Thermal Alteration of Peat Soil By Low-Severity Fire Reduces Net Carbon Loss to Microbial Respiration

Neal FlanaganCurtis Richardson

Hongjun WangSuzanne Hodgkins*

Duke University Wetland Center*Florida State University

Perception of Wetland Fires- Pocosins(Example Deep Peat Fire)

Photo Credit USFWS

Reality of Most Wetland Fires(Surface Fires)

Photo by Caleb Spiegel, US FWS

US FWS https://www.flickr.com/photos/usfwssoutheast/

Croatan NF, Spring 2017

More Typical Endpoint of Wetland Fires (Pocosin Prescribed Burn)

< 2 years

Severity and Seasonality of Wetland Fires

http://www.mtbs.gov/dataaccess.html

Low Severity -Areas where more than a small proportion of the site burned. Duff, woody debris typically exhibit some change [but is not consumed].

Monitoring Trends in Burn Severity (MTBS)

Question:Do these low-severity Fires Affect Peat Recalcitrance?

Possible Mechanisms for increase peat recalcitrance (i.e. reduce microbial respiration):

1. Selective removal of labile compounds by thermal alteration thus concentrating recalcitrant fractions

2. Alteration of physical structure of SOM• Hydrophobic/aromatic coating on soil aggregates• Physical protection of SOM

3. Formation of charcoal / Black Carbon

NC Pocosin Lakes NWR

Sites

FL Loxahatchee NWR

Fire Frequency

Soil Order

MN Marcell Experimental Forest

Los Amigos, Peru

Evergreen Shrub (Pocosin) Peatlands• High summer temperatures• Seasonal (summer) water table recession > 1.2 MFire Return Interval 1 to 6 years

Blackspruce-sphagnum peatlands• Cool summer temperatures• Seasonal (summer) water table recession 0.3 m• Fire Return Interval 50 to 150 years

Sawgrass- Waterlily- Tree island• High summer temperatures• Seasonal water table recession > 0.6 MFire Return Interval 4 to 10 years

Mauritia Palm Peatlands• Little water table variation > 0.3 MFire Return Interval 500 to 1000 years?

Methods• Monitored Prescribed Burn at Pocosin Lakes NWR

• Soil temperature, moisture and depth to water table• Collected soil and litter before and after fire

• Collected soil and litter from all sites (MN, NC, FL, PERU)• Performed simulated burns of surface layer in lab at DUWC

• Mimic temperature pattern of prescribed burn (450°C, 10 minutes)• Soil moisture 350%• Burned samples inoculated or “reseeded” with unburned soil slurry after

cooling

• Aerobic, near-saturated incubation of burned and unburned litter at 5, 15 and 25°C for six months

• Analyzed structure of burned and unburned soils using XPS, FTIR-ATR and SEM

Elemental Analysis of Bulk Chemistry

MNNCFLPERUPrescribe Fire 0-5 cm soil

Structural Changes to peatUnburned litter SEM

Coatings reminiscent of hydrochar

• Physical protection of SOM

Hydrochar

Engineered recalcitrant soil amendment with potential for carbon sequestrationWet organic matter heated to between 180–260°C for 5min to 2 hours in the presence of sub and super-critically heated waterResults in a process that resembles coalification and produces characteristic microspheres coating organic matter

NC UNBURNED

NameC=C-CC-C, C-HC-O, C-OHO-C-O, O=C0=C-O

Pos.284.2850284.6850285.9850287.4850288.8850

%Area26.8321.4231.9315.384.43

x 102

5

10

15

20

25

CPS

296 292 288 284 280Binding Energy (eV)

NC BURNED

NameC=C-CC-C,C-HC-O, C-OHO-C-O, O=CO=C-O

Pos.284.8914285.2914286.5914288.0914289.3914

%Area36.1620.7025.0717.150.93

x 102

51015202530354045

50

CPS

296 292 288 284 280Binding Energy (eV)

XPS Peat Surface Carbon Functional Groups

Aromatic, Aliphatic, Alcohol/ Ether, Carbonyl/acetal , Carboxylic

Soil Aggregate Alteration

Pocosin “coffee ground” soil aggregates

Pocosin unburned litter

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

MN NC FL PERU

Soil Surface Aromaticity Index (XPS)

unburned burned

Fire (simulated burn) Effect on Microbial Respiration 25℃

y = 7019.9x-0.45

R² = 0.9446

y = 5481.9x-0.29

R² = 0.95110

2000

4000

6000

8000

0 50 100 150 200 250

Emiss

ion

CO2

(µg.

g-1.

day-

1)

MNBurnedUnburned

y = 1491.1x-0.377

R² = 0.8955

y = 989.84x-0.179

R² = 0.6406

0

500

1000

1500

2000

0 50 100 150 200

Emiss

ion

CO2

(µg.

g-1.d

ay-1

) NC

Point = Mean ± SE (n=4)

y = 1253.5x-0.395

R² = 0.8486

y = 557.12x-0.11

R² = 0.7568

0

500

1000

1500

2000

0 50 100 150 200

Emiss

ion

CO2

(µg.

g-1.d

ay-1

)

Days

FL

y = 1780.1x-0.437

R² = 0.9012

y = 1157.1x-0.286

R² = 0.7306

0

500

1000

1500

2000

0 50 100 150 200

Emiss

ion

CO2

(µg.

g-1.d

ay-1

Days

PERU

0

500

1,000

1,500

0 365 730 1095

Cum

. CO

2Em

issi

on (m

g/g

) MN

Initial Carbon loss to fire

0

100

200

300

400

0 365 730 1095

Cum

. CO

2 Em

issi

on (m

g/g

) NCBurnedUnburned

0

100

200

300

400

0 365 730 1095Cu

m. C

O2

Emis

sion

(mg/

g )

Days

PERU

0

100

200

300

400

0 365 730 1095

Cum

. CO

2Em

issi

on (m

g/g)

Days

FL

Cumulative CO2 Emissions

1

1.5

2

2.5

3

PERU MN FL NC

Q10

BURNED UNBURNED

Effect of Fire on Temperature Kinetics of Decomposition

Mean ± SE, *P < 0.05 between burned and unburned within a given site† P < 0.1

**

*†

seasonal water table variability

Surface Chemistry vs. Temperature Sensitivity

Conclusions• Thermal alteration of peat by Low-severity fire may be wide-

spread in peatlands having regimes of frequent fire• Large changes to bulk carbon chemistry are not necessary to

influence recalcitrance to microbial respiration, • Low-severity fires can alter surface chemistry• Physical protection of SOM is supported by reduced Q10 and SEM

images• Direct organic matter losses to low-severity fire was offset by

reduced microbial respiration after 1 to 3 years • All is predicated on maintaining high soil moisture content

during typical fire season (drainage, climate change).