Prescribed fire and soil

disturbance effects on above

ground and belowground

processes in the NJ Pinelands.

Dennis Gray

Rutgers Pineland Station

Four Mile Road, New Lisbon NJ

The State University of New Jersey

http://marine.rutgers.edu/pinelands/station.htm

Ericaceous herb layer

graminoid herb layer

Lichen & moss herb layer

Alternate stable states

Disturbance has been postulated as the

mechanism that causes shifts between

Alternate Stable States

Pinelands Herb Layer

Plant Communities

Sandy soils Microbial communities Nutrient dynamics

Plant communities

Background

Liebig's law of the minimum

• Growth is limited by the availability of the

single necessary factor in least supply

relative to demand (light, water, nutrients)

• Over time depletion of the necessary

factor leads to reduced productivity and

vegetation tolerant of impoverished

conditions

Liebig's law of the minimum

• Macro nutrients & Micro nutrients

• Nitrogen is considered to be the element most

limiting to temperate and boreal forests

– Primary production fixes atmospheric nitrogen

in organic compounds

– Primary producers require larger quantities of

nitrogen than of other nutrients

• nitrogen is energetically expensive to obtain

Liebig's law of the minimum

• Phosphorus is the one major element that must be supplied by the parent material

• low atmospheric returns

• Sandy soils have small phosphorus reserves

• Phosphorus acquisition is moderately expensive energetically

• Vegetation tolerant of impoverished conditions

tightly cycle elemental nutrients in low supply

• Successful species in a community retain

& efficiently use limiting nutrients

Effects of Disturbance

• Hypotheses:

– Soil Disturbance will mobilize nutrients

• Damaged plants leak mineralization

– Fire will mobilize nutrients

• volatilization (C & N) & mineralization

– Mineralized nutrients may be incorporated

into plant and microbial biomass • fertilizer effect

Effects of Disturbance

• Hypotheses:

– Retention of mineralized nutrients is limited • sandy soils

– Mineralized nutrients not incorporated into

biomass are leached • stream eutrophication

– Significant disturbance community change

Fire

• High speed

decomposition

• Volatilization loss

of Carbon & Nitrogen

• Non-biological

mineralization

– Fertilization effect

– increased pH

– Leaching loss of

nutrients

Soil Disturbance

• Above ground

&

Belowground mortality

• biological

mineralization

– Fertilization effect

– increased pH

– Leaching loss of

nutrients

Experiment

• replicated chronic, low intensity disturbance

fire vs soil disturbance

• to determine:

– if canopy tree growth is affected

– if herbaceous layer plant community changes

– how soil nutrient availability is affected

488

2007

30 circular study plots established

≈ 2 m diameter (7 m-2 treatment area)

≈ 4 m-2 sample area

10 plots were maintained as un-manipulated controls

10 plots were spaded to a depth of 20 cm

concentric rings around the stem in 10 cm increments

10 plots were burned

Treatment Methods

• 10 burned plots

• Fire severity was low (propane torch)

– ericaceous stems killed by heat scorch

– standing dead carex biomass burned

– ≈ 50% of the litter layer consumed.

Treatment Methods

– 1 burn treatment tree & 1 spade treatment tree lost due to wind-throw

– Total of 28 samples

spading and prescribed fire treatments applied late Feb – early March 2007,2008, 2009, 2010 & 2011

Vegetation Analysis Methods

• Ground cover analysis

late summer, 2006 – 2011 – Graminoid ramet and plant stem count, by species

litter depth and % cover lichen + moss

• Summer 2011, following % cover, stem and ramet count the vegetation was clipped at the ground surface – leaf, stem and ramet mass determined

• Canopy tree growth performance was determined by annual DBH measures

• March 2012 canopy cover of 7 spade, fire & control treatment trees was determined by upward facing LIDAR acquisition

LIDAR acquisition

N

W

S

E Horizontal cover

4 transects, returns / meter

Vertical cover

Standard deviation of return height

Soil Analysis Methods

• Soil samples from 0-20 cm depth were obtained

– 3 months post fire (May 2007)

– 3 months post fire (May 2011)

– Early October 2007- 2010 • at the onset of senescence

• ≈ 7 months post fire treatment

Soil Analysis Methods

• Litter depth, soil bulk density

• Soil Nutrient analysis:

• extractable inorganic nitrogen (NH4 & NO3)

• extractable inorganic phosphate (PO4)

• Microbial Biomass Nitrogen (MBN).

Canopy Layer response

gypsy moth, Lymantria dispar

Gypsy moth larvae consumed the deciduous tree canopy in May 2007

Approximately 15 % of the oaks in the study area died as a result

DBH

2007 2008 2009 2010 201120.0

22.5

25.0

27.5

30.0

32.5

spade

fire

control

Treatment X TimeF (8,100) = 0.33

p = 0.9527

cm

Cover

Stem growth over time

verticle distrubution

sta

nd

ard

devia

tio

n

Control Fire Spade0

1

2

3

4Control

Fire

Spade

F(2,18) = 0.69

p = 0.5123

LiDARderived cover

%

Control Fire Spade0

20

40

60

80Control

Fire

Spade

F(2,18) = 2.36

p = 0.1234

Herbaceous Layer response

carex

Control Fire Spade0.000

0.025

0.050

0.075

0.100

0.125A

B

B

F(2,23)= 10.72

p = 0.0005

2011

g s

tem

-1

carex

Control Fire Spade0

5

10

15

20

25

30

35

F(2,25)= 0.35

p = 0.7056

2011

g m

-2

Control has more robust ramets but disturbance treatments have more ramets

carex ramets

2006 2007 2008 2009 2010 20110

50

100

150

200

250

300

350control

fire

spade

treatment X timeF(10,128)= 2.56

p = 0.0074

ste

ms m

-2

Spade and fire treatment

carex ramet count slopes are positive

Control treatment slope flat

huckleberry

control fire spade0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

F(2,21)= 9.45

p = 0.0012

A

B

B

2011

g s

tem

-1huckleberry leaf

control fire spade0

5

10

15

20

25

30

35

40

45

F(2,25)= 0.74

p = 0.4861

2011

g m

-2

Control stems are more robust

Fire treated plots produce more leaf / stem

huckleberry

control fire spade0.0

0.1

0.2

0.3

0.4

0.5

0.6

F(2,21)= 23.30

p < 0.0001

B

A

B

2011

leaf

/ p

lan

tg

new stem produces more leaf / stem

blueberry

control fire spade0.00

0.25

0.50

0.75

1.00

1.25

F(2,22)= 24.89

p < 0.0001

A

B

B

2011

g s

tem

-1

blueberry leaf

control fire spade0

1

2

3

4

F(2,25)= 6.24

p = 0.0063

A

AB

B

2011

g m

-2

Control stems are more robust

Disturbance treated plots produce more leaf / stem

blueberry

control fire spade0.0

0.1

0.2

0.3

0.4

0.5

0.6

F(2,22)= 24.89

p < 0.0001

A

A

B

leaf

/ p

lan

t g

huckleberry stems

2006 2007 2008 2009 2010 20110

10

20

30

40

50

60

70control

fire

spade

treatment X timeF(10,128)= 5.59

p < 0.0001

ste

ms m

-2

blueberry stems

2006 2007 2008 2009 2010 20110

2

4

6

8

10

12

14control

fire

spade

treatment X timeF(10,128)= 1.73

p = 0.0811

ste

ms m

-2

Spade treatment huckleberry

stem count decreased over time

Spade treatment blueberry stem

count decreased over time

moss & lichen

%

2006 2007 2008 2009 2010 20110

5

10

15Control

Fire

Spade

Treatment X Time F(10,128) = 1.28

p = 0.2477

oak

seed

lin

gs m

-2

2006 2007 2008 2009 2010 20110.0

0.5

1.0

1.5

2.0

2.5Control

Fire

Spade

Treatment X Time F(10,128) = 1.23

p = 0.2761

pine

seed

lin

gs m

-2

2006 2007 2008 2009 2010 20110.0

0.5

1.0

1.5

2.0Control

Fire

Spade

Treatment X Time F(10,128) = 2.62

p = 0.0062

No difference in oak regeneration

No difference in moss / lichen cover Pine regeneration greatest in Spade plots

Stem count

% change 2006 - 2011

Carex Huckleberry Blueberry 77 35 16 70 56 29 78 -49 -50

Control

Fire

Spade

Soil Nutrient / Microbial Biomass

Response

Soil Surface

litter depth

2006 2007 2008 2009 2010 20110.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5control

fire

spade

treatment X timeF(10,128)= 7.51

p < 0.0001

cm

Litter depth decreased in all treatments over time

Litter depth decrease was greater in the disturbance plots

Reduced stem mass resulted in reduced litter retention in disturbance plots

NH4- N

Control Fire Spade Control Fire Spade0

1

2

3

4

5

6

7

8 May2007

May2011

Treatment X Time F(2,25)= 9.11

p = 0.0011

g

g-1

NO3- N

Control Fire Spade Control Fire Spade0.0

0.1

0.2

0.3

0.4

May2007

May2011

Treatment X Time F(2,25)= 5.57

p = 0.0100g

g-1

MBN

Control Fire Spade Control Fire Spade0

10

20

30

40

50

60

70

80 May2007

May2011

Treatment X Time F(2,21)= 6.71

p = 0.0056

g

g-1

A

AB

B

A

B

AB

F (2,26) = 6.90

p = 0.0040

F (2,26) = 3.970

p = 0.0314

F (2,22) = 1.01

p = 0.3800

Heavy oak gypsy moth defoliation of oaks in 2007

Canopy opening from oak mortality & increase shrub layer growth

Increased nitrogen demand from increased shrub species

NH4 - N

2007 2008 2009 20100

1

2

3

4

5

6Control

Fire

Spade

Treatment X Time F(6,77)= 1.51

p = 0.1866

g

g-1

Spade treatment reduced nitrogen demand from the herb layer

Increased nitrification of ammonium by soil bacteria in spade plots

NO3 - N

2007 2008 2009 20100.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8Control

Fire

Spade

Treatment X Time F(6,77)= 4.36

p = 0.0008

g

g-1

Spade treatment slope less steep

Microbial biomass does not decline in control or fire treatment plots

MBN

2007 2008 2009 20100

10

20

30

40

50

60Control

Fire

Spade

Treatment X Time F(6,69) = 5.83

p < 0.0001

g

g-1

Only Spade treatment slope significantly negative

Spade treatment damages plant roots

phosphate lost from dying roots leached from the soil

PO4 - P

2007 2008 20090.00

0.25

0.50

0.75

1.00Control

Fire

Spade

Treatment X Time F(4,52)= 4.99

p = 0.0018

g

g-1

Only Spade treatment slope significantly negative

Conclusions

Canopy Layer:

• Canopy trees did not benefit or suffer as a

consequence of fire or soil disturbance

– Fire mineralized nutrients did not lead to increased

tree biomass

– Root disruption did not lead to reduced rates of

biomass growth or canopy cover after 5 years of

disturbance

Conclusions

Herb Layer: (After 5 years of treatment)

• Carex ramet count increased

in disturbance treatments

• Disturbance treatments decreased mass / ramet

• Fire increased huckleberry leaf mass

• both disturbance treatments increase blueberry leaf

mass

• Huckleberry & Blueberry stem count decreased due

to spade treatment

Conclusions

• Litter inputs reduced due to disturbance

– fewer & smaller stems retain less litterfall

• Litter dam effect

• Reduced litter inputs ultimately result in

reduced nutrient inputs

Conclusions

Soil Nutrients / Microbial Biomass:

• Ammonium declined across all treatments

– Herb layer response to gypsy moth defoliation ?

• Nitrate also declined, although less steeply for

spade treatment

– Reduced ammonium demand and greater nitrification ?

• MBN declined over time in spade treatment

– coupled with reduced herb layer indicates nitrogen

leaching from spade treatment plots

• Phosphate steeply declines in spade treatment

Discussion

• Growth is limited by the availability of a single

necessary factor in least supply relative to

demand (light, water, nutrients)

– Oak mortality due to gypsy moth defoliation increased

sunlight and reduced competition for water & nutrients

– Soil inorganic nitrogen stores declined irrespective of

treatment

• Nitrogen is considered to be the element most

limiting to temperate and boreal forests

• In the pine barrens

Yes & No

Tree & shrub

N uptake 93.5 kg / ha/ yr

60% of N recycled

Pineland Upland Forest

Pinelands Atmospheric N deposition 9.5 kg / ha/ yr

40% of N returned As litter &

root turnover 37.5 kg / ha / yr

Leaky Ecosystem

Uptake values USFS

Net N uptake 4.2 kg / ha/ yr

Fire in the Pines

RxB fuel consumption

~ 6370 kg / ha

N release

~59 kg / ha

average annual acres burned

By the NJ Forest Fire Service

(Div B 2002-2008)

~ 11000 acres

Annual N release by RxB

~266,000 kg

RxB return interval

~ 4-8 years

Pinelands N deposition

~ 9.5 kg / ha / year

Fuel consumption values USFS

Wildfire in the Pines

2007 Warren Grove wildfire

estimated fuel consumption

~ 18600 kg / ha

N release

~163 kg / ha

2007 Warren Grove wildfire

~ 15,500 acres burned

N release

~1,00,000 kg

Pinelands N deposition

~ 9.5 kg / ha / year

Wildfire return interval

~ 17 years

Fuel consumption values USFS

• Soil disturbance by spading yielded:

– Increased carex, decreased huckleberry & blueberry

– Reduced microbial biomass

• Indicating reduced C, N & P immobilization

• increased N leaching

– Reduced soil phosphorus

• C & N immobilization ceases at the point of P limitation

– Disturbance reduces P retention

– Extent of P retention in soils is unknown

• High intensity disturbance can significantly impact

– Ecosystem productivity

– Water quality down gradient

• Pinelands plant communities are essential for nutrient

retention & forest productivity

Conclusion

Thank You

Pinelands Research Series,

April 17, 2012

Pinelands Commission, Richard J. Sullivan Center

15 Springfield Road, New Lisbon, NJ