Soil Accumulation of Nitrogen and Phosphorus ...

Forest, Range & Wildland Soils

Soil Accumulation of Nitrogen and Phosphorus Following Annual Fertilization of Loblolly Pine and Sweetgum on Sandy Sites L. C. Kiser* Abraham Baldwin Agricultural College Tifton, GA 31793

T. R. Fox Virginia Polytechnic Institute and State University Blacksburg, VA 24061

A key question that arises following fertilization of forest plantations in the southeastern United States is whether fertilization has a long-term effect on nutrient availability and long-term site quality. Site quality could improve if mineral soil N and P pools are increased but is less likely if only forest floor pools are increased. Rapid forest floor decomposition and nutrient release following harvest likely lead to only rotation-length increases in nutrient availability. To examine the impact of repeated applications of N and P, forest floor and mineral soil N and P pools were measured at a 24-yr-old loblolly pine (Pinus taeda L.) plantation in North Carolina (Southeast Tree Research and Education Site [SETRES]) and 13-yr-old loblolly pine and sweetgum (Liquidambar styraciflua L.) plantations in Georgia (Mt. Pleasant) receiving annual incremental fertilizer applications totaling >1000 kg N ha−1 and >160 kg P ha−1. Mineral soil N pools varied only slightly. In contrast, fertilization increased forest floor N pools by 300 kg N ha−1 at SETRES. Long-term impacts on site N availability and site quality following these large N applications are unlikely. The forest floor P pool at SETRES was increased by 19.8 kg P ha−1. Fertilization also increased the mineral soil extractable P pool at SETRES to a depth of 1.5 m by 56 kg P ha−1 but not at Mt. Pleasant where P fertilization during the previous rotation may have already increased this pool. Observed increases in mineral soil P from fertilization indicated a potential for increased long-term site quality in terms of higher P availability for future rotations. Abbreviations: CEC, cation exchange capacity; Ct, control treatment; DAP, diammonium phosphate; F, fertilization treatment; FI, fertilization combined with irrigation; I, irrigation treatment; SETRES, Southeast Tree Research and Education Site.

F

ertilization is commonly used to increase growth in forest plantations in the southeastern United States because most soils in the region are deficient in both nitrogen (N) and phosphorus (P) (Allen, 1987; Albaugh et al., 2007; Fox et al., 2007). Within the southeastern United States, the Sandhills and Coastal Plain Regions comprise approximately 48 million hectares, and sandy-textured soils are common. Nitrogen and P deficiencies can be severe on sandy-textured soils indicated by the large growth response of loblolly pine (Pinus taeda L.) and slash pine (Pinus elliottii Engelm.) plantations to N and P fertilization (Albaugh et al., 2009). Fertilization increases foliar nutrient concentrations resulting in a temporary increase in photosynthetic capacity that increases leaf area (Gough et al., 2004). This increase in leaf area produces a growth response in fertilized trees (Albaugh et al., 1998). In southeastern U.S. plantation forestry, a single midrotation application of 225 kg N ha−1 and 28 kg P ha−1 is a common operational fertilization prescription (Fox et al., 2007). Elevated growth rates are typically sustained for 6 to 10 yr following a single application of 225 kg N ha−1 and Soil Sci. Soc. Am. J. doi:10.2136/sssaj2012.0118 Received 3 Apr. 2012. *Corresponding author ([email protected]). © Soil Science Society of America, 5585 Guilford Rd., Madison WI 53711 USA All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Soil Science Society of America Journal

28 kg P ha−1 (Fox et al., 2007; Albaugh et al., 2009). Sweetgum (Liquidambar styraciflua L.) is a plantation hardwood grown to a much more limited extent on these soils relative to loblolly pine and responds similarly to fertilization with N and P (Scott et al., 2004; Cobb et al., 2008). While fertilization with N and P increases growth of the current stand on nutrient deficient sandy-textured soils (Albaugh et al., 2008; Cobb et al., 2008; Albaugh et al., 2009), it remains uncertain whether the applied fertilizer will also improve long-term site quality. Long-term site quality could improve if fertilization increased mineral soil N and P pools and consequently improved long-term N and P availability. According to Miller (1981), fertilization will have limited impacts on long-term site quality in terms of N availability because N fertilization on sandy-textured soils only temporarily increases mineral soil N. This is in contrast to finer textured soils where fertilization has been found to result in sustained increases in mineral soil N (Binkley and Reid, 1985; Will et al., 2006). Miller (1981) found that while fertilization increased tree growth, the forest floor rather than the mineral soil accumulated N following fertilization of trees growing on sandy-textured soils. However, in the study by Miller (1981), N additions as high as 1500 kg ha−1 were added over a 3-yr period. One might anticipate that a more gradual annual application of N will produce different results. Following fertilization with N, mineral soil N availability is temporarily increased (Fig. 1). The duration of this increase has been reported to be approximately 5 mo (Mudano, 1986).

Fig. 1. Hypothesized forest floor and mineral soil N availability responses following fertilization of loblolly pine and sweetgum plantations growing on sandy-textured soils. Following fertilization (1), mineral soil N availability exhibits a short-term increase from the N fertilizer (2); N availability from forest floor decomposition exhibits a rotation-length short-term increase due to higher forest floor mass (3); after harvest (4), rapid forest floor decomposition results in a large spike in N availability known as the assart effect (5); N availability from forest floor decomposition returns to an equilibrium level (6), and mineral soil N availability exhibits another short-term increase derived from harvest-stimulated forest floor and mineral soil organic matter decomposition (7). Mineral soil N availability exhibits only short-term increases (2, 7) due to lack of NH4+ and NO3− retention in the sandy textured mineral soil. Forest floor and mineral soil N availability are not shown to scale. 6

Nitrogen from the fertilizer application in excess of tree uptake is not held in the mineral soil because of the low capacity of sandy-textured mineral soils to retain NH4+ and NO3−. Nitrogen accumulates in the forest floor of fertilized loblolly pine (Piatek and Allen, 2001; Sanchez, 2001) (Fig. 1). Accumulation of N in the forest floor would not likely increase long-term N availability because the forest floor rapidly releases accumulated N following harvest in the assart flush (Kimmins, 1997) (Fig. 1). Disturbance of the site from harvesting also increases N availability in the mineral soil (Matson and Vitousek, 1981; Vitousek and Matson, 1985). Fertilization increases the magnitude but not the duration of the assart flush (Fig. 1). Because of the rapid mineralization of N in the decomposing forest floor, N availability exceeds the N demand of newly planted seedlings in the next rotation. This lack of N captured by vegetation combined with the low capacity of sandy-textured soils to retain NH4+ and the rapid permeability of sandy-textured soils likely results in a considerable amount of NO3− leaching. Harvesting of a 22-yr-old loblolly pine plantation was reported by Vitousek and Matson (1985) to increase soil NO3− from 2 to 91 kg ha−1 in the first year and from 1 to 30 kg ha−1 in the second year. Four years after harvesting, soil NO3− levels were similar to an uncut stand (Vitousek et al., 1992). The instability of the forest floor N pool and the low capacity of sandy-textured mineral soils to retain NH4+ and NO3− suggest that accumulations of N in the forest floor from fertilization will result in only temporary increases in N availability rather than long-term effects on site quality that occur from increases in mineral soil N (Fig. 1). Following fertilization with P, mineral soil P availability is increased, and P in excess of tree uptake is held in the mineral soil because of the ability of acidic mineral soils to retain P as Al- and Fe-phosphates (Yuan et al., 1960) (Fig. 2). Accumulation of P also occurs in the forest floor (Harding and Jokela, 1994); however, this P is more readily released (Polglase et al., 1992) suggesting P availability from the forest floor is higher relative to N (Fig. 2). Polglase et al. (1992) found that as much as 15% of P in fresh loblolly pine litter was leached in 5 d as inorganic-P while no N was leached. Following harvest, the assart flush also results in the release of P that has accumulated in the forest floor over the rotation (Fig. 2). However, available P in excess of uptake by the newly planted seedlings in the next rotation is retained in the mineral soil. Unlike N, P fertilization is proposed to also benefit the site by increasing mineral soil P leading to a long-term increase in P availability (Pritchett and Comerford, 1982; Fox et al., 2011). A single application of P has been found to increase tree growth throughout the rotation and potentially in subsequent rotations across a range of forest soils and tree species (Ballard, 1978; Gentle et al., 1986; Comerford et al., 2002; Crous et al., 2007; Everett and Palm-Leis, 2009). Our objective was to measure accumulation of N and P in the forest floor and mineral soil following fertilization of loblolly pine and sweetgum plantations located on sandy-textured soils in the Coastal Plain. We evaluated two sites that received annual incremental fertilizer applications totaling over 1000 kg N ha−1 and 160 kg P ha−1 and exhibited growth increases of 56 to 100% Soil Science Society of America Journal

for loblolly pine and 400% for sweetgum (Albaugh et al., 1998, 2004, 2008; Cobb et al., 2008). It is our estimation that if the extreme cases of fertilization on these sites do not increase mineral soil N and P, then the much smaller operational rates will not. We hypothesized that these sites would respond to N fertilization with increases in forest floor N pools with little to no increase in mineral soil N pools. In contrast, P fertilization would increase both forest floor and mineral soil P pools. The implication of these trends in N and P pools is that fertilization will result in only short-term increases in N availability. In contrast, fertilization results in long-term increases in P by increasing the mineral soil pool suggesting an increase in site-quality in terms of P.

MATERIALS AND METHODS Site Descriptions The Southeast Tree Research and Education Site The Southeast Tree Research and Education Site (SETRES) was established in 1985 to investigate the effects of irrigation and fertilization on loblolly pine growing on an infertile, sandy soil. The experiment and tree growth response to treatments is described in detail by Albaugh et al. (1998, 2004, 2008). The site was located in the Sandhills Region in Scotland Co., NC, 17 km north of Laurinburg, NC (34°54′ N, 79°28′ W). Mean annual precipitation was 120 cm and the mean minimum and maximum temperatures were 10.4°C in January and 23.6°C in July, respectively. Before planting of loblolly pine in 1985, the site was

a native longleaf pine (Pinus palustris Mill.) forest. The stand was 24 yr old when sampling began for this study. The soil was mapped as the Wakulla series: a siliceous, thermic Psammentic Hapludult. While the soil was mapped as a Hapludult, the clay lenses giving this designation were weakly expressed. Soil structure was dominantly single grained. The A horizon extends from 10 to 15 cm in depth with a weak Bt horizon at ~60 cm, followed by multiple C horizons. Select mineral soil properties can be found in Table 1. The experimental design at SETRES was a 2 by 2 factorial randomized complete block with 4 blocks (n = 4, N = 16). In 1992, 50 by 50 m treatment plots containing interior 30 by 30 m measurement plots were established. Planting density was 1670 trees ha−1. Before initial treatment application, plots were thinned (1260 trees ha−1) and competing vegetation was controlled. Treatments included fertilization, no addition and optimal foliar nutrition based on target foliar nutrient concentrations (1.3% N, 0.10% P, 0.35% K, 0.12% Ca, 0.06% Mg, >12 ppm B [Albaugh et al., 2004]) and irrigation, and no addition and addition of 2.5 cm water wk−1 March to November. Treatment levels were control (Ct), irrigation (I), fertilization (F), and fertilization combined with irrigation (FI). Fertilization began in March 1992 and irrigation began in April 1993 and continues to the present. Fertilization was conducted each spring with a broadcast surface application of various fertilizer sources including granular urea, boron-coated urea, ammonium sulfate, diammonium phosphate (DAP), triple super phosphate, potassium chloride, dolomitic lime, Epsom salts, calcium sulfate, and borax (Albaugh et al., 2008). Irrigation water was applied to each treatment plot using multiple rotary sprinkler heads to assure uniform coverage. Total N and P applications from 1992 to 2008 were 1378 and 168 kg ha−1, respectively (Table 2).

Mt. Pleasant, Georgia A similar study designed to evaluate the effects of irrigation and fertilization on a number of pine and hardwood species was established in Mt. Pleasant, GA, on an infertile, sandy soil. Cobb et al. (2008) provide a detailed description of the experiment and tree growth response to treatments. The site was located in the southeastern Georgia Coastal Plain in Wayne Co. (31°23′ N,

Fig. 2. Hypothesized forest floor and mineral soil P availability responses following fertilization of loblolly pine and sweetgum plantations growing on sandy textured soils. Following fertilization (1), mineral soil P availability exhibits a long-term increase due to P fertilizer retention as Al- and Fe-phosphates (2); P availability from forest floor decomposition exhibits a rotation-length short-term increase due to higher forest floor mass (3); after harvest (4), rapid forest floor decomposition results in a large spike in P availability known as the assart effect (5); P availability from forest floor decomposition returns to an equilibrium level (6), and mineral soil P availability exhibits another increase derived from harvest-stimulated forest floor and mineral soil organic matter decomposition, and the long-term increase in mineral soil P availability continues (7). Mineral soil P availability exhibits a long-term increase (2, 7) due to retention of P as Al- and Fe-phosphates in the sandy-textured mineral soil. Forest floor and mineral soil P availability are not shown to scale.

www.soils.org/publications/sssaj

Table 1. Select properties of the 0 to 15 cm and 15 to 30 cm mineral soil depths at SETRES and Mt. Pleasant.† Property

SETRES

Mt. Pleasant 0–15 cm

pH (1:1, soil/water by vol) CEC, cmolc kg−1 Mehlich 3-Al, mg kg−1 Mehlich 3-Fe, mg kg−1

4.4 4 no data 135.3

5.3 3.9 383.2 93.4

15–30 cm 4.9 5.4 0.89 2.2 CEC, cmolc kg−1 no data 612.3 Mehlich 3-Al, mg kg−1 59.8 90.2 Mehlich 3-Fe, mg kg−1 †CEC, cation exchange capacity; SETRES, Southeast Tree Research and Education Site. pH (1:1, soil/water by vol)

6

Table 2. Nutrient additions at SETRES from 1992 to 2008.† Application Date

Year N

P

K

Ca

Mg

S

B

—————————— kg ha−1 —————————— Mar. 1992 224 56 112 0 0 0 0 Apr. 1992 0 0 0 134 56 0 0 June 1992 0 0 0 0 0 0 2 Mar. 1993 0 28 0 0 0 0 0 Apr. 1993 26 22 21 0 0 0 0 June 1993 0 0 92 0 56 120 0 Aug. 1993 56 0 0 0 0 0 0 Mar. 1994 112 0 0 0 0 0 0 Mar. 1995 56 28 56 24 34 74 0 May 1995 0 0 0 0 0 0 1 Mar. 1996 112 11 56 10 0 15 0 Apr. 1996 0 0 0 0 0 0 1 Apr. 1997 135 0 0 0 0 0 0 Mar. 1998 56 6 0 0 0 0 0 Mar. 1999 69 0 0 0 0 0 0 Mar. 2000 56 6 0 0 0 0 0 Apr. 2001 56 6 0 0 0 0 0 Apr. 2002 56 6 0 0 0 0 0 May 2002 0 0 56 0 0 0 0 Apr. 2003 56 0 0 0 0 0 0 Mar. 2004 56 0 0 0 0 64 0 Mar. 2005 84 0 0 0 0 0 1 Mar. 2006 56 0 0 0 0 64 0 Apr. 2007 56 0 0 0 0 0 0 Apr. 2008 56 0 0 0 0 0 0 Total 1378 168 393 168 146 337 6 †Granular fertilizer was applied to the surface on each application date. SETRES, Southeast Tree Research and Education Site.

81°43′ W). Mean annual precipitation was 127 cm and the mean minimum and maximum temperatures were 12.0°C in January and 26.0°C in July, respectively. Before establishment of the study in 1997, the site was a slash pine plantation that had been fertilized with P at some point in the previous rotations (Plum Creek Timber Co., personal communication, 2009). After harvesting of the previous pine stand, the site was cleared, disked, limed, and planted with separate plots containing loblolly pine, slash pine, sweetgum, and sycamore (Platanus occidentalis L.). Records of the quantity of lime applied were unavailable. This study includes data from only the loblolly pine and sweetgum plots. The stands were 13 yr old when sampling began for this study. The soil was mapped as the Klej series: a mesic, coated Aquic Quartzipsamment. During our sampling, redoximorphic features were found below 1 m in depth. Soil structure was dominantly single grained. Because of site preparation, the Ap horizon extended to approximately 30 cm in depth. Below the Ap horizon were a series of C horizons. Out of 90 soil pits dug for bulk density sampling, a poorly developed spodic horizon was found in two pits. Select mineral soil properties can be found in Table 1. The study was a randomized complete block design with 3 blocks (n = 3, N = 18). In 1997, 35 by 25.5 m treatment plots containing interior 22.2 by 13.5 m measurement plots were established. Planting density was 1790 trees ha−1. 6

Table 3. Nutrient additions at Mt. Pleasant from 1997 to 2006.† N

P

K

Cu

Mn

Mo

S

B

Fe

—————————— kg ha−1 —————————— 1997 113 17 62 0.0 0.0 0.0 0.0 0.0 0.0 1998 113 17 62 0.0 0.0 0.0 0.0 0.0 0.0 1999 113 17 62 0.0 0.0 0.0 0.0 0.0 0.0 2000 170 26 94 0.0 0.0 0.0 0.0 0.0 0.0 2001 51 8 28 0.0 0.0 0.0 0.0 0.0 0.0 2002 116 17 64 0.0 0.0 0.0 0.0 0.0 0.0 2003 116 17 64 0.0 0.0 0.0 0.0 0.0 0.0 2004 112 17 62 0.6 1.4 0.0 1.4 0.6 3.4 2005 112 17 62 0.7 1.6 0.5 0.0 0.7 3.8 2006 112 17 62 0.7 1.6 0.5 0.0 0.7 3.8 Total 1128 169 621 2.0 4.6 1.0 1.4 2.0 11.0 †From 1997 to 2003, soluble fertilizer was applied over a 28 wk period in the irrigation water. From 2004 to 2006, a single surface application of granular fertilizer was conducted in the spring.

Loblolly pine and sweetgum treatment levels included a control (Ct), irrigation (I), and 3 levels of fertilization combined with irrigation (FI). Only the high fertilization rate was evaluated in this study where annual nutrient additions averaged 113 N, 17 P, and 62 K kg ha−1 yr−1 (Table 2). Fertilization and irrigation treatments began in 1997. Irrigation was conducted by applying 3.05 cm water wk−1 with a drip system for 28 wk yr−1 (April to November) from 1997 to 2003. Fertilization from 1997 to 2003 was conducted by applying nutrients in solution in the irrigation water throughout the 28 wk of irrigation each year. Nitrogen was supplied as ammonium nitrate (NH4NO3). From 2004 to 2006, surface applications of N, P, and micronutrients were conducted in the spring. Nitrogen and P were applied as diammonium phosphate and urea. Fertilization was discontinued in late 2006, 3 yr before our sampling in 2009. Total N and P applications from 1997 to 2006 were 1128 and 169 kg ha−1 (Table 3), respectively, which were comparable to total fertilizer N and P applied at SETRES.

Sampling The forest floor and mineral soil were sampled at SETRES in April 2008, before the annual fertilization, and in March 2009 at Mt. Pleasant. At 18 (SETRES) and 15 (Mt. Pleasant) random locations within each plot, a 2-cm diameter push tube was used to sample the mineral soil to a depth of 60 cm. Each of the subsamples was divided into five depth classes (0–3.75, 3.75–7.5, 7.5–15, 15–30, and 30–60 cm). At five (SETRES) and three (Mt. Pleasant) random locations within each plot, a bucket auger was used to collect subsamples at 60 to 90, 90 to 120, and 120 to 150 cm. Subsamples were composited to produce one sample for each depth class per plot. Soils were air-dried at room temperature for 1 wk and sieved (2 mm). Bulk density determined at stand age 15 in a previous study at SETRES (Lee, 2002) was used to report nutrients in kg ha−1. In the study by Lee (2002), bulk density was determined in 0- to 7.5-, 7.5- to 15-, 15- to 30-, and 30- to 60-cm depth classes. Therefore, bulk density of the 0- to 7.5-cm depth class was used to approximate the 0- to 3.75-cm depth class and bulk density of the 30- to


Table 4. Mean mineral soil bulk density at SETRES and Mt. Pleasant.† Mt. Pleasant SETRES Depth (cm)

Ct

I

F

FI

Ct

0–3.75 1.25 1.00 1.15 1.15 0.94 3.75–7.5 1.25 1.00 1.15 1.15 1.01 7.5–15 1.36 1.16 1.28 1.28 1.29 15–30 1.38 1.21 1.32 1.32 1.49 30–60 1.45 1.37 1.42 1.42 1.49 60–90 1.45 1.37 1.42 1.42 1.50 90–120 1.45 1.37 1.42 1.42 1.48 120–150 1.45 1.37 1.42 1.42 1.48 † Bulk density (g cm−3); SETRES, Southeast Tree Research and Education Site.

60-cm depth class was used to approximate the 60- to 90-, 90- to 120-, and 120- to 150-cm depth classes (Table 4). Bulk density measurements were conducted at Mt. Pleasant in January 2010 (Table 4). Five subsample cores were collected in each plot for the depth classes described above to a depth of 60 cm (n = 15 per treatment–species combination). One core was collected in each plot for classes contained within the 60- to 120-cm depth (n = 3 per treatment–species combination). Bulk density of the 90- to 120-cm depth class was used to approximate the 120- to 150-cm depth class. At five (SETRES) and three (Mt. Pleasant) random locations within each plot, samples of the Oi, Oe, and Oa (when present) forest floor horizons were collected using a 0.07 m2 sampler (SETRES) and a 0.25-m2 sampler (Mt. Pleasant). Collection of forest floor material from a specific area allowed extrapolation of nutrient contents to kg ha−1. Subsamples were composited to produce one sample for each forest floor horizon per plot. Samples were oven-dried at 60°C for approximately 1 wk, weighed to determine oven-dry mass, and ground using a Thomas-Wiley Model 4 mill (Thomas Scientific, USA). Mass of Oa material was corrected with loss on ignition.

Loblolly Pine I

FI

Ct

Sweetgum I

FI

1.04 1.12 1.31 1.46 1.51 1.47 1.42 1.42

1.09 1.13 1.31 1.47 1.49 1.47 1.41 1.41

1.01 1.01 1.33 1.49 1.50 1.48 1.45 1.45

0.91 1.07 1.35 1.48 1.48 1.46 1.41 1.41

1.11 1.11 1.33 1.50 1.50 1.52 1.47 1.47

of 0.05 g kg−1 in the subsoil at both sites. Values below the detection limit were replaced with zero (Yess, 1993).

Statistical Analysis Forest floor and mineral soil variables were tested for treatment, species, sampling depth and horizon, and interaction effects with mixed model ANOVA (Littell et al., 1998) (PROC MIXED, SAS Institute Inc., Cary, NC, USA). Pairwise comparisons were conducted with the Bonferroni test (α = 0.10).

RESULTS SETRES Forest Floor Fertilization and irrigation did not affect forest floor C concentrations (Fig. 3). Fertilization increased N concentration in all forest floor horizons (Fig. 4). Phosphorus concentrations were also increased by fertilization (Fig. 5). Forest floor mass and C, N, and P pools were reported for the whole forest floor. Fertilization increased total forest floor mass by 16.7 Mg ha−1 as the mean difference of unfertilized and fertilized treatments (Table 5). Fertilization increased C, N, and P pools by 9.4 Mg ha−1, 300 kg ha−1, and 19.8 kg ha−1, respectively, as the mean difference of unfertilized and fertilized treatments (Table 5).

Chemical Analysis Mineral soil and forest floor samples were analyzed for total-C and total-N by dry combustion with a Vario MAX CNS analyzer (Elementar, Hanau, Germany). Forest floor P was determined by dry-ashing at 500°C followed by dissolution with 6 M HCl. Nutrient concentration of Oa material was corrected with loss on ignition. Mineral soil extractable P was determined with a Mehlich 3 extraction. Mineral soil total-P was determined on a subset of the surface mineral soil samples (0–15 cm) from both sites by digestion in sulfuric acid with a copper sulfate catalyst. Mineral soil extractable Al and Fe were determined by Mehlich 3 extraction. Phosphorus, Al, and Fe in solution was analyzed with inductively-coupled plasma atomic emission spectrophotometry on a Varian Vista MPX (Varian, Palo Alto, CA, USA). Mineral soil ph (1:1, soil/water by volume) and effective cation exchange capacity (CEC) were determined by Virginia’s Cooperative Extension Virginia Tech Soil Testing Laboratory. Mineral soil total-N concentrations fell below the method detection limit


SETRES Mineral Soil Fertilization, irrigation, and their interaction (F × I) affected mineral soil C concentration in the surface depths indicating C concentration was highest in the F treatment and lowest in the FI treatment (Fig. 6). Nitrogen concentration exhibited an FxI interaction effect indicating FI treatment N concentration was lower than in the F treatment in the mineral soil surface (Fig. 7). Fertilization increased P concentration in the surface depths (Fig. 8). Mineral soil C, N, and P pools were reported for the whole sampled profile from 0 to 150 cm. The C pool was highest in the F treatment (49.4 Mg ha−1) and lowest in the Ct treatment (38.3 Mg ha−1) (Table 5). The N pool in the Ct treatment (907 kg ha−1) was lower than in the other treatments (I 1107, F 1219, FI 1193 kg ha−1) (Table 5). Fertilization increased the P pool by 56.9 kg ha−1 as the mean difference of unfertilized and fertilized treatments (Table 5). Comparison of total-P and Mehlich 3-P concentrations in the 0- to 15-cm depth indicated that fertilization increased the Mehlich 3 extractable P fraction. For example,

6

Fig. 3. Mean forest floor total-C concentrations at Southeast Tree Research and Education Site (left panel) and Mt. Pleasant (right panel). Main and interaction effects were significant at α = 0.10 (*), 0.05 (**), and 0.01 (***). Effects: fertilization (F), irrigation (I), treatment (T), and species (S). Species: loblolly pine (L) and sweetgum (S).

Fig. 4. Mean forest floor total-N concentrations at Southeast Tree Research and Education Site (left panel) and Mt. Pleasant (right panel). Main and interaction effects were significant at α = 0.10 (*), 0.05 (**), and 0.01 (***). Effects: fertilization (F), irrigation (I), treatment (T), and species (S). Species: loblolly pine (L) and sweetgum (S).

Fig. 5. Mean forest floor total-P concentrations at Southeast Tree Research and Education Site (left panel) and Mt. Pleasant (right panel). Main and interaction effects were significant at α = 0.10 (*), 0.05 (**), and 0.01 (***). Effects: fertilization (F), irrigation (I), treatment (T), and species (S). Species: loblolly pine (L) and sweetgum (S).

6


Table 5. Mean forest floor mass and forest floor and mineral soil C, N, and P content at SETRES.† Location

Ct

I

F

FI

Forest Floor

Mass (Mg ha−1) 25.3b 25.3b

40.5a

43.6a

Forest Floor Mineral Soil

Total-C (Mg ha−1) 13.5b 13.8b 38.3c 39.0bc

22.2a 49.4a

23.8a 41.8b

Forest Floor Mineral Soil

Total-N (kg ha−1) 235b 249b 907b 1107a

519a 1219a

565a 1193a

Forest Floor

Total-P (kg ha−1) 14.0b 14.4b

35.6a

32.3a

Mehlich 3-P (kg ha−1) Mineral Soil 55.0c 58.2c 103.5b 123.5a †For the forest floor, values represent the sum of all horizons. For the mineral soil, values represent the sum of all depth classes from 0 to 150 cm. Means in row followed by different letters were significantly different (α = 0.10). Ct, control treatment; F, fertilization treatment; FI, fertilization combined with irrigation; I, irrigation treatment; SETRES, Southeast Tree Research and Education Site.

Mehlich 3-P concentration as a percent of total-P concentration averaged 4.6% for the unfertilized treatments and 26.3% for the fertilized treatments.

Mt. Pleasant Forest Floor Forest floor C concentration was generally higher in loblolly pine than in sweetgum (Fig. 3). In loblolly pine, C concentration in the Oi and Oe horizons was lowest in the Ct treatment. In sweetgum, C concentration of the FI treatment was highest in the Oi horizon and lowest in the Oe horizon. Forest floor N concentration tended to be higher in sweetgum than in loblolly pine and highest in FI treatments of both species in the Oi and Oe horizons (Fig. 4). The same pattern was generally found for P concentration except in the Oa horizon where P concentration was highest in the loblolly Ct treatment (Fig. 5). Forest floor mass and C, N, and P pools were reported for the whole forest floor. In loblolly pine, forest floor mass and C

and N pools did not differ among treatments (Table 6). In sweetgum, forest floor mass in the FI treatment was higher than the I treatment (Table 6). Forest floor mass decreased on the order of loblolly Ct > loblolly FI > sweetgum FI > loblolly I > sweetgum Ct > sweetgum I. Carbon, N, and P pools tended to be higher in loblolly pine than in sweetgum (Table 6). The N pool was highest in the FI treatments of both species (Table 6). However, in loblolly pine, the difference in N pools was not statistically significant (Table 6). In sweetgum, the N pool of the FI treatment was higher than the I treatment but not the Ct treatment (Table 6). In loblolly pine, the P pool of the FI treatment was higher than the I treatment but not the Ct treatment (Table 6). The same general pattern was found for sweetgum P pools (Table 6).

Mt. Pleasant Mineral Soil Treatment and species did not affect C, N, and P concentrations (Fig. 6, 7, and 8). Although not statistically significant, C, N, and P concentrations in the mineral soil surface were lowest in the loblolly I treatment. Mineral soil C, N, and P pools were reported for the whole sampled profile from 0 to 150 cm. Mineral soil C and N pools were not affected by treatment and species (Table 6). Across the site, C and N pools averaged 58.2 Mg ha−1 and 1658 kg ha−1, respectively (Table 6). Phosphorus pools differed by treatment and species, did not indicate an increase from the fertilization treatment, and were lowest in the loblolly I treatment (Table 6). Mehlich 3-P concentration as a percent of total-P concentration in the 0- to 15-cm depth across all treatment–species combinations was generally comparable to percentages observed in the SETRES fertilized treatments with the exception of the loblolly I treatment. In the loblolly I treatment, Mehlich 3-P concentration as a percentage of total-P concentration was consistently lower at 13.8% than the other treatments which averaged 27.4%, indicating irrigation increased mobility of this P pool in loblolly pine.

Fig. 6. Mean mineral soil total-C concentrations at Southeast Tree Research and Education Site (left panel) and Mt. Pleasant (right panel). Main and interaction effects were significant at α = 0.10 (*), 0.05 (**), and 0.01 (***). Effects: fertilization (F), irrigation (I), treatment (T), and species (S). Species: loblolly pine (L) and sweetgum (S).


6

Fig. 7. Mean mineral soil total-N concentrations at Southeast Tree Research and Education Site (left panel) and Mt. Pleasant (right panel). Main and interaction effects were significant at α = 0.10 (*), 0.05 (**), and 0.01 (***). Effects: fertilization (F), irrigation (I), treatment (T), and species (S). Species: loblolly pine (L) and sweetgum (S).

DISCUSSION Forest floor and mineral soil N pools were quantified at two sites to test the hypothesis that fertilization with N would increase forest floor N pools with little increase in mineral soil N pools. As anticipated, forest floor N was increased by fertilization at SETRES. Miller (1981) stated that N fertilization only benefits the trees and not the site because there is generally little impact of N fertilization on sandy-textured soils. While Miller (1981) did not detect an increase in mineral soil N, we did detect an increase in the mineral soil N pool at SETRES. However, the mineral soil N pool did not differ among the fertilized treatments (F and FI) and the I treatment (Table 5). Instead, the mineral soil N pool of the I, F, and FI treatments was higher than the control (Ct). We believe these results may indicate some potential for fertilization to increase mineral soil N. On the basis of these findings, our results were inconclusive in regards to Miller’s (1981) hypothesis. It is also important to note that pretreatment mineral soil N concentrations were not determined. Pretreatment conditions should be evaluated since potential increases

in mineral soil N from fertilization could be small and results confounded by site heterogeneity in mineral soil N. Miller (1981) also stated that increased mineral soil N is not likely unless the amount of nutrient applied is large in relation to the site capital. At SETRES, site N capital was approximately 1500 kg ha−1 in the unfertilized plots (taken from Albaugh et al. (2008) at age 21 and extrapolated to age 24). The 1378 kg ha−1 of N added was 92% of site N capital. In the study by Miller (1981), the highest amount of N fertilization was 140% of the site N capital in unfertilized stands. The amount of N added in relation to site N capital in the study by Miller (1981) was larger than at SETRES suggesting that the more gradual annual application of N rather than the amount added in relation to site capital may contribute more to increased mineral soil N. Increases in mineral soil N from fertilization have been reported at sites where the amount of N applied was much less than the site N capital relative to our study sites (13% Binkley and Reid, 1985; 40% Will et al., 2006). Mineral soil N accumulation at these sites was likely due to their finer textures resulting in retention of NH4+

Fig. 8. Mean mineral soil Mehlich 3-P concentrations at Southeast Tree Research and Education Site (left panel) and Mt. Pleasant (right panel). Main and interaction effects were significant at α = 0.10 (*), 0.05 (**), and 0.01 (***). Effects: fertilization (F), irrigation (I), treatment (T), and species (S). Species: loblolly pine (L) and sweetgum (S).

6


Table 6. Mean forest floor mass and forest floor and mineral soil C, N, and P content at Mt. Pleasant.†

to the mineral soil, the forest floor N pool is simply not stable enough to supply N over the long-term Loblolly Pine Sweetgum from one rotation to the next. Upon harvest, the Location Ct I FI Ct I FI forest floor rapidly decomposes and releases N that Mass (Mg ha−1) has accumulated during the life of the stand in the Forest Floor 16.6a 10.2abc 14.6ab 5.8bc 3.6c 10.8abc assart flush (Kimmins, 1997). At SETRES, which Total-C (Mg ha−1) was more representative of a rotation length than Forest Floor 8.7a 5.7ab 8.1a 2.7b 1.7b 4.9ab the 13-yr-old stands at Mt. Pleasant, the forest Mineral Soil 56.0 51.6 61.7 56.8 68.7 54.2 −1 floor had accumulated as much as 565 kg N ha− Total-N (kg ha ) 1. Since next rotation seedling nutrient demand Forest Floor 137ab 77abc 161a 54c 28d 128abc Mineral Soil 2065 1012 1833 1747 1883 1407 (Fox et al., 2007) and competing vegetation nutriTotal-P (kg ha−1) ent demand is not great enough to capture the N Forest Floor 11.7ab 7.3bcd 14.4a 4.3cd 2.3d 9.6abc released from the decomposing forest floor, much Mehlich 3-P (kg ha−1) of the N sequestered in the forest floor may be lost Mineral Soil 413.9a 125.8c 326.4a 173.3bc 304.0ab 294.0ab following harvest by nitrate leaching, particularly †For the forest floor, values represent the sum of all horizons. For the mineral soil, values on sandy-textured soils. The effects of N fertilizarepresent the sum of all depth classes from 0 to 150 cm. Means in row followed by tion on these sites imply that N fertilization will different letters were significantly different (α = 0.10). Ct, control treatment; FI, fertilization be necessary in each rotation to obtain higher procombined with irrigation; I, irrigation treatment. ductivity since incremental applications of over by clay (Mamo et al., 1993). The soil in a study by Binkley and Reid 1000 kg N ha−1 did not definitively increase mineral soil N. (1985) had andic properties and high organic matter content. The Fertilization increased both the forest floor and mineral soil soils in a study by Will et al. (2006), referred to in this discussion, P pools, supporting our hypothesis. The increase in the mineral were Ultisols with mineral soil C concentration similar to our sites. soil P pool suggests a potential for an increase in long-term site Nitrogen fertilization of Douglas-fir [Pseudotsuga menziesii (Mirb.) quality in terms of higher long-term P availability. This finding Franco] at Wind River was found to result in a long-term increase was consistent with the effects of P fertilization on long-term site in mineral soil N availability (Binkley and Reid, 1985). FertilizaP availability proposed by Pritchett and Comerford (1982) and tion with a broadcast application of 470 kg ha−1 of NH4NO3 Fox et al. (2011). The increase in mineral soil P from fertilizasustained a growth response at Wind River for 15 yr (Miller and tion at SETRES was also consistent with the findings of Harding Tarrant, 1983). The 470 kg ha−1 of N added was only 13% of the and Jokela (1994) in a study of P fertilization of slash pine growsite N capital. Mineral soil N availability in the 0- to 15-cm depth ing on a sandy soil. The implication of accumulations of P in the (56 mg kg−1) 15 yr after the NH4NO3 application was still double mineral soil is that fertilization with P may not be necessary at that of unfertilized plots (25 mg kg−1). The soil at Wind River had each rotation. A possible scenario arising from the application of higher organic matter than our sites (Tarrant and Miller, 1963) and 168 kg P ha−1 at SETRES doubling the mineral soil pool is that was a gravelly loam Andic Haplumbrept (Binkley and Reid, 1985) this pool may supply higher quantities of P to trees in the next indicating more active mineralogy. In a study by Will et al. (2006), rotation or rotations until it is gradually depleted over time and addition of 900 kg N ha−1 as DAP and NH4NO3 was 40% of the another P application is required to maintain higher productivsite N capital in unfertilized plots. Extractable NH4–N and NO3– ity. Everett and Palm-Leis (2009) conducted a study where they determined foliar P concentrations in second and third loblolly N in the 0- to 10-cm mineral soil depth was increased from 7.1 to pine rotations after P fertilization of the first rotation. They rec21.7 μg g−1 and total-N increased from 0.55 to 0.71 g kg−1. Carbon concentration was not increased by fertilization suggesting the ommended refertilizing stands fertilized with 45 kg P ha−1 in increase in mineral soil N was attributable to retention of mineral-N. the first rotation by age 3 in the second rotation to maintain opWe observed that mineral soil total N in the 0- to 3.75-cm depth at timal foliar P concentrations. The study by Everett and Palm-Leis SETRES was 0.50 g kg−1 in the Ct treatment and 0.67 g kg−1 in the differed from our study (169 kg ha−1) most notably by the lower F treatment, a change similar to that observed by Will et al. (2006). (45 kg ha−1) P application and also by soil texture (clayey versus Comparison of our results with those of Binkley and Reid (1985) sandy). Without conducting a study similar to the one by Everand Will et al. (2006) suggest a greater potential for fertilization to ett and Palm-Leis (2009), we cannot definitively say whether the increase mineral soil N at sites with more active mineralogy such as increase in mineral soil P we observed would supply quantities of clays and those with andic properties compared to the low-activity P to the next rotation sufficient to maintain higher productivity. mineralogy of soils dominated by quartz. However, a reasonable hypothesis would be that a threshold level When increases in N occur in the forest floor instead of the of P fertilization in the first rotation exists where the supply of mineral soil N pool, fertilization will not likely improve longP to the second rotation would be sufficient to maintain higher term site quality in terms of N. Increases in forest floor N pools productivity. This threshold level would follow a gradient of iniindicate the effects of N fertilization are likely only transient betial site P fertility and soil type. cause of the relative instability of the forest floor N pool. Relative


6

Mt. Pleasant mineral soil P was not increased by fertilizer P additions that began at the initiation of the study in 1997. However, Mehlich 3-P in the Mt. Pleasant control plots was higher than in the fertilized plots at SETRES, indicating the site had likely already undergone past P fertilizer applications. Before establishment of the study, the Mt. Pleasant site was a short-rotation pulpwood plantation that was probably fertilized in previous rotations (Plum Creek Timber Co., personal communication, 2009). In contrast, SETRES was a native longleaf pine forest that had not been fertilized. Past P applications at Mt. Pleasant were likely typical operational fertilization rates that range from 28 to 56 kg ha−1 (Albaugh et al., 2007). Our results indicated that the recent P fertilizer additions that have taken place since 1997 did not increase mineral soil P. However, high mineral soil P concentrations in the control plots relative to the SETRES control plots indicated the mineral soil has likely accumulated P from past applications, providing further support for our hypothesis that fertilization increases mineral soil P in sandy-textured soils. Our results indicated a trend for irrigation to decrease mineral soil P. Therefore, it is possible that further increases in mineral soil P may have been observed at Mt. Pleasant in a fertilization only treatment. The accumulation of P in the mineral soil was likely due to properties of the mineral soil promoting P accumulation. In a study by Yuan et al. (1960) on sandy-textured soils with similar pH and CEC of the soils in this study, over 80% of added P was retained by Al- and Fe-phosphates. Smith (1965) also suggested Al and Fe are largely responsible for absorbing P in acid forest soils. Ballard and Fiskell (1974) reported that ammonium oxalate extractable Al and Fe provided the best indices for accumulation of P in acid forest soils. In acidic soils, P saturation estimated from Mehlich 3-P was found to be highly correlated (r = 0.94) with P saturation estimated from ammonium oxalate extractable P (Kleinman and Sharpley, 2002). Results suggest that irrigation increased the mobility of mineral soil P at Mt. Pleasant. Within a species, mineral soil Mehlich 3-P concentration in the surface depths was lowest in the I treatments. Furthermore, in spite of the addition of 169 kg P ha−1, this P in the FI treatments was generally comparable to the Ct treatments. This effect was much greater in loblolly pine compared to sweetgum. In a study comparing forest floor leachate among forest floor types, eastern hemlock (Tsuga canadensis L. [Carr.]) leachate was more acidic at pH 3.9 than hardwood leachate at pH 4.1 to 4.9 because of higher organic acid concentrations (De Walle et al., 1985). Similar differences were suggested by the results to occur among loblolly pine and sweetgum. Al- and Fe-phosphates can be made available for plant uptake by dissolution and ligandexchange reactions with organic acids (Fox et al., 1990; Chen et al., 2008). We hypothesize that at Mt. Pleasant irrigation increased the quantity of forest floor leachate entering the mineral soil resulting in increased release of Al- and Fe-phosphate bound P through dissolution and ligand exchange. This effect was much more pronounced in loblolly pine than in sweetgum because of the higher acidity of loblolly pine forest floor leachate. This was supported by mineral soil P data indicating that in irrigated lob6

lolly pine, Mehlich 3-P as a percent of total-P was reduced from 25.1% to 13.8% compared to the control. Results suggest that mineral soil P accumulations may be reduced in areas with higher rainfall with this effect greater in pines relative to hardwoods. Another plausible explanation for the observed reduction of mineral soil P by irrigation is P leaching due to saturation of P sorption complexes. While sandy-textured mineral soils can retain P as Al- and Fe-phosphates, the capacity of these soils to retain P is limited relative to the P retention capacities of finer textured soils that contain greater quantities of Al and Fe (Ballard and Fiskell, 1974). At Mt. Pleasant, mineral soil P was much higher than at SETRES, and results indicated a trend for irrigation to reduce mineral soil P. Results suggest that large applications of P may saturate the relatively limited amount of P sorption complexes in sandy-textured mineral soils. Once P sorption complexes are saturated, P leaching can occur. This process was reported to occur in coarse textured soils because of high P applications from biosolids (Sukkariyah et al., 2007; Alleoni et al., 2008), dairy manure (Nair et al., 1998), and poultry manure (Sharpley et al., 2007). Results indicated that irrigation may have exerted a strong influence on forest floor decomposition at Mt. Pleasant. Unlike the response at SETRES, loblolly pine total forest floor mass was highest in the Ct treatment and not the FI treatment. This was unusual since foliage and litterfall production of the loblolly FI treatment was approximately 30% to 50% higher than the Ct and I treatments (Cobb et al., 2008). In addition, at Mt. Pleasant forest floor mass of the I treatment was lower than the Ct treatment for both loblolly pine and sweetgum. At SETRES, forest floor mass of the Ct and I treatments did not differ suggesting no decomposition response to irrigation. Results suggest forest floor decomposition of nonirrigated plots at Mt. Pleasant may have been limited by available moisture. This was also suggested by the loblolly pine Oa horizon proportions at Mt. Pleasant. While Oi and Oe horizon mass were highest in the Ct treatment, loblolly pine Oa horizon mass decreased on the order of FI (1.0 Mg ha−1) > I (0.7 Mg ha−1) > Ct (0.5 Mg ha−1) suggesting a more highly decomposed forest floor in the irrigated I and FI treatments.

CONCLUSIONS It remains uncertain whether N fertilization of loblolly pine and sweetgum growing on sandy-textured soils in the southeastern United States will result in an increase in long-term site quality in terms of higher mineral soil N. Forest floor N contents were, however, clearly increased by N fertilization suggesting a relatively temporary increase in site N. To adequately test the hypothesis that N fertilization will not increase mineral soil N in sandy-textured soils, pretreatment N concentrations must be determined. In contrast to the N response, and consistent with our hypotheses, P contents were increased by annual fertilization in the forest floor and mineral soil at both sites and for both loblolly pine and sweetgum. Unlike N, fertilization increased mineral soil P in a relatively stable form that can be made available for plant uptake over time suggesting the potential for higher long-term site productivity in terms of increased supply of P to trees in future rotations. Soil Science Society of America Journal

ACKNOWLEDGMENTS Funding for this research was provided by the National Science Foundation’s Center for Advanced Forestry Systems, the Forest Productivity Cooperative, and Virginia Polytechnic Institute & State University.

REFERENCES Albaugh, T.J., H.L. Allen, P.M. Dougherty, L.W. Kress, and J.S. King. 1998. Leafarea and above- and belowground growth responses of loblolly pine to nutrient and water additions. For. Sci. 44:317–328. Albaugh, T.J., H.L. Allen, P.M. Dougherty, and K.H. Johnsen. 2004. Long term growth responses of loblolly pine to optimal nutrient and water resource availability. For. Ecol. Manage. 192:3–19. doi:10.1016/j.foreco.2004.01.002 Albaugh, T.J., H.L. Allen, and T.R. Fox. 2007. Historical patterns of forest fertilization in the southeastern United States from 1969 to 2004. South. J. Appl. For. 31:129–137. Albaugh, T.J., H.L. Allen, and T.R. Fox. 2008. Nutrient use and uptake in Pinus taeda. Tree Physiol. 28:1083–1098. doi:10.1093/treephys/28.7.1083 Albaugh, T.J., H.L. Allen, T.R. Fox, C.A. Carlson, and R.A. Rubilar. 2009. Opportunities for fertilization of loblolly pine in the Sandhills of the southeastern United States. South. J. Appl. For. 33:129–136. Allen, H.L. 1987. Forest fertilizers: Nutrient amendment, stand productivity, and environmental impact. J. For. 85:37–46. Alleoni, L.R., S.R. Brinton, and G.A. O’Connor. 2008. Runoff and leachate losses of phosphorus in a sandy Spodosol amended with biosolids. J. Environ. Qual. 37:259–265. doi:10.2134/jeq2006.0302 Ballard, R., and J.G.A. Fiskell. 1974. Phosphorus retention in Coastal Plain forest soils: I. Relationship to soil properties. Soil Sci. Soc. Am. Proc. 38:250–255. doi:10.2136/sssaj1974.03615995003800020015x Ballard, R. 1978. Effect of first rotation phosphorus applications on fertilizer requirements of second rotation pine. N. Z. J. For. Sci. 8:135–145. Binkley, D., and P. Reid. 1985. Long-term increase of nitrogen availability from fertilization of Douglas-fir. Can. J. For. Res. 15:723–724. doi:10.1139/x85-117 Chen, C.R., L.M. Condron, and Z.H. Xu. 2008. Impacts of grassland afforestation with coniferous trees on soil phosphorus dynamics and associated microbial processes: A review. For. Ecol. Manage. 255:396–409. doi:10.1016/j. foreco.2007.10.040 Cobb, W.R., R.E. Will, R.F. Daniels, and M.A. Jacobson. 2008. Aboveground biomass and nitrogen in four short-rotation woody crop species growing with different water and nutrient availabilities. For. Ecol. Manage. 255:4032–4039. doi:10.1016/j.foreco.2008.03.045 Comerford, N.B., M. McLeod, and M. Skinner. 2002. Phosphorus form and bioavailability in the pine rotation following fertilization: P fertilization influences P form and potential bioavailability to pine in the subsequent rotation. For. Ecol. Manage. 169:203–211. doi:10.1016/S0378-1127(01)00680-6 Crous, J.W., A.R. Morris, and M.C. Schole. 2007. The significance of residual phosphorus and potassium fertilization in countering yield decline in a fourth rotation of Pinus patula in Swaziland. South. Hemisphere For. J. 69:1–8. doi:10.2989/SHFJ.2007.69.1.1.163 De Walle, D.R., G.C. Ribblett, J.D. Helvey, and J. Kochenderfer. 1985. Investigation of leachate chemistry from six Appalachian forest floor types subjected to simulated acid rain. J. Environ. Qual. 14:234–240. doi:10.2134/ jeq1985.00472425001400020016x Everett, C.J., and H. Palm-Leis. 2009. Availability of residual phosphorus fertilizer for loblolly pine. For. Ecol. Manage. 258:2207–2213. doi:10.1016/j. foreco.2008.11.029 Fox, T.R., N.B. Comerford, and W.W. McFee. 1990. Kinetics of phosphorus release from spodosols: Effects of oxalate and formate. Soil Sci. Soc. Am. J. 54:1441–1447. doi:10.2136/sssaj1990.03615995005400050038x Fox, T.R., E.J. Jokela, and H.L. Allen. 2007. The development of pine plantation silviculture in the southern United States. J. For. Oct./Nov.:337–347. Fox, T.R., B.W. Miller, R. Rubilar, J.L. Stape, and T.J. Albaugh. 2011. Phosphorus nutrition of forest plantations: The role of inorganic and organic phosphorus. In: E. Bünemann, A. Oberson, and E. Frossard, editors, Phosphorus in action: Biological Processes in Soil Phosphorus Cycling. Soil Biology Vol. 26. Springer, Berlin. p. 317–338. Gentle, S.W., F.R. Humphreys, and M.J. Lambert. 1986. Continuing response of Pinus radiata to phosphate fertilizers over two rotations. For. Sci. 32:822–829. Gough, C.M., J.R. Seiler, and C.A. Maier. 2004. Short-term effects of fertilization on loblolly pine (Pinus taeda L.) physiology. Plant Cell Environ. 27:876–886. doi:10.1111/j.1365-3040.2004.01193.x


Harding, R.B., and E.J. Jokela. 1994. Long-term effects of forest fertilization on site organic matter and nutrients. Soil Sci. Soc. Am. J. 58:216–221. doi:10.2136/ sssaj1994.03615995005800010032x Kimmins, J.P. 1997. Forest ecology: A foundation for sustainable management. 2nd ed. Prentice Hall, Upper Saddle River, NJ. Kleinman, P.J.A., and A.N. Sharpley. 2002. Estimating soil phosphorus sorption saturation from Mehlich-3 data. Commun. Soil Sci. Plant Anal. 33:1825– 1839. doi:10.1081/CSS-120004825 Lee, Y.S. 2002. Retention of fertilizer nitrogen by a pine forest ecosystem supported by a sandy, leachable soil. M.S. thesis, Duke Univ. Durham, NC. Littell, R.C., P.R. Henry, and C.B. Ammerman. 1998. Statistical analysis of repeated measures data using SAS Procedures. J. Anim. Sci. 76:1216–1231. Mamo, M., R.W. Taylor, and J.W. Shuford. 1993. Ammonium fixation by soil and pure clay minerals. Commun. Soil Sci. Plant Anal. 24:1115–1126. doi:10.1080/00103629309368864 Matson, P.A., and P.M. Vitousek. 1981. Nitrification potentials following clearcutting in the Hoosier National Forest, Indiana. For. Sci. 27:781–791. Miller, H.G. 1981. Forest fertilization: Some guiding concepts. Forestry 54:157– 167. doi:10.1093/forestry/54.2.157 Miller, R.E., and R.F. Tarrant. 1983. Long-term growth response of Douglas-fir to ammonium nitrate fertilizer. For. Sci. 29:127–137. Mudano, J.E. 1986. Assessment of soil nitrogen availability following nitrogen and phosphorus fertilization of a loblolly pine stand. M.S. thesis, North Carolina State Univ., Raleigh. Nair, V.D., D.A. Graetz, and K.R. Reddy. 1998. Dairy manure influences on phosphorus retention capacity of Spodosols. J. Environ. Qual. 27:522–527. doi:10.2134/jeq1998.00472425002700030007x Piatek, K.B., and H.L. Allen. 2001. Are forest floors in mid-rotation loblolly pine stands a sink for nitrogen and phosphorus? Can. J. For. Res. 31:1164–1174. doi:10.1139/x01-049 Polglase, P.J., N.B. Comerford, and E.J. Jokela. 1992. Leaching of inorganic phosphorus from litter of southern pine plantations. Soil Sci. Soc. Am. J. 56:573– 577. doi:10.2136/sssaj1992.03615995005600020037x Pritchett, W.L., and N.B. Comerford. 1982. Long-term response to phosphorus fertilization on selected southeastern Coastal Plain soils. Soil Sci. Soc. Am. J. 46:640–644. doi:10.2136/sssaj1982.03615995004600030038x Sanchez, F. 2001. Loblolly pine needle decomposition and nutrient dynamics as affected by irrigation, fertilization, and substrate quality. For. Ecol. Manage. 152:85–96. doi:10.1016/S0378-1127(00)00592-2 Scott, D.A., J.A. Burger, D.J. Kaczmarek, and M.B. Kane. 2004. Growth and nutrition response of young sweetgum plantations to repeated nitrogen fertilization on two site types. Biomass Bioenergy 27:313–325. doi:10.1016/j.biombioe.2004.02.003 Sharpley, A.N., S. Herron, and T. Daniel. 2007. Overcoming the challenges of phosphorus-based management in poultry farming. J. Soil Water Conserv. 62:375–389. Smith, A.N. 1965. Aluminum and iron phosphates in soils. J. Aust. Inst. Agric. Sci. 31:110–126. Sukkariyah, B., G. Evanylo, and L. Zelazny. 2007. Distribution of copper, zinc, and phosphorus in Coastal Plain soils receiving repeated liquid biosolids applications. J. Environ. Qual. 36:1618–1626. doi:10.2134/jeq2006.0558 Tarrant, R.F., and R.E. Miller. 1963. Accumulation of organic matter and soil nitrogen beneath a plantation of red alder and Douglas-fir. Soil Sci. Soc. Am. Proc. 27:231–234. doi:10.2136/sssaj1963.03615995002700020041x Vitousek, P.M., and P.A. Matson. 1985. Disturbance, nitrogen availability, and nitrogen losses in an intensively managed loblolly pine plantation. Ecology 66:1360–1376. doi:10.2307/1939189 Vitousek, P.M., S.W. Andariese, P.A. Matson, L. Morris, and R.L. Sanford. 1992. Effects of harvest intensity, site preparation, and herbicide use on soil nitrogen transformations in a young loblolly pine plantation. For. Ecol. Manage. 49:277–292. doi:10.1016/0378-1127(92)90141-U Will, R.E., D. Markewitz, R.L. Hendrick, D.F. Meason, T.R. Crocker, and B.E. Borders. 2006. Nitrogen and phosphorus dynamics for 13-year-old loblolly pine stands receiving complete competition control and annual N fertilizer. For. Ecol. Manage. 227:155–168. doi:10.1016/j.foreco.2006.02.027 Yess, M. 1993. U.S. Food and Drug Administration survey of methyl mercury in canned tuna. J. AOAC Int. 76:36–38. Yuan, T.L., W.K. Robertson, and J.R. Neller. 1960. Forms of newly fixed phosphorus in three acid sandy soils. Soil Sci. Soc. Am. J. 24:447–450. doi:10.2136/ sssaj1960.03615995002400060010x

6