Influence of Pine Straw Harvesting, Prescribed Fire, and Fertilization on a Louisiana Longleaf Pine Site

Thinning of the site to a target basal area of 65 ft/ac was conducted in June 1999 to maintain stands within the 50to 80-ft/ac basal area range recommended for red-cockaded woodpecker (Picoides borealis) habitat (US Fish and Wildlife Service [USFWS] 2009). Additionally, it was thought that by reducing stand stocking the remaining trees would have sufficient growing space to better respond to management. Before thinning, basal area on the subplots ranged from 103 to 111 ft/ac and averaged 107 ft/ac. The subplots were thinned from below to a residual basal area of 62–66 ft/ac and averaged 64 ft/ac. The criteria for selecting trees to be cut were (1) removal of scattered loblolly pine trees (Pinus taeda L.), (2) removal of longleaf pine trees in the intermediate crown class and ones with poor form or having diseased or injured boles, and (3) provide better spacing between residual trees. Treatment Establishment Four, 3.2-ac research blocks were installed in the spring and early summer of 1990 in a randomized complete block split-plot design with the four blocks as replicates. Blocking was based on initial overstory basal area and topography. In total, there were 32, 0.39-ac subplots (4 blocks by 2 main plots by 4 subplots per main plot). An interior 0.23-ac area within each 0.39-ac subplot was used for measurement and sampling purposes. In August 1990, the understory vegetation was rotary mowed to create uniform understory conditions and to facilitate plot establishment. The two main-plot treatments within each block were (1) no fertilization and (2) fertilization—plots were fertilized three times with 45 lb N and 50 lb P/ac broadcast evenly over the main plot as 250 lb/ac of diammonium phosphate in April 1991 and May 1997 and with 50 lb P and 72 lb K/ac as 250 lb/ac triple superphosphate and 250 lb/ac of potash, respectively, in April 2004. The four subplot treatments were control (C)—no prescribed fire after 1987 or rotary mowing after 1990; prescribed burned (PB)—subplots were burned in March 1991, February 1994, March 1997, January 2000, June 2002, and May 2004; prescribed burned and two straw harvests (PBH)—subplots were prescribed burned along with the PB subplots, and the pine straw was harvested in early 1992 and 1993; and annual straw harvest (AH)—subplots were prescribed burned in August 1991, rather than March 1991, and the pine straw was harvested 13 times from early 1992 through April 2006 (Table 1). All subplots were thinned as part of a standwide thinning in June 1999. Hurricane Lili passed through central Louisiana on Oct. 4, 2002. It spawned a small tornado or shearing wind that disproportionally felled longleaf pine trees across the study site. On the PBH subplots, six trees were lost (average dbh of 13 in.), whereas only three trees were lost on the other three subplot treatments (average dbh of 13 in.). No other natural phenomenon caused significant damage to the study. Pine Straw Harvesting and Dry Weight Determination The potential detriment to forest vegetation from one to two pine straw harvests over a 10-year period is a management concern of the US Forest Service. To address this concern, harvesting stopped on PBH subplots once it was apparent that there were no statistical differences in pine straw yields between PBH and AH subplots (Haywood et al. 1998). Prescribed burning continued as part of the PBH subplot treatment just as it would have normally continued on US Forest Service lands. The PBH and AH subplot treatments were rotary mowed in July 1991 (Table 1). After needle fall peaked from September through December (Haywood et al. 1996), the litter was collected in windrows with a tractor-mounted straight-bar rake. Large limbs and cones were removed, and the straw was mechanically baled in early 1992 and removed from the site and used for erosion control. The PBH and AH subplots were again rotary mowed in July 1992 and the straw was harvested in early 1993. Straw harvesting continued on the AH subplots with rotary mowing in July 1993 and 1994 and straw harvesting in April 1994 and January 1995. The bales were weighed and a subsample was taken to determine moisture content and dry matter production in pounds per acre for the 1992 through 1995 harvests (Haywood et al. 1998). In this early period, removal of all pine straw was attempted and some understory vegetation was uprooted. As a result, more forest floor material was removed than was added on a yearly basis as needle fall, and the mineral soil was eventually left bare on much of the harvested surface until new needle fall and herbaceous vegetation again covered the soil (Haywood et al. 1998). The loss of the forest floor is commonly observed after continual mechanical harvesting of pine straw. For the 1996, 1997, and 1998 harvests, AH subplots were raked to collect the straw before it was moved off the subplots and left in the surrounding woods because personnel were not available for baling and weighing bales (Table 1). All subplots were thinned in June 1999. Pine straw was not harvested in 1999 to allow a fuel bed Table 1. Chronological listing of management activities on the subplot treatments from August 1990 through April 2006. Dates Activity and subplot treatment August 1990 Rotary mowed C, PB, PBH, and AH March 1991 Prescribed burned PB and PBH July 1991 Rotary mowed PBH and AH August 1991 Prescribed burned AH March/April 1992 Harvested pine straw on PBH and AH July 1992 Rotary mowed PBH and AH March/April 1993 Harvested pine straw on PBH and AH July 1993 Rotary mowed AH February 1994 Prescribed burned PB and PBH April 1994 Harvested pine straw on AH July 1994 Rotary mowed AH January 1995 Harvested pine straw on AH July 1995 Rotary mowed AH February 1996 Raked pine straw off AH July 1996 Rotary mowed AH March 1997 Prescribed burned PB and PBH April 1997 Raked pine straw off AH July 1997 Rotary mowed AH April 1998 Raked pine straw off AH June 1999 Thinned C, PB, PBH, and AH January 2000 Prescribed burned PB, PBH, and AH October 2000 Rotary mowed AH January 2001 Raked pine straw off AH July 2001 Rotary mowed AH January 2002 Harvested pine straw on AH June 2002 Prescribed burned PB and PBH July 2002 Rotary mowed AH February 2003 Harvested pine straw on AH July 2003 Rotary mowed AH January 2004 Harvested pine straw on AH May 2004 Prescribed burned PB and PBH July 2004 Rotary mowed AH April 2005 Raked pine straw off AH July 2005 Rotary mowed AH April 2006 Harvested pine straw on AH The subplot abbreviations are C, control; PB, prescribed burned; PBH, prescribed burned and two straw harvests; and AH, annual straw harvest. 116 SOUTH. J. APPL. FOR. 33(3) 2009 to develop before the prescribed burn on the PB, PBH, and AH subplots in January 2000. Straw harvesting resumed on the AH subplot treatment with rotary mowing in October 2000 and the pine straw being raked off the subplots in January 2001 (Table 1). However, pine straw was again baled, weighed, and removed from the site after the 9th through 11th and 13th harvests in January 2002, February 2003, January 2004, and April 2006. In April 2005, the pine straw was raked off the subplots and not baled because of equipment failure. From 2001 through 2006, only current-year pine straw was removed and some material was left near trees and stumps, in depressions, and among understory vegetation. After the pine straw bales were weighed following the 2002, 2003, 2004, and 2006 harvests, a subsample of straw was taken from the center of each bale, and the subsamples were weighed and dried at 175°F for 72 hours in a forced-air oven to determine oven-dried weight. The wet and dry weights of the subsamples were used to determine percent dry matter in the samples, and the percent dry matter was used to calculate the dry weight (in pounds per acre) of the harvested pine straw. Prescribed Fire Prescribed burning was done with strip head fires, which were monitored to determine their intensity in 1991 and 1994 (Haywood et al. 1998). One month after the 1991 and 1994 prescribed burns, the percentage of crown scorch was estimated for each pine tree on the PB and PBH subplots. Crown scorch averaged 15 and 11% after the March 1991 and February 1994 burns, respectively. Fire intensity was not determined for the remaining four prescribed burns. The AH subplots were no longer protected from fire beginning in January 2000, but the fires only fingered into them and went out because the sparse fuel bed was not continuous enough to carry a fire. Longleaf Pine Sampling Total height and dbh of all overstory pine trees in the 0.23-ac measurement area of each plot were measured using a laser hypsometer (Criterion 400 Survey Laser; Laser Technology, Inc., Centennial, CO) and diameter tape, respectively. The trees were measured in December 1997 (18 months prethinning), September 1999 (3 months postthinning), and December 2006 (90 months postthinning). The outside-bark (o.b.) stem volume and green weight of stemwood per tree were calculated using the relationships of Baldwin and Saucier (1983). Data Analysis For the weights of pine straw harvested in 2002, 2003, 2004, and 2006, a randomized complete block design model was used to compare the two fertilization levels. For longleaf pine, dependent variables were outside-bark (o.b.) stem volume per tree and number of trees, basal area, volume, and green weight of stemwood per acre. The prethinning measurements made in December 1997 and the postthinning measurements made in September 1999 were subjected to analyses of variance for a randomized complete block splitplot design model to determine if thinning changed treatment effects. Additionally, an analysis of covariance with the December 2006 results as the dependent variable and the September 1999 (postthinning) measurements as the covariate was conducted. For all analyses of variance, orthogonal linear contrasts were used to compare subplot treatments: (1) control versus management (PB PBH AH), (2) prescribed fire (PB) versus pine straw harvesting (PBH AH), and (3) two harvests (PBH) versus annual harvest (AH). All main, subplot, and interaction effects were considered significant at P 0.05. Results and Discussion Pine Straw In earlier years (1992–1995), there were no differences in the weights of harvested pine straw (Haywood et al. 1998). Despite the thinning in June 1999, fertilized plots continued to produce no more pine straw than the unfertilized plots in 2002 and 2003 (Table 2). However, the harvest in January 2004 on the fertilized plots was significantly greater than the harvest on the unfertilized plots by a difference of 279 lb/ac. Although plots were again fertilized in April 2004, the 331-lb/ac difference in pine straw harvested in April 2006 was not significant between the two fertilization levels. Therefore, over 13 annual harvests in 15 years, fertilization did not have a consistent yearly effect on pine straw yields in this direct-seeded longleaf pine stand. In contrast, Chastain et al. (2007) reported that applying granular fertilizer (86 lb N, 38 lb P, and 71 lb K/ac) increased production of pine straw in a 22-year-old longleaf pine stand by 26% over a 3-year period. Longleaf Pine Earlier in this study, management practices affected longleaf pine growth from 1991 through 1994, but these responses were no longer statistically significant after the 1995 growing season (Haywood et al. 1998). This trend continued, and after the 1997 growing season, the fertilization treatment still did not significantly affect stand stocking, o.b. volume per acre, or volume per tree (Table 3). Basal area per acre in 1997 was significantly greater on the unfertilized plots than on the fertilized plots, but the difference was not biologically important. Additionally, none of the four subplot treatments significantly affected any of the longleaf pine variables in 1997. Across the study and 18 months before thinning, longleaf pine averaged 128 trees/ac, 107 ft/ac of basal area, and 3,875 ft or 120 green tn/ac of stemwood, and stem volume averaged 31 ft/tree. The number of pine trees harvested in June 1999 averaged 60 trees/ac. Basal area was reduced by 43 ft/ac, and 1,431 ft or 44 green tn/ac of stemwood was harvested. Three months after thinning, there were no statistical differences in the residual pine variables between the two fertilization treatments or among the four subplot treatments. Across the study in September 1999 longleaf Table 2. Oven-dried weight of pine straw harvested annually from 2002 through 2006 (9th through 13th harvests). Main plot treatments January 2002 February 2003 January 2004 April 2005 April 2006