Water and contaminants transfer in the chalk critical zone
Undergound quarry of Saint Martin le Noeud
Danièle Valdes1,*, Ningxin Chen1, Marc Dumont1, Christelle Marlin2, Hélène Blanchoud1,3, Cyrille Fauchard4, Fabrice Alliot1,3, Ayoub Saydy4,Emmanuel Aubry1, J. Guillemoteau5, Roger
Guérin1, Pierre Ribstein1.
*corresponding author : [email protected] Université, CNRS, EPHE, UMR 7619 METIS, F-75005 Paris, France
2 Université Paris-Sud/Université Paris-Saclay, CNRS, UMR 8148 GEOPS, , F-91405 Orsay, France3 EPHE, PSL Research University, UMR METIS 7619 (UPMC, CNRS, EPHE), F-75005 Paris, France
4 CEREMA, Direction Territoriale Normandie Centre, F-76000 Rouen, France5Universität Potsdam, Institut für Geowissenschaften, D-14476 Potsdam-Golm, Germany
2
Aquifère
Output : groundwaters
Processus de transferts Processus géochimiques
INTRODUCTION
Input : rainfall + NO3 + pesticides
CHALK AQUIFER Important resource for drinking water
UZ
chalk
?
QUESTION: Water and contaminant transfers in the chalk critical zone?
Focus on agricultural contaminants
Underground Quarry of Saint Martin le Nœud
- Direct access in the aquifer- Limit Unsaturated /saturated zones
SZ
STUDY SITE: underground quarry
B
A
BEAUVAIS
PARIS
chalk
PERCOLATIONOutput of UZ
UNDERGROUND LAKE Groundwater
BEAUVAIS
UZ: unsaturatedzone
SZ: saturatedzone
A B
16 underground lakes17m < depth <30m
Land Use: agriculture Fertilizers : nitrate, pesticides (atrazine)(time series available from 1960)
200m
STUDY SITE: underground quarry
CHARACTERIZATION OF THE CRITICAL ZONE Ground Conductivity (measured by CMD)= proxy of the clay with flints’s thickness
ERT profiles in contrasted areas
Very thick layer of clay : about 10 m
Very thin layer of clay : about 10 m
Spatial variation of the clay-with flints thickness spatial variation of the infiltration processes preferential infiltration points in the depressions with clay ? Storage of water in surface in the areas with high thickness of clay?
CHARACTERIZATION OF THE CRITICAL ZONE
UZ thickness(m)
Ground conductivity
(mS/m) Proxy of clay
thickness
Distance (m)
Back of the QuarryEntrance of the quarry
0 m 1200 m
LUCARNESBLEU PEDRO
Group 1 Group 2 Group 3
UZ thickness ≈ 30 m ≈ 30m 17 – 24m
Clay with flints thickness +/- + -
Spatial variation of the unsaturated zone characteristics
GROUNDWATERS: Data acquisition
Percolation
Lake
High frequency measurements (every hour)- Percolating flow- Lake level - Conductivity and temperature of lake
Sampling of percolation and lake every 2 months
Chemical analysis: - Major ions (K+, Na+, Ca2+, Mg2+, Cl-, SO4
2-, NO3-, HCO3-)
- Pesticides (atrazine and metabolites)
Beaker installed to collect Percolation sample
Rainfall data : daily time-series of Beauvais Station
16 sites (percolation + lake)
Raingauge to record Percolation flow
2D Graph 1
X Data
2012 2013 2014 2015 2016 2017 2018 2019 2020
Rain
fall
(m
m/m
on
th)
0
50
100
150
200
Rainfall (mm/month)
BLEU
2012 2013 2014 2015 2016 2017 2018 2019 2020
Lake w
ate
r le
vel (c
m)
-50
0
50
100
150
Perc
ola
tio
n f
low
(m
m/d
)
0
20
40
60
80
Co
nd
(m
S/c
m)
0,3
0,4
0,5
0,6
0,7
Col 61 vs Col 62
Col 63 vs Col 64
Col 65 vs Col 66
Col 67 vs Col 68
Col 69 vs Col 72
Col 63 vs Col 74
LUCARNES
2012 2013 2014 2015 2016 2017 2018 2019 2020
Lake w
ate
r le
vel (c
m)
-10
0
10
20
Perc
ola
tio
n f
low
(m
m/d
)
0
200
400
600 Co
nd
(m
S/c
m)
0,4
0,5
0,6
0,7
0,8
Col 91 vs Col 92
Col 93 vs Col 94
Col 95 vs Col 96
Col 97 vs Col 98
Col 99 vs Col 100
Col 99 vs Col 102
Col 93 vs Col 104
PEDRO
2012 2013 2014 2015 2016 2017 2018 2019 2020
Lake w
ate
r le
vel (c
m)
-200
-100
0
100
Perc
ola
tio
n f
low
(m
m/d
) 0
10
20
30C
on
d (
mS
/cm
)
0,50
0,55
0,60
0,65
0,70
0,75
0,80
Col 105 vs Col 106
Col 107 vs Col 108
Col 109 vs Col 110
Col 113 vs Col 114
Col 113 vs Col 116
Col 107 vs Col 118
GROUNDWATERS: Temporal approach
Lake: Dh(/year) ≈ 100 cmPerco discontinuous, summer: high , winter: ∅
Electrical conductivity: [lake] > [perco]high seasonal variations (~200uS/cm/y)
Lake: Dh(/year) ≈ 15 cmPercolation continuous, during all the year
Electrical conductivity: [perco] > [lake]low seasonal variations (<100uS/cm/y)
Lake: Dh(/year) ≈ 100 cmPercolation seasonal: summer: high
winter:stops or decreases
Electrical conductivity: [lake] > [perco]medium seasonal variations (~100uS/cm/y)
Hydrodynamic functionnings: response time: 2-3 months ≈ 100m/y velocity or celerity ?spatial differencies
Groundwater quality: spatial differencies
lakes ≠ percolation origin of the 2 different types of groundwater?
Spatial variation of the transfer processes processes controlled by the ZNS properties?
Approximate periodof effective rainfall
Hydrodynamics and elec. conductivity
time LBLE vs Heau LLUC
date prélevt vs Heau(cm)
time hobo PLUC vs Qperco hobo PLUC
time tinytag PLUC vs Qperco tiny PLUC
date prélevt vs Qperco(ml/j)
date prélevt vs NO3 perco (mg/l)
date prélevt vs NO3 lake (mg/l)
Lake water level
Percolation flow
Percolation sample
time LBLE vs Heau LLUC
date prélevt vs Heau(cm)
time hobo PLUC vs Qperco hobo PLUC
time tinytag PLUC vs Qperco tiny PLUC
date prélevt vs Qperco(ml/j)
date prélevt vs NO3 perco (mg/l)
date prélevt vs NO3 lake (mg/l) time LBLE vs Heau LLUC
date prélevt vs Heau(cm)
time hobo PLUC vs Qperco hobo PLUC
time tinytag PLUC vs Qperco tiny PLUC
date prélevt vs Qperco(ml/j)
date prélevt vs NO3 perco (mg/l)
date prélevt vs NO3 lake (mg/l)
Lake sampletime LBLE vs Heau LLUC
date prélevt vs Heau(cm)
time hobo PLUC vs Qperco hobo PLUC
time tinytag PLUC vs Qperco tiny PLUC
date prélevt vs Qperco(ml/j)
date prélevt vs NO3 perco (mg/l)
date prélevt vs NO3 lake (mg/l)
GROUNDWATERS: Temporal approach
[NO3]lake ≈ 30mg/l, evolution: ↑[NO3]perco ≈ 20mg/l evolution: ↑↑ , [NO3]lake > [NO3]perco
[NO3]lake ≈ 50mg/l, evolution: ↔[NO3]perco ≈ 70mg/l , evolution (1,5mg/l /y)[NO3]perco > [NO3]lake
[NO3]lake ≈ 90mg/l, evolution: ↔[NO3]perco ≈ 100mg/l , evolution: ↑↑[NO3]perco > [NO3]lake
Very high spatial variation of the [NO3] concentrationImportant differencies between lakes and percolationGlobally, increase in the concentrations, in both lakes and percolationsGlobally decrease of the [NO3] concentration in the lakes when the water level of lake increases
Recharge with recent surface water (less contaminated)? Nitrate storage in the unsaturated zone of chalk?
2D Graph 1
X Data
2012 2013 2014 2015 2016 2017 2018 2019 2020
Rain
fall (
mm
/mo
nth
) 0
50
100
150
200
Rainfall (mm/month)
BLEU
2012 2013 2014 2015 2016 2017 2018 2019 2020
Lake w
ate
r le
vel (c
m)
-50
0
50
100
150
Perc
ola
tio
n f
low
(m
l/d
)
0
20
40
60
80
[NO
3]
(mg
/l)
0
10
20
30
40
50
Col 47 vs Col 48
Col 49 vs Col 50
Col 51 vs Col 52
Col 53 vs Col 54
Col 55 vs Col 58
Col 49 vs Col 60
LUCARNES
2012 2013 2014 2015 2016 2017 2018 2019 2020
Lake w
ate
r le
vel (c
m)
-10
0
10
20
Perc
ola
tio
n f
low
(m
l/d
)
0
200
400
600
[NO
3]
(mg
/l)
40
50
60
70
80time LBLE vs Heau LLUC
date prélevt vs Heau(cm)
time hobo PLUC vs Qperco hobo PLUC
time tinytag PLUC vs Qperco tiny PLUC
date prélevt vs Qperco(ml/j)
date prélevt vs NO3 perco (mg/l)
date prélevt vs NO3 lake (mg/l)
PEDRO
2012 2013 2014 2015 2016 2017 2018 2019 2020
Lake w
ate
r le
vel (c
m)
-200
-150
-100
-50
0
50
100
Perc
ola
tio
n f
low
(m
l/d
)
0
10
20
30
[NO
3]
(mg
/l)
70
80
90
100
110
120
Col 75 vs Col 76
Col 77 vs Col 78
Col 79 vs Col 80
Col 83 vs Col 84
Col 83 vs Col 86
Col 77 vs Col 88
Nitrate concentrationstime LBLE vs Heau LLUC
date prélevt vs Heau(cm)
time hobo PLUC vs Qperco hobo PLUC
time tinytag PLUC vs Qperco tiny PLUC
date prélevt vs Qperco(ml/j)
date prélevt vs NO3 perco (mg/l)
date prélevt vs NO3 lake (mg/l)
Lake water level
Percolation flow
Percolation sample
time LBLE vs Heau LLUC
date prélevt vs Heau(cm)
time hobo PLUC vs Qperco hobo PLUC
time tinytag PLUC vs Qperco tiny PLUC
date prélevt vs Qperco(ml/j)
date prélevt vs NO3 perco (mg/l)
date prélevt vs NO3 lake (mg/l) time LBLE vs Heau LLUC
date prélevt vs Heau(cm)
time hobo PLUC vs Qperco hobo PLUC
time tinytag PLUC vs Qperco tiny PLUC
date prélevt vs Qperco(ml/j)
date prélevt vs NO3 perco (mg/l)
date prélevt vs NO3 lake (mg/l)
Lake sampletime LBLE vs Heau LLUC
date prélevt vs Heau(cm)
time hobo PLUC vs Qperco hobo PLUC
time tinytag PLUC vs Qperco tiny PLUC
date prélevt vs Qperco(ml/j)
date prélevt vs NO3 perco (mg/l)
date prélevt vs NO3 lake (mg/l)
GROUNDWATERS: contamination profile in the UZ - Chen (2019) PhD thesis
Sites
LUCARNES
BLEU
PEDRO
[NO3] in the percolation (mg/l) from 2014 to 2018
Storage of contaminants in the UZ of chalk
Nitrate vertical profile goes down . 0,5 m/y < Vtransfer < 2 m/y
Peak of nitrate at 18m of depth
Higher is the UZ thickness, lower is [NO3]Lower is the UZ thickness, higher is [NO3]
Chen et al. (2019) – J of Hydrol, Chen (2019) PhD thesis
Percolation samples at different depths allow to show the vertical profile of contamination in the UZ Chalk
GROUNDWATERS : Pesticides
UZ
GR
OU
ND
WA
TER
S
Nitrate and pesticides in the LAKES
Dat
a fr
om
20
16
to
20
17
EC
.(m
S/c
m)
UZ
thic
knes
s(m
)
Results- Important differences of pesticides
concentrations in the lakes- Very high concentrations in atrazine and DEA in
the lakes (specially in the group 2)- DAR ≈ 3 for groups 1 and 3
DAR >> 3 for group2 important degradationprocesses
Pesticides :- Focus on atrazine and desethylatrazine (DEA) - DEA is a metabolite resulting from
degradation of atrazine (probably nearsurface)
- DAR= degradation atrazine ratio DAR =atrazine/DEA
Focus on the lakes
UZ
GR
OU
ND
WA
TER
S
Nitrate and pesticides in the LAKES
Dat
a fr
om
20
16
to
20
17
EC
.(m
S/c
m)
UZ
thic
knes
s(m
)
Focus on the lakes
HydrodynamicsPERCOLATION
LAKE
GROUNDWATERS / UZ characteristics
Group 1 Group 2 Group 3
UZ thickness ≈ 30 m ≈ 30m 17 – 24m
Clay with flints thickness +/- + -
[NO3] - + +
[Atrazine] - + +
[DEA] - ++ +Degradation
(DAR) - ++ -Percolation
flow + ++ +Water level
variation + - +Vtransfer in UZ (Chen et al. 2019) - + -
UZ
GR
OU
ND
WA
TER
S
Chen et al. (2019) showed Vtransfer is higher undera thick clay-with-flints layer
Focus on the lakes
GROUNDWATERS / UZ characteristics
G1 G2 G3
UZ thickness ≈ 30 m ≈ 30m 17 – 24m
Clay with flints thickness +/- + -
[NO3] - + +
[Atrazine] - + +
[DEA] - ++ +Degradation
(DAR) - ++ -Percolation
flow + ++ +Water level
variation + - +Vtransfer in UZ (Chen et al. 2019) - + -
UZ
GR
OU
ND
WA
TER
S
Comparison Groups 1 and 2 :About same UZ thickness, clay thickness higher for Group 2
Under a thick clay-with fints layer, at the same depth
- Contamination is highermore rapid transfer
- DAR is highermore degradation processes on the surface
- Percolation is more important and continuous- Lake’s water level variation is lower
continuous recharge (storage of water in UZ)
- Transfer velocities are higher
Thick clay layer: LUCARNES Group 2Thin clay layer: BLEU Group 1:
chalk
PerchedgroundwaterClay-with flints
Concentratedinfiltration
Diffuse infiltration
chalk
Clay-with flints
Diffuse infiltration
Diffuse infiltration : low tranfer velocitySeasonal rechargeVariation of percolation flow and water level
Storage of water: perched groundwaterallowing degradtion processes
Diffuse and concentrated infiltration (highertransfer velocityContinous recharge from perched groundwater to water table
allowing continuous percolationinducing low variation of lake’s water level
CONCLUSION: TRANSFER PROCESSES
The clay-with-flints layer controls the transfer processes