The source of trace elements in groundwater in sandy aquifers

1
The source of trace elements in groundwater in
sandy aquifers

• Marc J.M. Vissers
• Faculty of geosciences

2
Why trace elements in groundwater

• Geochemistry
• Redistribution of trace elements (ore and natural
  anomalies)
• Global biogeochemical cycle
• Environmental science
• Atmospheric pollution / acidification
• Agricultural pollution / acidification
• Consumption (direct and indirect)

3
This talk Environmental geochemistry

• Study area and processes
• Present a 3-step approach for interpretation
• 1 Equilibrium modeling approach
• 2 Coprecipitation- codissolution approach
• 3 New Steady-state input approach

4
Study area and processes Map of the study area

• Sandy, unconsolidated aquifer, with ice-pushed
  ridge in the east
• Mainly Agricultural land use, eastern part
  cultivated in the 1920s
• 10 Borings, total of 244 mini screens

5
Study area and processes Cross-section of the
study area

• Filtrated over 0.45µm, analyzed on ICP-MS
• Sampled in 1989 (no trace elements), 1996 (½),
  and 2002 (all)
• Randomly analyzed on gt 70 (mostly inorganic)
  parameters

6
70 elements for 10 wells x 25 screens 1
Equilibrium modeling approach 2
Codissolution-coprecipitation approach 3 New
Steady-state input approach
7
1 Equilibrium modeling Theory and Assumptions

• Using CHEAQS and WATEQP
• Al3(aq) 3OH-(aq) ?? AlOH3(s) Solid phase
• Al3(aq) F-(aq) ?? AlF2-(aq) Speciation
• Equilibrium modeling assumes
• chemical equilibrium (also redox and pH)
• pure phases
• transport in dissolved phase only

8
1 Equilibrium modeling Results

• Pure phase saturation explains
• Sulfate Barium (barite)
• Carbonate Calcium and apparently iron and
  manganese in reduced zone
• Hydroxides Aluminum, manganese in acid zone
• Iron / Calcium / pH Phosphorous (vivianite and
  apatite)
• Phosphates REY in acid water
• Pure phase Uranium (uraninite) in reduced water
• Depending on local conditions!

9
1 Equilibrium modeling Summary

• Not many elements are controlled by saturation,
  so one may conclude
• Source-term limitation
• Source-term limitation may be sedimentary and /
  or input-determined.

10
2 Coprecipitation-codissolution Assumptions
and theory

• Codissolution
• Ca(1-x)SrxCO3 ?(1-x)Ca2 xSr2 CO32-
• Congruent, and main source
• Where x is the fraction of a TRACE ELEMENT in a
  MAJOR ELEMENT PHASE
• Can (and should be) verified using mineral data
• Coprecipitation
• When saturation of a major element phase is
  reached through increasing concentrations or
  changing redox or pH conditions, the opposite
  reaction may occur

11
2 Coprecipitation-codissolution Bulk sediment
geochemistry
12
2 Codissolution Example 1
13
2 Codissolution / Coprecipitation Example 2

• Al-Be and Al-Ga (also Al-REE) is observed
  codissolution real dissolution?

water
Dutch soil
14
2 Coprecipitation Example
Different source, but relation
15
2 Coprecipitation-codissolution Results

• Codissolution
• Ca Sr (carbonates and feldspar, and clay)
• K Rb (from clay mineral as identified from
  observed ratios)
• Fe As (iron (oxy-) hydroxides)
• Mn Mo (manganese hydroxides?)
• Clay (Ca-Mg-Sr) Cd-Tl (maybe Pb)
• Al Ga / Be / REY
• Zr Hf
• Coprecipitation
• Fe Mn
• Al REY / Be?
• Fe/S As

16
3 A novel approach

• But what about the normal background (e.g. Cu,
  Pb, Li, etc) and unexplained anomalies (e.g. Zn,
  Co).
• ? INPUT SOURCE LIMITATION

17
3 Steady-state input approach Assumptions

• Atmospheric deposition has been relatively
  constant in the Holocene, and the sediments have
  become saturated with these TE
• Concentrations should be constant with depth
• Differences in evaporative concentration ? ratio
  TE/CE should be constant with depth
• X-Na Me(aq) ?? X-Me Na(aq)
• seemingly conservative behavior!
• The start of the Anthropocene has caused
  changes!
• Geochemical processes cause changes!

18
3 Steady-state input approach Results
Absolute concentrations match Evap.
Rain Salland Rain Sweden
Seawater
19
3 Steady-state input approach Results
Absolute concentrations match Evap.
20
Boron
21
3 Steady-state input approach Lithium
normalizing on Sodium (Na)
2 log units
2 log units
22
3 Steady-state input approach Lithium, Cobalt,
Nickel, Rubidium, and Copper
23
Element EQ CD-CP SEQSSI ratio SEQSSI Other Details
Li CD 15103 Na X Low-pH weathering, slow ubiquitous IDIS4
Be CD Low-pH weathering
B 2.4103 Na
Al X Gibbsite
P X Apatite, Vivianite
V 28104 -
Mn X CP EQ Mn(hydr)oxides/rhodochrosite, CP siderite
Fe X Siderite
Co CD 2104 Ca X Low-pH weathering, mobilization in reduced acid GW
Ni CD 5104 Na X Low-pH weathering, mobilization in reduced acid GW
Cu 5104 Na/Ca
Zn CD 3103 Ca X Low-pH weathering, mobilization in reduced acid GW
As CD X CD Fe oxyhydroxides Sedimentary control in A3
Rb CD 5104 Na X Low-pH weathering, slow ubiquitous IDIS4
Sr CD Calcite and Al-silicates
Mo X Redox-control
Cd CD 12106 Ca Low-pH weathering
Cs CD 5106 Na X Low-pH weathering, slow ubiquitous IDIS4
Ba X CD EQ Barite, CD Calcite and Al-silicates
U X X EQ Uraninite, S Mobilisation at Mn redox boundary
REY CD Low-pH weathering
Ga CD Low-pH weathering
Sb Behaviour similar to U
Tl CD Low-pH weathering
Pb 1105 Ca
Zr Mobilization on organic complexation
Hf CD Zircon
Vissers, M.J.M., 2005, Patterns of groundwater
quality, NGS335
24
Conclusions

• The steady-state input approach significantly
  increases the understanding of trace element
  behavior in the subsurface
• Anomalies can be identified
• Anomalously high weathering releasing Be, Cd, Tl,
  Ga, Co, Ni
• Kinetic incongruent dissolution, releasing Li,
  Rb, Cs
• Mobilization in specific redox environments, Zn,
  Co, Ni
• Diffuse atmospheric / agricultural pollution
• The true baseline concentrations can be
  predicted!

m.vissers_at_geo.uu.nl
25
Conclusions

• For many elements rain is the main source.
• Apart from breakthrough of K and Rb, also Cu, Pb
  and many other elements are observed to be
  anthropogenically enriched in groundwater
• Groundwater enrichment factors of many trace
  elements vary from 1 (Lithium) to more than 100
  (Co, Ni, Zn)

m.vissers_at_geo.uu.nl
26
?