffusion 800 1000 decay curve 0 5000 20 25 30 35 40 45 50 55 60 14 C (pmC) Summary: The combined contribution of isotope exchange with the aquifer rocks, diffusive exchange with aquitards, dispersion and eventually underground production resolves the discrepancy between 39 Ar and 14 C ages: The resulting age span ranges between recent and 500-700 years (and not up to 2000 years as indicated by 14 C data)

Age spectra 20 39 Ar ages Uncorrected 15 80±84 10 5 0 0 100 200 300 400 500 600 700 800 20 14 C- 39 Ar tracer ages 15 Corrected 10 240±270 5 0 0 200 400 600 800 20

Age spectra 20 39 Ar ages Uncorrected 15 80±84 10 5 0 0 100 200 300 400 500 600 700 800 20 14 C- 39 Ar tracer ages 15 Corrected 10 240±270 5 0 0 200 400 600 800  Similar age range 20 years  Discrepancy at the lower age 15 196±90 limit 10 Numerical particle tracking ages 5 0 0 100 200 300 400 500 600 700 800

Stable isotopes d 18 O- d 2 H d 18 O (‰) d 18 O (‰) -8.7 -8.6 -8.5 -8.4 -8.3 -8.2 -8.1 -8.0 -7.9 -8.60 -8.50 -8.40 -8.30 -8.20 -8.10 -8.00 -7.90 -53 0 18 d O d C      ‰ / °C 0.56 T 0.8 C Time 10  18 d O -54 m     ‰ (Schrag, 1996) 0.28 h 150 m 20 100 50 yr mean depth below surface (m) (Poage , 1996) -55 30 40 d 2 H (‰) -56 50 300 yr -57 60 70 -58 80 600 yr -59 90 100 -60 IPCC 2001, Mann et al, 1999