Using 137Cs and 210Pbex and other sediment source

fingerprints to quantify fine sediment sources in

forested catchments in Chile

International Conference on Managing Soil for Food Security and Climate Change Adaption and Mitigation

IAEA, Vienna, Austria, 23-27 July 2012

Paulina Schuller*, Desmond E. Walling**, Andrés Iroumé*,

Alejandra Castillo* and César Quilodrán

*Universidad Austral de Chile, Valdivia, Chile

**University of Exeter, Exeter, UK

UACh IAEA-CN-191-50

• Forest export increased by 70% and accounts for ~10%

of total export in Chile

• Forest companies are involved in certification

processes and committed to meet international

environmental requirements, adopting management

practices to mitigate erosion impacts

Against this background, a study of the impact of

forestry operations on sediment output from forested

catchments and the effectiveness of potential mitigation

measures is being undertaken in South-Central Chile

OBJECTIVE

To quantify the relative

contribution of fine

sediment sources to the

total sediment yield

measured at the outlet of

forest plantation

catchments

• before and after clear-

cutting operations

• under contrasting rainfall

conditions

OC

EAN

O P

AC

ÍFIC

O

400 km

AN

DES

MO

UN

TAIN

RA

NG

E

Nacimiento NAT & NAC

Los Ulmos LUT & LUC

PP= 1200 mm y-1

PP= 2500 mm y-1

STUDY AREAS: Paired forest catchments

8 and 13 ha

8 and 20 ha

Radioisotopes were tested as fingerprints for this purpose:

FRN 137Cs, 210Pbex, (7Be)

ERN 40K, 226Ra, SOM, Nt were tested to further improve the

source discrimination provided by FRN

22

6R

a

21

0P

b

7B

e

13

7C

s

40K

22

6R

a

21

0P

bex

7B

e 1

37C

s

40K

Gamma spectrometry

Sediment source fingerprinting technique based

primarily on FRNs

Z1 Forest

slopes

Z3 Stream channel

(bed and bank) Z2 Forest roads

cut slopes

Defined homogeneous

landscape zones to identify

potential fine sediment

sources

Weir Time-integrating Discharge monitoring

trap samplers station

TARGET ZONE

V-notch weir

Time integrating

trap samplers

Discharge monitoring

station (automatically

collected integrated water samples)

Outlet of each catchment

0

30

60

90

120

150

180

1 11 21 31 41 51 61 71 81 91 101 111Su

sp

en

ded

sed

imen

t

(kg

ha

-1)

Period

0

20

40

60

80

100

1 11 21 31 41 51 61 71 81 91 101 111

Ru

no

ff (

mm

)

0

50

100

150

200

250

1 11 21 31 41 51 61 71 81 91 101 111

Pre

cip

itati

on

(m

m)

0

30

60

90

120

150

180

1 11 21 31 41 51 61 71 81 91 101 111

Period

0

20

40

60

80

100

1 11 21 31 41 51 61 71 81 91 101 111

0

50

100

150

200

250

1 11 21 31 41 51 61 71 81 91 101 111

The precipitation (mm), runoff (mm) and suspended sediment load (kg ha-1)

recorded at the outlet in Nacimiento catchments during consecutive weekly

measurement periods (16-09-2009 to 31-12-2011). The left and right arrows

indicate the date of the harvest and reforestation periods, respectively, in NAT.

NAT NAC

~320 kg ha-1

Earthquake: 27-02-2010, 8.8 magnitude

0

100

200

300

400

500

600

700

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31

Su

sp

en

de

d

se

dim

en

t (

kg

ha

-1)

Period

0

100

200

300

400

500

600

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31

Ru

no

ff(m

m)

0

200

400

600

800

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31

Pre

cip

ita

tio

n (m

m)

0

100

200

300

400

500

600

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31

0

100

200

300

400

500

600

700

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31

Period

0

200

400

600

800

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31

LUT LUC

The precipitation (mm), runoff (mm) and suspended sediment load

(kg ha-1) recorded at the outlet in Los Ulmos catchments during

consecutive monthly periods (17-07-2009 to 31-12-2011). The black

arrows indicate the date of the final harvest period in LUT catchment.

Z1 (Forested slopes)

Z2 (Roads)

Z3 (Stream bed and bank)

ZPC

Weir

7Be (Road))

ZPC

Plantation

Roads

Streams

Legend

NAC NAT

SOURCE FINGERPRINTS: To ensure that the results obtained were

representative of the sediment sources involved, multiple representative

composite samples of different potential source material were collected

from the upper 1-cm soil layer in Z1, Z2, Z3, in the control catchments and

in the treatment catchment before and after harvest operations

Z1 (Forest slopes)

Z2 (Roads)

Z3 (Stream bed and bank)

ZPC

Weir

7Be (Roads))

ZPC Buffer

zone

Forest

plantation

Roads

Streams

Legend

LUT LUC

During all observation periods, monthly samples of fine sediment

output were provided by the suspended sediment trap samplers,

supplemented by sediment collected from the weir pools

The <63 μm fractions of the source materials and target sediment

were used for radionuclide, SOM and Nt assay, to take account of

contrasts in grain size composition between source material and

target fine sediment

Radionuclide mass activity concentration was determined in 80

mL samples sealed in Petri dishes using an extended range Ge

detector of 53% relative efficiency

No particle size correction was performed, because no

correlation was found between the fingerprints concentrations

and particle size composition

Fingerprints concentration associated with the <63 μm

fraction of source material

NAT Pre-harvest

Z1 15 6,2 ± 0,8 32 ± 3 390 ± 25 13 ± 1

Z2 3 3 ± 1 18 ± 8 395 ± 24 10 ± 4

Z3 7 1,7 ± 0,2 16 ± 1 539 ± 10 10 ± 1

Mean ± se

Source

zones

Number of

analyzed

points

137Cs (Bq kg

-1)

210Pbex (Bq kg

-1)

40K (Bq kg

-1) SOM (%)

Mean ± se Mean ± se Mean ± se

LUT Pre-harvest

Z1 10 14 ± 2 76 ± 11 101 ± 20 23 ± 2

Z2 5 3 ± 1 13 ± 4 240 ± 110 9,9 ± 0,5

Z3 6 4,6 ± 0,4 9 ± 3 364 ± 38 11 ± 1

40K (Bq kg

-1)Source

zones

Number of

analyzed

points

137Cs (Bq kg

-1)

210Pbex (Bq kg

-1)

Mean ± se Mean ± se

SOM (%)

Mean ± se Mean ± se

… 226Ra, Nt

… 226Ra, Nt

Identification of a composite fingerprint set of source

material properties capable of discriminating between the

potential sources

Two stage procedure:

• Kruskal–Wallis (KW) test to identify the fingerprint

properties which were able to discriminate between the

potential sources

• Multivariate discriminant function analysis (DFA) to

select the optimum sub-set of fingerprint properties from

those identified as potential properties in the first stage

The optimum composite fingerprint for discriminating sediment

source types in each catchment during each observation period

Catchment Period Fingerprints

selected

Correctly classified

source samples (%)

NAC 06/08/09 - 18/07/11 137Cs 210Pbex93,3

NAT pre-harvest 06/08/09 - 24/03/10 210Pbex40K 80,8

NAT post-harvest 24/03/10 - 09/12/10 137Cs 40K 85,3

NAT post-reforestation 09/12/10 - 18/07/11 137Cs 40K 81,1

LUC 01/10/09 - 20/06/11 137Cs 210Pbex82,7

LUT pre-harvest 10/09/09 - 19/03/10 137Cs 210Pbex83,3

LUT post-harvest 19/03/10 - 20/06/11 137Cs SOM 71,9

Estimating the relative contribution of each potential source to the

suspended sediment samples collected at the catchment outlet (Collins,

A.L. et al. 2010)

A multivariate mixing model was used minimizing the sum of the

squares of relative errors using Monte Carlo analysis

Ci = mass concentration of fingerprint property (i) in catchment outlet

time-integrated fine sediment sample collected at the outlet

Ps = sediment relative contribution from source zone (s)

Ssi = mean mass concentration of fingerprint property (i) in sediment

source zone (s)

n = number of fingerprint properties comprising the optimum composite

fingerprint set (n = 2)

m = number of sediment source zones (m = 3)

2

1 1

/

n

i

i

m

s

sisi CSPC 11

m

s

sP

NAT post-harvest

Fo

res

t

slo

pe

F

ore

st

ro

ad

s

S

tre

am

c

ha

nn

el

%

co

ntr

ibu

tio

n

0

10

20

30

40

50

60

70

NAT post-reforestation

F

ore

st

s

lop

e

F

ore

st

ro

ad

s

S

tre

am

c

ha

nn

el

%

co

ntr

ibu

tio

n

0

10

20

30

40

50

60

70

NAC

F

ore

st

s

lop

e

F

ore

st

ro

ad

s

S

tre

am

c

ha

nn

el

%

co

ntr

ibu

tio

n

0

10

20

30

40

50

60

70

NAT pre-harvest

F

ore

st

s

lop

e

F

ore

st

ro

ad

s

S

tre

am

c

ha

nn

el

%

co

ntr

ibu

tio

n

0

10

20

30

40

50

60

70

Preliminary results:

Nacimiento catchments

NAC

F

ore

st

s

lop

e

F

ore

st

ro

ad

s

S

tre

am

c

ha

nn

el

%

co

ntr

ibu

tio

n

0

10

20

30

40

50

60

70

Confidence limit

Upper

Mean

Lower

Relative contribution of the fine

sediment source zones to the

target sediment, based on load

weighted contributions

LUT pre-harvest

F

ore

st

s

lop

e

F

ore

st

ro

ad

s

S

tre

am

c

ha

nn

el

%

co

ntr

ibu

tio

n

0

20

40

60

80

100

LUT post-harvest

F

ore

st

s

lop

e

F

ore

st

ro

ad

s

S

tre

am

c

ha

nn

el

%

co

ntr

ibu

tio

n

0

20

40

60

80

100

LUC

F

ore

st

s

lop

e

F

ore

st

ro

ad

s

S

tre

am

c

ha

nn

el

%

co

ntr

ibu

tio

n

0

20

40

60

80

100

LUT post-harvest-1

F

ore

st

s

lop

e

F

ore

st

ro

ad

s

S

tre

am

c

ha

nn

el

%

co

ntr

ibu

tio

n

0

20

40

60

80

100

LUT post-harvest-2

F

ore

st

s

lop

e

F

ore

st

ro

ad

s

S

tre

am

c

ha

nn

el

%

co

ntr

ibu

tio

n

0

20

40

60

80

100

Relative contribution in

Los Ulmos catchments

78% SL

Pp 1390 mm

22% SL

Pp 1270 mm

Comparison between fingerprints measured values in target sediment samples

and the calculated values using the estimated relative source contributions

NAC

LUC

5

6

7

8

9

10

06

/07

/09

14

/10

/09

22

/01

/10

02

/05

/10

10

/08

/10

18

/11

/10

26

/02

/11

06

/06

/11

14

/09

/11

13

7C

s (B

q k

g-1)

Measured Cs values

Estimated Cs values

10

15

20

25

30

35

40

45

06

/07

/09

14

/10

/09

22

/01

/10

02

/05

/10

10

/08

/10

18

/11

/10

26

/02

/11

06

/06

/11

14

/09

/11

21

0P

bex

(B

q k

g-1) Measured Pb values

Estimated Pb values

0

1

2

3

06

/07

/09

14

/10

/09

22

/01

/10

02

/05

/10

10

/08

/10

18

/11

/10

26

/02

/11

06

/06

/11

14

/09

/11

13

7C

s (B

q k

g-1) Measured Cs values

Estimated Cs values

0

5

10

15

20

25

30

06

/07

/09

14

/10

/09

22

/01

/10

02

/05

/10

10

/08

/10

18

/11

/10

26

/02

/11

06

/06

/11

14

/09

/11

21

0P

bex

(B

q k

g-1)

Measured Pb values

Estimated Pb values

This study provides important benefits

• Demonstrates the application of sediment source fingerprinting

procedures in Chile based primarily on FRNs 137Cs and 210Pbex

• Provides a basis for expanding such work in the future into other

regions, agricultural areas and studies linked to other sediment-

related environmental problems

• Provides improved understanding of sediment source types and

their relative contribution to the sediment yields for forested

catchments in south-central Chile, particularly during critical periods

of disturbance

• Contributes to the decision-making process for implementing

improved and cost-effective sediment management practices, in

order to reduce the impact of plantation forest clearcutting on water

quality

• Provides important empirical information to support forest

certification procedures and plantation sustainability

Acknowledgements

FONDECYT 1090574

FAO/IAEA CRP D1.20.11, CHI-15331

IAEA RLA 05/051

Gobierno Regional de Los Ríos, Chile

Forestal Mininco S.A.

CEFOR UACh, DID-UACh