This study examines the effects of long-term prescribed burning treatments on vegetation structure and composition, productivity, and nutrient cycling in upland oak savanna and woodland vegetation. The basis for the study is an ongoing, experimental prescribed burning program begun in 1964 at Cedar Creek, and a similar program operating since 1962 on the adjacent Helen Allison Savanna property (owned by The Nature Conservancy). These prescribed burning programs are designed to subject upland oak communities (and some old fields) to different burn frequencies and patterns of burning, with the ultimate objectives of 1) restoring and maintaining the historically important savanna and open woodland vegetation, and 2) providing information about the effects of different burning patterns on vegetation structure and composition. This study addresses the latter of these two purposes and expands on it by also investigating possible influences of fire on resource availability (nutrients, water, and light) and net primary productivity. This study represents a continuation and expansion of experiments 015 and 094.
Experiment Design
from: Peterson, D. W., and P. B. Reich. 2001. Prescribed fire in oak savanna: Fire frequency effects on stand structure and dynamics. Ecological Applications 11:914-927.
Field Operations: Prescribed Burning
A prescribed burning program was initiated at CCESR in 1964 to restore and maintain oak savanna and woodland vegetation and to test the effectiveness of different prescribed burning treatments. An area of about 210 hectares was divided into 14 management units of 2.4 to 30 hectares each. Each unit was assigned to one of seven burn frequency treatments, ranging from annual burns to complete fire exclusion.
The first burn was conducted in 1964, and all units not designated for fire protection had been burnt at least once by 1969. Prescribed fires are generally conducted in the spring using a form of the strip-headfire method. Firebreaks are disked in the spring prior to burning to remove most fuels. In areas where permanent fire breaks do not exist, the grass layer is mowed to produce a 1-2 meter wide fuel break. A backfire is started along the lee side of the burn unit, to widen the firebreak. Flanking fires are then lit along the sides of the units and monitored until a sufficiently wide blackline is produced. Finally, a headfire is started along the windward edge of the unit to complete the burn.
Two additional burn units west of East Bethel Blvd. received prescribed burns beginning in 1987 and 1992. Additional burn units were added to the program in 1996, including 3 units along the west side of East Bethel Blvd., and one unit (409) that is a portion of one of the original "no burn" units. The Nature Conservancy began a similar prescribed burning experiment in 1962 on their 20 hectare Helen Allison Savanna property, which is adjacent to the Cedar Creek burn units. Eight rectangular 1-ha burn units were outlined on the east end of the property in a 4x2 design. The remaining 12 hectares were managed with protection from fire until 1987, when this unit was added to the burning program for conservation purposes. Prescribed fires are generally conducted in the spring using a form of the strip-headfire method. Firebreaks are disked in the spring prior to burning to remove most fuels. In areas where permanent fire breaks do not exist, the grass layer is mowed to produce a 1-2 meter wide fuel break. A backfire is started along the lee side of the burn unit, to widen the firebreak. Flanking fires are then lit along the sides of the units and monitored until a sufficiently wide blackline is produced. Finally, a headfire is started along the windward edge of the unit to complete the burn.
For dates of burns in E133 see table, Burns/Year in the Method Tables. Table E133 Plot Records Table links changes in burn units designations over time, from the continuation and expansion of experiments 015 and 094.
Sampling: Oak seedling survival and demographics
This portion of the study is designed to improve our understanding of the influence of fire frequency and canopy cover on the spatial patterns, growth, and mortality/survival rates of oak seedlings. Belt transects have been established in burn units representing the complete range of burn frequencies. Oak stems, both new germinants and older seedlings/sprouts have been tagged for long-term demographic studies.
Sampling: Permanent Plots
A series of permanent plots have been established in the burn units to facilitate repeated sampling of tree, shrub, and herbaceous community composition. Plots are sampled every 5-6 years to monitor changes in structure and composition. The first twelve of these were established in CCNHA Fish Lake burn units in 1984 as part of experiment 015. Additional plots were added in 1990 as part of experiment 094, including four additional plots in the Fish Lake burn units, five in the "Davis grid" savanna area west of East Bethel boulevard, and three in the Allison Savanna burn units. Finally, three new plots were added in 1995, two in old fields and one in burn unit 104.
Each plot consists of four parallel 50-meter transects placed 25 meters apart, outlining an area 50m x 75m. Points for quadrat sampling of herb and shrub cover are placed at 10 meter intervals along each transect for a total of 24 sample points per plot. (Note: the two old field plots, 22 and 28, were laid out as two 110-meter transects placed 25 meters apart, again producing 24 sample points.) Endpoints of each transect are marked with fenceposts (T-posts). The fencepost on the front end of each transect (points 1, 7, 13 and 19) are labeled with aluminum tags identifying the experiment (E133), plot number, and transect number. Interior sample points on each transect were marked by rebar posts starting in 1995 to facilitate true repeated sampling at each point. The points are arranged along the four transects in the following format. The points that have been most frequently sampled within the plots are 3, 4, 8, 9, 10, 11, 14, 15, 16, 17, 21, 22.
See table “Permanent Plots” in the Methods Tables for spatial orientation of the transects and points.
Sampling: Trees
All trees within the 50m x 75m area were tagged in 1995 using numbered round aluminum tags. Tags were nailed to the tree at a height of 1.3-1.7 meters and facing the front of the plot (the long side with labeled fenceposts). Trees on many of the plots were previously tagged as part of experiment 094. These foil tags appear to have been highly susceptible to fire and animal damage, however, and were missing from a large proportion of the trees. Old and new tag numbers were matched up for the vast majority of trees, however, so individual trees can still be followed over time. The aluminum nails from the old tags are valuable, however, because the E094 diameter measurements were made 1 centimeter above this nail. The 1995 measurements were made at "breast height" (about 1.35 meters) as determined by the field crews, so there may be some inconsistencies between 1990 and 1995 measurements. Trees were surveyed in 1995 for species identification, status (alive, dead, snag, fallen), and diameter at breast height (dbh). Previous surveys also recorded approximate tree height and crown vigor (ordinal scale, 1-5) on some plots. On many of the plots, trees were also mapped to the nearest 0.1 meter using the "Topcon Total Station" surveying equipment. Since readings were taken to the edge of the tree, the true accuracy is probably somewhere between 0.1 and 0.4 meters. The mapping process included all tagged trees (>= 5 cm) and the understory vegetation sample points.
Treatment Tables
Download treatment tables (zip file)
Data
Select Publications
Prescribed fire in oak savanna: fire frequency effects on stand structure and dynamics
DW Peterson, PB Reich
2001 Ecological Applications 11 (3), 914-927
Fire suppression and ecosystem carbon storage
D Tilman, P Reich, H Phillips, M Menton, A Patel, E Vos, D Peterson, ...
2000 Ecology 81 (10), 2680-2685
Fire and vegetation effects on productivity and nitrogen cycling across a forest–grassland continuum
PB Reich, DW Peterson, DA Wedin, K Wrage
2001 Ecology 82 (6), 1703-1719
Shocks to the system: community assembly of the oak savanna in a 40‐year fire frequency experiment
J Cavender-Bares, PB Reich
2012 Ecology 93 (sp8), S52-S69
Disease and fire interact to influence transitions between savanna–forest ecosystems over a multi‐decadal experiment
Pellegrini AFA, AM Hein, J Cavender‐Bares, RA Montgomery, AC Staver, F Silla, SE Hobbie, PB Reich
2021 Ecology Letters 24 (5), 1007-1017