- Email: [email protected]
Artificially regenerating longleaf pine in canopy gaps: initial survival and growth during a year of drought
Forest Ecology and Management 180 (2003) 25–36
Artificially regenerating longleaf pine in canopy gaps: initial survival and growth during a year of drought Dante Arturo Rodrı´guez-Trejoa,1, Mary L. Duryeab,*, Timothy L. Whitec, Jeff R. Englishc, John McGuired a
Divisio´n de Ciencias Forestales y del Ambiente, Universidad Auto´noma Chapingo, Chapingo, Edo. de Me´xico, C.P. 56230, km 38.5 Carretera, Texcoco, Mexico, Mexico b Institute of Food and Agricultural Sciences, University of Florida, P.O. Box 110200, Gainesville, FL 32611-0200, USA c School of Forest Resources and Conservation, University of Florida, 367 Newins-Ziegler Hall, P.O. Box 110410, Gainesville, FL 32611-0410, USA d Joseph W. Jones Ecological Research Center, Newton, GA 31770, USA Received 27 April 2002; accepted 25 October 2002
Abstract Nitrogen fertilization in the nursery, along with altering the configuration of forest gaps, may improve the reforestation success of longleaf pine seedlings. During the very droughty 1998 growing season in Florida and Georgia, survival was higher under the forest canopy than in small (0.10 ha, 36 m diameter) and large (1.6 ha, 144 m diameter) canopy gaps. In the large gaps, survival of containerized seedlings was higher along the edges, particularly the SW edge. Shade from adult trees and the nurse effect of shrubs increased survival, while grass competition reduced survival. During dry years part of the ‘‘exclusionary zone’’ along the edge of canopy gaps (SW sector) may serve as a ‘‘survival zone’’, at least in the short term. A model using oval-shaped gaps oriented from NW to SE, with an area of 0.25 ha is proposed to maximize the survival and growth of artificially regenerated longleaf pine seedlings. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Pinus palustris; Disturbances; Exclusionary zone; Nitrogen; Fertilization; Restoration
1. Introduction In the southeastern United States, natural disturbances of varying types (hurricanes, lightning, forest fires, microgaps by wildlife), play key roles in maintaining * Corresponding author. Tel.: þ1-352-392-1784; fax: þ1-352-392-4965. E-mail addresses: [email protected] (D.A. Rodrı´guez-Trejo), [email protected] (M.L. Duryea), [email protected] (T.L. White). 1 Tel.: þ595-95-21500x5468; fax: þ595-95-41957.
the structure and function of longleaf pine (Pinus palustris Mill.) ecosystems (Gordon et al., 1997; Gilliam and Platt, 1999). For longleaf pine, the dominant tree species in the longleaf pine ecosystem, these disturbances also play important roles in regeneration (Wahlenberg, 1946; Platt, 1998). Disturbances create forest gaps where increased availability of light, water and nutrient levels may increase regeneration success. Regeneration in gaps has been studied in tropical (e.g., Denslow, 1980; Martı´nez-Ramos et al., 1988; Dirzo et al., 1992), temperate (e.g. Gray and Spies, 1996) and semiarid (e.g. Branch et al., 1996) environments.
0378-1127/02/$ – see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 1 2 7 ( 0 2 ) 0 0 5 5 7 - 1
26
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
Forest gaps, the product of many types of disturbances, have recently become recognized as ecologically important features in temperate ecosystems (Whitmore, 1978, 1989; Oliver and Larson, 1990), driving the forest cycle through open, growth and closed phases (Whitmore, 1989). Longleaf pine has been largely recognized as shade intolerant, with its regeneration primarily concentrated in forest gaps (Wahlenberg, 1946; Platt et al., 1988). However, there is a lack of understanding of the ecosystem responses to overstory disturbances (Palik et al., 1997). Mitchell et al. (1996) note the need for better understanding of the community responses to disturbance regimes to improve forest ecosystem management. The objective of this research was to assess differences and trends in survival, morphology, growth and interactions with understory plants of bare-root and containerized longleaf pine seedlings planted in different growing conditions represented by three different gap types, and in different locations within large gaps. During this experiment (1998), the southeastern US experienced a severe drought (NOAA, 2000) that represented ‘‘extreme’’ conditions. At the planting site, annual precipitation between 1997 and 2000 varied little, with values between 1025 and 1205 mm. However, 1998 had more months (6) with low precipitation (<50 mm) than the other years. The number of months with precipitation <50 mm was 2 for 1997, 1 for 1999 and 3 for 2000 (International Paper Company, unpublished data).
2. Methodology 2.1. Plant production Two seedling types were compared: (1) bare-root seedlings produced in a Florida nursery and (2) containerized seedlings produced in a Georgia nursery. Seed was collected from International Paper Company’s second-growth forest land near Bainbridge, Georgia (seed lot ICHLL93). In the bare-root nursery, with sandy well-drained La Hogue series soil, classified as (mesic aquic argiudolls), seed was sown in October 1996 and seedlings were lifted in February 1998. Needles were clipped, and the seedlings fertilized and irrigated as scheduled for the production of
longleaf pine. Fall fertilization treatments consisted of 0, 2, or 4 extra applications (no N, low N and high N) of 167 kg/ha of ammonium nitrate every other week. Containerized seedlings were grown in a media of peat moss, vermiculite and perlite in polyethylene multi-pot trays with 40 cavities each (571 per m2) and a cell volume of 107 cm3. Seeds were spring-sown (March 1997). The needles were not clipped, and the seedlings were fertilized and irrigated as scheduled for longleaf production in this nursery. The fall fertilization treatments consisted of 0, 2, and 4 extra applications of ammonium nitrate solution (100 ppm N) every other week. 2.2. Morphological and nitrogen analysis of nursery seedlings Ten variables were measured (shoot diameter, height, root length and the biomass of shoot, root and total biomass) or calculated (shoot:root ratio, shoot relative dry weight and root relative dry weight). Foliar nitrogen (N) was determined by the micro Kjeldhal method (Bremner and Mulvaney, 1982). The containerized seedlings had 0.69, 0.88 and 0.87% foliar nitrogen for no N, low and high N, respectively. Bare-root seedlings had 0.81, 0.97 and 1.11% foliar nitrogen, respectively, for these treatments. 2.3. Plantation and experimental design Longleaf pine seedlings were planted in February 1998 at International Paper Company’s second-growth longleaf pine forest near Bainbridge, Georgia. Soils at this forest are the Norfolk and Orangeburg series (fine loamy, kaolinitic, thermic Typic Kandiuluts). The experimental design was a factorial split plot, with six sites (blocks) serving as plots and gaps acting as subplots. Every site contained three types of gaps: large 1.6 ha gap (144 m diameter), small 0.1 ha gap (36 m diameter), and intact canopy (no gap). The total number of gaps was 18 (6 sites 3 gaptypes). Thesub–sub-plots were the cardinal directions, the sub–sub–sub-plots were the seedling stock, and the sub–sub–sub–subplots corresponded to nursery N fertilization. Canopy gaps were created by clearcutting in 1996 in a secondgrowth longleaf pine forest that was 23 m in height. In the large gaps, 10 seedlings of each stock type, nursery N treatment, cardinal direction (NE, SE, SW,
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
NW) from the center, distance from the edge, and site were planted (2 m 2 m), for a subtotal of 4320 seedlings (6 sites 2 stock types 3 N levels 4 directions 3 distances 10 seedlings). In the small gaps, 10–13 seedlings per stock type, nursery N treatment, cardinal direction and site were planted, for a subtotal of 1704 seedlings. In the forest, 15 seedlings per stock type, nursery N treatment, cardinal direction and site were planted, for a sub-total of 2160 seedlings. In total, 8184 seedlings were planted in 120 sub–sub– sub–sub-plots. Containerized seedlings were planted with a planting dibble and the bare-root seedlings with a planting shovel. The roots of bare-root seedlings were pruned with a straight edge if they exceeded 30 cm in length at planting. The distance between trees within a plot was 1 m. No herbicide, irrigation or fertilization was applied following planting.
on a tripod 1 m above the floor and oriented north. The film negatives were converted to a digital format and imported from automated digitized imaging using Hemiview1 (Delta-T Devices Ltd., Burwell, Cambridge, UK). Hemiview1 was also used to calculate several indices of solar radiation based on canopy openness, using an index of the sun’s position and projected path of movement over a 6 month period. A gap light index (GLI) (sensu Canham, 1988) was calculated for each location and gap treatment (McGuire et al., 2001). 2.7. Statistical analysis The relative importance of the variables analyzed was evaluated using the linear statistical model yijklmn ¼ ai þ bj þ gk þ dl þ Zm þ ðabÞij þ ðagÞik þ ðadÞil þ ðaZÞim þ ðbgÞjk þ ðbdÞjl þ ðbZÞjm
2.4. Survival, morphology and growth Survival was recorded 1 year after planting. Also, two to three seedlings for each N treatment and seedling type per plot were harvested for morphological analysis after 1 year. Harvested seedlings were temporarily stored in a large cooler with ice and then at 2 8C in a refrigerator. Measurements (e.g. diameter, stem length, dry weight, etc.) were obtained from the seedlings. Then roots and shoots were separated, placed in an oven at 70 8C for 2 weeks, and weighed. 2.5. Characterization of interactions with understory Competing vegetation was characterized in every plot. Plant life forms (forbs, grasses, and shrubs), density and plant cover (%) by life form were observed ocularly from 120 one-square meter plots, in the referred sub–sub–sub–sub-plots. 2.6. Light Light data were obtained from a series of studies at the same sites in Georgia conducted by McGuire et al. (2001). Hemispherical photographs were taken, with the photographs recorded on 400-speed black and white film during uniformly cloudy days or mornings prior to sunrise according to Easter and Spies (1994). An 8 mm fisheye lens and camera assembly were leveled skyward
27
þ ðgdÞkl þ ðgZÞkm þ ðdZÞlm þ ðabgÞijk þ ðabdÞijl þ ðabZÞijm þ ðagdÞikm þ ðagZÞikm þ ðadZÞilm þ ðbgdÞjkl þ ðbgZÞjkm þ ðbdZÞjlm þ ðgdZÞklm þ ðabgdÞijkl þ ðabgZÞijkm þ ðabdZÞijlm þ ðagdZÞiklm þ ðbgdZÞjklm þ ðabgdZÞijklm þ Eijklmn
(1)
where a is the effect of the site (block) (random effect), b the effect of the gap (fixed effect), g the effect of the direction (fixed effect), d the effect of the seedling type (fixed effect), Z the effect of the N treatment (fixed effect), the combinations of factors represent the respective interactions, (ab) the whole plot error effect N(0, s2W ), and E the split-plot error effect (0, s2). Because the large gaps were the only planting sites with a variable for distance from the center of the gap, those also were analyzed with the same type of model, excluding the factor gap type and including the factor distance with three fixed effect levels. The hypotheses being tested were: Effects of individual factors were: Ho :
m1 ¼ m2 ¼ m3
all gaps main-effect means equal Ho :
m1 ¼ m2 ¼ m3 ¼ m4
all directions main-effect means equal
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
28
Ho :
m1 ¼ m2
all stock types main-effect means equal Ho :
m1 ¼ m2 ¼ m3
all nitrogen treatments main-effect means equal Interactions were: Ho : allðbgÞjk ¼ 0 no interaction among gap and direction And similarly for the other interactions. Seedling survival and all morphological variables were analyzed 1 year following planting. The data were processed using SAS (Statistical Analysis System1, version 6.12) analysis of variance (Proc Mixed), calculating the degrees of freedom with the Satterthwaite option ðddfm ¼ satterthÞ (Littell et al., 1999). Means and standard errors were obtained using the least squares means (lsmeans) statement. For the case of the pooled analysis for each variable, after running the program with Proc Glm, the three-way, four-way, and five-way interactions with a p-value larger than 0.25 were omitted to conduct an analysis with Proc Mixed using a reduced model to increase the power of the test. Growth was analyzed with Proc Mixed, adding the variables time 0 (planting moment), and time 1 (1 year after planting). Bud status (broken or non broken) was analyzed with the non-parametric chi-square (w2) test. To predict survival we used multiple regression models by first obtaining Pearson’s correlation coefficients among survival and all of the field variables (e.g. GLI, grass cover, total cover, etc.). Light data were obtained for three of the six blocks and our models and correlations with GLI included the data from these three blocks. Variables were selected by magnitude of their coefficient with survival and had to be independent of other variables, to prevent multicolinearity. The residuals of every selected independent variable were also checked to see if there was a trend. Survival was transformed using the arcsin function before analyses.
3. Results 3.1. Survival In both the pooled and reduced analyses of survival the interaction between seedling type and gap type
was significant ðP ¼ 0:0059Þ. Initial survival for containerized seedlings in the forest, small and large gaps was 35.1, 23.4, and 15.4%, respectively, while survival of bare-root seedlings was 12.3, 9.7, and 8.9%. In all of the cases containerized seedlings (24.6%) exhibited better survival than bare-root seedlings (10.3%) ðP ¼ 0:0002Þ. The N treatments did not have an effect on short-term survival ðP ¼ 0:5670Þ. Containerized seedlings exhibited negative correlations between initial survival (dependent variable) and GLI (independent variable) (r 2 ¼ 0:721, P ¼ 0:0001), with the following multiple regression model (P ¼ 0:0001, r 2 ¼ 0:583): S ¼ 80:983 0:862ðGLIÞþ 0:031ðWCÞ 0:221ðGCÞ þ 0:062ðFCÞ þ 0:005ðGLIGCÞ þ 0:012ðGLIFCÞ 0:033ðGCFCÞ
(2)
where S is the survival (%), GLI the gap light index, WC the shrub cover (%), GC the grass cover (%), FC the forb cover (%), GLIGC the interaction among GLI and GC, GLIFC the interaction among GLI and FC, and GCFC the interaction among GC and FC. Survival was negatively correlated with GLI for bare-root seedlings (r 2 ¼ 0:330, P ¼ 0:0100), but the linear regression model was not significant ðP ¼ 0:251Þ. 3.2. Morphological variables In the containerized seedlings, stem diameter after 1 year was smaller in the forest (10.8 mm) when compared with small gaps (11.3 mm) and large gaps (11.7 mm) ðP ¼ 0:0013Þ. The shoot:root ratio was influenced by the interaction of gap by N treatment ðP ¼ 0:0066Þ. In the forest low N was different from no N but not from high N. In the small gaps, shoot:root ratio in high N was higher than no N and low N. In the large gaps, no clear differences among N fertilizations were found. The relative shoot weight also exhibited a gap by N treatment interaction ðP ¼ 0:0069Þ. In the forest, higher levels of N corresponded to higher relative shoot weights, but no differences among N treatments were observed in the large gaps (Fig. 1). The relative root weight responded accordingly also ðP ¼ 0:0069Þ. In the case of bare-root seedlings, after 1 year in the field, shoot dry weight was larger at high N levels (6.3, 8.1, and 8.5 g, for no N, low N and high N).
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
29
Fig. 1. Mean (standard error) relative shoot weight for containerized seedlings in different gap types and N treatments (0, 2 and 4).
3.3. Bud break Non-parametric (w2) analysis in a pooled comparison showed that a larger proportion of bare-root seedlings (67.5%) were breaking buds in February, in comparison with 57.8% for containerized seedlings ðP ¼ 0:006Þ. Bud break timing of containerized seedlings was
influenced by gap type: 70.4% for forest, 67.5% for small gaps, and 43.5% for large gaps ðP ¼ 0:001Þ. 3.4. Biomass Biomass comparisons were made for dry weight at planting time compared with dry weight 1 year later
Fig. 2. Average survival of containerized seedlings in large 1.6 ha gaps (144 m diameter) 1 year after planting by cardinal direction and distance. The values with different letters exhibited statistically significant differences.
30
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
for the three nursery N treatments. In the containerized seedlings, root (2.5, 3.3 g, P ¼ 0:0023) and total dry weights (5.7, 6.8 g, P ¼ 0:0032) were different between the initial planting time and 1 year later. No variable responded to N treatment. For bare-root seedlings, root dry weight increased with time (8.0, 9.5 g, P ¼ 0:0010). All the other variables showed no change or exhibited a reduction in biomass related to mortality of tissue. For instance, the total seedling biomass exhibited no difference between planting time and 1 year later (18.6, 17.1 g, P ¼ 0:1252). Only shoot weight had differences due to N treatments (7.9, 9.3, 10.1 g, for no N, low and high N, P ¼ 0:0300). 3.5. Large gaps 3.5.1. Survival The containerized seedlings had higher survival than bare-root seedlings, and survival was influenced by the cardinal direction by distance interaction in large gaps, with highest survival at the southern edges of the gaps ðP ¼ 0:0020Þ. Seedlings planted at the edge of gaps had the highest survival, particularly in the SW cardinal direction (Fig. 2). In all directions (except in the NE), seedlings planted in the center had the lowest survival. The Pearson’s correlation coefficient among survival and distance was 0.72 ðP ¼ 0:0001Þ. The survival of bare-root seedlings was independent of distance from the center of the gap, with average survivals of only 6.0, 7.9, and 12.4% for center, middle, and edge areas of the gap, respectively, ðP ¼ 0:2388Þ. There was also no effect on survival for cardinal direction across the gap, N treatment, or any of the interactions. In the pooled analysis across seedling type, survival was affected only by the interaction between cardinal direction and distance ðP ¼ 0:0132Þ. The Pearson’s correlation coefficient for survival and distance was 0.53 ðP ¼ 0:0001Þ. 3.5.2. Morphological variables Root length was influenced by the distance by N treatment interaction ðP ¼ 0:0408Þ. In the center of the gap, no differences were found among N levels (22.0, 20.6 and 24.8 cm for levels 0, 2 and 4, respectively). In the middle part of the gap, high N was lower than the control (25.1, 21.7, 18.5 cm, for levels 0, 2
and 4). At the edge, high N was higher than the other two levels (22.6, 23.3, 27.8 cm, for levels 0, 2 and 4). 3.5.3. Bud break Bud break timing did not vary with distance, direction nor N treatment (w2-test). 3.5.4. Interactions with understory plants Total plant cover ðgrasses þ forbs þ shrubsÞ was also affected by the interaction of cardinal direction with distance ðP ¼ 0:0226Þ. In northern quadrants, the edges had the lowest total vegetation cover (NE: 65.8, 70.8, 60% for center, middle, and edge; NW: 71.7, 70.8, 61.7% for center, middle and edge). In all the directions, except for the SE, the central or the middle areas of the gaps had the highest total plant cover (SE: 69.2, 65.8, 74.2%; SW: 80.8, 63.3, 65.0%). In the nonparametric analysis, the only association found was forbs increasing with distance from the center of the gap (P ¼ 0:042, w2-test). For containerized seedlings there was a negative correlation between survival and grass cover r ¼ 0:34, P ¼ 0:0071). Bare-root seedling survival was negatively correlated with grass cover r ¼ 0:25, P ¼ 0:0497), and total plant cover r ¼ 0:28, P ¼ 0:0284).
4. Discussion 4.1. Survival of seedlings Survival was higher for containerized seedlings than for bare-root seedlings, as has been shown in other studies (e.g., Barber and Smith, 1996; Barnett et al., 1996). However, survival was higher under the forest canopy than in small or large gaps contradicting other studies. For instance, natural longleaf pines had higher survival in areas of low adult density (Wahlenberg, 1946; Platt et al., 1988; Boyer, 1993; Grace and Platt, 1995; Mosser, 1997) because of its shade-intolerance and competition intolerance. One possible reason for these contradictory results is the 1998 climate. The present research was conducted during a very dry and hot year. Precipitation in the spring for the period April–June was the lowest in 104 years in Florida (and southern Georgia). During this period, many localities in this region received less than half of normal rainfall (generally under 150 mm) (NOAA,
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
31
2000). Also 110 days of maximum daily temperatures averaging 36 8C were observed (International Paper Co., unpublished data), so the types of sites that in normal years adversely affect survival, exhibited better growing conditions during this hot dry year. An analysis of the main environmental factors may help explain this finding. 4.2. Light The GLI decreased from large gaps to the forest (84.2, 67.3, 48.2%), and from the center to the edge of gaps. Light availability in the southern portions of gaps was lower than in the northern parts (McGuire et al., 2001), and was inversely correlated to survival. Despite the reported intolerance of longleaf pine to shade, photosynthetically useful light also reaches shady areas. The forest floor at gap edges receives both diffuse and direct light, while the forest (nongaps) receives mainly diffuse light, and the gap center has mainly direct light. Diffuse radiation, relatively abundant at the edge of a gap, is relatively rich in blue wave lengths, the most effective in photosynthesis (Smith et al., 1997). So, even if shade does not represent the highest light condition for longleaf pine seedlings, growth may still occur there. 4.3. Temperature and humidity The greatest extremes in temperature in gaps occur when gap diameter is roughly equal to 1.5 times the height of surrounding trees, because of the combined effect of shade and ventilation (Geiger et al., 1995). In the central portion of large gaps, with diameters three times the height of surrounding trees, the environmental conditions are approximately the same conditions that would be present in much larger openings (Minckler and Woerheide, 1965). In our study, the large gap diameter was six times the height of the surrounding trees, so the conditions in the center were similar to much larger gaps. In these large gaps, survival increased from the center to edge, with the highest value (40%) in the SW-edge sector. In the northern hemisphere, in general, the highest surface temperatures and lowest fuel moisture are on SW slopes of mountains (Schroeder and Buck, 1970). Similarly, Randall and Johnson (1998) found for Pseudotsuga menziesii (Mirb.)
Fig. 3. Reshaping forest gaps to be oval-shaped patches oriented from NW to SE in their longitudinal axis may maximize the survival and growth of longleaf pine regeneration in dry years (matching oval-shaped 0.25 ha gaps with the SW edge of the 1.6 ha circular gaps, where survival was higher).
Franco, Abies procera Rehd and Pinus flexilis James, that S and SW aspects had reduced first and/or third year survival. Schneider et al. (1998) showed that the NE-facing slopes were generally more cool and moist and more favorable for seedlings of P. menziesii (Mirb.) Franco and Pinus ponderosa Dougl. Ex Laws. survival. A gap is the inverse of a small mountain. In a gap, the absence of the SW slope makes the radiation generally reside for more time in the year in the NE sector giving the SW sector more humidity (Fig. 3). Increases in temperature lead to lower relative humidity, and higher amounts of incident light are associated with higher temperatures. 4.4. Interactions with understory plants In this experiment, there was a dual effect of other plants and trees on longleaf pine seedling survival. Large trees could have a negative or a positive effect on the seedlings, depending on the conditions. The term ‘‘competition’’ in the present study is defined as the direct interaction of two plants attempting to exploit the same limiting resources (Tilman, 1987; Connell, 1990). In normal years, fine root competition from adult trees affects young seedling survival, but in droughty years shade may actually reduce evapotranspiration and thereby improve short-term survival. In a circular gap, competition factors controlled by adult trees at the edge, such as diffuse solar radiation, air movements and outgoing radiation, should operate in a nearly concentric pattern and factors related to the
32
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
slanting rays of direct solar radiation will follow a crescentic pattern (Smith et al., 1997). The competitive effects for water and nutrients created by the fine roots of adult trees may extend as far as 16 m into gaps (Boyer, 1963; Brockway and Outcalt, 1998). Over the long term, natural longleaf pine regeneration becomes clustered near the center of gaps, encircled by a wide zone from which seedlings are generally excluded due to the direct effects of fine root competition and indirect effects on needle accumulation as fine fuel (Brockway and Outcalt, 1998). In our study during droughty conditions, grasses had a negative effect (competition effect), and shrubs produced a positive effect (nurse effect) on longleaf pine seedlings. Shade protects longleaf pine seedlings from high soil temperatures and desiccating conditions (Myers, 1992). Grass cover alone is less effective than scrub oak cover in maintaining lower soil temperatures. When moderate, the shade from oaks favors seedling survival by retarding soil moisture loss and preventing soil overheating (Wahlenberg, 1946). In our multiple linear regression for estimating survival of containerized seedlings, GLI and grass cover negatively affected survival. In this study, it was hypothesized that GLI would be positively related to survival, but unusually dry conditions caused a negative correlation that was highly significant. Many grasses have morphological and physiological characteristics, such as a fibrous root system and C4 photosynthesis, which make them strong competitors. Some may overtop pine seedlings, reducing the availability of light (Minogue et al., 1991; Morris et al., 1993). Grasses can deplete moisture from the rooting zone of newly establishing conifers (Cleary et al., 1982). For instance, Pinus radiata D. Don doubled its uptake of N after the removal of pasture competition. Shrubs tend to deplete moisture also, but usually from deeper in the soil profile (Cleary et al., 1982). In our model, survival of containerized seedlings was positively related to shrub vegetation. There are several examples in the literature of nurse trees aiding initial survival. Wahlenberg (1946) and Cleary et al. (1982) observe that temperature extremes are moderated by brush, with air and soil temperatures under brush 11–17 8C lower than in adjacent openings. Hydraulic lift of deep soil water to the surface may also be active in this system, as has been seen for other
droughty soils. Artemisia tridentata roots cause shallow water depletion during the day, but by night the liquid absorbed from soil by deeper roots is transported to and lost from roots in upper and drier soil layers (Richards and Caldwell, 1987; Caldwell and Richards, 1989; Caldwell et al., 1998). If this type of water movement was caused by shrubs in our experiment, water taken up by shrubs and even adult trees may have become available to the longleaf pine seedlings. It is important to recognize that site differences and weather variation, will alter the competition at a site from year to year. For instance, seedlings of Pinus strobus L. had N as the most important resource limiting growth one year, but the next year light was the most important limiting factor (Elliott and Vose, 1995). Similarly, the survival patterns observed among gaps and within a large gap will vary because the limiting factors (e.g., temperature and water availability) will change from year to year. 4.5. Morphological variables and biomass In the present study, bud break in the spring was earlier at high N for bare-root seedlings as it was for P. menziesii (Mirb.) Franco (Margolis and Waring, 1986). Also, bud break in containerized seedlings occurred first in large gaps, then in small gaps, and then in the forest. Similarly, van Pelt and Franklin (1999) found (for P. menziesii (Mirb.) Franco and Tsuga heterophylla (Raf.) Sarg. seedlings) a concentric pattern of bud break occurring earliest towards gap centers and latest at the edges, with increasing direct light and length of the growing season at the center. In large gaps, root length of containerized seedlings tended to be shorter at higher N under high temperatures (middle area), but seedlings with the high N fertilization treatment had the longest root length at the edge under lower temperatures. This may be explained by the fact that under water stress, nutrient stress, or increased temperatures, cell expansion is more adversely affected than photosynthesis, thus producing excess carbohydrates that trees store (Landsberg and Gower, 1997). The proportion of carbohydrates allocated to root growth is higher when water and nutrients are limiting (Comeau and Kimmins, 1989; Landsberg and Gower, 1997).
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
In bare-root seedlings, N fertilization in the nursery continued to result in higher shoot biomass and lower relative tap root weight in the field. In containerized seedlings, the shoot:root ratio was lower at high N treatments in small gaps, consistent with previous studies (Waring and Running, 1998). Growth rates are correlated to N status under an appropriate nutritional ˚ gren and Ingestad, 1987; Waring, 1989). A balance (A decrease in N supply will result in a decrease in N in roots and shoots. Yet reduced N availability produces a relative increase in C allocation to roots (Landsberg and Gower, 1997). Lambers et al. (1998) mention cytokinins as the cause of this behavior. In high N environments there is a high rate of cytokinin synthesis in roots, which is exported to the leaves. As a result, photosynthetic capacity and leaf expansion rate increases, requiring a high rate of sugar consumption. The low rate of sugar export results in a slow root growth rate. Conversely, under low N conditions, the same relationship results in rapid root growth rates. 4.6. Growth Palik et al. (1997) note that growth of longleaf pine seedlings increased significantly in the center of gaps. Boyer (1963) mentions that growth of seedlings declines as overstory density increases. In our study during this droughty year, biomass loss of bare-root seedlings in the center and middle of large gaps and a related loss of vigor was observed. In the case of containerized seedlings, there was higher growth at the edge, most likely due to greater shading and protection from excessive temperatures and probably higher water availability. In containerized seedlings, buds broke first in the forest, then in small gaps, and lastly in large gaps, following a gradient of less harsh to more harsh conditions. Khan et al. (1996) found similar results in P. menziesii, where both very low and very high levels of humidity delayed bud development. 4.7. Management implications Palik et al. (1997) proposed the use of circular gaps larger than 0.14 ha (42 m diameter) to maximize longleaf pine seedling growth, because N and light availability are maximized. Similarly, Brockway and
33
Outcalt (1998) recommend creating gaps at least 40– 50 m in diameter (0.13–0.20 ha) to minimize intraspecific root competition between adult pines and seedlings. However, McGuire et al. (2001) observed that regeneration of longleaf pine may be maximized in gaps as small as 0.10 ha (36 m diameter). Building on those results and the results of our research, where the highest survival was at the SW edges of large circular gaps, we propose an approach using oval-shaped gaps to maximize both survival and growth. The longitudinal axis of the oval would be oriented from NW to SE, in such a way that includes most of the SW edge part of a large gap (Fig. 3). Increasing the width of the oval will decrease competition from adult pines, increasing seedling survival and growth during non-droughty years. The size of these oval gaps should range from 0.25 (the surface area of the SW sector) to 0.50 ha. Further research is needed for testing the efficacy of oval gaps. Some of the models for estimating future global warming for a significant part of the longleaf pine range, including the area where this study was conducted, indicate a 3–5 8C increase in mean annual surface air temperature, from 73 to þ73 mm change in mean annual precipitation, and less soil moisture (related to a larger evaporation and transpiration) (The Hadley Centre, 2002; United States Environmental Protection Agency, 2002). This means that one of the probable future scenarios for part of the longleaf pine ecosystem range is more drought. If that were the case, the results of work like this one may be more relevant. In such climatically altered environments, with less soil humidity, the temporary survival zone may gain more importance. The potentially higher available humidity in the SW edge may imply less root development of adult trees, implying less competition, and higher eventual survival to the seedlings. The exclusion zone exists. However, it is hardly identifiable when trees are older. There are no mature stands originated from gaps, surrounded by the 16 m wide band. Eventually, at least some trees survive in the exclusion zone, making that band not always evident across time. One of the mechanisms for that particular recruitment may be related to the results of the present work. If droughts become more frequent, it may be advisable to plant in the SW edges of gaps.
34
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
5. Conclusions and recommendations According to the literature, longleaf pine seedlings survive better in the central area of gaps, beyond a 16 m wide belt at the edge where fine root competition from adult trees, shade from the overstory and more intense surface fires exclude regeneration. However, in our research using artificial regeneration, we found that during an extremely dry year (1998), the shadier area near the gap edge in the SW sector was the best area for short-term seedling survival. During extremely dry conditions, the center of the gap is less favorable for seedling survival. In normal years, seedlings survive better in gaps than under canopies (Boyer, 1963; Brockway and Outcalt, 1998). However, in the especially dry year for this study the opposite occurred. Several shoot and root morphological traits were shown to be improved by fall N nursery fertilization, even 1 year after planting. Also, bud break was earliest at the sites with less harsh conditions (forest), followed by those with harsher conditions. Apparently, some of the understory plants (grasses) negatively affected the survival of seedlings because of direct competition, while shrubs ameliorated environmental temperature extremes, increasing short-term survival. The availability of the many environmental resources in gap regeneration operate not only in a spatial scale, but also in a time scale, and their effect may change radically when the most crucial environmental variables (e.g., water availability) change. When this happens, the micro-sites that are normally harsh may become the best for short-term seedling establishment during extreme drought events. During those extreme years in large gaps, the approximately 16 m wide exclusion zone could be at least temporarily re-named the ‘‘survival zone’’, particularly in the SW portion. Oval-shaped gaps with a surface area larger or equal to 0.25 ha and oriented from NW to SE, may maximize both survival and growth of longleaf pine seedlings. Because it improves performance-related traits at planting sites under difficult conditions (e.g., drought, competition), N nursery fertilization is recommended. Among those traits there may be faster bud break that may represent an advantage in a competitive environment. Also, a larger shoot dry weight may provide more resistance against physical agents, plus plant
biomass has been related to good field performance. Some of these improved traits may mean that Nfertilized longleaf will have a shorter grass stage. However, N did not increase stem diameter in our study, and longleaf pine starts to leave the grass stage after reaching 2.5 cm diameter (Boyer, 1990). We recommend further research with N fertilization. In addition, comparing or modeling gap sizes, shapes and orientations across forest sites in the south would be very beneficial.
Acknowledgements The authors thank two anonymous reviewers, and Henry Gholz, Doria Gordon and Robert J. Mitchell for their invaluable comments during the research. We also wish to thank Dave Noletti and Steve Gilly for logistic support, and the Divisio´ n de Ciencias Forestales, Universidad Auto´ noma Chapingo, CONACYT and The Florida-Mexico Institute for supporting the Ph.D. studies of the first author. This is Florida Agricultural Experiment Station Journal #R-09113 of the Institute of Food and Agricultural Sciences, University of Florida, Gainesville, FL. References ˚ gren, G.I., Ingestad, T., 1987. Root:shoot ratio as a balance A between nitrogen productivity and photosynthesis. Plant Cell Environ. 10, 579–586. Barber, B., Smith, P., 1996. Comparison of first year survival between container-grown and bare-root longleaf pine seedlings outplanted on a site in southeast Texas. In: Kush, J.S. (Comp.), Proceedings of the First Longleaf Alliance Conference on Longleaf Pine: A Regional Perspective of Challenges and Opportunities, September 17–19. The Longleaf Alliance, Mobile, AL, pp. 50–51. Barnett, J.P., Haywood, J.D., Sword, M.A., 1996. Artificial regeneration of longleaf pine; reevaluation technique. In: Kush, J.S. (Comp.), Proceedings of the First Longleaf Alliance Conference; Longleaf Pine: A Regional Perspective of Challenges and Opportunities, September 17–19. The Longleaf Alliance, Mobile, AL, pp. 53–55. Boyer, W.D., 1963. Development of longleaf pine seedlings under parent trees. Forestry Experimental Station Research Paper SO4. USDA Forest Service, South, New Orleans, LA, 5 pp. Boyer, W.D., 1990. Pinus palustris Mill. In: Burns, R.M., Honkala, B.H. (Eds.), Silvics of North America, vol. 1. Conifers. Forestry Service Agricultural Handbook 654. USDA, pp. 405– 412.
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36 Boyer, W.D., 1993. Long-term development of regeneration under longleaf pine seed tree and shelterwood stands. South. J. Appl. For. 17, 10–15. Branch, L., Villarreal, D., Hierro, J.L., Portier, K.M., 1996. Effects of local extinction of the plains vizcacha (Lagostomus maximus) on vegetation patterns in semi-arid scrub. Oecologia 106, 389–399. Bremner, J.M., Mulvaney, C.S., 1982. Nitrogen-total. In: Methods of Soil Analysis, Part 2. Chemical and Microbiological Properties. Agronomy Monograph No. 9. N.P., pp. 595–607. Brockway, D.G., Outcalt, K.W., 1998. Gap-phase regeneration in longleaf pine wire grass ecosystems. For. Ecol. Manage. 106, 125–139. Caldwell, M.M., Richards, J.H., 1989. Hydraulic lift: water efflux from upper roots improves effectiveness of water uptake by deep roots. Oecologia 79, 1–5. Caldwell, M.M., Dawson, T.E., Richards, J.H., 1998. Hydraulic lift: consequences of water efflux from the roots of plants. Oecologia 113, 151–161. Canham, C.D., 1988. Growth and canopy architecture of shadetolerant trees response to canopy gaps. Ecology 6, 786–795. Cleary, B.D., Greaves, R.D., Hermann, R.K., 1982. Regenerating Oregon’s forests. Oregon State University, USDA Cooperative Extension Service, Corvallis, OR, 282 pp. Comeau, P.G., Kimmins, J.P., 1989. Above and below-ground biomass and production of lodgepole pine on sites with differing soil moisture. Can. J. For. Res. 19, 447–454. Connell, J.H., 1990. Apparent versus ‘‘real’’ competition in plants. In: Grace, J.B., Tilman, D. (Eds.), Perspectives on Plant Competition. Academic Press, San Diego, pp. 9–26. Denslow, J.S., 1980. Gap partitioning among tropical rain forest trees. Biotropica 12, 47–55. Dirzo, R., Horvitz, C.C., Quevedo, H., Lo´ pez, M.A., 1992. The effects of gap size and age on the understory herb community of a tropical Mexican rain forest. J. Ecol. 80, 809–822. Easter, M.J., Spies, T.A., 1994. Using hemispherical photography for estimating photosynthetic photon flux density under canopies in gaps in Douglas-fir forests of the Pacific northwest. Can. J. For. Res. 24, 2050–2058. Elliott, K.J., Vose, J.M., 1995. Evaluation of the competitive environment for white pine (Pinus strobus L.) seedlings planted on prescribed burn sites in the southern Appalachians. For. Sci. 41, 513–530. Geiger, R.R., Aron, H., Todhunter, P., 1995. The Climate Near the Ground, 5th ed. Vieweq, Braunschweig, Germany. Gilliam, F.S., Platt, W.J., 1999. Effects of long-term fire exclusion on tree species composition and stand structure in an oldgrowth Pinus palustris (longleaf pine) forest. Plant Ecol. 140, 15–26. Gordon, D.R., Provencher, L., Hardesty, J.L., 1997. Measurement scales and ecosystem management. In: Pickett, S.T.A., Ostfeld, R.S., Shachak, M., Likens, G.E. (Eds.), The ecological basis of conservation: heterogeneity, ecosystems and biodiversity. Chapman & Hall, New York, pp. 262–273. Grace, S.L., Platt, J.W., 1995. Effects of adult tree density and fire on the demography of pregrass stage juvenile longleaf pine (Pinus palustris Mill.). J. Ecol. 83, 75–86.
35
Gray, A.N., Spies, T.A., 1996. Gap size, within-gap position and canopy structure effects on conifer seedling establishment. J. Ecol. 84, 635–645. Khan, S.R., Rose, R., Hasse, D.L., Sabin, T.E., 1996. Soil water stress: its effects on phenology, physiology, and morphology of containerized Douglas-fir seedlings. New Forests 12, 19–39. Lambers, H., Chapin III, F.S., Pons, T.L., 1998. Plant Physiological Ecology. Springer, New York, 540 pp. Landsberg, J.J., Gower, S.T., 1997. Applications of Physiological Ecology to Forest Management. Academic Press, San Diego, 354 pp. Littell, R.C., Milliken, G.A., Stroup, W.W., Wolfinger, R.D., 1999. SAS System for Mixed Models. SAS Institute, Cary, NC, 633 pp. Margolis, H.A., Waring, R.H., 1986. Carbon and nitrogen allocation patterns of Douglas-fir seedlings fertilized with nitrogen in autumn. II. Field performance. Can. J. For. Res. 16, 903–909. Martı´nez-Ramos, M., Sarukha´ n-Kermes, J., Pinero, D., 1988. The demography of tropical trees in the context of forest gap dynamics. Symp. Br. Ecol. Soc. 28, 293–313. McGuire, J.P., Mitchell, R.J., Moser, E.B., Pecot, S.D., GjestadDean, H., Hedman, C.W., 2001. Gaps in a gappy forest: plant resources, longleaf pine regeneration, and understory response to tree removal in longleaf pine savannas. Can. J. For. Res. 31, 765–778. Minckler, L.S., Woerheide, J.D., 1965. Reproduction of hardwoods 10 years after cutting as affected by site and opening size. J. For. 63, 103–107. Minogue, P.J., Cantrell, R.L., Griswold, H.C., 1991. Vegetation management after plantation establishment. In: Duryea, M.L., Dougherty, P.M. (Eds.), Forest Regeneration Manual. Kluwer Academic Publishers, The Netherlands, pp. 335–358. Mitchell, R.J., Palik, B.J., Kirkman, L.K., Hedman C.W., Pecot, S.D., McGuire, Gjerstad, D.H., 1996. Gaps in longleaf pine forests: a tool for ecosystem management. In: Kush, J.S. (Comp.), Proceedings of the First Longleaf Alliance Conference on Longleaf Pine: A Regional Perspective of Challenges and Opportunities, September 17–19. The Longleaf Alliance, Mobile, AL, 127 pp. Morris, L.A., Moss, S.A., Garbett, W.S., 1993. Competitive interference between selected herbaceous and woody plants and Pinus taeda L. during two growing seasons following planting. For. Sci. 30, 166–187. Mosser, W.K., 1997. Artificial establishment of longleaf pine under varying overstories. Annual Report. Tall Timbers Research Station, pp. 20–21. Myers, R.L., 1992. Scrub and high pine. In: Myers, R.L., Ewel, J.J. (Eds.), Ecosystems of Florida. University of Central Florida Press, Orlando, FL, pp. 150–193. NOAA (National Oceanic and Atmospheric Administration), 2000. Home page: http://www.ncdc.noaa.gov/ol/climate/research/1998/ jun/jun98.html#recent. Oliver, C.D., Larson, B.C., 1990. Forest Stand Dynamics. Wiley, New York, 467 pp. Palik, B.J., Mitchell, R.J., Houseal, G., Pederson, N., 1997. Effects of canopy structure on resource availability and seedling responses in longleaf pine ecosystem. Can. J. For. Res. 27, 1458–1464.
36
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
Platt, W.J., 1998. Southeastern pine savannas. In: Anderson, R.C., Fralish, J.S., Baskin, J.M. (Eds.), Savannas, barrens, and rock outcrop plant communities of North America. Cambridge University Press, Cambridge, pp. 23–51. Platt, W.J., Evans, G.W., Rathbun, S.L., 1988. The population dynamics of a long-lived conifer (Pinus palustris). Am. Nat. 131, 491–525. Randall, W., Johnson, G.R., 1998. The impact of environment and nursery on survival and early growth of Douglas-fir, noble fir and white pine: a case study. West. J. Appl. For. 13, 137–143. Richards, J.H., Caldwell, M.M., 1987. Hydraulic lift: substantial nocturnal water transport between soil layers by Artemisia tridentata roots. Oecologia 73, 486–489. Schneider, W.G., Knowe, S.A., Harrington, T.B., 1998. Predicting survival of planted Douglas-fir and ponderosa pine seedlings on dry, low elevation sites in northwestern Oregon. New For. 15, 139–159. Schroeder, M.J., Buck, C., 1970. Fire weather. Forestry Service Agricultural Handbook 360. USDA, Washington, DC. Smith, D.M., Larson, B.C., Kelty, M.J., Ashton, P.M.S., 1997. The Practice of Silviculture: Applied Forest Ecology, 9th ed. Wiley, New York, 537 pp. The Hadley Centre, 2002. http://www.metoffice.com/research/ hadleycentre/index.html.
Tilman, D., 1987. The importance of mechanisms of interspecific competition. Am. Nat. 129, 769–774. United States Environmental Protection Agency, 2002. Global warming. State impacts. Georgia. http://www.epa.gov/globalwarming/impacts/stateimp/georgia/index.html. van Pelt, R., Franklin, V., 1999. Response of understory trees to experimental gaps in old-growth Douglas-fir forests. Ecol. Appl. 9, 504–512. Wahlenberg, W.G., 1946. Longleaf pine: its use, ecology, regeneration, protection, growth and management. CL. Pack Forestry Foundation and USDA Forestry Service, Washington, D.C., 426 pp. Waring, R.H., 1989. Ecosystems: fluxes of matter and energy. In: Cherrett, J.M., Bradshaw, A.D., Grubb, P.J., Goldsmith, F.B., Krebs, J.R. (Eds.), Ecological Concepts: The Contribution of Ecology to an Understanding of the Natural Word. Blackwell, Oxford, pp. 17–41. Waring, R.H., Running, S.W., 1998. Forest Ecosystems, 2nd ed. Academic Press, San Diego, 370 pp. Whitmore, T.C., 1978. Gaps in the forestry canopy. In: Tomlinson, P.B., Zimmerman, M.H. (Eds.), Tropical Trees as Living Systems. Cambridge University Press, New York, pp. 639–655. Whitmore, T.C., 1989. Canopy gaps and the two major groups of forest trees. Ecology 70, 536–538.
Artificially regenerating longleaf pine in canopy gaps: initial survival and growth during a year of drought Dante Arturo Rodrı´guez-Trejoa,1, Mary L. Duryeab,*, Timothy L. Whitec, Jeff R. Englishc, John McGuired a
Divisio´n de Ciencias Forestales y del Ambiente, Universidad Auto´noma Chapingo, Chapingo, Edo. de Me´xico, C.P. 56230, km 38.5 Carretera, Texcoco, Mexico, Mexico b Institute of Food and Agricultural Sciences, University of Florida, P.O. Box 110200, Gainesville, FL 32611-0200, USA c School of Forest Resources and Conservation, University of Florida, 367 Newins-Ziegler Hall, P.O. Box 110410, Gainesville, FL 32611-0410, USA d Joseph W. Jones Ecological Research Center, Newton, GA 31770, USA Received 27 April 2002; accepted 25 October 2002
Abstract Nitrogen fertilization in the nursery, along with altering the configuration of forest gaps, may improve the reforestation success of longleaf pine seedlings. During the very droughty 1998 growing season in Florida and Georgia, survival was higher under the forest canopy than in small (0.10 ha, 36 m diameter) and large (1.6 ha, 144 m diameter) canopy gaps. In the large gaps, survival of containerized seedlings was higher along the edges, particularly the SW edge. Shade from adult trees and the nurse effect of shrubs increased survival, while grass competition reduced survival. During dry years part of the ‘‘exclusionary zone’’ along the edge of canopy gaps (SW sector) may serve as a ‘‘survival zone’’, at least in the short term. A model using oval-shaped gaps oriented from NW to SE, with an area of 0.25 ha is proposed to maximize the survival and growth of artificially regenerated longleaf pine seedlings. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Pinus palustris; Disturbances; Exclusionary zone; Nitrogen; Fertilization; Restoration
1. Introduction In the southeastern United States, natural disturbances of varying types (hurricanes, lightning, forest fires, microgaps by wildlife), play key roles in maintaining * Corresponding author. Tel.: þ1-352-392-1784; fax: þ1-352-392-4965. E-mail addresses: [email protected] (D.A. Rodrı´guez-Trejo), [email protected] (M.L. Duryea), [email protected] (T.L. White). 1 Tel.: þ595-95-21500x5468; fax: þ595-95-41957.
the structure and function of longleaf pine (Pinus palustris Mill.) ecosystems (Gordon et al., 1997; Gilliam and Platt, 1999). For longleaf pine, the dominant tree species in the longleaf pine ecosystem, these disturbances also play important roles in regeneration (Wahlenberg, 1946; Platt, 1998). Disturbances create forest gaps where increased availability of light, water and nutrient levels may increase regeneration success. Regeneration in gaps has been studied in tropical (e.g., Denslow, 1980; Martı´nez-Ramos et al., 1988; Dirzo et al., 1992), temperate (e.g. Gray and Spies, 1996) and semiarid (e.g. Branch et al., 1996) environments.
0378-1127/02/$ – see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 1 2 7 ( 0 2 ) 0 0 5 5 7 - 1
26
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
Forest gaps, the product of many types of disturbances, have recently become recognized as ecologically important features in temperate ecosystems (Whitmore, 1978, 1989; Oliver and Larson, 1990), driving the forest cycle through open, growth and closed phases (Whitmore, 1989). Longleaf pine has been largely recognized as shade intolerant, with its regeneration primarily concentrated in forest gaps (Wahlenberg, 1946; Platt et al., 1988). However, there is a lack of understanding of the ecosystem responses to overstory disturbances (Palik et al., 1997). Mitchell et al. (1996) note the need for better understanding of the community responses to disturbance regimes to improve forest ecosystem management. The objective of this research was to assess differences and trends in survival, morphology, growth and interactions with understory plants of bare-root and containerized longleaf pine seedlings planted in different growing conditions represented by three different gap types, and in different locations within large gaps. During this experiment (1998), the southeastern US experienced a severe drought (NOAA, 2000) that represented ‘‘extreme’’ conditions. At the planting site, annual precipitation between 1997 and 2000 varied little, with values between 1025 and 1205 mm. However, 1998 had more months (6) with low precipitation (<50 mm) than the other years. The number of months with precipitation <50 mm was 2 for 1997, 1 for 1999 and 3 for 2000 (International Paper Company, unpublished data).
2. Methodology 2.1. Plant production Two seedling types were compared: (1) bare-root seedlings produced in a Florida nursery and (2) containerized seedlings produced in a Georgia nursery. Seed was collected from International Paper Company’s second-growth forest land near Bainbridge, Georgia (seed lot ICHLL93). In the bare-root nursery, with sandy well-drained La Hogue series soil, classified as (mesic aquic argiudolls), seed was sown in October 1996 and seedlings were lifted in February 1998. Needles were clipped, and the seedlings fertilized and irrigated as scheduled for the production of
longleaf pine. Fall fertilization treatments consisted of 0, 2, or 4 extra applications (no N, low N and high N) of 167 kg/ha of ammonium nitrate every other week. Containerized seedlings were grown in a media of peat moss, vermiculite and perlite in polyethylene multi-pot trays with 40 cavities each (571 per m2) and a cell volume of 107 cm3. Seeds were spring-sown (March 1997). The needles were not clipped, and the seedlings were fertilized and irrigated as scheduled for longleaf production in this nursery. The fall fertilization treatments consisted of 0, 2, and 4 extra applications of ammonium nitrate solution (100 ppm N) every other week. 2.2. Morphological and nitrogen analysis of nursery seedlings Ten variables were measured (shoot diameter, height, root length and the biomass of shoot, root and total biomass) or calculated (shoot:root ratio, shoot relative dry weight and root relative dry weight). Foliar nitrogen (N) was determined by the micro Kjeldhal method (Bremner and Mulvaney, 1982). The containerized seedlings had 0.69, 0.88 and 0.87% foliar nitrogen for no N, low and high N, respectively. Bare-root seedlings had 0.81, 0.97 and 1.11% foliar nitrogen, respectively, for these treatments. 2.3. Plantation and experimental design Longleaf pine seedlings were planted in February 1998 at International Paper Company’s second-growth longleaf pine forest near Bainbridge, Georgia. Soils at this forest are the Norfolk and Orangeburg series (fine loamy, kaolinitic, thermic Typic Kandiuluts). The experimental design was a factorial split plot, with six sites (blocks) serving as plots and gaps acting as subplots. Every site contained three types of gaps: large 1.6 ha gap (144 m diameter), small 0.1 ha gap (36 m diameter), and intact canopy (no gap). The total number of gaps was 18 (6 sites 3 gaptypes). Thesub–sub-plots were the cardinal directions, the sub–sub–sub-plots were the seedling stock, and the sub–sub–sub–subplots corresponded to nursery N fertilization. Canopy gaps were created by clearcutting in 1996 in a secondgrowth longleaf pine forest that was 23 m in height. In the large gaps, 10 seedlings of each stock type, nursery N treatment, cardinal direction (NE, SE, SW,
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
NW) from the center, distance from the edge, and site were planted (2 m 2 m), for a subtotal of 4320 seedlings (6 sites 2 stock types 3 N levels 4 directions 3 distances 10 seedlings). In the small gaps, 10–13 seedlings per stock type, nursery N treatment, cardinal direction and site were planted, for a subtotal of 1704 seedlings. In the forest, 15 seedlings per stock type, nursery N treatment, cardinal direction and site were planted, for a sub-total of 2160 seedlings. In total, 8184 seedlings were planted in 120 sub–sub– sub–sub-plots. Containerized seedlings were planted with a planting dibble and the bare-root seedlings with a planting shovel. The roots of bare-root seedlings were pruned with a straight edge if they exceeded 30 cm in length at planting. The distance between trees within a plot was 1 m. No herbicide, irrigation or fertilization was applied following planting.
on a tripod 1 m above the floor and oriented north. The film negatives were converted to a digital format and imported from automated digitized imaging using Hemiview1 (Delta-T Devices Ltd., Burwell, Cambridge, UK). Hemiview1 was also used to calculate several indices of solar radiation based on canopy openness, using an index of the sun’s position and projected path of movement over a 6 month period. A gap light index (GLI) (sensu Canham, 1988) was calculated for each location and gap treatment (McGuire et al., 2001). 2.7. Statistical analysis The relative importance of the variables analyzed was evaluated using the linear statistical model yijklmn ¼ ai þ bj þ gk þ dl þ Zm þ ðabÞij þ ðagÞik þ ðadÞil þ ðaZÞim þ ðbgÞjk þ ðbdÞjl þ ðbZÞjm
2.4. Survival, morphology and growth Survival was recorded 1 year after planting. Also, two to three seedlings for each N treatment and seedling type per plot were harvested for morphological analysis after 1 year. Harvested seedlings were temporarily stored in a large cooler with ice and then at 2 8C in a refrigerator. Measurements (e.g. diameter, stem length, dry weight, etc.) were obtained from the seedlings. Then roots and shoots were separated, placed in an oven at 70 8C for 2 weeks, and weighed. 2.5. Characterization of interactions with understory Competing vegetation was characterized in every plot. Plant life forms (forbs, grasses, and shrubs), density and plant cover (%) by life form were observed ocularly from 120 one-square meter plots, in the referred sub–sub–sub–sub-plots. 2.6. Light Light data were obtained from a series of studies at the same sites in Georgia conducted by McGuire et al. (2001). Hemispherical photographs were taken, with the photographs recorded on 400-speed black and white film during uniformly cloudy days or mornings prior to sunrise according to Easter and Spies (1994). An 8 mm fisheye lens and camera assembly were leveled skyward
27
þ ðgdÞkl þ ðgZÞkm þ ðdZÞlm þ ðabgÞijk þ ðabdÞijl þ ðabZÞijm þ ðagdÞikm þ ðagZÞikm þ ðadZÞilm þ ðbgdÞjkl þ ðbgZÞjkm þ ðbdZÞjlm þ ðgdZÞklm þ ðabgdÞijkl þ ðabgZÞijkm þ ðabdZÞijlm þ ðagdZÞiklm þ ðbgdZÞjklm þ ðabgdZÞijklm þ Eijklmn
(1)
where a is the effect of the site (block) (random effect), b the effect of the gap (fixed effect), g the effect of the direction (fixed effect), d the effect of the seedling type (fixed effect), Z the effect of the N treatment (fixed effect), the combinations of factors represent the respective interactions, (ab) the whole plot error effect N(0, s2W ), and E the split-plot error effect (0, s2). Because the large gaps were the only planting sites with a variable for distance from the center of the gap, those also were analyzed with the same type of model, excluding the factor gap type and including the factor distance with three fixed effect levels. The hypotheses being tested were: Effects of individual factors were: Ho :
m1 ¼ m2 ¼ m3
all gaps main-effect means equal Ho :
m1 ¼ m2 ¼ m3 ¼ m4
all directions main-effect means equal
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
28
Ho :
m1 ¼ m2
all stock types main-effect means equal Ho :
m1 ¼ m2 ¼ m3
all nitrogen treatments main-effect means equal Interactions were: Ho : allðbgÞjk ¼ 0 no interaction among gap and direction And similarly for the other interactions. Seedling survival and all morphological variables were analyzed 1 year following planting. The data were processed using SAS (Statistical Analysis System1, version 6.12) analysis of variance (Proc Mixed), calculating the degrees of freedom with the Satterthwaite option ðddfm ¼ satterthÞ (Littell et al., 1999). Means and standard errors were obtained using the least squares means (lsmeans) statement. For the case of the pooled analysis for each variable, after running the program with Proc Glm, the three-way, four-way, and five-way interactions with a p-value larger than 0.25 were omitted to conduct an analysis with Proc Mixed using a reduced model to increase the power of the test. Growth was analyzed with Proc Mixed, adding the variables time 0 (planting moment), and time 1 (1 year after planting). Bud status (broken or non broken) was analyzed with the non-parametric chi-square (w2) test. To predict survival we used multiple regression models by first obtaining Pearson’s correlation coefficients among survival and all of the field variables (e.g. GLI, grass cover, total cover, etc.). Light data were obtained for three of the six blocks and our models and correlations with GLI included the data from these three blocks. Variables were selected by magnitude of their coefficient with survival and had to be independent of other variables, to prevent multicolinearity. The residuals of every selected independent variable were also checked to see if there was a trend. Survival was transformed using the arcsin function before analyses.
3. Results 3.1. Survival In both the pooled and reduced analyses of survival the interaction between seedling type and gap type
was significant ðP ¼ 0:0059Þ. Initial survival for containerized seedlings in the forest, small and large gaps was 35.1, 23.4, and 15.4%, respectively, while survival of bare-root seedlings was 12.3, 9.7, and 8.9%. In all of the cases containerized seedlings (24.6%) exhibited better survival than bare-root seedlings (10.3%) ðP ¼ 0:0002Þ. The N treatments did not have an effect on short-term survival ðP ¼ 0:5670Þ. Containerized seedlings exhibited negative correlations between initial survival (dependent variable) and GLI (independent variable) (r 2 ¼ 0:721, P ¼ 0:0001), with the following multiple regression model (P ¼ 0:0001, r 2 ¼ 0:583): S ¼ 80:983 0:862ðGLIÞþ 0:031ðWCÞ 0:221ðGCÞ þ 0:062ðFCÞ þ 0:005ðGLIGCÞ þ 0:012ðGLIFCÞ 0:033ðGCFCÞ
(2)
where S is the survival (%), GLI the gap light index, WC the shrub cover (%), GC the grass cover (%), FC the forb cover (%), GLIGC the interaction among GLI and GC, GLIFC the interaction among GLI and FC, and GCFC the interaction among GC and FC. Survival was negatively correlated with GLI for bare-root seedlings (r 2 ¼ 0:330, P ¼ 0:0100), but the linear regression model was not significant ðP ¼ 0:251Þ. 3.2. Morphological variables In the containerized seedlings, stem diameter after 1 year was smaller in the forest (10.8 mm) when compared with small gaps (11.3 mm) and large gaps (11.7 mm) ðP ¼ 0:0013Þ. The shoot:root ratio was influenced by the interaction of gap by N treatment ðP ¼ 0:0066Þ. In the forest low N was different from no N but not from high N. In the small gaps, shoot:root ratio in high N was higher than no N and low N. In the large gaps, no clear differences among N fertilizations were found. The relative shoot weight also exhibited a gap by N treatment interaction ðP ¼ 0:0069Þ. In the forest, higher levels of N corresponded to higher relative shoot weights, but no differences among N treatments were observed in the large gaps (Fig. 1). The relative root weight responded accordingly also ðP ¼ 0:0069Þ. In the case of bare-root seedlings, after 1 year in the field, shoot dry weight was larger at high N levels (6.3, 8.1, and 8.5 g, for no N, low N and high N).
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
29
Fig. 1. Mean (standard error) relative shoot weight for containerized seedlings in different gap types and N treatments (0, 2 and 4).
3.3. Bud break Non-parametric (w2) analysis in a pooled comparison showed that a larger proportion of bare-root seedlings (67.5%) were breaking buds in February, in comparison with 57.8% for containerized seedlings ðP ¼ 0:006Þ. Bud break timing of containerized seedlings was
influenced by gap type: 70.4% for forest, 67.5% for small gaps, and 43.5% for large gaps ðP ¼ 0:001Þ. 3.4. Biomass Biomass comparisons were made for dry weight at planting time compared with dry weight 1 year later
Fig. 2. Average survival of containerized seedlings in large 1.6 ha gaps (144 m diameter) 1 year after planting by cardinal direction and distance. The values with different letters exhibited statistically significant differences.
30
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
for the three nursery N treatments. In the containerized seedlings, root (2.5, 3.3 g, P ¼ 0:0023) and total dry weights (5.7, 6.8 g, P ¼ 0:0032) were different between the initial planting time and 1 year later. No variable responded to N treatment. For bare-root seedlings, root dry weight increased with time (8.0, 9.5 g, P ¼ 0:0010). All the other variables showed no change or exhibited a reduction in biomass related to mortality of tissue. For instance, the total seedling biomass exhibited no difference between planting time and 1 year later (18.6, 17.1 g, P ¼ 0:1252). Only shoot weight had differences due to N treatments (7.9, 9.3, 10.1 g, for no N, low and high N, P ¼ 0:0300). 3.5. Large gaps 3.5.1. Survival The containerized seedlings had higher survival than bare-root seedlings, and survival was influenced by the cardinal direction by distance interaction in large gaps, with highest survival at the southern edges of the gaps ðP ¼ 0:0020Þ. Seedlings planted at the edge of gaps had the highest survival, particularly in the SW cardinal direction (Fig. 2). In all directions (except in the NE), seedlings planted in the center had the lowest survival. The Pearson’s correlation coefficient among survival and distance was 0.72 ðP ¼ 0:0001Þ. The survival of bare-root seedlings was independent of distance from the center of the gap, with average survivals of only 6.0, 7.9, and 12.4% for center, middle, and edge areas of the gap, respectively, ðP ¼ 0:2388Þ. There was also no effect on survival for cardinal direction across the gap, N treatment, or any of the interactions. In the pooled analysis across seedling type, survival was affected only by the interaction between cardinal direction and distance ðP ¼ 0:0132Þ. The Pearson’s correlation coefficient for survival and distance was 0.53 ðP ¼ 0:0001Þ. 3.5.2. Morphological variables Root length was influenced by the distance by N treatment interaction ðP ¼ 0:0408Þ. In the center of the gap, no differences were found among N levels (22.0, 20.6 and 24.8 cm for levels 0, 2 and 4, respectively). In the middle part of the gap, high N was lower than the control (25.1, 21.7, 18.5 cm, for levels 0, 2
and 4). At the edge, high N was higher than the other two levels (22.6, 23.3, 27.8 cm, for levels 0, 2 and 4). 3.5.3. Bud break Bud break timing did not vary with distance, direction nor N treatment (w2-test). 3.5.4. Interactions with understory plants Total plant cover ðgrasses þ forbs þ shrubsÞ was also affected by the interaction of cardinal direction with distance ðP ¼ 0:0226Þ. In northern quadrants, the edges had the lowest total vegetation cover (NE: 65.8, 70.8, 60% for center, middle, and edge; NW: 71.7, 70.8, 61.7% for center, middle and edge). In all the directions, except for the SE, the central or the middle areas of the gaps had the highest total plant cover (SE: 69.2, 65.8, 74.2%; SW: 80.8, 63.3, 65.0%). In the nonparametric analysis, the only association found was forbs increasing with distance from the center of the gap (P ¼ 0:042, w2-test). For containerized seedlings there was a negative correlation between survival and grass cover r ¼ 0:34, P ¼ 0:0071). Bare-root seedling survival was negatively correlated with grass cover r ¼ 0:25, P ¼ 0:0497), and total plant cover r ¼ 0:28, P ¼ 0:0284).
4. Discussion 4.1. Survival of seedlings Survival was higher for containerized seedlings than for bare-root seedlings, as has been shown in other studies (e.g., Barber and Smith, 1996; Barnett et al., 1996). However, survival was higher under the forest canopy than in small or large gaps contradicting other studies. For instance, natural longleaf pines had higher survival in areas of low adult density (Wahlenberg, 1946; Platt et al., 1988; Boyer, 1993; Grace and Platt, 1995; Mosser, 1997) because of its shade-intolerance and competition intolerance. One possible reason for these contradictory results is the 1998 climate. The present research was conducted during a very dry and hot year. Precipitation in the spring for the period April–June was the lowest in 104 years in Florida (and southern Georgia). During this period, many localities in this region received less than half of normal rainfall (generally under 150 mm) (NOAA,
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
31
2000). Also 110 days of maximum daily temperatures averaging 36 8C were observed (International Paper Co., unpublished data), so the types of sites that in normal years adversely affect survival, exhibited better growing conditions during this hot dry year. An analysis of the main environmental factors may help explain this finding. 4.2. Light The GLI decreased from large gaps to the forest (84.2, 67.3, 48.2%), and from the center to the edge of gaps. Light availability in the southern portions of gaps was lower than in the northern parts (McGuire et al., 2001), and was inversely correlated to survival. Despite the reported intolerance of longleaf pine to shade, photosynthetically useful light also reaches shady areas. The forest floor at gap edges receives both diffuse and direct light, while the forest (nongaps) receives mainly diffuse light, and the gap center has mainly direct light. Diffuse radiation, relatively abundant at the edge of a gap, is relatively rich in blue wave lengths, the most effective in photosynthesis (Smith et al., 1997). So, even if shade does not represent the highest light condition for longleaf pine seedlings, growth may still occur there. 4.3. Temperature and humidity The greatest extremes in temperature in gaps occur when gap diameter is roughly equal to 1.5 times the height of surrounding trees, because of the combined effect of shade and ventilation (Geiger et al., 1995). In the central portion of large gaps, with diameters three times the height of surrounding trees, the environmental conditions are approximately the same conditions that would be present in much larger openings (Minckler and Woerheide, 1965). In our study, the large gap diameter was six times the height of the surrounding trees, so the conditions in the center were similar to much larger gaps. In these large gaps, survival increased from the center to edge, with the highest value (40%) in the SW-edge sector. In the northern hemisphere, in general, the highest surface temperatures and lowest fuel moisture are on SW slopes of mountains (Schroeder and Buck, 1970). Similarly, Randall and Johnson (1998) found for Pseudotsuga menziesii (Mirb.)
Fig. 3. Reshaping forest gaps to be oval-shaped patches oriented from NW to SE in their longitudinal axis may maximize the survival and growth of longleaf pine regeneration in dry years (matching oval-shaped 0.25 ha gaps with the SW edge of the 1.6 ha circular gaps, where survival was higher).
Franco, Abies procera Rehd and Pinus flexilis James, that S and SW aspects had reduced first and/or third year survival. Schneider et al. (1998) showed that the NE-facing slopes were generally more cool and moist and more favorable for seedlings of P. menziesii (Mirb.) Franco and Pinus ponderosa Dougl. Ex Laws. survival. A gap is the inverse of a small mountain. In a gap, the absence of the SW slope makes the radiation generally reside for more time in the year in the NE sector giving the SW sector more humidity (Fig. 3). Increases in temperature lead to lower relative humidity, and higher amounts of incident light are associated with higher temperatures. 4.4. Interactions with understory plants In this experiment, there was a dual effect of other plants and trees on longleaf pine seedling survival. Large trees could have a negative or a positive effect on the seedlings, depending on the conditions. The term ‘‘competition’’ in the present study is defined as the direct interaction of two plants attempting to exploit the same limiting resources (Tilman, 1987; Connell, 1990). In normal years, fine root competition from adult trees affects young seedling survival, but in droughty years shade may actually reduce evapotranspiration and thereby improve short-term survival. In a circular gap, competition factors controlled by adult trees at the edge, such as diffuse solar radiation, air movements and outgoing radiation, should operate in a nearly concentric pattern and factors related to the
32
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
slanting rays of direct solar radiation will follow a crescentic pattern (Smith et al., 1997). The competitive effects for water and nutrients created by the fine roots of adult trees may extend as far as 16 m into gaps (Boyer, 1963; Brockway and Outcalt, 1998). Over the long term, natural longleaf pine regeneration becomes clustered near the center of gaps, encircled by a wide zone from which seedlings are generally excluded due to the direct effects of fine root competition and indirect effects on needle accumulation as fine fuel (Brockway and Outcalt, 1998). In our study during droughty conditions, grasses had a negative effect (competition effect), and shrubs produced a positive effect (nurse effect) on longleaf pine seedlings. Shade protects longleaf pine seedlings from high soil temperatures and desiccating conditions (Myers, 1992). Grass cover alone is less effective than scrub oak cover in maintaining lower soil temperatures. When moderate, the shade from oaks favors seedling survival by retarding soil moisture loss and preventing soil overheating (Wahlenberg, 1946). In our multiple linear regression for estimating survival of containerized seedlings, GLI and grass cover negatively affected survival. In this study, it was hypothesized that GLI would be positively related to survival, but unusually dry conditions caused a negative correlation that was highly significant. Many grasses have morphological and physiological characteristics, such as a fibrous root system and C4 photosynthesis, which make them strong competitors. Some may overtop pine seedlings, reducing the availability of light (Minogue et al., 1991; Morris et al., 1993). Grasses can deplete moisture from the rooting zone of newly establishing conifers (Cleary et al., 1982). For instance, Pinus radiata D. Don doubled its uptake of N after the removal of pasture competition. Shrubs tend to deplete moisture also, but usually from deeper in the soil profile (Cleary et al., 1982). In our model, survival of containerized seedlings was positively related to shrub vegetation. There are several examples in the literature of nurse trees aiding initial survival. Wahlenberg (1946) and Cleary et al. (1982) observe that temperature extremes are moderated by brush, with air and soil temperatures under brush 11–17 8C lower than in adjacent openings. Hydraulic lift of deep soil water to the surface may also be active in this system, as has been seen for other
droughty soils. Artemisia tridentata roots cause shallow water depletion during the day, but by night the liquid absorbed from soil by deeper roots is transported to and lost from roots in upper and drier soil layers (Richards and Caldwell, 1987; Caldwell and Richards, 1989; Caldwell et al., 1998). If this type of water movement was caused by shrubs in our experiment, water taken up by shrubs and even adult trees may have become available to the longleaf pine seedlings. It is important to recognize that site differences and weather variation, will alter the competition at a site from year to year. For instance, seedlings of Pinus strobus L. had N as the most important resource limiting growth one year, but the next year light was the most important limiting factor (Elliott and Vose, 1995). Similarly, the survival patterns observed among gaps and within a large gap will vary because the limiting factors (e.g., temperature and water availability) will change from year to year. 4.5. Morphological variables and biomass In the present study, bud break in the spring was earlier at high N for bare-root seedlings as it was for P. menziesii (Mirb.) Franco (Margolis and Waring, 1986). Also, bud break in containerized seedlings occurred first in large gaps, then in small gaps, and then in the forest. Similarly, van Pelt and Franklin (1999) found (for P. menziesii (Mirb.) Franco and Tsuga heterophylla (Raf.) Sarg. seedlings) a concentric pattern of bud break occurring earliest towards gap centers and latest at the edges, with increasing direct light and length of the growing season at the center. In large gaps, root length of containerized seedlings tended to be shorter at higher N under high temperatures (middle area), but seedlings with the high N fertilization treatment had the longest root length at the edge under lower temperatures. This may be explained by the fact that under water stress, nutrient stress, or increased temperatures, cell expansion is more adversely affected than photosynthesis, thus producing excess carbohydrates that trees store (Landsberg and Gower, 1997). The proportion of carbohydrates allocated to root growth is higher when water and nutrients are limiting (Comeau and Kimmins, 1989; Landsberg and Gower, 1997).
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
In bare-root seedlings, N fertilization in the nursery continued to result in higher shoot biomass and lower relative tap root weight in the field. In containerized seedlings, the shoot:root ratio was lower at high N treatments in small gaps, consistent with previous studies (Waring and Running, 1998). Growth rates are correlated to N status under an appropriate nutritional ˚ gren and Ingestad, 1987; Waring, 1989). A balance (A decrease in N supply will result in a decrease in N in roots and shoots. Yet reduced N availability produces a relative increase in C allocation to roots (Landsberg and Gower, 1997). Lambers et al. (1998) mention cytokinins as the cause of this behavior. In high N environments there is a high rate of cytokinin synthesis in roots, which is exported to the leaves. As a result, photosynthetic capacity and leaf expansion rate increases, requiring a high rate of sugar consumption. The low rate of sugar export results in a slow root growth rate. Conversely, under low N conditions, the same relationship results in rapid root growth rates. 4.6. Growth Palik et al. (1997) note that growth of longleaf pine seedlings increased significantly in the center of gaps. Boyer (1963) mentions that growth of seedlings declines as overstory density increases. In our study during this droughty year, biomass loss of bare-root seedlings in the center and middle of large gaps and a related loss of vigor was observed. In the case of containerized seedlings, there was higher growth at the edge, most likely due to greater shading and protection from excessive temperatures and probably higher water availability. In containerized seedlings, buds broke first in the forest, then in small gaps, and lastly in large gaps, following a gradient of less harsh to more harsh conditions. Khan et al. (1996) found similar results in P. menziesii, where both very low and very high levels of humidity delayed bud development. 4.7. Management implications Palik et al. (1997) proposed the use of circular gaps larger than 0.14 ha (42 m diameter) to maximize longleaf pine seedling growth, because N and light availability are maximized. Similarly, Brockway and
33
Outcalt (1998) recommend creating gaps at least 40– 50 m in diameter (0.13–0.20 ha) to minimize intraspecific root competition between adult pines and seedlings. However, McGuire et al. (2001) observed that regeneration of longleaf pine may be maximized in gaps as small as 0.10 ha (36 m diameter). Building on those results and the results of our research, where the highest survival was at the SW edges of large circular gaps, we propose an approach using oval-shaped gaps to maximize both survival and growth. The longitudinal axis of the oval would be oriented from NW to SE, in such a way that includes most of the SW edge part of a large gap (Fig. 3). Increasing the width of the oval will decrease competition from adult pines, increasing seedling survival and growth during non-droughty years. The size of these oval gaps should range from 0.25 (the surface area of the SW sector) to 0.50 ha. Further research is needed for testing the efficacy of oval gaps. Some of the models for estimating future global warming for a significant part of the longleaf pine range, including the area where this study was conducted, indicate a 3–5 8C increase in mean annual surface air temperature, from 73 to þ73 mm change in mean annual precipitation, and less soil moisture (related to a larger evaporation and transpiration) (The Hadley Centre, 2002; United States Environmental Protection Agency, 2002). This means that one of the probable future scenarios for part of the longleaf pine ecosystem range is more drought. If that were the case, the results of work like this one may be more relevant. In such climatically altered environments, with less soil humidity, the temporary survival zone may gain more importance. The potentially higher available humidity in the SW edge may imply less root development of adult trees, implying less competition, and higher eventual survival to the seedlings. The exclusion zone exists. However, it is hardly identifiable when trees are older. There are no mature stands originated from gaps, surrounded by the 16 m wide band. Eventually, at least some trees survive in the exclusion zone, making that band not always evident across time. One of the mechanisms for that particular recruitment may be related to the results of the present work. If droughts become more frequent, it may be advisable to plant in the SW edges of gaps.
34
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
5. Conclusions and recommendations According to the literature, longleaf pine seedlings survive better in the central area of gaps, beyond a 16 m wide belt at the edge where fine root competition from adult trees, shade from the overstory and more intense surface fires exclude regeneration. However, in our research using artificial regeneration, we found that during an extremely dry year (1998), the shadier area near the gap edge in the SW sector was the best area for short-term seedling survival. During extremely dry conditions, the center of the gap is less favorable for seedling survival. In normal years, seedlings survive better in gaps than under canopies (Boyer, 1963; Brockway and Outcalt, 1998). However, in the especially dry year for this study the opposite occurred. Several shoot and root morphological traits were shown to be improved by fall N nursery fertilization, even 1 year after planting. Also, bud break was earliest at the sites with less harsh conditions (forest), followed by those with harsher conditions. Apparently, some of the understory plants (grasses) negatively affected the survival of seedlings because of direct competition, while shrubs ameliorated environmental temperature extremes, increasing short-term survival. The availability of the many environmental resources in gap regeneration operate not only in a spatial scale, but also in a time scale, and their effect may change radically when the most crucial environmental variables (e.g., water availability) change. When this happens, the micro-sites that are normally harsh may become the best for short-term seedling establishment during extreme drought events. During those extreme years in large gaps, the approximately 16 m wide exclusion zone could be at least temporarily re-named the ‘‘survival zone’’, particularly in the SW portion. Oval-shaped gaps with a surface area larger or equal to 0.25 ha and oriented from NW to SE, may maximize both survival and growth of longleaf pine seedlings. Because it improves performance-related traits at planting sites under difficult conditions (e.g., drought, competition), N nursery fertilization is recommended. Among those traits there may be faster bud break that may represent an advantage in a competitive environment. Also, a larger shoot dry weight may provide more resistance against physical agents, plus plant
biomass has been related to good field performance. Some of these improved traits may mean that Nfertilized longleaf will have a shorter grass stage. However, N did not increase stem diameter in our study, and longleaf pine starts to leave the grass stage after reaching 2.5 cm diameter (Boyer, 1990). We recommend further research with N fertilization. In addition, comparing or modeling gap sizes, shapes and orientations across forest sites in the south would be very beneficial.
Acknowledgements The authors thank two anonymous reviewers, and Henry Gholz, Doria Gordon and Robert J. Mitchell for their invaluable comments during the research. We also wish to thank Dave Noletti and Steve Gilly for logistic support, and the Divisio´ n de Ciencias Forestales, Universidad Auto´ noma Chapingo, CONACYT and The Florida-Mexico Institute for supporting the Ph.D. studies of the first author. This is Florida Agricultural Experiment Station Journal #R-09113 of the Institute of Food and Agricultural Sciences, University of Florida, Gainesville, FL. References ˚ gren, G.I., Ingestad, T., 1987. Root:shoot ratio as a balance A between nitrogen productivity and photosynthesis. Plant Cell Environ. 10, 579–586. Barber, B., Smith, P., 1996. Comparison of first year survival between container-grown and bare-root longleaf pine seedlings outplanted on a site in southeast Texas. In: Kush, J.S. (Comp.), Proceedings of the First Longleaf Alliance Conference on Longleaf Pine: A Regional Perspective of Challenges and Opportunities, September 17–19. The Longleaf Alliance, Mobile, AL, pp. 50–51. Barnett, J.P., Haywood, J.D., Sword, M.A., 1996. Artificial regeneration of longleaf pine; reevaluation technique. In: Kush, J.S. (Comp.), Proceedings of the First Longleaf Alliance Conference; Longleaf Pine: A Regional Perspective of Challenges and Opportunities, September 17–19. The Longleaf Alliance, Mobile, AL, pp. 53–55. Boyer, W.D., 1963. Development of longleaf pine seedlings under parent trees. Forestry Experimental Station Research Paper SO4. USDA Forest Service, South, New Orleans, LA, 5 pp. Boyer, W.D., 1990. Pinus palustris Mill. In: Burns, R.M., Honkala, B.H. (Eds.), Silvics of North America, vol. 1. Conifers. Forestry Service Agricultural Handbook 654. USDA, pp. 405– 412.
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36 Boyer, W.D., 1993. Long-term development of regeneration under longleaf pine seed tree and shelterwood stands. South. J. Appl. For. 17, 10–15. Branch, L., Villarreal, D., Hierro, J.L., Portier, K.M., 1996. Effects of local extinction of the plains vizcacha (Lagostomus maximus) on vegetation patterns in semi-arid scrub. Oecologia 106, 389–399. Bremner, J.M., Mulvaney, C.S., 1982. Nitrogen-total. In: Methods of Soil Analysis, Part 2. Chemical and Microbiological Properties. Agronomy Monograph No. 9. N.P., pp. 595–607. Brockway, D.G., Outcalt, K.W., 1998. Gap-phase regeneration in longleaf pine wire grass ecosystems. For. Ecol. Manage. 106, 125–139. Caldwell, M.M., Richards, J.H., 1989. Hydraulic lift: water efflux from upper roots improves effectiveness of water uptake by deep roots. Oecologia 79, 1–5. Caldwell, M.M., Dawson, T.E., Richards, J.H., 1998. Hydraulic lift: consequences of water efflux from the roots of plants. Oecologia 113, 151–161. Canham, C.D., 1988. Growth and canopy architecture of shadetolerant trees response to canopy gaps. Ecology 6, 786–795. Cleary, B.D., Greaves, R.D., Hermann, R.K., 1982. Regenerating Oregon’s forests. Oregon State University, USDA Cooperative Extension Service, Corvallis, OR, 282 pp. Comeau, P.G., Kimmins, J.P., 1989. Above and below-ground biomass and production of lodgepole pine on sites with differing soil moisture. Can. J. For. Res. 19, 447–454. Connell, J.H., 1990. Apparent versus ‘‘real’’ competition in plants. In: Grace, J.B., Tilman, D. (Eds.), Perspectives on Plant Competition. Academic Press, San Diego, pp. 9–26. Denslow, J.S., 1980. Gap partitioning among tropical rain forest trees. Biotropica 12, 47–55. Dirzo, R., Horvitz, C.C., Quevedo, H., Lo´ pez, M.A., 1992. The effects of gap size and age on the understory herb community of a tropical Mexican rain forest. J. Ecol. 80, 809–822. Easter, M.J., Spies, T.A., 1994. Using hemispherical photography for estimating photosynthetic photon flux density under canopies in gaps in Douglas-fir forests of the Pacific northwest. Can. J. For. Res. 24, 2050–2058. Elliott, K.J., Vose, J.M., 1995. Evaluation of the competitive environment for white pine (Pinus strobus L.) seedlings planted on prescribed burn sites in the southern Appalachians. For. Sci. 41, 513–530. Geiger, R.R., Aron, H., Todhunter, P., 1995. The Climate Near the Ground, 5th ed. Vieweq, Braunschweig, Germany. Gilliam, F.S., Platt, W.J., 1999. Effects of long-term fire exclusion on tree species composition and stand structure in an oldgrowth Pinus palustris (longleaf pine) forest. Plant Ecol. 140, 15–26. Gordon, D.R., Provencher, L., Hardesty, J.L., 1997. Measurement scales and ecosystem management. In: Pickett, S.T.A., Ostfeld, R.S., Shachak, M., Likens, G.E. (Eds.), The ecological basis of conservation: heterogeneity, ecosystems and biodiversity. Chapman & Hall, New York, pp. 262–273. Grace, S.L., Platt, J.W., 1995. Effects of adult tree density and fire on the demography of pregrass stage juvenile longleaf pine (Pinus palustris Mill.). J. Ecol. 83, 75–86.
35
Gray, A.N., Spies, T.A., 1996. Gap size, within-gap position and canopy structure effects on conifer seedling establishment. J. Ecol. 84, 635–645. Khan, S.R., Rose, R., Hasse, D.L., Sabin, T.E., 1996. Soil water stress: its effects on phenology, physiology, and morphology of containerized Douglas-fir seedlings. New Forests 12, 19–39. Lambers, H., Chapin III, F.S., Pons, T.L., 1998. Plant Physiological Ecology. Springer, New York, 540 pp. Landsberg, J.J., Gower, S.T., 1997. Applications of Physiological Ecology to Forest Management. Academic Press, San Diego, 354 pp. Littell, R.C., Milliken, G.A., Stroup, W.W., Wolfinger, R.D., 1999. SAS System for Mixed Models. SAS Institute, Cary, NC, 633 pp. Margolis, H.A., Waring, R.H., 1986. Carbon and nitrogen allocation patterns of Douglas-fir seedlings fertilized with nitrogen in autumn. II. Field performance. Can. J. For. Res. 16, 903–909. Martı´nez-Ramos, M., Sarukha´ n-Kermes, J., Pinero, D., 1988. The demography of tropical trees in the context of forest gap dynamics. Symp. Br. Ecol. Soc. 28, 293–313. McGuire, J.P., Mitchell, R.J., Moser, E.B., Pecot, S.D., GjestadDean, H., Hedman, C.W., 2001. Gaps in a gappy forest: plant resources, longleaf pine regeneration, and understory response to tree removal in longleaf pine savannas. Can. J. For. Res. 31, 765–778. Minckler, L.S., Woerheide, J.D., 1965. Reproduction of hardwoods 10 years after cutting as affected by site and opening size. J. For. 63, 103–107. Minogue, P.J., Cantrell, R.L., Griswold, H.C., 1991. Vegetation management after plantation establishment. In: Duryea, M.L., Dougherty, P.M. (Eds.), Forest Regeneration Manual. Kluwer Academic Publishers, The Netherlands, pp. 335–358. Mitchell, R.J., Palik, B.J., Kirkman, L.K., Hedman C.W., Pecot, S.D., McGuire, Gjerstad, D.H., 1996. Gaps in longleaf pine forests: a tool for ecosystem management. In: Kush, J.S. (Comp.), Proceedings of the First Longleaf Alliance Conference on Longleaf Pine: A Regional Perspective of Challenges and Opportunities, September 17–19. The Longleaf Alliance, Mobile, AL, 127 pp. Morris, L.A., Moss, S.A., Garbett, W.S., 1993. Competitive interference between selected herbaceous and woody plants and Pinus taeda L. during two growing seasons following planting. For. Sci. 30, 166–187. Mosser, W.K., 1997. Artificial establishment of longleaf pine under varying overstories. Annual Report. Tall Timbers Research Station, pp. 20–21. Myers, R.L., 1992. Scrub and high pine. In: Myers, R.L., Ewel, J.J. (Eds.), Ecosystems of Florida. University of Central Florida Press, Orlando, FL, pp. 150–193. NOAA (National Oceanic and Atmospheric Administration), 2000. Home page: http://www.ncdc.noaa.gov/ol/climate/research/1998/ jun/jun98.html#recent. Oliver, C.D., Larson, B.C., 1990. Forest Stand Dynamics. Wiley, New York, 467 pp. Palik, B.J., Mitchell, R.J., Houseal, G., Pederson, N., 1997. Effects of canopy structure on resource availability and seedling responses in longleaf pine ecosystem. Can. J. For. Res. 27, 1458–1464.
36
D.A. Rodrı´guez-Trejo et al. / Forest Ecology and Management 180 (2003) 25–36
Platt, W.J., 1998. Southeastern pine savannas. In: Anderson, R.C., Fralish, J.S., Baskin, J.M. (Eds.), Savannas, barrens, and rock outcrop plant communities of North America. Cambridge University Press, Cambridge, pp. 23–51. Platt, W.J., Evans, G.W., Rathbun, S.L., 1988. The population dynamics of a long-lived conifer (Pinus palustris). Am. Nat. 131, 491–525. Randall, W., Johnson, G.R., 1998. The impact of environment and nursery on survival and early growth of Douglas-fir, noble fir and white pine: a case study. West. J. Appl. For. 13, 137–143. Richards, J.H., Caldwell, M.M., 1987. Hydraulic lift: substantial nocturnal water transport between soil layers by Artemisia tridentata roots. Oecologia 73, 486–489. Schneider, W.G., Knowe, S.A., Harrington, T.B., 1998. Predicting survival of planted Douglas-fir and ponderosa pine seedlings on dry, low elevation sites in northwestern Oregon. New For. 15, 139–159. Schroeder, M.J., Buck, C., 1970. Fire weather. Forestry Service Agricultural Handbook 360. USDA, Washington, DC. Smith, D.M., Larson, B.C., Kelty, M.J., Ashton, P.M.S., 1997. The Practice of Silviculture: Applied Forest Ecology, 9th ed. Wiley, New York, 537 pp. The Hadley Centre, 2002. http://www.metoffice.com/research/ hadleycentre/index.html.
Tilman, D., 1987. The importance of mechanisms of interspecific competition. Am. Nat. 129, 769–774. United States Environmental Protection Agency, 2002. Global warming. State impacts. Georgia. http://www.epa.gov/globalwarming/impacts/stateimp/georgia/index.html. van Pelt, R., Franklin, V., 1999. Response of understory trees to experimental gaps in old-growth Douglas-fir forests. Ecol. Appl. 9, 504–512. Wahlenberg, W.G., 1946. Longleaf pine: its use, ecology, regeneration, protection, growth and management. CL. Pack Forestry Foundation and USDA Forestry Service, Washington, D.C., 426 pp. Waring, R.H., 1989. Ecosystems: fluxes of matter and energy. In: Cherrett, J.M., Bradshaw, A.D., Grubb, P.J., Goldsmith, F.B., Krebs, J.R. (Eds.), Ecological Concepts: The Contribution of Ecology to an Understanding of the Natural Word. Blackwell, Oxford, pp. 17–41. Waring, R.H., Running, S.W., 1998. Forest Ecosystems, 2nd ed. Academic Press, San Diego, 370 pp. Whitmore, T.C., 1978. Gaps in the forestry canopy. In: Tomlinson, P.B., Zimmerman, M.H. (Eds.), Tropical Trees as Living Systems. Cambridge University Press, New York, pp. 639–655. Whitmore, T.C., 1989. Canopy gaps and the two major groups of forest trees. Ecology 70, 536–538.