Influence of variable organic matter retention on nutrient availability in a 10-year-old loblolly pine plantation

Forest Ecology and Management 259 (2010) 1480–1489

Contents lists available at ScienceDirect

Forest Ecology and Management journal homepage: www.elsevier.com/locate/foreco

Influence of variable organic matter retention on nutrient availability in a 10-year-old loblolly pine plantation J.L. Zerpa a,*, H.L. Allen a, R.G. Campbell b, J. Phelan a, H. Duzan b a b

North Carolina State University, 3108 Jordan Hall, Raleigh, NC 27695-8002, USA Weyerhaeuser, Co., New Bern, NC 28560, USA

A R T I C L E I N F O

A B S T R A C T

Article history: Received 8 August 2009 Received in revised form 4 January 2010 Accepted 10 January 2010

The effects of varying forest floor and slash retention at time of regeneration were evaluated 10 years after the establishment of a loblolly pine plantation near Millport, Alabama. Treatments included removing, leaving unaltered, or doubling the forest floor and slash material. Forest floor and litterfall mass and nutrient concentrations, available soil N, foliar nutrient concentrations and stand yield were all impacted by the treatments. Forest floor mass and nutrient contents in the doubled treatment were significantly greater than the other two treatments. The doubled treatment accumulated 25, 45 and 350% more forest floor mass and 56, 56, and 310% more N than the control treatment in the Oi, Oe, and Oa layers, respectively. The other nutrients followed similar patterns. Potentially mineralized NO3-N in the mineral soil was also significantly higher in the doubled treatment. The positive effect of doubling the forest floor on soil N availability was reflected in greater foliage production, 30% more litterfall and 25% more stand yield for this treatment. This study shows that increasing the forest floor retention has resulted in increased nutrient availability and improved tree growth. ß 2010 Elsevier B.V. All rights reserved.

Keywords: Available nitrogen Forest floor Pinus taeda

1. Introduction Intensively managed pine plantations provide an efficient way to produce the wood and fiber required to satisfy the demands of our growing society (Sedjo, 2001). The productivity of these plantations strongly depends on the ability of the soils to provide essential nutrients. However, productivity of most loblolly pine plantations in the southeast USA is limited by low soil nutrient availability (Fox et al., 2007). To overcome these limitations, fertilizer application has become common practice to increase leaf area and stemwood production (Albaugh et al., 2007). With the resulting greater leaf area, litterfall is increased, and a larger forest floor typically accumulates since increases in litterfall are not matched by similar increases in forest floor decomposition and nutrient release (Gurlevik et al., 2003). Nitrogen (N) accumulations contained in the forest floor of 100, 300, and up to 700 kg-N ha1 have been reported for loblolly pine plantations in the southeast US at ages 15 years (Switzer and Nelson, 1972), 22 years (Tew et al., 1986), and 34 years (Markewitz et al., 1998), respectively. Similarly, forest floor N and phosphorus (P) accumulations were reported to be 1.7–3.2 and 1–1.7 times greater than the aboveground biomass N and P contents, respectively (Tew et al., 1986;

* Corresponding author. Tel.: +1 540 231 7250/919 889 2639; fax: +1 919 515 6193. E-mail addresses: [email protected], [email protected] (J.L. Zerpa). 0378-1127/$ – see front matter ß 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.foreco.2010.01.024

Markewitz et al., 1998). Increases of forest floor mass and N content due to fertilization in the range of 200% and 400%, respectively are not uncommon in very responsive sites of the Southeast US (Rojas, 2005). Nutrient accumulation in the forest floor, at levels comparable or greater than those occurring in the above-ground biomass highlight the importance of the forest floor as a source or sink of nutrients (Piatek and Allen, 2001) for current and subsequent rotations. Nutrient cycling studies have shown that the forest floor mineralizes (Switzer and Nelson, 1972; Jorgensen et al., 1980; Covington, 1981), as well as retains nutrients through immobilization (Vitousek and Matson, 1985; Piatek and Allen, 2001), making the forest floor both a sink and a source of nutrients depending on the nutrient, tissue type (e.g. branches, foliage), and time since deposition. Climatic factors such as temperature and moisture have explained most of the differences in decomposition (Carey et al., 1982; Cortina and Vallejo, 1994) especially in recently deposited material (McHale et al., 1998; Rustad and Fernandez, 1998) at the regional level. However, at the local level, litter quality (Berg et al., 1993; De Santo et al., 1993; Piatek and Allen, 2001), lignin content (Berg, 1986; Scott and Binkley, 1997; Sariyildiz and Anderson, 2003), and the type of colonizing fungal species during microbial succession (Cox et al., 2001) have been associated with differences in decomposition rates. Based on these factors, the amount and quality of forest floor in a stand are expected to influence its decomposition and nutrient dynamics, thus the nutritional status of the stand as a whole.

J.L. Zerpa et al. / Forest Ecology and Management 259 (2010) 1480–1489

Forest floor removal was common with past site preparation practices in loblolly pine plantations such as shearing, piling, and burning, and was found to have either no impact (Vitousek and Matson, 1985; Fox et al., 1986; Li et al., 2003) or negatively influence (Burger and Pritchett, 1984) nutrient availability. Currently, more widely used practices include strip shearing, bedding, and hardwood control with herbicides, all of which retain the forest floor on site. However, little information is available concerning the effects of forest floor retention on nutrient availability and productivity in loblolly pine plantations. Unfortunately in many residue management studies, the inclusion of tillage (bedding or disking) in several but not all treatment combinations confounds the interpretation of organic matter removal and/or retention. Furthermore, few reported studies (Smith et al., 2000; Mendham et al., 2003; Tutua et al., 2008 for plantations of Pinus radiata, Eucalyptus globulus, and Pinus elliotti  Pinus caribaea hybrid, respectively) and none in loblolly pine have included treatments with organic matter additions above levels originally on the site, a condition which would more closely mimic forest floor accumulations obtained with current fertilization practices. We hypothesized that on nutrient-poor sites supporting loblolly pine, the retention of organic matter could have positive effects on growth and productivity if nutrients retained on site contribute to nutrient levels in the soil. Specific objectives of our study were to assess the effects of varying organic matter retention treatments imposed at time of regeneration on: forest floor and litterfall mass, nutrient concentrations and contents, available N in the mineral soil, foliar nutrition and stand yield 10 years after treatment imposition.

1481

2.2. Forest floor sampling and analysis Ten years after treatment imposition, the forest floor was collected from five randomly located points per plot using a 30.5 cm diameter round sampler. The forest floor layer was cut until the mineral soil was reached. For each sampling location, the forest floor was separated in the field into three layers, Oi, Oe, and Oa (Guthrie and Witty, 1982). These correspond to the litter (L), fermentation (F), and humus (H) classification of forest floor layers, respectively (Kendrick, 1959). Samples were bulked by layer, providing three forest floor samples per plot. Forest floor samples were oven dried at 70 8C to a constant mass. The lost-on-ignition method (Nelson and Sommers, 1996) was used to determine the ash-free weight of the Oi, Oe, and Oa layers and to correct for any mineral soil fraction that might have been collected along with the forest floor. Based on the area of the forest floor sampler, these mass estimates were scaled to a per hectare basis. Oven dry samples of the Oi, Oe, and Oa layers were ground to pass through a 1 mm mesh sieve and analyzed for N and carbon (C) concentration using the CHN elemental analyzer (CE InstrumentsNC 2100, CE Elatech Inc., Lakewood, NJ). Phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), manganese (Mn), sulfur (S), boron (B), copper (Cu), and zinc (Zn) concentrations were determined by digesting 0.8 g of ground, oven-dry material with nitric acid (Zarcinas et al., 1987) followed by analysis using an inductively coupled plasma atomic emission spectrometer (IPSAES, Varian ICP, Liberty Series 2, Varian analytical instruments, Walnut Creek, CA). 2.3. Mineral soil sampling and analysis

2. Materials and methods 2.1. Site and study description The study site was located in the Upper Coastal Plain physiographic province in Lamar County near the town of Millport, Alabama (338320 22.8700 N, 88870 7.5300 W). Mean annual temperature (1971–2000) is 15.9 8C with mean monthly temperatures ranging from 4.6 8C in January to 26.3 8C in July. Mean annual precipitation is 1,398 mm with a fairly uniform distribution throughout the year. September is the driest month with 85 mm, and January is the wettest month with 157 mm (NOAA, 2003). The soils are deep, well-drained Ruston soil series (fine-loamy, siliceous, semiactive, thermic Typic Paleudults). The A horizon is sandy loam with an average depth of 23 cm over a clay loam Bt horizon. The study was established in 1994 by Weyerhaeuser Co. after harvesting a 34-year-old loblolly pine plantation (site index of 17 m at age 25). Twelve 0.16 ha plots were established in a randomized complete block design with 3 treatments and 4 blocks. Blocking was done against the slope to account for possible site differences. The three treatments imposed after harvest and immediately before planting the current rotation included removed treatment, doubled treatment and control. In the removed treatment, all forest floor, understory, and slash material, comprised mostly of small branches, were removed using rakes and tarps. In the doubled treatment, all the material coming from the removed treatment plots was uniformly applied to the plots. On average, 118.2 Mg ha1 of dry matter where displaced from the removed treatments to the doubled treatments of which, 72%, 9%, and 19% were forest floor, understory, and pine slash respectively. In the control treatment, the understory was cut and left on site along with the forest floor and the slash material. Loblolly pine seedlings were planted at 4.3 m  3 m spacing in each 121-tree plot (11  11 trees) and only the inner 49 trees (7  7 trees) were considered for measurement purposes, leaving the trees in the treated perimeter as buffers.

A-horizon samples were collected immediately after the forest floor collections using the same 5 randomly located points per plot. The thickness of the A-horizon was measured in each sampling point to obtain a plot average. The samples were composited by plot in the field, put in plastic bags and transported in coolers with ice to the laboratory where they were immediately sieved through 2 mm mesh to remove the coarser fraction. No coarse fraction was detected in this fineloamy textured soil. These samples were then stored in the fridge at 4 8C until further analysis. Three bulk density samples were also collected from the A-horizon in each plot using the core method (Grossman and Reinsch, 2002). Soil variables were scaled to a per hectare basis using the depth of the A-horizon and the bulk density of the soil. A 28-day aerobic incubation was used as an index of potential net N mineralization in the mineral soil (Hart et al., 1994b). Five 10-g sub samples of each mineral soil sample were weighed and prepared for this incubation; one was used for moisture content determinations, two were used for the N extraction values at time zero, and the two remaining samples were left to incubate at field moisture content and 25 8C for 28 days. Changes in the moisture content of the incubated samples were monitored every other day and deionized water was added when the moisture contents in the samples dropped by more than 5% below their initial levels. Soil samples, at time zero, were extracted in 35 ml of 2 M KCl by shaking at high speed for one hour and centrifuging for 15 min at 4000 rpm. The centrifuged solution was filtered using Fisherbrand G8 glass fiber filters and the filtered solutions were analyzed for inorganic N with a Lachat Autoanalyzer (Quick-Chem 8000, Zellweger Analytics, Inc., Milwaukee, WI). After 28 days, the same extraction procedure was used for the incubated samples. Potential N net mineralization was calculated by subtracting the time zero averaged values of NO3-N and NH4+-N from the incubated average values.

J.L. Zerpa et al. / Forest Ecology and Management 259 (2010) 1480–1489

1482

2.4. Foliar and litterfall sampling and analyses Foliar samples were collected in January 2005 from the upper third of the live crown of 5 dominant or codominant trees in each plot. A total of 100 complete and healthy fascicles (20 fascicles from each selected tree) were collected from the first flush produced during the previous (10th) growing season. The samples were prepared and analyzed with the same methodology used for the forest floor samples. Litterfall was collected using five—1 m2 littertraps randomly located in each plot. The litterfall was collected bimonthly from these traps from April 2004 To March 2005, during the tenth year of growth, and was summed to produce estimates of annual litterfall. The collected litter was oven dried at 70 8C to a constant mass, and nutrient concentrations were determined using the same preparation and analysis methodology, as described for the forest floor material. The ash content of these samples was determined by the loss-on-ignition technique and used to adjust the weights and nutrient concentrations to an ashfree basis. Retranslocation for each nutrient was estimated using the following equation: R ð%Þ ¼

NutðfoliarÞ  NutðlitterÞ NutðfoliarÞ  100

(1)

where R (%) is the estimated percentage of retranslocation, Nut(foliar) is the foliar nutrient concentration, and Nut(litter) is the litterfall nutrient concentration. These retranslocation estimates were approximations because the nutrient data were from different foliage cohorts, age 9 for litterfall and age 10 for foliage and the dry mass losses that could occur between the stages of green needles and litterfall were not taken into account. Two decomposition indexes were also calculated. One index described litter decomposition in the Oi layer and was calculated as litterfall mass/Oi layer mass and the other estimated decomposition in the Oe layer and was defined as the ratio of Oi layer mass to Oe layer mass. 2.5. Stand yield Diameter at breast height and total tree height were measured in February 2005 at age 10. Using these data, individual tree volume was calculated with the volume equation of Burkhart (1977). 3. Data analysis Analyses of variance (ANOVA) using the general linear model (SAS, 2005) were performed to test for treatment effects on mass, nutrient concentrations, and nutrient contents of the forest floor and litterfall, nutrient concentrations of foliage, and potential N mineralization of the mineral soil, nutrient retranslocation, decomposition indexes, and stand yield. Given the nonlinear response observed on the forest floor mass, a polynomial model: Yij = m + b1ti + b2ti2 + bj + eij was used to determine if the treatment responses were sufficiently explained by a linear function for the variables: forest floor mass, forest floor nutrient concentration, and forest floor nutrient content. For this model, Yij is the jth observation of the ith treatment, m is the overall mean, b1 and b2 are the regression coefficients for the linear and quadratic terms, ti is the treatment effect of the ith treatment, bj is the block effect of the jth replicate, and eij is the random error. For all the other the variables, the model used was: Yij = m + ti + bj + eij where Yij is the jth observation of the ith treatment, m is the overall mean, ti is the treatment effect of the ith treatment, bj is the block effect of the jth replicate, and eij is the random error.

The Tukey’s Studentized range (HSD) test was used for treatment means comparisons. The Mitscherlich equation, a curvilinear asymptotic model with the form: Y = a(1  eb(x+c)) was fitted using PROC NLIN (SAS, 2005) to model the nonlinear relationships between wood production and forest floor total mass, Oa layer mass and potential nitrification. For this model (Y) was the volume at age 10 (m3 ha1), (X) was either forest floor total mass, Oa layer mass, or potential nitrification, a was the asymptotic wood production and b was the shape parameter, and c the intercept. 4. Results 4.1. Forest floor Forest floor mass on the doubled treatment was significantly greater than on the control and removed treatments for the Oa layer and all layers combined (Tables 1 and 3). Total forest floor mass increased non-linearly with the amount of forest floor and slash initially retained (Table 3 and Fig. 1). The doubled treatment had a forest floor mass of 19 Mg ha1; a 96% increase over the control (9.7 Mg ha1) and it had 25, 45, and 350% more mass in the Oi, Oe, and Oa layers, respectively, thus this non-linear accumulation is largely related to the differences in the accumulation of Oa across treatments. Forest floor masses in the removed and control treatments were very similar and averaged 4.8, 2.9, and 1.7 Mg ha1 for the Oi, Oe, and Oa layers, respectively. The two indexes of forest floor decomposition were not affected by treatment, indicating similar rates of decomposition for litterfall and Oi layer materials as they decomposed to Oe layer material. For the three treatments combined, the litterfall/Oi layer ratio averaged 1.05 and the Oi layer/Oe layer ratio averaged 1.63. Forest floor C concentrations averaged 50% and were not significantly different across treatments or forest floor layers (Table 1). Nitrogen was the only element that consistently showed significant linear or quadratic concentration responses to treatment in the different forest floor layers (Table 3). For the Oi layer, N concentration in the doubled treatment was 0.64%. This is 25 and 31% greater than the control and removed treatments, respectively. For the Oe and Oa layers, the Tukey range (HSD) test for mean comparisons did not show significant differences at P = 0.05 (Table 1). For all layers combined, the weighted average N concentration in the doubled treatment was 31% and 62% greater than in the control and removed treatments respectively (Tables 1 and 3). The significantly different N but constant C concentrations resulted in significantly different C:N ratios among treatments and

Fig. 1. Forest floor mass in a 10-year-old loblolly pine plantation regenerated under different forest floor and slash retention treatments. Mean comparisons shown by layer. Means with the same letter are not significantly different by Tukey’s Studentized range test (P = 0.05).

Table 1 Treatment means for forest floor mass and nutrient concentrations by layer in a 10-year-old loblolly pine plantation regenerated under different levels of forest floor and slash retention. Litter layer (Oi)

Fermentation layer (Oe)

Humus layer (Oa)

All layers

R

C

D

R

C

D

R

C

D

R

C

D

4.83 a (0.56)

4.84 a (0.53)

6.03 a (0.44)

2.85 a (0.21)

2.95 a (0.29)

4.26 a (0.59)

1.50 b (0.58)

1.95 b (0.17)

8.75 a (1.61)

9.18 b (1.17)

9.74 b (0.77)

19.05 a (2.43)

50.6 a (0.3) 0.49 b (0.03) 0.49 a (0.03) 0.63 a (0.02) 5.3 ab (0.4) 1.01 a (0.03) 0.69 a (0.05) 795 a (106) 9.7 a (0.2) 2.0 a (0.1) 24 a (2) 103 a (6)

50.6 a (0.2) 0.51 b (0.01) 0.42 a (0.03) 0.51 b (0.02) 5.6 a (0.3) 1.04 a (0.03) 0.69 a (0.01) 790 a (62) 10.1 a (0.3) 2.1 a (0.1) 28 a (1) 99 a (2)

51.0 a (0.3) 0.64 a (0.02) 0.48 a (0.02) 0.63 a (0.03) 4.7 b (0.3) 1.05 a (0.06) 0.76 a (0.02) 716 a (68) 10.2 a (0.2) 2.5 a (0.3) 26 a (3) 80 b (3)

51.1 a (0.3) 0.97 a (0.06) 0.80 a (0.03) 0.83 a (0.03) 6.4 a (0.3) 1.04 a (0.05) 1.20 a (0.08) 812 a (150) 13.5 a (0.2) 7.2 b (1.0) 42 a (3) 54 a (4)

50.3 a (0.5) 1.09 a (0.06) 0.80 a (0.01) 0.90 a (0.03) 7.6 a (0.6) 1.09 a (0.09) 1.34 a (0.08) 964 a (119) 13.2 a (0.8) 9.5 a (0.5) 50 a (1) 47 ab (2)

50.9 a (0.3) 1.19 a (0.08) 0.87 a (0.09) 0.92 a (0.07) 6.8 a (0.7) 1.07 a (0.08) 1.42 a (0.10) 1088 a (131) 13.9 a (0.5) 8.8 ab (1.2) 48 a (4) 43 b (3)

41.2 a (2.0) 1.58 a (0.04) 1.60 a (0.11) 1.82 a (0.09) 5.5 a (0.3) 1.57 a (0.21) 2.63 a (0.19) 1201 a (399) 34.2 a (3.5) 54.2 a (16.1) 102 a (4) 26 a (1)

50.4 a (3.5) 2.08 a (0.12) 1.43 a (0.08) 1.57 a (0.11) 8.0 a (1.1) 1.75 a (0.18) 2.62 a (0.20) 1588 a (163) 40.5 a (5.4) 39.3 a (11.5) 126 a (12) 24 a (2)

43.3 a (4.9) 1.82 a (0.21) 1.27 a (0.19) 1.17 a (0.27) 6.9 a (0.4) 1.22 a (0.15) 2.11 a (0.74) 1427 a (167) 36.2 a (4.4) 37.3 a (14.5) 105 a (3) 24 a (2)

49.4 a (0.4) 0.81 b (0.05) 0.74 a (0.01) 0.86 a (0.02) 5.7 a (0.3) 1.08 a (0.04) 1.14 a (0.05) 837 a (148) 14.4 b (0.5) 11.7 a (2.6) 41 b (4) 75 a (4)

50.5 a (0.9) 1.00 ab (0.03) 0.73 a (0.03) 0.84 a (0.03) 6.6 a (0.4) 1.20 a (0.06) 1.27 a (0.02) 1006 a (84) 17.2 ab (1.2) 11.8 a (2.4) 54 ab (3) 68 a (2)

47.9 a (2.0) 1.31 a (0.14) 0.93 a (0.07) 0.94 a (0.10) 6.1 a (0.1) 1.14 a (0.08) 1.54 a (0.32) 1127 a (117) 22.6 a (1.9) 19.0 a (5.3) 67 a (4) 47 b (3)

Mass (Mg ha1)

Treatments: R = Removed, C = Control, D = Doubled. Standard error shown in parenthesis (n = 4); means followed by the same letters are not significantly different by Tukey’s Studentized range (HSD) test (P = 0.05).

Table 2 Treatment means for forest floor nutrient content by layer in a 10-year-old loblolly pine plantation regenerated under different levels of forest floor and slash retention. Litter layer (Oi)

Contents C (kg ha1) N (kg ha1) P (kg ha1) K (kg ha1) Ca (kg ha1) Mg (kg ha1) S (kg ha1) Mn (kg ha1) B (kg ha1) Cu (kg ha1) Zn (kg ha1)

Fermentation layer (Oe)

Humus layer (Oa)

All layers

R

C

D

R

C

D

R

C

D

R

C

D

2446 a (288) 24 b (4) 2.4 a (0.3) 3.0 ab (0.3) 25 a (3) 4.9 b (0.5) 3.3 a (0.5) 3.8 a (0.6) 0.047 a (0.006) 0.010 a (0.001) 0.12 a (0.02)

2453 a (273) 25 b (3) 2.0 a (0.3) 2.5 b (0.3) 27 a (2) 5.0 b (0.4) 3.3 a (0.3) 3.8 a (0.4) 0.049 a (0.006) 0.010 a (0.002) 0.14 a (0.01)

3075 a (219) 39 a (4) 2.9 a (0.1) 3.8 a (0.2) 28 a (1) 6.3 a (0.4) 4.6 a (0.4) 4.3 a (0.4) 0.062 a (0.005) 0.015 a (0.001) 0.16 a (0.01)

1454 a (104) 28 b (3) 2.3 a (0.2) 2.4 a (0.2) 18 b (1) 2.9 a (0.2) 3.4 b (0.3) 2.2 b (0.3) 0.038 a (0.003) 0.020 b (0.003) 0.12 b (0.02)

1485 a (152) 32 ab (3) 2.4 a (0.3) 2.7 a (0.4) 22 ab (2) 3.2 a (0.4) 3.9 ab (0.4) 2.9 ab (0.6) 0.039 a (0.004) 0.028 ab (0.003) 0.15 ab (0.01)

2167 a (294) 50 a (7) 3.6 a (0.5) 3.8 a (0.5) 28 a (2) 4.4 a (0.3) 6.0 a (0.7) 4.5 a (0.5) 0.059 a (0.008) 0.036 a (0.004) 0.20 a (0.02)

627 b (262) 24 b (10) 2.2 b (0.7) 2.6 b (0.9) 8 b (3) 2.0 b (0.5) 3.9 a (1.6) 1.4 b (0.4) 0.049 b (0.015) 0.083 a (0.039) 0.15 b (0.06)

992 b (130) 41 b (5) 2.8 b (0.3) 3.1 b (0.4) 15 b (2) 3.4 b (0.3) 5.0 a (0.3) 3.2 b (0.6) 0.081 b (0.017) 0.075 a (0.021) 0.25 b (0.03)

4027 a (1202) 168 a (51) 11.7 a (3.1) 10.5 a (2.7) 61 a (13) 11.1 a (2.7) 20.9 a (8.6) 13.0 a (3.4) 0.303 a (0.039) 0.311 a (0.097) 0.93 a (0.19)

4527 b (562) 76 b (14) 6.8 b (0.9) 8.0 b (1.2) 52 b (4) 9.8 b (0.9) 10.7 a (1.9) 7.4 b (1.2) 0.134 b (0.022) 0.113 a (0.038) 0.39 b (0.09)

4930 b (456) 98 b (10) 7.1 b (0.7) 8.2 b (0.8) 64 b (4) 11.6 b (0.6) 12.3 a (0.9) 9.9 b (1.3) 0.169 b (0.024) 0.113 a (0.020) 0.53 b (0.05)

9267 a (1595) 257 a (60) 18.2 a (3.5) 18.1 a (2.8) 117 a (14) 21.8 a (3.3) 31.4 a (9.5) 21.7 a (4.1) 0.424 a (0.044) 0.362 a (0.098) 1.29 a (0.20)

J.L. Zerpa et al. / Forest Ecology and Management 259 (2010) 1480–1489

Concentration C (%) N (%) P (g kg1) K (g kg1) Ca (g kg1) Mg (g kg1) S (g kg1) Mn (mg kg1) B (mg kg1) Cu (mg kg1) Zn (mg kg1) C:N ratio

Standard error shown in parenthesis (n = 4); means followed by the same letters are not significantly different by Tukey’s Studentized range (HSD) test (P = 0.05). Treatments: R = Removed, C = Control, D = Doubled.

1483

J.L. Zerpa et al. / Forest Ecology and Management 259 (2010) 1480–1489

1484

also among forest floor layers. The Oi and all layers combined had significantly lower C:N ratios in the doubled than in the other two treatments (Tables 1 and 3). On average C:N ratios dropped from 94 to 48 and finally to 25 as forest floor material decomposed from the Oi to the Oe and finally to the Oa layer (Table 1). Forest floor nutrient contents followed similar trends as for forest floor mass (Table 2). Summing all forest floor layers, the N content was 257, 98, and 76 kg ha1 for the doubled, control and removed treatments, respectively (Table 2). Nitrogen contents in the doubled treatment were 60, 67, and 417% higher than the average N content in the other two treatments for the Oi, Oe, and Oa layer, respectively. The analyses in Table 3 show significant linear responses to treatment for C, N, K, Mg, S, B, and Cu content in the Oi layer and for all elements in the Oe, Oa, and in all layers combined. These results are slightly different from the mean comparisons shown in Table 2 because of the higher control on the type I error rate that Tukey’s HSD range test places on the mean comparisons. The blocking effect was in general non-significant for the forest floor variables (Table 3).

NH4+-N dominated the extractable-N fractions in the initial samples and NO3-N dominated in the incubated samples (Table 4). Initial NO3-N and NH4+-N values showed no significant differences among treatments (Table 4). However, potential nitrification showed a 400% increased in the doubled (12.77 kg NO3-N ha1) as compared with the control (3.23 kg NO3-N ha1) and was significantly higher that the potential nitrification of the removed (0.07 kg NO3-N ha1) treatments (Table 4). Potential nitrification values on the removed treatments were not significantly different from 0. Potential ammonification values did not differ from zero and consequently showed no significant differences among treatments (Table 4). Summing the potential ammonification and nitrification values resulted in significantly higher potential N mineralization on the doubled as compared to the removed treatment (Table 4). Mineral soil in the doubled treatment mineralized 9.87 kg NO3-N + NH4+-N ha1 28 days1 which accounts for 0.7% of the total N pool in the A-horizon of the mineral soil. During the incubation period the removed treatment immobilized 0.73 kg NO3-N + NH4+-N ha1 28 days1.

4.2. Mineral soil

4.3. Foliage, litterfall, retranslocation, and stand yield

Mineral soil properties in the A horizon were not significantly different across treatments. Total C and N, exchangeable P and Mn, exchangeable cations (Ca2+, K+, Mg2+, and Na+) in the A-horizon of the mineral soil averaged 21,000, 1500, 20, 75, 6, 5, 8, and 0.6 kg ha1 respectively. The pH, bulk density, and depth of the Ahorizon had averages of 4.8, 1.3 g cm3, and 23 cm, respectively (Table 4).

No significant differences among treatments were found in foliar nutrient concentrations (Table 5). Foliar concentrations averaged 1.45% for N and 1.2, 4.7, 2.4, and 1.1 g kg1 for P, K, Ca and Mg, respectively. Litterfall mass was significantly increased (30%) for the doubled (6565 kg ha1) as compared to the control (5058 kg ha1) and removed (4807 kg ha1) treatments (Table 5). Litterfall N and B concentrations in the doubled

Table 3 Summary of statistical significance (Pr > F) from ANOVA analyses on forest floor mass, concentration, and content in a 10-year-old loblolly pine plantation regenerated under different levels of forest floor and slash retention. Trt(lin), Trt(quad) are first and second degree polynomial functions describing the treatment response. Values in bold are significant at the P = 0.05 level. Layer

Source

Massa (kg ha1)

Concentrationa C (%)

N (%)

P (g kg1)

K (g kg1)

Ca (g kg1)

Mg (g kg1)

S (g kg1)

Mn (mg kg1)

B (mg kg1)

Cu (mg kg1)

Zn (mg kg1)

Oi

Block Trt (lin) Trt (quad)

0.0194 0.0257 0.1453

0.2904 0.2781 0.6034

0.2550 0.0024 0.0674

0.0977 0.8033 0.0213

0.9087 0.9981 0.0121

0.0209 0.0580 0.0431

0.3994 0.4925 0.7899

0.1827 0.1021 0.3455

0.0510 0.3422 0.6182

0.4525 0.1796 0.5691

0.4949 0.1431 0.4218

0.9731 0.5009 0.3322

Oe

Block Trt (lin) Trt (quad)

0.2847 0.0337 0.2241

0.9245 0.8057 0.2031

0.1831 0.0351 0.9063

0.4482 0.3734 0.5990

0.7583 0.2904 0.7388

0.0808 0.5112 0.0922

0.0080 0.5055 0.4277

0.2538 0.0808 0.7785

0.2447 0.1516 0.9247

0.2761 0.5586 0.5149

0.0075 0.0450 0.0351

0.3515 0.1884 0.2361

Oa

Block Trt (lin) Trt (quad)

0.1774 0.0008 0.0205

0.1993 0.6557 0.0782

0.1965 0.2188 0.0432

0.7351 0.1718 0.9742

0.4119 0.0344 0.7470

0.6065 0.2506 0.0956

0.7394 0.2747 0.2006

0.4798 0.4477 0.6754

0.5624 0.5888 0.4530

0.8474 0.8001 0.4333

0.4047 0.4209 0.7158

0.8982 0.7822 0.0783

All

Block Trt (lin) Trt (quad)

0.0488 0.0006 0.0159

0.3232 0.4082 0.2601

0.2176 0.0033 0.5177

0.6262 0.0319 0.1241

0.4240 0.3796 0.4403

0.5105 0.3877 0.1017

0.5871 0.6003 0.3189

0.3813 0.1694 0.7426

0.1278 0.0716 0.8432

0.9746 0.0111 0.5130

0.5295 0.2215 0.4810

0.7679 0.0031 0.9818

Layer

Source

Ratioa

Contenta

C:N

C (kg ha1)

N (kg ha1)

P (kg ha1)

K (kg ha1)

Ca (kg ha1)

Mg (kg ha1)

S (kg ha1)

Mn (kg ha1)

B (kg ha1)

Cu (kg ha1)

Zn (kg ha1)

Oi

Block Trt (lin) Trt (quad)

0.1925 0.0030 0.1279

0.0208 0.0248 0.1445

0.0780 0.0070 0.0786

0.3247 0.1181 0.0541

0.0673 0.0313 0.0086

0.0778 0.2589 0.9557

0.0143 0.0057 0.1070

0.1695 0.0371 0.1762

0.0998 0.3204 0.5850

0.0314 0.0223 0.2779

0.7685 0.0410 0.2512

0.3392 0.0802 0.9871

Oe

Block Trt (lin) Trt (quad)

0.0914 0.0212 0.5084

0.2922 0.0337 0.1977

0.3303 0.0099 0.2452

0.5031 0.0275 0.1963

0.4159 0.0271 0.3557

0.9295 0.0145 0.7176

0.8703 0.0267 0.3223

0.3878 0.0085 0.2403

0.6684 0.0179 0.4945

0.2155 0.0256 0.1590

0.3149 0.0094 0.8917

0.2357 0.0083 0.4444

Oa

Block Trt (lin) Trt (quad)

0.7933 0.4676 0.7514

0.2233 0.0084 0.1307

0.2563 0.0093 0.1452

0.2632 0.0073 0.0888

0.3000 0.0101 0.1073

0.1534 0.0009 0.0442

0.3765 0.0058 0.1395

0.2683 0.0393 0.2355

0.4413 0.0065 0.1500

0.5044 0.0005 0.0275

0.3630 0.0332 0.1408

0.2368 0.0016 0.0556

All

Block Trt (lin) Trt (quad)

0.2860 0.0002 0.0643

0.0770 0.0036 0.0691

0.1979 0.0061 0.1189

0.1944 0.0043 0.0511

0.1351 0.0029 0.0369

0.0636 0.0003 0.0384

0.2449 0.0031 0.1000

0.2108 0.0229 0.1904

0.3015 0.0052 0.1535

0.2567 0.0003 0.0185

0.3554 0.0238 0.1334

0.1487 0.0009 0.0514

a

Variable.

J.L. Zerpa et al. / Forest Ecology and Management 259 (2010) 1480–1489

1485

Table 4 Treatment means and summary of statistical significance (Pr > F) from ANOVA analyses for mineral soil variables and available nitrogen from the A-horizon in a 10-year-old loblolly pine plantation regenerated under different levels of forest floor and slash retention. Treatments

Removed

Control

Doubled

Source (Pr > F) Block

Trt

Mineral soil Total C (kg ha1) Total N (kg ha1) Mehlich-3 extractable P (kg ha1) Mehlich-3 extractable Mn (kg ha1) NH4Cl extractable Ca (kg ha1) NH4Cl extractable K (kg ha 1) NH4Cl extractable Mg (kg ha1) NH4Cl extractable Na (kg ha1) pH Bulk density (g cm3) Depth A-horizon (cm)

19563 a (1973) 1352 a (128) 18.9 a (3.9) 72.5 a (42.1) 7.2 a (2.5) 4.3 a (1.4) 7.5 a (1.8) 0.6 a (0.1) 4.81 a (0.04) 1.25 a (0.04) 21.2 a (2.2)

24514 a (4327) 1704 a (224) 11.7 a (1.3) 67.5 a (14.9) 7.2 a (1.5) 7.2 a (1.5) 8.9 a (0.6) 0.7 a (0.1) 4.81 a (0.03) 1.32 a (0.01) 23.5 a (2.2)

19063 a (1135) 1406 a (123) 30.3 a (8.6) 85.3 a (27.9) 4.7 a (1.2) 4.7 a (1.2) 6.8 a (1.1) 0.6 a (0.1) 4.89 a (0.04) 1.30 a (0.01) 24.3 a (4.2)

0.2034 0.5090 0.3184 0.5764 0.4793 0.2833 0.6025 0.7727 0.5628 0.3059 0.7365

0.2862 0.3488 0.1043 0.9212 0.5645 0.2904 0.5600 0.4266 0.3038 0.1698 0.7937

Available N Initial KCl extractable NO3 (kg ha1) Initial KCl extractable NH4+ (kg ha1) Pot. NO3 mineralization (kg ha1 28 days1) Pot. NH4+ mineralization (kg ha1 28 days1) Total N mineralization (kg ha1 28 days1)

0.50 a (0.19) 5.17 a (1.80) 0.07 b (0.25) 0.66 a (1.56) 0.73 b (1.61)

0.00 3.84 3.23 2.65 5.88

0.38 a (0.15) 5.00 a (0.51) 12.77 a (4.00) 2.90 a (1.94) 9.87 a (2.35)

0.5992 0.2603 0.3071 0.1993 0.8534

0.1168 0.6196 0.0195 0.1652 0.0278

a (0.00) a (0.49) ab (1.73) a (2.60) ab (1.11)

Standard error shown in parenthesis (n = 4); means followed by the same letters are not significantly different by Tukey’s Studentized range (HSD) test (P = 0.05). P-values in bold are significant at the P = 0.05 level.

treatment were also significantly higher than in the control and removed treatments. In contrast, P concentrations in the removed were higher than in the control treatment, and Ca concentrations in the control and removed treatments were higher than in the doubled treatment (Table 5). Litterfall nutrient contents (C, N, Mg, S, and B) were significantly increased in the doubled as compared to the other two treatments (Table 5). Foliar concentrations were higher than litter concentrations for all nutrients except for Ca, Mg, Mn, and B which are known to be very immobile elements. For the other nutrients, the estimated retranslocation rates were 68%, 69%, 32%, 41% and 28%, for N, K, S, Cu and Zn, respectively. Retranslocation was only affected by treatment for P. The 50% P retranslocation in the removed treatment was significantly lower than the control (67%) treatment (P = 0.040). Phosphorus retranslocation for the doubled treatment was 63%. The blocking effect was significant for litterfall mass, litterfall C, N, K, Ca, Mg, S, and B contents, and P and K foliar concentrations (Table 5). Ten-year stand volume in the double treatment was significantly greater than in the control and removed treatments (Table 5). Volume increased by 25 and 36% in the doubled (109 m3 ha1) as compared to the control (87 m3 ha1) and the removed (80 m3 ha1) treatments, respectively (Table 5). Stand density at age 10 averaged 727 trees ha1 across all treatments. The blocking effect was significant for stand volume (Table 5). 4.4. Relationships among variables Using the Mitschelich equation; total forest floor mass, Oa layer mass, and potential nitrification accounted for 66, 66, and 61% of the variation in stand volume respectively. 5. Discussion The 9.4 Mg ha1 of forest floor mass (Table 1) and 87 kg ha1 of N (Table 2) found on the removed and control treatments in this 10-year-old loblolly pine plantation are comparable to those reported by Larsen et al. (1976) for a 13-year-old loblolly pine plantation established in the hilly costal plain of Alabama. The 19.1 Mg ha1 and 257 kg ha1 of N content found on the doubled treatment are more typical of the 20+ year plantations reported by Shepard (1985).

The non-linear pattern of forest floor mass and nutrient accumulation observed across treatments (Fig. 1 and Table 2) is interesting because the amount of forest floor material retained at time of planting was linearly related to treatment (0, 1, and 2, for the removed, control, and doubled treatments, respectively). This is especially true for the Oa layer mass where the doubled treatment was >400% the control and removed treatments. Several factors may account for this non-linear pattern including: (1) slower decomposition of the retained material from the previous rotation for the doubled treatment, (2) slower decomposition rates of this rotation’s litterfall, and (3) greater inputs of litterfall. Unfortunately, decomposition of the retained material was not assessed at the beginning of the study. However, decomposition models for regions with high summer temperatures and well distributed seasonal precipitation, such as the Southeast US, show relatively fast decomposition rates (Meentemeyer, 1978; Moorhead et al., 1999) and accelerated litter decomposition has been previously reported after harvest (Gadgil and Gadgil, 1978; Bengston, 1981; Pritchett and Fisher, 1987; Blumfield et al., 2004). Therefore, it is unlikely that the retained forest floor from the previous rotation would still have a direct effect on forest floor accumulation 10 years after treatment. The lack of treatment differences in the decomposition indexes for Oi and Oe layers indicate that initial decomposition rates of this rotation’s litterfall were similar across treatments and could not account for the nonlinear pattern of forest floor accumulation. In contrast, litterfall was over 30% greater on the doubled treatment (Table 5) as compared to the control and the removed treatments. We expect that increased litter production accounted for much of the observed non-linear pattern in forest floor accumulation and nutrient content. Interestingly, none of the other published studies that include a ‘‘doubling’’ treatment (Smith et al., 2000; Mendham et al., 2003; Tutua et al., 2008) have shown significant positive long term effects on litterfall and forest floor accumulation. This highlights the need to account for site and species differences through replications across the landscape rather than within a site, and study installations using species with different litter qualities. A possible explanation for the lack of differences in forest floor accumulation between the control and removed treatments may be due to rapid forest floor decomposition on the control treatment during the early stand development. This decomposition could

J.L. Zerpa et al. / Forest Ecology and Management 259 (2010) 1480–1489

1486

Table 5 Treatment means and summary of statistical significance (Pr > F) from ANOVA analyses for foliar nutrient concentrations, litterfall mass, nutrient concentration, and content, and stand yield in a 10-year-old loblolly pine plantation regenerated under different levels of forest floor and slash retention. Treatment

Source

Removed

Control

Doubled

Block

Trt

Litterfall (kg ha1)

4807 b (689)

5058 b (343)

6565 a (351)

0.0104

0.0049

Foliar nutrient concentrations N (%) P (g kg1) K (g kg1) Ca (g kg1) Mg (g kg1) S (g kg1) Mn (mg kg1) B (mg kg1) Cu (mg kg1) Zn (mg kg1)

1.44 a (0.07) 1.25 a (0.03) 4.83 a (0.10) 2.23 a (0.10) 1.13 a (0.06) 0.88 a (0.03) 524 a (52) 9.75 a (1.25) 5.50 a (0.87) 38 a (3)

1.43 a (0.03) 1.20 a (0.06) 4.70 a (0.26) 2.55 a (0.22) 1.15 a (0.03) 0.95 a (0.03) 581 a (52) 10.50 a (0.65) 5.25 a (0.95) 43 a (1)

1.47 a (0.03) 1.25 a (0.03) 4.50 a (0.12) 2.28 a (0.20) 1.15 a (0.10) 0.93 a (0.03) 497 a (53) 11.00 a (1.08) 5.25 a (0.48) 44 a (2)

0.3805 0.0029 0.0106 0.9930 0.4547 0.8195 0.1369 0.5314 0.2113 0.8972

0.8008 0.1250 0.1047 0.5699 0.9610 0.2746 0.4028 0.7156 0.9565 0.2512

Litterfall nutrient concentrations C (%) N (%) P (g kg1) K (g kg1) Ca (g kg1) Mg (g kg1) S (g kg1) Mn (mg kg1) B (mg kg1) Cu (mg kg1) Zn (mg kg1)

52.93 a (0.32) 0.44 b (0.01) 0.62 a (0.04) 1.56 a (0.10) 4.92 a (0.20) 1.06 a (0.03) 0.61 a (0.01) 906 a (99) 9.15 b (0.27) 2.89 a (0.16) 25.49 a (3.20)

53.10 a (0.18) 0.44 b (0.02) 0.39 b (0.05) 1.39 a (0.07) 4.69 a (0.09) 1.07 a (0.01) 0.59 a (0.01) 861 a (74) 9.13 b (0.29) 3.03 a (0.29) 31.82 a (1.77)

53.13 a (0.07) 0.51 a (0.03) 0.46 ab (0.04) 1.39 a (0.04) 3.93 b (0.15) 1.10 a (0.03) 0.66 a (0.02) 705 a (82) 10.03 a (0.19) 2.87 a (0.14) 30.88 a (1.84)

0.1787 0.0609 0.5686 0.6646 0.3295 0.3902 0.9080 0.1245 0.1252 0.7512 0.7247

0.7086 0.0170 0.0294 0.2799 0.0070 0.5549 0.1101 0.1661 0.0308 0.8645 0.2586

Litterfall nutrient contents C (kg ha1) N (kg ha1) P (kg ha1) K (kg ha1) Ca (kg ha1) Mg (kg ha1) S (kg ha1) Mn (kg ha1) B (kg ha1) Cu (kg ha1) Zn (kg ha1)

2546 b (369) 21 b (3) 2.9 a (0.3) 8 a (1) 23 a (2) 5.1 b (0.7) 2.9 b (0.4) 4.2 a (0.2) 0.044 b (0.007) 0.014 a (0.001) 0.129 a (0.035)

2687 b (190) 22 b (2) 2.0 a (0.3) 7 a (1) 24 a (2) 5.5 b (0.4) 3.0 b (0.3) 4.3 a (0.4) 0.046 b (0.004) 0.016 a (0.002) 0.162 a (0.018)

3484 a (188) 34 a (4) 3.1 a (0.3) 9 a (1) 26 a (1) 7.3 a (0.5) 4.3 a (0.3) 4.6 a (0.4) 0.067 a (0.005) 0.019 a (0.002) 0.201 a (0.002)

0.0099 0.0061 0.3507 0.0368 0.0199 0.0016 0.0035 0.1693 0.0077 0.0969 0.1460

0.0050 0.0020 0.0822 0.0582 0.2535 0.0007 0.0006 0.6319 0.0025 0.0696 0.0816

Stand yield dbh (cm) Height (m) Volume (m3 ha1)

16.0 b (1.0) 10.9 a (0.5) 80 b (13)

17.4 ab (0.4) 11.1 a (0.2) 87 b (6)

18.8 a (0.6) 11.3 a (0.2) 109 a (7)

0.0323 0.0683 0.0127

0.0155 0.4711 0.0134

Standard error shown in parenthesis (n = 4); means followed by the same letters are not significantly different by Tukey’s Studentized range (HSD) test (P = 0.05). P-values in bold are significant at the P = 0.05 level.

have released large amounts of available nutrients at a time when the root systems of the young trees were not well enough developed to capture these nutrients. After this initial release of available nutrients, the removed and the control treatments should have had similar levels of nutrient availability. In contrast, the doubled treatment may still have had sufficient forest floor to continue decomposing providing needed nutrients for stand growth when demand for nutrients increased (Allen et al., 1990). The forest floor mass from the Oi and Oe layers not only exhibited a significant linear response to treatment (Table 3) but also significantly better quality as indicated by higher N concentrations and lower C:N ratios (Table 1). The higher N concentration in the litterfall (Table 5) may be in part responsible for the improved quality of the Oi layer. The N contained in the forest floor accounted for only 5% of the combined N of the forest floor and A horizon in the removed and control treatments and 15% of the N in the doubled treatment. No treatment differences were found in the total C and N pools or in extractable P or basic cations in the mineral soil (Table 4) despite the significant treatment differences on carbon and nutrient content found in the forest floor (Tables 2 and 3). Other studies where post-harvest residue retention has been manipulated have

shown similar results where little (Olsson et al., 1996; Mendham et al., 2003) to no effect (Stone and Elioff, 1998; Tiarks et al., 2003) of residue retention on mineral soil C and nutrient pools have been found. In order to understand the contributions of the forest floor and mineral soil N pools to N availability, several fluxes were quantified using commonly accepted methods such as extractions of fresh soil and aerobic incubations. Significant treatment effects were only found for potential N mineralization principally due to potential nitrification effects (Table 4). These results are in agreement with studies where the retention of harvest residues has shown a more pronounced effect on N fluxes (O’Connell et al., 2000) than on N pools (Mendham et al., 2003) in the mineral soil. The low values of extractable soil NO3-N observed are typical of loblolly pine systems (Piatek and Allen, 1999; Gurlevik et al., 2004) and they apparently reflect the preference that microbial populations have for NH4+ since it is a more energy efficient, already reduced form of N. If the microbial populations are provided with high and constant C inputs as occurs in undisturbed forests, they tend to immobilize most of the available NH4+ and little NO3 is produced (Davidson et al., 1992; Hart et al., 1994a). Other reasons for the low levels of NO3-N observed in the soil extractions could be attributed to low

J.L. Zerpa et al. / Forest Ecology and Management 259 (2010) 1480–1489

Fig. 2. Relationship between stand volume at age 10 and total forest floor mass for a loblolly pine plantation regenerated under different forest floor and slash retention treatments.

nitrifier populations and low initial NH4+ levels which lead to low nitrification rates. The high extractable NO3-N values following the 28-day lab incubation on the doubled treatment (Table 4) may have resulted from eliminating C inputs that would have been present from the large forest floor mass and root systems in the field. The reduction in C inputs would reduce the NH4+ demand from heterotrophic microbes and leaving it available for oxidation by the nitrifiers (Davidson et al., 1992; Hart et al., 1994a). Litterfall for these homogeneous plantations is mainly composed of foliage, thus the positive effect of the doubled treatment on soil N availability was well reflected by the greater foliage production (indicated by litterfall) and stand yield found in this treatment (Table 5). In addition, this linkage is indicated by the strong relationship among stand yield, litterfall, and potential nitrification (Figs. 2 and 6). The relationship between stand volume and total forest floor mass was curvilinear (Fig. 3) suggesting no additional increase in volume with increasing forest floor accumulation above 15 Mg ha1. This relationship was very similar when the Oa layer mass, instead of total forest floor mass, was used (Fig. 4), which indicates the important contributions of this layer to productivity on this site. The positive linear relationship between potential nitrification and Oa layer mass (Fig. 5) suggests increasing levels of nitrogen availability with greater humus accumulation. Based on the

Fig. 3. Relationship between stand volume at age 10 and Oa layer mass for a loblolly pine plantation regenerated under different forest floor and slash retention treatments.

1487

Fig. 4. Relationship between potential nitrification in the mineral soil and Oa layer mass for a 10-year-old loblolly pine plantation regenerated under different forest floor and slash retention treatments. 95% confidence intervals represented by dotted lines.

Fig. 5. Relationship between stand volume at age 10 and annual litterfall for a loblolly pine plantation regenerated under different forest floor and slash retention treatments. 95% confidence intervals represented by dotted lines.

Fig. 6. Relationship between stand volume at age 10 and potential nitrification in the mineral soil for a loblolly pine plantation regenerated under different forest floor and slash retention treatments.

1488

J.L. Zerpa et al. / Forest Ecology and Management 259 (2010) 1480–1489

asymptotic relationship between volume and forest floor mass, it appears that the high levels of available nitrogen (4 kg ha1 28 days1) present with high forest floor masses may not be fully used by the stand (Fig. 6). The lack of significant treatment effects on foliar nutrient concentrations was not surprising as mid rotation pine plantations exhibiting a range in productivity typically have similar foliar nutrient concentrations (Valentine and Allen, 1990). Available nutrient levels are most commonly reflected in the production of foliage and not its concentration (Vose and Allen, 1988). Additionally, by year 10, sufficient crown development has occurred that retranslocated nutrients such as N, P, and K play an important role in meeting nutrient demand (Switzer and Nelson, 1972). Greater amounts of available N in the mineral soil on the doubled treatment apparently resulted in a feed forward effect with greater growth and in turn greater litterfall and accumulation of new forest floor material (Fig. 1). This in turn has resulted in long term increases in soil available N through the 10th year following plantation establishment (Fig. 5). The native soil fertility at this site was relatively low which may explain the positive tree growth responses to increased post-harvest residue retention. Studies where these same type of treatments have been imposed on better sites show less (Smith et al., 2000) or more transient effects (Mendham et al., 2003). Thus the initial fertility of the site along with the magnitude and quality of the retained material seem to been good predictors of the differences in stand productivity across treatments (Saint-Andre et al., 2008). As the stand continues to grow it would be interesting to follow up with a study to quantify nutrient pools and fluxes at the individual tree and stand level to gain a better understanding of the organic matter retention effect on tree growth and site productivity (Laclau et al., 2003). The fact that the doubled forest floor treatment was still positively impacting stand productivity and nitrogen availability after 10 years suggests the potential for long term increases in site nutrient availability when the larger forest floor masses found in today’s fertilized stands are retained. References Albaugh, T.J., Allen, H.L., Fox, T.R., 2007. Historical patterns of forest fertilization in the southeastern United States from 1969 to 2004. South J. Appl. For. 31, 129– 137. Allen, H.L., Dougherty, P.M., Campbell, R.G., 1990. Manipulation of water and nutrients—practice and opportunity in Southern U.S. pine forests. For. Ecol. Manage. 30, 437–453. Bengston, G.W., 1981. Nutrient conservation in forestry: a perspective. South J. Appl. For. 5, 50–58. Berg, B., 1986. Nutrient release from litter and humus in coniferous forest soils—a mini review. Scand. J. For. Res. 1, 359–369. Berg, B., Berg, M., Bottner, P., Box, E., Breymenyer, A., Calvo de Anta, R., Couteaux, M.M., Gallardo, A., Escudero, A., Kartz, W., Madeira, M., Malkonen, E., Meentemeyer, V., Munoz, F., Piussi, P., Remacle, J., Virzo de Santo, A., 1993. Litter mass loss rates in pine forests of Europe and Eastern United States; some relationships with climate and litter quality. Biogeochemistry 20, 127–159. Blumfield, T.J., Xu, Z., Mathers, N.J., Saffigna, P.G., 2004. Decomposition of nitrogen15 labeled hoop pine harvest residue in Subtropical Australia. Soil Sci. Soc. Am. J. 68, 1751–1761. Burger, J.A., Pritchett, W.L., 1984. Effects of clearfelling and site preparation on nitrogen mineralization in a southern pine stand. Soil Sci. Soc. Am. J. 48, 1432– 1437. Burkhart, H.D., 1977. Cubic-foot volume of loblolly pine to any merchantable top limit. South J. Appl. For. 1, 7–9. Carey, M.L., Hunter, I.R., Andrew, I., 1982. Pinus radiata forest floors: factors affecting organic matter and nutrient dynamics. N. Zeal. J. For. Sci. 12, 36–48. Cortina, J., Vallejo, V.R., 1994. Effects of clearfelling on forest floor accumulation and litter decomposition in a radiata pine plantation. For. Ecol. Manage. 70, 299–310. Covington, W.W., 1981. Changes in forest floor organic matter and nutrient content following clear cutting in northern hardwoods. Ecology 62, 41–48. Cox, P., Wilkinson, S.P., Anderson, J.M., 2001. Effects of fungal inocula on the decomposition of lignin and structural polysaccharides in Pinus sylvestris litter. Biol. Fertil. Soils 33, 246–251. Davidson, E.A., Hart, S.C., Firestone, M.K., 1992. Internal cycing of nitrate in soils of a mature coniferous forest. Ecology 73, 1148–1156.

De Santo, A.V., Berg, B., Rutigliano, F.A., Aleani, A., Fioretto, A., 1993. Factors regulating early stage decomposition of needle litters in five different coniferous forests. Soil Biol. Biochem. 25, 1423–1433. Fox, T.R., Burger, J.A., Kreh, R.E., 1986. Effects of site preparation on nitrogen dynamics in the southern Piedmont. For. Ecol. Manage. 15, 241–256. Fox, T.R., Allen, H.L., Albaugh, T.J., Rubilar, R., Carlson, C.A., 2007. Tree nutrition and forest fertilization of pine plantations in the southern United States. South J. Appl. For. 31, 5–11. Gadgil, R.L., Gadgil, P.D., 1978. Influence of clearfelling on decomposition of Pinus radiata litter. N. Zeal. J. For. Sci. 8, 213–224. Grossman, R.B., Reinsch, T.G., 2002. The solid phase, bulk density and linear extensibility. In: Dane, J.H., Topp, G.C. (Eds.), Methods of Soil Analysis. Part 4—Physical Methods. Soil Science Society of America, Madison, WI, pp. 201– 228. Gurlevik, N., Kelting, D.L., Allen, H.L., 2003. The effects of vegetation control and fertilization on net nutrient release from decomposing loblolly pine needles. Can. J. For. Res. 33, 2491–2502. Gurlevik, N., Kelting, D.L., Allen, H.L., 2004. Nitrogen mineralization following vegetation control and fertilization in a 14-year-old loblolly pine plantation. Soil Sci. Soc. Am. J. 68, 272–281. Guthrie, R.L., Witty, J.E., 1982. New designations for soil horizons and layers and the new Soil Survey Manual. Soil Sci. Soc. Am. J. 46, 443–444. Hart, S.C., Nason, G.E., Myrold, D.D., Perry, D.A., 1994a. Dynamics of gross nitrogen transformations in an old growth forest: the carbon connection. Ecology 75, 880–891. Hart, S.C., Stark, J.M., Davidson, E.A., Firestone, M.K., 1994b. Nitrogen mineralization, immobilization, and nitrification. In: Weaver, R.W., Angle, J.S., Bottomley, P.S. (Eds.), Methods of Soil Analysis. Part 2. Microbiological and Biochemical Properties. Soil Science Society of America, Madison, WI, pp. 985–1018. Jorgensen, J.R., Wells, C.G., Metz, L.J., 1980. Nutrient changes in decomposing loblolly pine forest floor. Soil Sci. Soc. Am. J. 44, 1307–1314. Kendrick, W.B., 1959. The time factor in the decomposition of coniferous leaf litter. Can. J. Bot. 37, 907–912. Laclau, J.P., Deleporte, P., Ranger, J., Bouilleti, J.P., Kazotti, G., 2003. Nutrient dynamics throughout the rotation of Eucalyptus clonal stands in Congo. Ann. Bot. 91, 879–892. Larsen, H.S., Carter, M.C., Gooding, J.W., Hynk, D.M., 1976. Biomass and nitrogen distribution in four 13-year-old loblolly pine plantations in the Hilly Coastal Plain of Alabama. Can. J. For. Res. 6, 187–194. Li, Q.C., Allen, H.L., Wilson, C.A., 2003. Nitrogen mineralization dynamics following the establishment of a loblolly pine plantation. Can. J. For. Res. 33, 364–374. Markewitz, D., Richter, D.D., Allen, H.L., Urrego, J.B., 1998. Three decades of observed soil acidification in the calhoun experimental forest: has acid rain made a difference? Soil Sci. Soc. Am. J. 62, 1428–1439. McHale, P.J., Mitchell, M.J., Bowles, F.P., 1998. Soil warming in a northern hardwood forest: trace gas fluxes and leaf litter decomposition. Can. J. For. Res. 28, 1365– 1372. Meentemeyer, V., 1978. Macroclimate the lignin control of litter decomposition rates. Ecology 59, 465–472. Mendham, D.S., O’Connell, A.M., Grove, T.S., Rance, S.J., 2003. Residue management effects on soil carbon and nutrient contents and growth of second rotation Eucalyptus. For. Ecol. Manage. 181, 357–372. Moorhead, D.L., Currie, W.S., Rastetter, E.B., Parton, W.J., Harmon, M.E., 1999. Climate and litter quality controls on decomposition: an analysis of modeling approaches. Global Biogeochem. Cycles 13. Nelson, D.W., Sommers, L.E., 1996. Total carbon, organic carbon, and organic matter. In: SSSA (Ed.), Methods of Soil Analysis. Part 3. Chemical Methods. Soil Science Society of America, Inc., Madison, WI, pp. 1004–1005. NOAA, 2003. Climatography of the United States. Report No. 81. National Oceanic and Atmospheric Administration, Ashville, NC, pp. 1–12. O’Connell, A.M., Grove, T.S., Mendham, D.S., Rance, S.J., 2000. Effects of site management in Eucalyptus plantations in southwestern Australia. In: Nambiar, E.K.S., Tiarks, A., Cossalter, C., Ranger, J. (Eds.), Site Management Productivity in Tropical Plantation Forests: A Progress Report. Center for International Forestry Research, Bogor, Indonesia, pp. 61–71. Olsson, B.A., Staaf, H., Lundkvist, H., Bengtsson, J., Rosen, K., 1996. Carbon and nitrogen in coniferous forest soils after clear-felling and harvests of different intensity. For. Ecol. Manage. 82, 19–32. Piatek, K.B., Allen, H.L., 1999. Nitrogen mineralization in a pine plantation fifteen years after harvesting and site preparation. Soil Sci. Soc. Am. J. 63, 990–998. Piatek, K.B., Allen, H.L., 2001. Are forest floors in mid-rotation stands of loblolly pine (Pinus taeda) a sink for nitrogen and phosphorus? Can. J. For. Res. 31, 1164–1174. Pritchett, W.L., Fisher, R.F., 1987. Properties and Management of Forest Soils. Wiley, New York. Rojas, J.C., 2005. Factors influencing responses of loblolly pine stands to fertilization. Dissertation. Dep. Forestry. North Carolina State University, Raleigh, p. 147. Rustad, L., Fernandez, I., 1998. Soil warming: consequences for foliar litter decay in a spruce-fir forest in Maine, USA. Soil Sci. Soc. Am. J. 62, 1072–1080. Saint-Andre, L., Laclau, J.P., Deleporte, P., Gava, J.L., Goncalves, J.L.M., Mendham, D.S., Nzila, J.D., Smith, C., du Toit, B., Xu, D.P., Sankaran, K.V., Marien, J.N., Nouvellon, Y., Bouillet, J.P., Ranger, J., 2008. Slash and litter management effects on Eucalyptus productivity: a synthesis using a growth and yield modelling approach. In: Nambiar, E.K.S. (Ed.), Site Management and Productivity in Tropical Plantation Forests. Center for International Forestry Research (CIFOR), Proceedings of Workshops in Piracicaba (Brazil) 2004 and Bogor (Indonesia) 2006, pp. 173–189.

J.L. Zerpa et al. / Forest Ecology and Management 259 (2010) 1480–1489 Sariyildiz, T., Anderson, J.M., 2003. Interactions between litter quality, decomposition and soil fertility: a laboratory study. Soil Biol. Biochem. 35, 391– 399. SAS, 2005. The SAS System for Windows Release 9.1.3. SAS Institute Inc., Cary, NC. Scott, N.A., Binkley, D., 1997. Foliage litter quality and annual net N mineralization: comparison across North America forest sites. Oecologia 111, 151–159. Sedjo, R.A., 2001. The role of forest plantations in the world’s future timber supply. Forest Chron. 77, 221–225. Shepard, J.P., 1985. Relationships Between Nutrient Cycling and Net Primary Productivity in Loblolly Pine Plantation Ecosystems of The East Gulf Coastal Plain. Department of Forestry. Mississippi State University, p. 195. Smith, C.T., Lowe, A.T., Skinner, M.F., Beets, P.N., Schoenholtz, S.H., Fang, S.Z., 2000. Response of radiata pine forests to residue management and fertilisation across a fertility gradient in New Zealand. For. Ecol. Manage. 138, 203–223. Stone, D.M., Elioff, J.D., 1998. Soil properties and aspen development 5 years after compaction and forest floor removal. Can. J. Soil Sci. 78, 51–58. Switzer, G., Nelson, L., 1972. Nutrient accumulation and cycling in loblolly pine (Pinus taeda L.) plantation ecosystems: the first twenty years. Soil Sci. Soc. Am. Proc. 36, 143–147. Tew, D.T., Morris, L.A., Allen, H.L., Wells, C.G., 1986. Estimates of nutrient removal, displacement and loss resulting from harvest and site preparation of a Pinus

1489

taeda plantation in the Piedmont of North-Carolina. For. Ecol. Manage. 15, 257– 267. Tiarks, A., Elliott-Smith, M., Stagg, R., 2003. Effects of logging residue management of the growth and nutrient distribution of a Pinus taeda plantation in Central Louisiana, USA. In: Nambiar, E.K.S., Ranger, J., Tiarks, A., Toma, T. (Eds.), Proceedings of the Fifth Workshop on Site Management and Productivity in Tropical Plantations Forests: Impact on Soils and Options for Management Over Successive Rotations, Guangzhou, China, pp. 151–162. Tutua, S.S., Xu, Z.H., Blumfield, T.J., Bubb, K.A., 2008. Long-term impacts of harvest residue management on nutrition, growth and productivity of an exotic pine plantation of sub-tropical Australia. For. Ecol. Manage. 256, 741–748. Valentine, D.W., Allen, H.L., 1990. Foliar responses to fertilization identify nutrient limitation in Loblolly-Pine. Can. J. For. Res. 20, 144–151. Vitousek, P.M., Matson, P.A., 1985. Intensive harvesting and site preparation decrease soil nitrogen availability in young plantations. South J. Appl. For. 9, 120– 125. Vose, J.M., Allen, H.L., 1988. Leaf-area, stemwood growth, and nutrition relationships in Loblolly-Pine. Forest Sci. 34, 547–563. Zarcinas, B.A., Cartwright, B., Spouncer, L.R., 1987. Nitric-acid digestion and multielement analysis of plant-material by inductively coupled plasma spectrometry. Commun. Soil Sci. Plan. 18, 131–146.