Using fire to restore shortleaf pine (Pinus echinata) ecosystems severely impacted by southern pine beetle (Dendroctonus frontalis) Oral presentation to Berry College, Rome, GA April 2007

Katherine J. Elliott and James M. Vose, Jennifer D. Knoepp, and Barton D. Clinton Coweeta Hydrologic Laboratory, Southern Research Station, Forest Service, US Department of Agriculture Presentation Outline

¾Background ¾Fire History ¾Coweeta’s fire studies

¾Restoration of shortleaf pine (Pinus echinata) Was fire important in the history of the southern Appalachians?

• Native American history • Early explorers • Paleoecology • Fire scars • Plant & ecosystem traits Native American History

“… in those days they said the mountains were like a park. Yeah. You could walk up in the mountains and there were huge trees everywhere and there was grass even in between the trees and not that much brush. And so it was like a walking park, you know, and now if you go up there there’s just a lot of brush, a lot of different kinds of things.” Cherokee elder (Cooley 2004, MS Thesis)

“Now, now that’s the way the old people [built] a fire and they set fire all, all over .. in April. And sometimes in May when there ain’t too much wind..” Cherokee elder (Cooley 2004) Early explorer’s accounts

¾William Bartram’s journey through the Little Tennessee River valley of Georgia and North Carolina in the 1770s

¾Andrew Ellicott’s similar route in 1812

¾Robert Love’s survey of Macon and Swain counties, North Carolina in 1820 40

30 Pine 20 % Pollen 10

0 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 5 2

4 Ragweed 3 Charcoal

2 Charcoal % Pollen 1

0 0 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 Year AD redrawn from Clark et al. (unpublished) Desoto's First White Full White exploration Settlement Settlement Soil charcoal and carbon dating

Average fire return intervals 1,000 to 4,000 BP – 123 yrs downslope – 258 yrs ridgetop – 90 yrs 400 to 1,000 BP – 26 yrs 100 to 400 BP – 35 yrs

“Fire History Among Southern Appalachian Forest Types.” Norman L. Christensen, Principal Investigator Post-logging and land clearing slash fires expose mineral soil

“….in 1908…86% of the acreage in the southern Appalachians was cleared….”

“….practically all of it, whether cut or not had been burned…..”

U.S. Secretary of State, Wilson Plant & Ecosystem Traits

• Serotinous cones • Sprouting • Thick bark • Buried seed dormancy • Basal meristems • Flammability (pyrogenic) • Nutrient conservation ~600 HUMAN FIRE REGIME TIMELINE Periodic, lowintensitysurface fires

Native American Burning Wildland fire

1800 Land clearing (see Brose et al. 2001) (see Brose et stand replacing High intensity, & woods burning Post-logging slash fires & woods burning 1900 Forest Protection & Fire Exclusion 1940 Prescribed Fire Ecological Consequences of Fire Exclusion

¾Changes in species composition

¾Changes in nutrient pool sizes and cycling rates - productivity

¾Increased probability of severe and intense disturbances Changes in species composition

¾Loss of fire dependent (or enhanced) species ¾e.g., pitch pine, table mountain pine, oaks

¾Expansion of fire sensitive species ¾ e.g., white pine, red maple Nutrient Pools and Cycling

¾Decrease in biological fixation

¾Decrease in N-cycling rates

¾Shift in the amount and distribution of nutrients among pools ¾re-calcitrant shrub layer Increased Probability of Severe & Intense Disturbance

¾Fuel build-up

¾Fuel ladders

¾Interactions with insects & disease southern pine beetle Coweeta Fire Studies

¾ Fire History (Clark and Lynch 2002, Jurgelski 2003, Cooley 2004, Gragson 2005, Christensen 2006)

¾ Xeric mixed pine/hardwood (5 CJFR 1993; 22 others) pitch pine, hardwoods; mountain laurel understory

¾ Dry mixed pine/hardwood (Hubbard et al. 2004; Elliott and Vose 2005a, 2005b, Clinton and Vose 2006) shortleaf pine, Virginia pine, hardwoods; white pine understory

¾ Mesic, mixed-hardwoods (Elliott et al. 2004, Clinton et al. 1998) mixed hardwoods, herbaceous understory

¾ Mesic, mixed-oak (in progress JFSP) Quercus (rubra, prinus, alba); herbaceous understory Summary of responses

Parameter Response References

Diversity + Elliott et al. 1999, 2004, 2005

Nitrogen cycling + Knoepp & Swank 1993, 1995; Vose et al. 1999; Knoepp et al. 2004

Water quality 0 Clinton et al. 2003, Vose et al. 2005; Elliott & Vose 2005

Erosion 0 Vose et al. 2005, Elliott & Vose 2005

Small mammals +/0 Ford et al. 1999, Crossley et al 1999, & insects Love et al. 2007 Conceptual Model of Forest Succession & Management Options in the Upper Conasauga River Basin

Restoration treatments (prescribed fire, silviculture)

Community Composition

Overstory Shortleaf pine/oaks Virginia pine White pine (Pinus echinata/Quercus) (Pinus virginiana) (Pinus strobus)

Fire suppression Fire suppression High grading

Understory Bluestem grasses White pine White pine/ fire - (Andropogons, Schizachyrium) sensitive hardwoods

Desired future Undesired current Undesired future condition condition condition

Successional sequence

Problem

• In the southern Appalachians and Cumberland Plateau, over one million acres (with a timber value loss of 1.5 billion dollars) have been impacted by the most recent southern pine beetle (SPB) (Dendroctonus frontalis) outbreak from 1999-2003 Objectives

1. to compare fuel load reduction methods in pine/hardwood forests heavily impacted by southern pine beetle induced tree mortality – overstory felling →material left on site → Rx fire – Rx fire only – no treatment 2. to evaluate the effects of further restoration treatments including planting shortleaf pine and seeding native grasses on ecosystem structure and function All sites had standing dead and down trees due to the SPB 30.000 800

20.000 Biomass Density 10.000

) 600 -1 ) -1 8.000

6.000 400

4.000 Density (stems ha

Wood biomass (Mg ha Wood biomass 200

2.000

0.000 0 Live Live pine Dead Dead pine Live Live pine Dead Dead pine

Fig. 1. Pre-treatment (2005) aboveground biomass and density of live and dead trees (branch + bole). Biomass was estimated from species specific allometric equations. A cut + burn site after cutting Sampling Methods

Forest In situ soil floor nitrogen sample mineralization

Diameter (dbh) of Vegetation shortleaf pine (herbs, shrubs (20 m x 20 m plot) & trees)

Coarse Soil wood & solution fuel load lysimeter Prescribed fires were hand-lit March 2006 600 Cut + burn site 500

400

300 C) at 30 cm o

200 Maximum soil temperature at 10 cm = 22oC

100 Temperature ( * 0

1350 1355 1415 1435 Time Fig. 2. Flame temperature response curve. High intensity, moderate severity fire Consumption of fine fuels (small wood < 7.5 cm diameter, and forest floor litter layer) Burn only site -- smoldering at base of tree leads to input of large wood to forest floor 80000 Pre-burn 70000 Post-burn 60000 New 50000 Consumed

) 40000 -1 30000 20000 10000 Mass (kg ha Mass 0 Burn Cut+Burn

Fig. 3. Down fuel load (≥7.5 cm diameter) and consumption. A site preparation burn Rx Ignition hand-lit • backing fire along ridges, • head fire from the lower slopes

Rx result - high intensity short duration moderate severity

Fires consumed nearly all fine fuels (litter and small wood [1- 100 hr]) and 23-31% of the larger fuels (1000 hr) Results 50000

40000

Pre-burn

) Post-burn -1 30000 Consumed

20000 Mass (kg ha

10000

0 Burn Cut+burn Burn Cut+burn ------litter ------humus ------Fig. 4. Forest floor litter (Oi) and humus (Oe+Oa) mass pre- and post-burn. Table 1. Understory woody species richness (number of species per plot), density (stems ha-1), and basal area (m2 ha-1) for the prescribed burn treatments Richness Density Basal area Treatment Pre Post Pre Post Pre Post Burn 6.4 6.5 16,500 32,250 a 2.78 1.87 (0.9) (0.7) (3466) (5870) (0.65) (0.43) Cut + burn 8.6 7.4 27,375 33,875 ab 4.99 2.04 (1.0) (0.7) (4160) (4480) (0.79) (0.30) Reference 7.6 8.5 12,450 14,550 b 1.94 2.56 (0.9) (0.7) (2330) (2757) (0.42) (0.64)

Woody stems < 5.0 cm dbh, >0.5 m height. Standard errors are in parentheses. 60 60 Cover Richness Pre-burn 2005 Post-burn 2006 p = 0.0607 50 50

40 40

30 30

Percent cover p = 0.0002 20 20

10 10 Species richness (# of species/plot)

0 0 Burn only Cut + Burn Reference

Fig. 5. Percent cover and species richness of herbaceous layer species before and the first growing season after the prescribed fires at Ocoee, TN. Significant differences were evaluated by repeated measures anaysis of variance (Proc Mixed, SAS 2002). Summary

¾ Before the prescribed fires, all sites had a large amount of dead wood (Fig 1) due to tree mortality from the southern pine beetle.

¾ Even though fire intensity was high (Fig 2, flame temperature), fire duration was short resulting in moderate severity burns.

¾ In the burn only treatment, the fire burned the base of standing dead trees, which contributed large wood mass to the forest floor. More new wood was contributed to down material than consumed (Fig 3).

¾ The cut+burn treatment reduced down wood mass (Fig 3). Summary (cont.) ¾ The forest floor litter layer was consumed, but a large proportion of the humus layer remained intact (Fig 4).

¾ Understory wood density was high before the treatments due to increased number of saplings following the pine mortality. Density increased again after treatments (Table 1).

¾ Herbaceous layer percent cover and species richness increased after the cut+burn treatment, but no significant change with burn only (Fig 5).

¾ We are continuing sampling and analyses to evaluate changes in vegetation composition and diversity, carbon and nitrogen pools, soil and soil water chemistry, and success of planted pine and bluestem grasses. Acknowledgements

„ Cherokee National Forest for conducting the prescribed burns.

„ Field Staff: Patsy Clinton, Neal Muldoon, Chris Sobek, Jim Gruhala, Duane Foster

„ Funding provided in part from Joint Fire Science Program, Co-PIs James M. Vose and Katherine J. Elliott.