Tree Coring to Assess Subsurface Volatile Organic … 1 Tree Coring to Assess Subsurface Volatile Organic Compound Contamination By Don A. Vroblesky (USGS, Columbia, SC)

1

1

Tree Coring to Assess Subsurface Volatile Organic Compound ContaminationBy Don A. Vroblesky (USGS, Columbia, SC)

2

User’s guide to tree coring for VOCsIntroduction (Advantages and

Limitations)Methodology• Tree core collection• Analysis• Quality control

Historical Perspectives and Technical Considerations• Historical perspectives• Technical rationale for

method aspects.• Factors influencing VOCs

in tree coresAppendix 1: Case StudiesAppendix 2: Gas chromatography

methodology

http://pubs.water.usgs.gov/sir2008-5088

3

Basic concept: •Tree roots take up water, gasses, and associated contaminants from the subsurface. •The contaminants move up the trunk. •Tree coring provides a sample of the ground water beneath the tree.

4

Application

• Rapid and inexpensive reconnaissance tool for determining whether subsurface VOCsare present or for areally mapping subsurface VOCs.

• Useful for directing drilling efforts.• Used at a variety of sites as tool in places

where contamination is suspected but little or no subsurface information is available.

5

What types of VOCs are appropriate?

• Typically VOCs with a log Kow of 0.5-3• Previous investigations have found:

– PCE, TCE, cDCE, VC– PCA, TCA, CT– Benzene, toluene, ethylbenzene, xylene

isomers, MTBE

6

Advantages (1 of 3)

• Adantages:1. Can provide information in areas where

access is limited (cultural influences, vegetation cover)

2. Applicable to a wide variety of common ground-water VOC contaminants (chlorinated solvents and petroleum hydrocarbons)

3. Presence of VOCs in tree cores is a strong indicator of subsurface VOCs.

7

Advantages (2 of 3)4. Sometimes can be used to detect VOCs in

soil gas.5. Can detect VOCs at relatively low

concentrations. Schumacher and others, (2004) detected PCE in tree cores where ground-water PCE was as low as 8 μg/L.

6. Rapid, simple method.7. Can be field analyzed or analyzed later.8. If analyzed later, the samples can be stored

without refrigeration.

References of tree coring related to soil gas: Struckhoff, 2003; Shumacher and others, 2004; Struckhoff and others, 2005; Vroblesky and others, 2006; 2008.

7

8

Advantages (3 of 3)9. Inexpensive method. Borers cost a few

hundred dollars and can be used indefinitely.

10. Minimum amt. of field equipment required. (Coring tool and bottles)

11.Sometimes tree-core parent/daughter VOC ratios can indicate areas of subsurface dechlorination.

12.Trees sample a large areal volume.

9

Limitations (1 of 2)1. Absence of VOC detection in tree cores

does not mean there are no VOCs in the subsurface.

2. Tree-core VOC concentrations provide only a generalized quantification of subsurface VOC concentrations.

3. Trees readily recover from coring, but care should be taken to avoid excess coring of an individual tree to reduce stress to the tree

10

Limitations (2 of 2)4. Under some conditions, rainfall and winter

dormancy can reduce VOC concentrations in tree cores.

5. A number of tree-related VOCs elute early on a typical short-column field gas chromatograph, potentially obscuring early eluting VOCs, such as vinyl chloride.

11

Methodology for Collection of Tree Cores for VOC analysis

12

Mark the tree to facilitate returning to it.• “engraved” aluminum tag, tree

paint, wooden stake, flagging, etc., depending on esthetic issues and how long the tag needs to last.

Take notes•Location info: date, time, site, tree ID, etc.•Tree info: species, stress indicators, diameter, etc.•Core info: height of core, side of tree, unusual characteristics, etc.

13

Tree CoringCollect cores from about the same height for the sake of comparison.

In general, a core near the ground can have higher VOC concs. than a core farther up the trunk, but for ease of collection, a simple approach is to use mean breast height (about 4.3 ft from the base).

13

14

How deep to core?B. Diffuse-porouswoodA.Nonporouswood C. Ring-porouswood

•Water transport takes place in the outermost part of the trunk, so you don’t have to core deep (not more than 2 to 3 inches).

•In ring-porous wood, such as oak, nearly all of the transpiration water moves through the outermost growth ring.

•In nonporous trees, such as pines, water transport is through several growth rings.

•In general, core length should be consistent across a particularstudy site and should be about the length of the sample jar being used.

15

Teeth on the core extractor

16

Core removal from core barrel: When you reach the coring depth, insert the

extractor, turn the coring tool ½ turn outward to break the core off, and pull it out.

17

Transfer the tree core to a vial

18

Cap the vial with Teflon–lined seal

19

20 mL

Or use standard VOA vial (just remember that they are made for water, not gases)

Crimp cap on vial

40 mL

20

Tree core in vial

In general, the vials tend to begin fogging up within seconds of capping, indicating vapor transfer to the headspace.

21

Post-coring tree care• Sealing the tree core hole probably is not needed

and sometimes can be harmful.• The open hole can allow decay and disease, but

plugging the hole does little to reduce discoloration or decay.

• In general, healthy trees (particularly conifers) respond well to coring by self sealing and creating chemical barriers.

• So a practical approach is to leave the core hole unsealed.

22

Quality Control

• Collect 10% duplicate samples (a second core slightly below the first core).

• Collect an air blank sample.• Collect samples of background trees for

each species to assure that the “contaminant” identification is not simply a natural VOC associated with the tree.

23

23

Increasing equilibration time

Tree T2b

0.08

0.25 0.4 26

0

10000

20000

30000

40000 Tree T1

2 4 26

TCE,

ppb

/v

0500

10001500200025003000

Equilibration time, in hours

4/18

/200

65/

3/20

07 Tree T100

0.1

0.13 29

29.5

0102030405060

Tree T802

0.08

0.08

1 24 27

0

200

400

600

800

1000

Day 0 Day 1

How long should the vials equilibrate before analysis?

Vroblesky and others, 4th Int. PhytotechnologiesConference, Denver, CO., September 24-28, 2007.

TCE in unheated cores over time

In some cases the tree VOC HSA concentration did not substantially change from 5 minutes after sampling to the next day. In other cases, the VOC concentration dramatically increased. The differences could be due to many factors, including the ambient temperature and how hot the core barrel became from friction during coring.

24

24

Comparison of tree-core equilibration after 1 day and 19 days

Vial number1 2 3 4 5 6 7 8 9Tr

ichl

oroe

then

e, in

par

tspe

r bill

ion

by v

olum

e

0

2000

4000

6000

8000

10000 24 hours19 Days

Vroblesky and others, 4th Int. Phytotechnologies Conference, Denver, CO., September 24-28, 2007.

In a comparison test, TCE concentrations did not increase enough over 19 days to make it worth waiting more than 24 hours.

25

Sample analysis approaches• Headspace analysis (HSA)

– Simple, cheap, probably sensitive enough for most shallow TCE ground-water plumes

• Methanol extraction: – Produces higher TCE concentrations than HSA, but costs more. – Extracts more than just the volatiles.

• Activated carbon (contact Joel Burken)– Very sensitive, but more expensive than HSA– Uses the corehole rather than the core.

• Field colorimetric gas-detector tubes– Very quick and easy, but useful only at high concentrations

26

Headspace sampling of vial

27

Inject the headspace into a gas chromatograph

28

Typical chromatograph of tree core showing chlorinated solvents

There typically are early eluting chromatographic peaks associated with natural volatile compounds in the tree.

28

29

29

Heaters that can be used in the field with a power inverter and a battery

Block heater Water bath

60-70 C for about 5 minutes


Equilibration time sometimes can be substantially shortened by heating the sample in the field.

30

30

Comparison of field heated and analyzed tree cores to unheated tree cores analyzed after 24 hours

TCE(r2=0.936)

1 10 100 100010000

100000Fiel

d he

ated

, ppb

/v

1

10

100

1000

10000

100000PCE

0 20 40 60 80 100 120 1400

20406080

100120140

24-hr stabilization, ppb/v

South Carolina Connecticut

1:1 correspondence

Vroblesky and others, 200, USGS SIR2007-5212.

31

31

Comparison of field heated to unheated tree cores 5 min. after collection

(NWS Charleston SWMU17)


Typicalreporting limit

02468

1012

TCE

, ppb

/v

020406080

100120140

0

400

800

1200

Not

det

ecte

d

Tree T100 Tree T802

Unheated Heated (70 oC)

Tree T807

32

Field analysis

• Field analysis can be used to direct the tree coring effort and optimize plume mapping.

• Heating the cores can increase the potential for VOC detection.

• Don’t heat them too much or you will volatilize tree constituents that complicate the chromatogram, potentially have vapor loss, and possibly stink up the place.

33

Colorimetric device

For very high concentrations (>1 ppmv), a simple colorimetric approach can be used in the field to detect VOCs in tree cores.

33

34

Gastube from tree core

Chlorinated VOCs are presentVroblesky and others, 4th Int. PhytotechnologiesConference, Denver, CO., September 24-28, 2007.

This is a colorimetric response from a tree core done within a few minutes of core collection showing that chlorinated VOCs are present.

34

35

Comparison to field GC

We did a field comparison of our modified ColorTec method to what we would find using a typical field gas chromatograph.

35

36

Comparison of colorimetric to field GC tree-core analysis after heating

Field GC CVOCs, in ppm/v0 10 20 30 40 50

Col

or T

ec C

VOC

s, in

ppm

/v

0

2

4

6

8

r2=0.95

Difference in magnitude is because colorimetric is calibrated for water


We got a very linear correlation between the field GC results and the simple colorimetric method at high tree-core VOC concentrations (>1 ppmv). The scale on the Color Tec axis is simply a relative scale. At

36

37

Comparison of field GC to colorimetric approach

• Good detection and relative correlation to field GC at GC vapor concentrations >1 ppmv, although the gas tubes have a different reporting scale.

• In the range of 0.7-1 ppmv (field GC), the values are at the detection limit for the colorimetric.

• CVOC concentrations <0.7 ppmv (field GC) are not detectable with the colorimetric approach. Field GCs are sensitive down to about 0.01 ppmv for TCE.

• The colorimetric method is inexpensive and easy for field analysis where tree-core concentrations are high (>1 ppm/v). (~$1500 for the reusable kit, about $5.65/analysis)


37

38

38

Selected factors influencing VOC

concentrations in tree cores

39

39

TCE concentrations in tree cores are partly a function of TCE concentrations in ground water

(Carswell Golf Course, TX)

Oak trees

Vroblesky and others, 2004, GWMR, v. 24, p. 124-138

40

40

TCE concentrations around the trunk near the edge of a plume

10,000 μg/L 1,000 μg/L

100 μg/L

TCE in gw

Forest

(SWMU12, NWS Charleston, 2000)

1,975 ppb 1,925 ppb

6,025 ppb5,087 ppb


Concentrations in this tree were substantially higher on the side facing the main body of contamination. This may not always be true because in some trees, upward movement of water spirals around the tree.

41

41

Differences in TCE concentrations relative to water-table depth

29-39 μg/L200 μg/L

99 ppbv2,191 ppbv

24 ft

PJ1

PJ3

Air Force Plant PJKS, CO (1999)


41

More intimate contact between the tree roots and the contaminated water can produce substantially higher tree-VOC concentrations than in a tree whose roots are more vertically distant from the contaminated water. This is partly because the tree in less intimate contact with the contaminated water probably has other sources of water (such as recharge capture).

42

42

Comparison of TCE in adjacent cottonwood and willow.

Carswell Golf Course


TCE

(nM

/L)

0

100

200

300

400

500

600

700

Tree 3Cottonwood

Tree 2Willow

These trees are adjacent to each other and have similar TCE concentrations, despite being different species. However, they are both diffuse porous, so they probably move water in similar ways.

43

43

Species comparison: TCE in adjacent trees, SRS, 1998

TCE

(nM

/L)

0

1000

2000

3000

4000

oak pine

TCE

(nM

/L)

0500

1000150020002500

T11

T12

T7

T25

T34

oak pine

TCE

(nM

/L)

0200400600800

1000

T13

T32

pines

TCE

(nM

/L)

0200400600800

1000T31

T30T29

T33T14

T32

T14

OakPineCypressSweet gum

sweet gums

oak pine cypress

Vroblesky et al.,1999, ES&T, v. 33, p. 510-515.

These graphs represent clusters of closely spaced trees of differing species. In general, pines and cypress contained more TCE than oaks and sweet gum.

44

Trees have higher VOC concentrations in the summer than in the winter

44

45

45

TCE in Tree Cores, nM/L

Hei

ght o

f tre

e co

re, i

n m

eter

s

0

5

10

15

20

2,000 4,000 6,000 8,000 10,000

July 1997

South Trunk(August 1997)

North Trunk (August 1997)

Savannah River Site, SC


Concentrations can vary with height up a tree trunk

46

46

Eastern Cottonwood, Carswell Golf Course, Texas

47

47

VOCs up tree after rainstorm, Carswell, TXTCE

ppbv60

0

800

1000

c-1,2-DCE

300

500

700

Hei

ght o

f cor

e (ft

)

0

10

20

30

40

50

60After Rainstorm After Rainstorm

ppbv

48

48

VOCs up tree before and after rainstorm, Carswell TX

TCE

ppbv60

0

800

1000

c-1,2-DCE

300

500

700

Hei

ght o

f cor

e (ft

)

0

10

20

30

40

50

60After Rainstorm After Rainstorm

Before Rainstorm

Before Rainstorm

ppbv

49

49

Cottonwood instrumented for irrigation test: Carswell Golf Course, TX

50

50

M a x im u m T r a n s p ir a t io n , in k g /h r0 1 0 0 2 0 0 3 0 0 4 0 0

Hei

ght,

in ft

0

2

4

6

8

1 0

1 2

M a x im u m T r a n s p ir a t io n , in k g /h r0 2 0 0 4 0 0 6 0 0 8 0 0 1 0 0 0

Hei

ght,

in ft

0

2

4

6

8

1 0

1 2

D a y 1D a y 2 : ir r ig a te dD a y 3 : ir r ig a te dD a y 4

D a y 2 : P o s t ir r ig a t io n

D a y 1 : P r e ir r ig a t io n

Maximum Transpiration During Artificial Irrigation of Cottonwood

Clinton and others, 2004, Int. J. Remed., v. 6, p. 239-252

51

51 TCE (ppb v/v)600 800 1000 1200

Hei

ght o

f tre

e co

re, i

n fe

et

0

2

4

6

8

10

12

14Pre

IrrigationPost

Irrigation

Day1

am

Day 1 pm

Day2

Day3

Day4

Avg. TCE in Tree CoresBefore and After Artificial Irrigation


52

52

Tree

Tree corer

Flame

Zeikus and Ward, 1974, Methane formation in living trees: A microbial origin: Science, v. 184, no. 4142

Methanogenicbacteria can infest the interior wood of visibly healthy trees in poorly drained soils, creating methanogenicconditions within the tree

Photograph of a flame extruding from a tree corer

53

53

TCE and cDCE in the inner and outer tree core (AF Plant PJKS, Colorado)


TCE dehalogenation under methanogenic conditionsTCE cis-1,2-DCE

Methanogenic

This tree spit methane and water from the tree corer once the corer penetrated the inner part of the tree. Analysis of the cores showed that the inner core contained a substantial amount of methane relative to the outer core. The methanogenic conditions of the inner core were conducive to reductive dechlorination of TCE. The TCE and cDCE results from the inner core also implies inner-trunk dechlorination activity.

54

Chlorinated solvent degradation by trees

• Newman and others (1977) found that cells from poplar trees were capable of transforming and mineralizing TCE without microbes.

• Additional evidence for VOC degradation by trees includes the presence of oxidative transformation products in tree tissue (several researchers)

54

55

Field Examples

56

56

Ground-water TCE and cDCE plume under a swamp,

Savannah River Site, SC

57

57

Vroblesky and others, 1999, ES&T, v. 33, p. 510-515

58

58

cDCE and TCE in Tree Cores, SRS, Jan/Feb 1998

Swamp

Source

Source

Vroblesky and others, 1999, ES&T, v. 33, p. 510-515

59Distance along ditchbed, in feet

0 200 400 600

TC

E, i

n pp

b v/

v

0

100

200

300

400

cDC

E, i

n pp

b v/

v

0

10

20

30

40

TCEcDCE

UpstreamDownstream

Results of passive vapor samplers beneath sediment in the ditch confirmed the plume extent defined by the tree cores.

59

6060

cis-1,2-Dichloroethene in Tree Cores, England AFB, LA, 3/11/03

Big Bayou

Landfill

This is another site showing that trees in a swamp were capable of seeing chlorinated solvents beneath the swamp, even though the standing water was relatively uncontaminated.

60

61

61

Air Force Plant PJKS, Colorado

62

62

Air Force Plant PJKS, Colorado

63

63

DI-1 Tributary

East Fork

Brush Creek

Alluvial Aquifer

TCE in GW


63

64

64

SWMU12, NWS Charleston, SC

65

65

1 ft of tight soil overlying about 10 ft of plastic clay (shown below). Well 12MW-29S,

SWMU12

66

66

Clay

Clayey sand

Sand

Sand

Land surface

15

10

10

5

2020 30

0

Dep

th (f

t)

Feet

VOCs

NWS Charleston, SC.About 6 to 10 ft of tight clay above the aquifer

67

67

TCE in ground water (μg/L), SWMU12, NWS Charleston

Forest

Swamp


68

68

Forest

Swamp

TCE in ground water (μg/L), and cored trees SWMU12


69

69

TCE in ground water (μg/L), and trees containing TCE

Forest

Swamp


70

70

Forest

Swamp

Tree 30

Tree 2

Tree 32

Locations of selected trees, SWMU12, Charleston SC


71

71 Vroblesky and others, 2004, GWMR, v. 24, p. 124-138

Changes in TCE over time in trees,

SWMU12, Charleston SC

Tree SC30

2000 2001 2002 2003

TCE

in tr

ee c

ores

,in

ppb

vol

10

100

1000

Tree SC2

Tree SC32

Trees

72

72

SWMU17, NWS Charleston

73

73

SWMU17, NWS Charleston,SC

400 feet

SWMU17

74

74

2001, 2002 wells SWMU17

110 ppb PCE

31,000 ppb TCE

Data from Tetra Tech NUS

190 ppb PCE61 ppb PCE

75

75

Potential source areas, SWMU17

Trenches

Shallow pit

76

76

Water levels, low tide, SWMU17


77

77

Water levels, high tide, SWMU17


78

78

SWMU17, NWS Charleston SC

79

79

Locations of cored trees


80

80

PCETCEcDCEND

VOCs from tree cores


81

81

PCE

TCE

PCETCEcDCEND

Plumes inferred from tree cores

PCE

TCE


82

82

Locations of temporary wells

New Temporary wells in red


83

83

TCE>100 ppbPCE>100 ppb

TCE>1,000 ppb

Plumes determined by well sampling

Data from Columbia Technologies

84

84

TCE>100 ppbPCE>100 ppb

TCE>1,000 ppb

GW plumes detected in wells and tree cores

85

Merriam Manufacturing Company Property, CT (tree cores and soil gas)

86

Merriam Manufacturing Company Property information

• Types of trees and shrubs– Maple, Catalpa, Mulberry, Sumac– Mostly small trees and shrubs

• (diam=2.2 to 4 inches) (one tree was 40 inches)• Depth to water= 15 to 20 ft• Rooting depth of Mulberry

– 1-2 ft in well drained soil (Peper, 1998)– 7.9 ft (but 70% of roots were in the top 2 ft) (Bunger and

Thompson (1938)– 3.9 ft (but 70% of roots were in the top 2 ft)(Olson and Fletcher,

1999; Olson and others, 2001)– Mulberry TCE probably from vadose contamination (Struckoff and

others, 2005); less likely to be hydraulic lift (Richards and Caldwell, 1987)

87

Comparison of field-analyzed 111TCA in 2007 trees and shrubs to 2004 soil-gas 111TCA

Vroblesky and others, 2008, USGS SIR 2007-5212

88

Comparison of field-analyzed TCE in 2007 trees and shrubs to 2004-2006 soil-gas TCE

Vroblesky and others, 2008, USGS SIR 2007-5212

Analyzing the tree onsite with field heating provided sufficient concentrations to identify zones of relatively high and low VOC concentrations.The tree-core VOC concentrations in the shallow-rooted species probably were derived from horizons shallower than the water table.

88

89

Merriam Manufacturing Company Property

• Analyzing the tree onsite with field heating provided sufficient concentrations to identify zones of relatively high and low VOC concentrations.

• The tree-core VOC concentrations in the shallow-rooted species probably were derived from horizons shallower than the water table.

90

90

SUMMARY

• Tree coring is a simple, rapid, inexpensive reconnaissance tool for examining VOC contamination in ground water

• It has application in a variety of environments

91

After viewing the links to additional resources, please complete our online feedback form.

Thank You

Links to Additional Resources

Feedback Form

Tree Coring to Assess Subsurface Volatile Organic … 1 Tree Coring to Assess Subsurface Volatile Organic Compound Contamination By Don A. Vroblesky (USGS, Columbia, SC)

Documents