SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Adegbidi, H. G.
Right arrow Articles by Barros, N. F.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Adegbidi, H. G.
Right arrow Articles by Barros, N. F.
Agricola
Right arrow Articles by Adegbidi, H. G.
Right arrow Articles by Barros, N. F.
Related Collections
Right arrow Ecosystem Management
Right arrow Plant and Soil Interactions
Right arrow Forest Soils
Published in Soil Sci. Soc. Am. J. 68:596-604 (2004).
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

DIVISION S-7—FOREST & RANGE SOILS

Root Development of Young Loblolly Pine in Spodosols in Southeast Georgia

H. G. Adegbidia, N. B. Comerford*,b, E. J. Jokelac and N. F. Barrosd

a 165 Boulevard Hébert, Université de Moncton—Campus d'Edmundston, Edmundston, New Brunswick E3V 2S8 Canada
b Soil and Water Science Dep., 2169 McCarty Hall, Univ. of Florida, P.O. Box 110290, Gainesville, FL 32611-0290
c School of Forest Resources and Conservation, Univ. of Florida, Gainesville, FL 32611
d Departamento do Solos, Universidade Federal de Viçosa, Viçosa, Minas Gerais 36570-000 Brazil

* Corresponding author (nbc{at}mail.ifas.ufl.edu).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Determining fine-root dynamics is fundamental to forest soil nutrient management yet root development of fast-growing loblolly pine (Pinus taeda L.) is poorly documented. The objectives of this study were to (i) investigate the spatial and temporal root development of loblolly pine; (ii) evaluate the relationship between root length, number of roots exiting a trench face, and root mass densities; and (iii) determine if there is a relationship between fine root and foliage mass as well as root and shoot mass during the early stages of stand development. Thirteen forest sites in southeastern Georgia covering ages 1 to 4 yr old were used. Roots temporal and spatial distributions were investigated using a trench method. The value of NX (# roots cm–2) was measured in August/September during the first 4 yr of stand development. Root density depth distributions fit a natural logarithm relationship with soil depth. An empirical model for root development over time was developed. A two-dimensional evaluation of root development showed that roots were present in 13 to >60% of the soil volume from Year 1 to Year 4. Regressions between root length density, LV (cm root cm–3 soil), and NX were weak until root mass and soil depth were included. Lastly, it was shown that the ratio of fine root mass/foliage mass was stable after the establishment phase, as was the ratio of root to shoot.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
FOREST TREE EXPLORATION of the soil profile for water and nutrients is a function of the growth and development of the root system and its associated mycorrhizal component. A depletion zone forms around fine roots and their extramatrical mycorrhizal hyphal strands, with the size of the depletion zone being a function of soil properties and time (Tinker and Nye, 2000, p. 133). The soil in this depletion zone represents the soil volume from which the root system is extracting a particular nutrient. Therefore, temporal and spatial patterns of root development define the soil volume from which a forest stand extracts resources and, therefore define the magnitude of soil resources available during any given time period.

Providing sufficient nutrients to support the growth requirements of a forest stand is the basis of forest soil nutrient management. Given the obvious importance of fine-root dynamics, the paucity of even simple fine-root depth and lateral distribution data for important commercial forest species has restricted a comprehensive understanding of stand development dynamics and accurate estimates of net primary production. Such information, especially in relation to varying stages of stand development, is completely absent for most tree species, even though the root system can represent 10 to 40% of total tree biomass (Santantonio et al., 1977; Fogel, 1983).

A mechanistic understanding of root dynamics that determines their distribution in the soil over time involves the development of quantitative relationships between soil characteristics and (i) allocation of fixed C to soil compartments, (ii) root mortality and turnover, and (iii) morphology of the root system that is developed. However, in the absence of such information, empirical observation represents a valid approach for understanding aspects of root dynamics.

The focus of this research is on loblolly pine that is growing rapidly because of competition control and nutrient amendments. Loblolly pine root distribution is influenced by soil characteristics such as nutrient level (Adams et al., 1989), water regime (Hallgren et al., 1991; Ludovici and Morris, 1996; Torreano and Morris, 1998; Albaugh et al., 1998), CO2 and temperature (Larigauderie et al., 1994; King et al., 1996). Although observational data can only be extended to similar soil, climatic, and plant conditions, they are useful because they add to the growth database of the tree species, stimulate thought and answer questions for specific soil and climatic conditions.

Predicting the supply of nutrients to a forest stand is partly a function of the amount of tree roots and their distribution throughout the soil profile (Smethurst and Comerford, 1993). Nutrient supply is most critical during the establishment and early developmental stages of stand growth when the root system is colonizing the soil profile. Therefore, the first objective of this study was to investigate spatial and temporal development of loblolly pine fine roots during the establishment phase and to develop an empirical growth model that characterizes their development.

One reason why there is such a scarcity of information concerning forest root development is the difficulty in measuring fine root length and biomass. Kendall and Moran (1963) and Melhuish and Lang (1968) recommended an easy method for field investigations of fine roots. They suggested that counts of roots exiting the face of a soil trench could be converted to root length densities using a simple algebraic expression: LV = 2NX, where LV is root length density (cm cm–3) and NX is the root count exiting a trench face (number of roots cm–2). Attempts to verify this approach have met with varying success (Bennie et al., 1977; Drew and Saker, 1980; Bland, 1989; Escamilla et al., 1991; Lopez-Zamora et al., 2002). While the equation may not always be appropriate, a correlation between LV and NX exists and can be used to make field investigations of root distributions less demanding (Escamilla et al., 1991; Lopez-Zamora et al., 2002). Therefore, a second objective of this study was to evaluate the relationship between fine root NX, LV, and root mass (M; g root cm–3 soil).

While root data are scarce, a more extensive database exists on aboveground growth dynamics of loblolly pine plantations on southeastern Coastal Plain Spodosols (Devan and Burkhart, 1982; Sprinz and Burkhart, 1987; Liu et al., 1989; Glover et al., 1989; Britt et al., 1991; Dean and Baldwin, 1996; Quicke et al., 1999; Martin and Shiver, 2002). An important question is whether it is possible to relate aboveground to belowground growth, especially over time. Root–shoot relationships have been the focus of several previous loblolly pine investigations (Monk, 1966; Harris et al., 1977; Pehl et al., 1984; Johnson, 1990; Van Lear and Kapeluck, 1995; King et al., 1999). Johnson (1990) observed that the root system comprised 18% of total biomass for 1-yr-old seedlings. More mature forest trees have ranged from 22% (25-yr-old trees in the western Gulf Coastal Plain [Pehl et al., 1984]) to 36% (9-yr-old loblolly in the sandhills of North Carolina [King et al., 1999]). No study has documented the change in root/shoot ratios of fast-growing loblolly pine stands during the establishment and early development phases. Likewise, we are not aware of published works on the relationship between the mass of fine roots and the foliage mass during this stage of growth. Therefore, our last objective was to evaluate the relationships between fine roots and foliage, and between tree root biomass and shoot biomass.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Study Sites
A previously published description of the study sites can be found in Adegbidi et al. (2002). This study was conducted on 10 fast-growing loblolly pine plantations that varied in age from 1 to 4 yr, planted at 1.8 by 3.7 m spacing and were growing in the lower Coastal Plain region of southern Georgia. All sampling was accomplished over two growing season. Ages 1 and 2 were represented by three spatially unique replicate stands that were sampled at age 1 and resampled at age 2. Age 3 was represented by three additional spatially unique stands; while age 4 data were from four more spatially unique forest stands (Table 1). Forest stands were a combination of operational plantations and an experimental study site. All operational plantation sites were selected based on their age, treatment history, and uniform survival to give plots of the dimensions noted below. All stands were growing on Spodosols that were underlain by an argillic horizon. Soils were classified as sandy, siliceous, thermic Ultic Alaquods and Oxyaquic Alorthods (Table 1). Mean annual precipitation ranged from 1250 to 1275 mm, while mean annual temperature averaged about 19.5°C. During the 2-yr study period, droughty conditions persisted and annual precipitation levels in 1999 and 2000 were approximately 8 and 33% below the long-term recorded averages, respectively. Sampling occurred in the early fall of 1999 and 2000.


View this table:
[in this window]
[in a new window]
  		Table 1. Characteristics of the experimental sites established in fast-growing loblolly pine plantations. All stands were planted at initial stand densities of 1495 stems ha–1.

 
Pre- and postestablishment operations were similar for all the sites used in this study, and followed the standard operating procedures of the company that owned the land. Before planting, sites were chopped, burned, disked, and bedded. Genetically improved 1-yr-old loblolly pine seedlings of the same family were planted at a 1.8 by 3.7 m spacing (1495 trees per ha). Understory plant competition was controlled during the first two growing seasons using a combination of broadcast and directed spray applications of herbicides applied at labeled rates (e.g., hexazinone, sulfometuron, imazypyr, glyphosate). During the first growing season, all sites were fertilized with N and P applied at elemental rates of 30 and 33 kg ha–1, respectively. At the beginning of the third growing season, additional applications of N, P, K and B were applied at elemental rates approximating 90, 25, 40, and 1.5 kg ha–1, respectively. Fertilizer was broadcast by helicopter. At each site three replicate areas of 261 m2 (14.5 by 18 m) were established to collect tree inventory data (Adegbidi et al., 2002), root and soil samples, resulting in three replicate subplots at each site or nine plots for each age class. The exception was that for age 4, where four replicate sites with three subplots per site were sampled to yield 12 plots.

Soil pits were excavated with a backhoe and used to estimate tap root mass and coarse root and fine root content. Trees were preselected to represent the range in tree size encountered within the stand and were used to develop biomass equations by age class (Adegbidi et al., 2002), and soil pits were located adjacent to three felled trees at each site (i.e., one tree per replicate subplot per site), resulting in nine soil pits per age class. Soil pits (0.7 m wide, 1.9 m long, 1.0 m deep) were begun adjacent (20 cm from base of tree) to the trees and in the middle of the planting row. The 1.9-m dimension of the pit was oriented perpendicular to the planting row and extended to approximately half the distance between planting rows. The 0.7-m dimension was in the direction of the next tree in the planting row.

As soil was removed by the backhoe, it was searched by hand and with rakes for coarse roots (>2 mm diam.). These roots were collected, transported to a nearby facility, washed, and weighed the same day that they were sampled. Coarse root subsamples were transported back to the laboratory, oven-dried at 70°C, and weighed to determine moisture contents. Moisture contents then were used to calculate coarse root dry weights. Estimates of coarse root biomass (Mg ha–1) were based on the calculated dry weights and the surface area dimensions of the soil pit.

To measure taproot biomass, after the fine root measurement was finished, the soil was dug away from the taproot, a chain was wrapped around the taproot and coarse roots were severed 20 cm from where they exited the taproot. The other end of the chain then was attached to the bucket of a backhoe that lifted the taproot from the pit. The taproot was transported to a nearby facility, washed, and weighed. Several 2.5-cm thick disks then were cut from different parts of the taproot, weighed, and put in a plastic bag on ice until they were transported to the laboratory where they were dried at 70°C and weighed. A moisture content conversion factor was calculated and the taproot dry weight was calculated.

To estimate the fine-root distribution, the face of the trench nearest the felled tree was cleaned with a shovel. A 10 by 10 cm grid was outlined on that face. Within that grid, six columns (10 cm wide to a depth of 1 m), separated by 20-cm intervals, were outlined (n = 60, 100 cm2 cells per soil pit). The number of fine roots (i.e., <2 mm diam.) exiting the trench face of each cell was tallied similar to the profile wall method described in Böhm (1979) and Lopez-Zamora et al. (2002). To see the roots more clearly, a magnifying glass was employed. As each grid cell was measured, a sharp pencil was used to ‘rough up’ the surface, which assisted in root identification and counting. In addition, a water bottle with a fine mist spray was used to wet the surface, which highlighted roots for better identification.

Approximately seven cells, representative of the range in observed root count data, were selected for more detailed root sampling by pressing a 10 by 10 by 10-cm metal core into the soil face. The seven cells were selected as described by Lopez-Zamora et al. (2002). In short, the frequency distribution of roots per cell was plotted for each pit and samples representing the mean, one standard deviation and two standard deviations were selected. In selecting the samples an attempt was made to get a distribution of samples from all soil depths. The soil from the cells was removed and transported to a nearby workstation where the fine roots were gently washed and separated. All fine root samples were frozen until root length and biomass measurements could be made in the laboratory. At that time the roots were thawed, each sample was laid out on a white plastic background and covered with transparent glass, and the image was scanned using an Envisions ENV 8800S Scanner (Envisions, Burlington, CA). The resulting files were analyzed for root length via the software program GSROOT Version 5.2 (PP Systems, Inc., Haverhill, MA). After measuring root length, each sample was oven-dried at 70°C and its dry weight recorded. That way, root number density NX (number of roots cm–2), root length density Lv (cm cm–3), and root mass density M (g cm–3) were obtained for each core sample. For each site, which included three soil pits, a total of 21 fine root cores were processed. Regression relationships then were developed to estimate fine root length (Escamilla et al., 1991; Lopez-Zamora et al., 2002) and fine root biomass (Lopez-Zamora et al., 2002) from the root number exiting the trench face. The shoot and foliage data used in the root/shoot and root/foliage ratio evaluations came from previously published work by Adegbidi et al. (2002).

All statistical analyses for fine-root distribution with soil depth and time were performed using elementary (nonaveraged) NX (number of roots cm–2) data collected from the trench face. The analysis of depth distribution of fine roots was accomplished for each age using a completely randomized ANOVA design where soil depth and replicate site were the main effects. Distance from the tree was used as a covariate. The purpose of this analysis was to determine if the depth distributions among sites in an age class were similar. Newman-Keuls post-hoc comparisons were used to separate means of significant effects.

Fine-root distribution throughout the soil profile was visualized with a two-dimensional contour plot that used a least squares routine to calculate contours in time and space. Weighing the entire graph and then weighing the portion of the graph with roots at the higher densities determined the percentage of soil volume colonized by root densities >0.01 roots cm–2. An empirical model for root distribution in time and space was developed using a general stepwise regression approach, where NX (number of roots cm–2 trench face) was expressed as a function of plantation age, distance from the tree stem, and depth in the soil profile.

We were also interested in the average fine root density distributions with soil depth for each age class. An empirical regression model was fitted to these data for each age class, using all data for that age class:

[1]
where NX is as defined above, x is the soil depth in centimeters and a and b are constants. Then the constants a and b were regressed against age to provide a generalized regression equation of the form in Eq. [1] to include age.

Correlations between the variables NX, LV, and mass for fine roots and the prediction of LV from NX and M were developed using all data collected from soil samples removed from the trench for all age classes. The relationship of fine root LV as a function of NX and mass were developed with these data using a general stepwise regression. Length/mass ratios were calculated for all soil sample data above and regressed against the distance from the stump. All statistics were run using Statistica '99 Edition (StatSoft, Inc., 1999) and all tests of significance for all analyses in all equations, tables, and figures were based on P values < 0.05.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Fine Root Development over Time and Space
At the end of Year 1, root development occurred primarily in the surface 25 cm of soil. Below the 35-cm depth, root densities were low and statistically equivalent among replicate sites (Table 2, Fig. 1a) . At ages 2 and 3 yr, root densities were still similar at and below 25 cm, with differences in root density occurring only at the 5- and 15-cm depths (Table 2, Fig. 1b,c). By age 4 yr, however, there was a tendency for the replicate study sites to diverge in root development, even though the general logarithmic decline with depth remained a common trend among all sites.


View this table:
[in this window]
[in a new window]
  		Table 2. Separation of mean NX values by soil depth and site within each age class using a Newman-Keuls test. The ANOVA indicated that the site x soil depth interaction for Year 1 was not significant, so the mean separation was done by depth for the means of all sites. Years 2 to 4 indicated a significant depth by site interaction, so mean separations were calculated for soil depth x site means.

 


View larger version (17K):
[in this window]
[in a new window]
  		Fig. 1. Fine-root density distribution with soil depth for fast growing loblolly pine on the Georgia lower Coastal Plain for plantations aged (a) 1 yr, (b) 2 yr, (c) 3 yr, and (d) 4 yr. Statistical differences among replications and depths are in Table 2.

 
Maximum observed rooting depth was 85, 85, 85, and 95+ cm at ages 1, 2, 3, and 4 yr, respectively (Fig. 1). However, total soil volume containing roots progressively increased with age. Using the depth and lateral distribution data illustrated in Fig. 2 , the proportion of the soil trench face occupied by fine roots at a density >0.01 roots cm–2 was 20, 31, 36, and 61% of the trench face area at ages 1, 2, 3, and 4 yr, respectively. At the end of the first year, fine roots had not yet occupied the entire surface-soil volume as they did in Years 2 to 4 (Fig. 2). As expected, fine-root density increased in the surface soil with increasing age and a root-density growing front with depth developed over time (Fig. 2).



View larger version (44K):
[in this window]
[in a new window]
  		Fig. 2. Two-dimensional contour plots of loblolly pine root distributions during the first 4 yr of development. The contours are NX (number of roots cm–2) plotted concurrently by horizontal distance from the stem and soil depth. All root samples were taken in the early fall.

 
One empirical modeling approach for fine-root development is to build an equation that predicts root density for any soil compartment in the profile. The result of this approach is provided below:

[2]
where NX is as described above, y is distance from the tree stem in centimeters, age is plantation age in years, and x is soil depth in centimeters.

Another empirical approach is to predict average root density at any soil depth to describe an average depth distribution profile. Regressions of NX versus the natural logarithm of soil depth were fit to the average root densities for each depth of each replication of each age group. In addition, all data for each age were fit to the same curve form (Table 3). The natural logarithm of soil depth adequately described the root distribution pattern in a majority of cases. When all sites within an age class were combined, R2 values ranged from 0.88 to 0.96, again showing a good fit to the natural logarithm of soil depth within each age group. Fast-growing young loblolly pine average fine-root depth distribution, and fine-root depth distributions at any distance from the tree stem, fit the generalized curve of Gerwitz and Page (1974). Their work showed that over 70% of documented plant-root depth distributions were best described by a logarithmic decrease in root amount with soil depth. In the absence of a mechanistic approach to predict root amount in a soil profile, this style of empirical equation is useful for capturing the spatial variability of root development under specific management and soil conditions. When these data also include repeated measurements in time, as are presented here, it is possible to regress the coefficients of these equations against the time variable and to derive an empirical model in both space and time.


View this table:
[in this window]
[in a new window]
  		Table 3. Regressions of root density with soil depth using the average density at each soil depth for all replicate sites at each age. The curve form is NX = aLn(x) + b, where NX is number of roots cm–2 of trench face, x is the soil depth in centimeters, and a and b are constants.

 
A simple linear relationship described both coefficients between ages 1 and 3 (R2 of 0.78 for coefficient a and 0.73 for coefficient b, Fig. 3) .



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 3. Regression of the a and b coefficients in the curve NX = aln(x) + b against plantation age. The equation describes loblolly pine root distribution with depth, using coefficients representing the fine-root distribution at ages 1 to 3.

 

At age 4 only one of the three replicate sites fit the pattern described by the 1- to 3-yr data. The other two sites were distinctly different; therefore, 4-yr-old old sites were not included in the following analysis. It is unknown why the depth patterns diverged among replicate sites at age 4, which is near crown closure (Adegbidi et al., 2002). One will note from Table 2 that variability among sites of similar age appears to be increasing. At Year 1 there was no site x depth interaction, while there was at ages 2 to 4. The mean separations at ages 3 and 4 were more complex than those at the earlier ages. The overall pattern is the same, but the surface root development diverges.

An empirical model for root density was developed by substituting the equations for coefficients a and b (shown above) on tree age into Eq. [1].


[3]
where NX is number of roots cm–2 of soil, age is the age of the plantation (early fall) in years, and x is the soil depth in centimeters. The variable age ranges from 1 to 3 yr and x ranges from 5 to 95 cm. When NX predicted from this equation was plotted against the NX values used to develop the equation, the resulting regression had a slope of 1.0, an intercept of 0.00, and an R2 = 0.80. The above equation appears to be a good fit of the data without any apparent bias and does an adequate job of describing early root development of fast-growing loblolly pine on these Coastal Plain Spodosols; however, this equation's utility under other growth conditions will require independent testing.

Nutrient management of forest soils is the process of supplying the proper nutrients in the amounts and forms necessary to maintain a desired level of growth. Mechanistic nutrient uptake modeling is a nutrient management tool that has the potential to make fertilizer prescriptions specific for both sites and stages of stand development. However, this approach requires knowledge of the root-development pattern since nutrient bioavailability is partially defined by the plant root system (Comerford, 1998). Mechanistic modeling of root development should be a preferred approach for defining root development over time and soil depth. In the absence of such models, however, the data and equations developed via the present study could be used in uptake models applied to loblolly pine growing on Coastal Plain Spodosols.

Relationship between Fine Root NX, LV, and Mass
There was a weak correlation between NX and LV (R2 = 0.38) and NX and M (R2 = 0.24), with neither stand age, distance from the stem nor soil depth significantly improving these simple correlations. However, the prediction of LV from NX was improved when both M and soil depth were incorporated in the model. The resulting equation was:

[4]
where LV is root length density measured in cm root cm–3 soil, NX is as defined above, M is root mass in g root cm–3, and x is soil depth in centimeters. Mass appears to be an important covariate because the length/mass ratio of fine roots was a function of distance from the tree (Fig. 4) . Fine roots within 50 cm of the stem had as much as 180% more root length per gram of root than those located further away. Assuming the same specific density of fine roots, then roots closer to the stem exhibited smaller diameters.



View larger version (13K):
[in this window]
[in a new window]
  		Fig. 4. Relationship between the length/mass ratio of fine roots for fast-growing loblolly pine as a function of distance from the tree stump. Data are from rapid developing loblolly pine plantations between the ages of 1 to 4 yr, growing on Coastal Plain Spodosols. Each point is the mean of all soil samples at all depths for that distance from the stem.

 
The root length density and NX are expected to exhibit a significant interrelationship (Melhuish and Lang, 1968). The low correlation between these variables was surprising, but the changing LV/M ratio with distance and the significance of soil depth in the above equation would seem to explain why the simple correlation was not strong. It is notable that the coefficient of NX in that equation was 1.46 ± 0.23 (95% confidence interval). The high end of the coefficient's confidence interval (1.69) approaches 2, which is the theoretical value suggested by Melhuish and Lang (1968). On similar soils Escamilla et al. (1991) tested the same relationship and calculated a value of 1.0 for 7-yr-old slash pine (Pinus elliottii Engelm.) roots in the surface soil. Most recently Lopez-Zamora et al. (2002) tested this relationship for Melaleuca quinquenervia (Cav.) S.T. Blake. Their coefficients for NX were not close to 2, though they did show a significant relationship between NX and LV for a Spodosol in south Florida, and showed that a credible relationship between NX and LV is still useful in defining a more efficient root-measurement procedure. Measuring NX is much less time consuming than sampling for and measuring LV (Vepraskas and Hoyt, 1988), and trends in NX are similar to those of LV and M (Escamilla et al., 1991). Thus, measuring NX and relating it to LV through subsampling, as described here and by Lopez-Zamora et al. (2002), should promote faster less expensive root studies. This should in turn make available more root data for important commercial forest species.

Ratios of Fine-Root Mass/Foliage Mass and Root Mass/Shoot Mass
A significant body of literature on aboveground loblolly pine growth and development is available. If well-documented relationships could be developed between above- and belowground tree components, they might prove useful for inferring belowground development as well. That is the objective of the following analysis.

After the first year of root development, there was approximately an equal amount of fine-root biomass to foliage biomass. Beyond Year 1, that ratio was significantly reduced, and remained stable between Years 2 and 4 (Fig. 5) . The fine root/foliage mass ratio stabilized in the range 0.13 to 0.18, with no significant differences found among years. Note that these ratios within any given age group also had a narrow confidence interval. The observed ratios are similar to ratios obtained by Albaugh et al. (1998) in fertilized (0.12) or fertilized and irrigated (0.13) trees of age 12 yr that were in a stage likened to full site occupancy (above and belowground) by the authors. Adegbidi et al. (2002) have shown that leaf area index (LAI) peaked at age 3 yr in the plantations of the current study. Such observations and the consistency of the ratio of fine root/foliage suggest that fine root might also be near full site occupancy. Such relationships need careful testing on a range of site, ages, and management conditions before they can become useful management tools, but these data point to the potential utility.



View larger version (10K):
[in this window]
[in a new window]
  		Fig. 5. Ratios between mass of fine roots and foliage biomass for ages 1 to 4 yr of rapidly developing loblolly pine growing on Coastal Plain Spodosols in southeastern Georgia.

 
Root/shoot ratios exhibited a similar trend. At the end of Year 1 the ratio was approximately 1, and then remained stable for the next 3 yr in the range 0.30 and 0.39 with no statistical difference among Years 2 to 4 (data not shown). These values are in the same range as those seen in the few loblolly pine studies that documented both aboveground and belowground biomass (Table 4). One would expect root/shoot ratios to be sensitive to soil parameters such as N and P bioavailability. However, it remains to be seen if the trends observed in seedling, pot, and greenhouse studies (Kim et al., 2002; Sigurdsson et al., 2001; Brunner and Brodbeck, 2001; Graham, 2001; Hawkins et al., 2000) can be convincingly replicated under fast-growing field conditions. A basic question is whether the change that may occur is detectable given the inherent variability of field measurements for rooting. Nonetheless, the consistency of the root/shoot ratio across studies (Table 4) suggests that after the early stage of plantation development, biomass partitioning between above- and belowground in loblolly pine might be under a strong influence of ontogenetic factors rather than site factors. However, these observations should be considered preliminary and must be tested on other sites under a range of environmental conditions.


View this table:
[in this window]
[in a new window]
  		Table 4. Live standing biomass (Mg ha–1) and root/shoot biomass ratios for loblolly pine plantations.

 
By the end of the first year the entire root system measured approximately 1.2 Mg ha–1 and had increased to 14.0 Mg ha–1 by age 4 yr (Fig. 6) . During the first year, fine roots represented the highest proportion of belowground biomass (60%), whereas starting at age 2 taproots became the dominant component (accounting for 50% of the belowground biomass or greater). Fine-root biomass was 0.7 to 0.8 Mg ha–1 during Years 1 to 3 and increased to 1.2 Mg ha–1 at Year 4. These data for root biomass are similar to those of Albaugh et al. (1998) for fertilized 12-yr-old loblolly pine, who found 12.4 to 13.5 Mg ha–1 for total belowground biomass, including 0.7 to 0.9 Mg ha–1 in fine roots. Site characteristics and management has stimulated the growth of this stand so that developmentally it is similar to older stands (this study and Adegbidi et al., 2002) in both total root biomass and root/shoot ratios.



View larger version (20K):
[in this window]
[in a new window]
  		Fig. 6. Root system and root-system-component biomass accumulation during the first 4 yr of development. The percentages represent the portion of total belowground biomass that each component constitutes. Confidence intervals (95%) are presented for each estimate.

 
In summary, root distribution patterns during the establishment and early developmental stages of growth were documented and empirical models were developed which should prove useful for the continued development of nutrient uptake models. It was shown that roots, by age 4 yr, have permeated over 60% of the soil, and the pattern of root occupancy was demonstrated. A meaningful relationship between LV and NX was investigated and the use of NX as a root measurement parameter was outlined. Lastly, it was shown that fine-root mass was related to foliage mass and that the relationship was stable after the establishment phase, as was the ratio of root/shoot biomass. These results add to our knowledge of loblolly pine root development, which has been ignored in the majority of studies investigating pine growth and development. While many of the data are observational and empirical, they have given a useful picture of loblolly pine root development that to date has been lacking.


   ACKNOWLEDGMENTS
 
This research was supported by a grant from the U.S. Department of Energy (contract No. DE-FC36-99GO10415; AF&PA Agenda 2020 program). Special thanks are extended to International Paper Company for providing experimental sites and fieldwork logistics. We are also grateful to B. Garbett, R. Scott, J. Sullivan, J. English, A. Coveney, and M. McLeod for technical assistance. Journal Series Paper No. R-09670 of the Florida Agriculture Experiment Station.

Received for publication January 7, 2003.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
S. G. Haile, P. K. R. Nair, and V. D. Nair
Carbon Storage of Different Soil-Size Fractions in Florida Silvopastoral Systems
J. Environ. Qual., August 8, 2008; 37(5): 1789 - 1797.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Adegbidi, H. G.
Right arrow Articles by Barros, N. F.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Adegbidi, H. G.
Right arrow Articles by Barros, N. F.
Agricola
Right arrow Articles by Adegbidi, H. G.
Right arrow Articles by Barros, N. F.
Related Collections
Right arrow Ecosystem Management
Right arrow Plant and Soil Interactions
Right arrow Forest Soils

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome