| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Direction de la Recherche Forestière, Forêt Québec, min. des Ressources Naturelles du Québec, 2700 rue Einstein, Sainte-Foy, QC, Canada G1P 3W8
b Université du Québec, INRS-Géoressources, 880 ch. Ste-Foy, bur. 840, C.P. 7500. QC, Canada G1V 4C7
* Corresponding author (Daniel.Houle{at}mrn.gouv.qc.ca)
Received for publication November 7, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Abbreviations: CECe, cationic exchangeable capacity • OM, organic matter • EA, exchangeable acidity • BSe, base saturation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
However, there are many concerns in the use of dendrogeochemistry for environmental monitoring and the historical reconstruction of soil chemistry (for a review, see Watmough, 1997). Changes in sapwood chemistry, which could occur many years after wood formation, is the most prevalent problem reported in the literature (Lepp, 1975; Barnes et al., 1976; Symeonides, 1979; Baes and Ragsdale, 1981; Baes, 1985; McClenahen et al., 1989; Smith and Shortle, 1996; Watmough, 1997). Although the term radial translocation is widely used to describe the change in concentration of a given element within growth rings formed in past years, most studies did not demonstrate the existence of a lateral movement from the bark to the heartwood for most elements. Sapwood by definition is potentially still conducting sap, and tissue formed before now could simply be directly affected by vertical flow across the entire sapwood band without radial translocation. In consequence, we will use the term lateral re-equilibration throughout the manuscript instead of radial translocation when appropriate.
It has been demonstrated for many tree species that 137Cs, which was introduced in the environment by explosions of nuclear weapons in the beginning of the 1960s, could be found in tree rings formed well before that time (Brownridge, 1984). However, Cs appears to be one of the most mobile elements. Element mobility in xylem is known to be affected, among others, by ion solubility, sap pH, and bonding in the xylem matrix (Cutter and Guyette, 1993). For example, McClenahen et al. (1989) reported evidence that lateral re-equilibration in the sapwood of Al, Si, Fe, and Ca was less pronounced than for most other elements in tuliptree (Liriodendron tulipifera L.). In a review of the dendrogeochemical approach, Watmough (1997) suggested that lateral re-equilibration of elements is more associated with coniferous rather than with deciduous trees.
Another important prerequisite in dendrogeochemical studies is a good comprehension of the variation in sapwood cation binding properties within a given tree. Momoshima and Bondietti (1990) showed that in red spruce (Picea rubens Sarg.), cation binding capacities decreased from the tree center to the bark. This process may create confusing dendrochemical profiles of basic cation concentrations that may be misinterpreted as a reduction of Ca or Mg soil availability through time. Nevertheless, Bondietti et al. (1990) suggested that anomalies in Ca and Mg concentrations in the stemwood of red spruce—which formed in the mid-1900s in New England, Tennessee, and North Carolina—reflected regional cation mobilization in the rooting zone, caused by rapid increases in acidic deposition of S and N. This situation may occur in the beginning of the soil acidification process when H+ causes basic cation leaching from soil exchange sites, thereby temporarily increasing their concentrations in soil solutions. In the medium-term, however, acid deposition is known to increase the leaching losses of Ca, Mg, and K (MacDonald et al., 1992; Likens et al., 1996; Houle et al., 1997; Markewitz et al., 1998; Lawrence and Huntington, 1999) from the rooting zone and to be responsible for the mobilization of inorganic Al and Fe (Robarge and Johnson, 1992). It has been demonstrated that varying soil acidity conditions, and consequently basic cation and Al availability, can influence the concentrations of divalent cations, such as Ca, Mg, and Mn in xylem tissue of various tree species (Frelich et al., 1989; Ohmann and Grigal, 1990; DeWalle et al., 1991; Guyette et al., 1992; Kashuba-Hockenberry and DeWalle, 1994; Côté and Camiré, 1995; Mohamed et al., 1997), emphasizing the possibility that dendrogeochemistry may be used as a proxy tool to document historical changes in soil chemistry in relation to acidic deposition.
Sapwood chemistry also changed appreciably for many tree species after watershed-scale soil acidification [addition of (NH4)2SO4] at two forest sites in West Virginia and one in Maine (DeWalle et al., 1999). Many conclusions could be drawn using this type of experimental approach because the exact timing and intensity of the causative agent is well documented. DeWalle et al. (1999) reported that sapwood ratios such as Ca/Mn and Mg/Mn were the best indices of soil acidification. However, the same authors concluded that tree sampling across the entire treated watershed was not an ideal experimental approach and suggested the use of random fertilization of soil around single trees from the same stand, to minimize the effect of soil spatial variability.
Such an experimental design was previously used in a liming experiment within a sugar maple (Acer saccharum Marsh.) stand at the Lake Clair Watershed (LCW) in Québec. In the LCW, liming of forest soils resulted in a strong improvement of nutrient status, vigor, and growth of the treated trees (Moore et al., 2000). To determine soil and tree xylem chemistry response to liming 4 yr after the application, 15 experimental units were randomly chosen and sampled from five liming treatments (0, 0.5, 2, 5, and 20 Mg ha-1) of this experiment. The specific objectives of the present study were: (i) to determine the effects of liming on soil chemistry; (ii) to determine whether the effect of liming on annual growth ring chemistry could be chronologically detected; and (iii) to assess the relationships between tree ring chemistry and soil chemistry.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Sampling
In August 1994, dolomitic lime was applied on forest soils from an experimental section located at the border of the LCW. This soil liming experiment was established to determine the nutrient, vigor, and growth response of sugar maple to amendment (Moore et al., 2000). Briefly, the investigators selected 98 sugar maples that were randomly distributed among eight CaMg(CO3)2 treatments (0, 0.5, 1, 2, 5, 10, 20, and 50 Mg ha-1). Dolomitic lime was applied evenly within a 5-m radius around each tree. For the purpose of the present study, three trees were randomly chosen from five treatments (0, 0.5, 2, 5, and 20 Mg ha-1) for a total of 15 trees. The trees initially selected to be included in the experimental design have comparable crown canopy and DBH (31 cm ± 2.7, 95% confidence interval). Care was taken to select trees with no major trunk defects or strong dieback rate, and selected trees had to be at least 15 m apart. In February 1999, a tree-ring (10 mm diam) core was extracted at breast height of each tree, placed in a sterilized plastic bag, and frozen upon arrival at the laboratory. All core handling in the field and laboratory was done using unpowdered surgical gloves.
In August 1999, soils were sampled 2.5 m from the center of each tree at each of the four cardinal points. The organic horizon was sampled quantitatively by extracting a known surface area (17.3 cm2) down to the mineral soil. The first 15 cm of the B horizon was sampled nonquantitatively. The samples were pooled for each tree, air dried, weighed, and sieved to 2 mm before chemical analysis.
Laboratory Procedures
Soil pH was measured with water using a soil/solution ratio of 1:2.5 (wt:wt). Organic matter content of forest floor samples was determined by loss-on-ignition (organic C = 0.58 x OM; Nelson and Sommers, 1982), and the organic C content of ground mineral samples (0.5 mm) by wet oxidation (Yeomans and Bremner, 1988). Exchangeable cations (K, Ca, Mg, Mn, Al, and Fe) were extracted with an unbuffered NH4Cl (1 M, 12 h) solution, and measured using inductively coupled plasma emission spectrophotometry (ICP). Exchangeable acidity (EA) was determined by summing the H+ (measured by pH probe), Al, Mn, and Fe concentrations of the extract. Effective cation exchange capacity (CECe) was computed as the sum of base cations and EA. Finally, effective base saturation (BSe) was estimated by the ratio of total basic cations on CECe.
Each tree-ring core was carefully examined under magnification with a binocular and divided into annual growth increments using stainless steel chisels and knives. The divisions were in accordance with the previously established master chronology of sugar maple on this site (Duchesne et al., 2002). The growth ring samples were dried in a dessicator at 30°C, weighed, and dissolved in closed Savilex Teflon bombs using 5 mL of ultrapure concentrated HNO3 for 24 h on a hot plate. Following digestion, the solutions were brought to 50 mL volume with deionized water and analyzed for K, Ca, Mn, and Mg by ICP. Aluminum and Fe concentrations were too low for reliable measurement.
Statistical Analyses
One soil replicate showing outlier values for many elements was rejected before statistical analysis. Levene's test for homogeneity of variance was performed on soil data set (Zar, 1974). The null hypothesis of homogeneity was accepted for all elements. Analysis of variance was performed on each element separately. Orthogonal polynomial contrasts were evaluated for the linear and quadratic effects of lime treatments on elemental concentrations of organic and mineral layers (Mize and Schultz, 1985). Calculation of the linear and quadratic coefficients for unequally spaced time intervals was made according to Robson (1959). Adjusted treatment means were calculated considering the unbalanced design.
Time trends of element concentrations in growth rings were investigated by a mixed ANOVA model, in which lime level and ring formation year were the fixed effects and tree was a random effect. Variance and covariance over time were modeled with an autoregressive structure. Linear and quadratic effects were computed. Correlation analysis was used to test relationships between the xylem and the soil chemical composition. After observation of statistical departure from normality by standard kurtosis test, average soil and bulk xylem data were log-transformed before analysis to ensure normality. All statistical analyses were performed using SAS version 8.01 (SAS Inst., 1999).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
|
The impact of liming on the forest floor and mineral B horizon was particularly detectable on the following soil indices that are deemed to be closely related to maple foliar nutrient status (Ouimet and Camiré, 1995): Mg saturation on soil exchange sites, and the exchangeable Ca/EA, Ca/Mg, and K/Mg ratios. For both soil layers, Mg saturation and the Ca/EA ratio increased with liming rate, while the Ca/Mg and K/Mg ratio decreased (p 
0.008).
Tree-Ring Cations
No significant temporal concentration trends were generally observed for the analyzed elements in the tree rings (Fig. 1)
. Only K showed a quadratic temporal trend without consideration to lime treatment (p 0.001). However, the lime treatments significantly changed the bulk xylem Mg, and Mn concentrations as well as the Mg/Mn and Ca/Mn ratio (Fig. 2)
. Quadratic effects were detected for Mg (p = 0.002) and Mg/Mn ratio (p = 0.008) while linear effects were detected for Mn (p = 0.013) and Ca/Mn ratio (p 0.001).
|
|
|
-0.68), and positively correlated to Mg, BSe, Ca/Al, and Mg/Al. Again the signs of the relationships were reversed with the Mg/Mn and Ca/Mn ratios. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
However, dolomitic lime applications have caused imbalances between K and Mg in the soil, inducing foliar K deficiencies (Moore et al., 2000). Soil Mg saturation increased, while the Ca/Mg and K/Mg ratios in both soil horizons decreased with liming rate, passing below (or above in the case of Mg/CECe) the norms (Mg/CECe 10%, Ca/Mg 
4, and K/Mg 1) established by Ouimet and Camiré (1995). Liming improved the value of the Ca/EA ratio in the top 15 cm B horizon, but only the 20 Mg ha-1 liming rate approached the threshold norm (Ca/EA ratio 1; Ouimet and Camiré, 1995). It is interesting to note that the best growth response was obtained at this liming rate (Moore et al., 2000). In addition, the very low CA/EA ratio (0.08) in the control mineral soils is in agreement with the Ca deficiency observed for the control trees (Moore et al., 2000). These results support the effectiveness of these soil indices as indicators of sugar maple nutrient status.
The increase in OM concentration and C/N ratio in the top 15 cm mineral B horizon at lower liming rates is somewhat surprising. The OM increase from 147.6 to 223.7 g kg-1 observed for the 0.5 Mg ha-1 treatment would represent 11400 g OM m-2 for a 15-cm soil layer. No trace of dwelling earthworms (Lumbricus terrestris), which could have incorporated OM from the surface layer into the mineral soil, have been found at the sites. Increased dissolved organic carbon (DOC) leaching from the forest floor to the upper mineral soil in response to liming, as reported by Persson et al. (1995) and Andersson et al. (1999), may explain <10% of the 76 g kg-1 OM augmentation [leachate flux is about 1040 mm at LCW (Houle et al., 2002), and its average DOC concentration is 10 mg L-1 (unpublished data, 2002)]. Another explanation would be the increased fine root production that is usually observed following improvement of soil fertility (Adams and Hutchinson, 1992; Bakker and Nys, 1999). Assuming a fine root production of 230 g m-2 in the first 15 cm of mineral soil and a turnover rate of 0.9 yr-1 (Burke and Raynal 1994), this would yield an increase in OM concentration of about only 7 g kg-1 over 5 yr. Indeed, we could not find any explanation for this OM increase, and further observations are needed to document the OM change in the mineral layer following the lower lime rate treatments. The lower OM concentration and C/N ratio found in the top 15 cm B horizon with the 20 Mg ha-1 liming rate suggests this treatment increased biological activity and organic matter decomposition as also evidenced by the forest floor OM content which, at this liming rate, tended to be reduced after 5 yr.
The significant Mn increase in the forest floor and the increasing trend observed in the top 15 cm B horizon (p = 0.083) with increasing liming rate was also surprising. This result may be explained by two mechanisms. First, soil exchangeable Mn concentrations may have increased because of the increased forest floor CECe, although Mn solubility in this soil layer may have decreased because of pH change. Second, soil Mn concentrations may have increased because of biological excretion by the aboveground vegetation (see below). This second mechanism is likely the one that can explain the observations on soil and tree wood Mn concentrations.
Tree-Ring Response
With the exception of the significant K enrichment close to the bark (not related to lime treatment), which is consistent with the mobility of this ion in plant tissues and its association with active metabolism (McClenahen et al., 1989), no other temporal trends of xylem basic cation concentrations or ratios were observed during the 1978–1998 period. Therefore, there was no clear temporal response to liming, although the analysis showed a strong influence of liming on the average xylem concentrations of Mg and Mn. Results for Mg and Mn in the present study are in good agreement with findings of DeWalle et al. (1991) and Guyette et al. (1992), who reported that sapwood Mg concentrations were higher and Mn lower in less-acidic compared with acidic soils for several tree species. Ratios of Ca and Mg to Mn also significantly increased with the liming rate (Table 2).
The lack of significant time trend in the concentrations of any elements with respect to lime treatments, in combination with significant changes in average xylem concentrations (Mg, Mn, Ca/Mn, and Mg/Mn), suggest that a lateral reequilibration of elements within the sapwood has occurred. Since only the last 20 growth rings (20 yr) were sampled, it is difficult to conclude about the extent of the cation reequilibration backward. However, it is known that sugar maple sapwood can extend through more than 30 annual growth rings for trees older than 40 yr (Raulier, 1997). These findings preclude the use of sugar maple trees to date historical soil chemical changes, at least by using total wood Ca or Mg concentrations or even ratios of the latter to Mn. These results are also in good agreement with the findings of Cutter and Guyette (1993), who classified various tree species for dendrochemistry mainly based on heartwood and sapwood characteristics. They concluded that sugar maple is not suitable for general use in dendrochemistry because of a large number of rings in the sapwood relative to the total number of rings, a high potential for reequilibration by solution flow or diffusion, and very permeable wood. DeWalle et al. (1999) also observed changes in wood chemistry back to more than 20 yr following a (NH4)2SO4 treatment using red maple (Acer rubrum L.), yellow poplar (Liriodendron tulipifera L.), yellow birch, red spruce, sugar maple, and American beech. However, the same authors concluded that Japanese larch [Larix kaempferi (Lamb.) Carr.] was a good candidate in dating past soil changes because of the very low lateral reequilibration observed for most analyzed elements. Similar low reequilibration potentials have been reported for Japanese cedar [Cryptomeria japonica (L.f.) D. Don] (Momoshima, 1995) and eastern red cedar (Juniperus virginiana L.) (Guyette et al., 1992). Despite the high reequilibration of most studied elements in sugar maple, preliminary data suggested that Al concentration in growth rings of sugar maple could be a good indicator of past soil acidification (Laflèche et al., unpublished data, 2002) owing to the low lateral reequilibration of this element in sapwood. Mohamed et al. (1997) provided evidence that xylem Al content in sugar maple was related to Al availability in acidifying soils, also suggesting that xylem Al concentrations may be used as a proxy environmental monitoring tool.
Soil–Tree Rings Relationship
The significant changes in Mg, Mn, and in the Ca/Mn and Mg/Mn ratios in the xylem of the maple trees after treatments were likely produced by major soil chemistry changes. However, the observed changes were not necessarily expected since the xylem concentration of any given element was not significantly related to the amount of the same element in the humus or mineral layers of the soils (Table 2). Clearly, xylem Ca/Mn and Mg/Mn ratios were the best predictors of the acid–base status in humus (and to a lesser extent for the top 15 cm B horizon). For example, the Mg/Mn ratio in xylem explained up to 85% of the humus CECe variability. Other strong correlations were also observed with soil Fe, pH, Ca, Mg, and even with the Ca/Al and Mg/Al ratios. These results support the conclusion of DeWalle et al. (1999) that bulk sapwood Mg/Mn and Ca/Mn ratios were sensitive indicators of the tree response to experimental watershed acidification for many species. Because Al and Fe concentrations were not detectable in xylem, we cannot compare the usefulness of the Ca/Mn and Mg/Mn ratios with ratios using Al. However, other studies reported that ratios using xylem Mn were better indices of soil acidification than ratios using Al (DeWalle et al., 1991; DeWalle et al., 1999).
Surprisingly, Mn concentrations in xylem were reduced four-fold in the 20 Mg ha-1 treatment relative to the control, even though exchangeable Mn increased slightly in limed soils. A Mn impoverishment in the wood of scarlet oak (Quercus coccinea Muenchh.) after liming was also observed by Kashuba-Hockenberry and DeWalle (1994), but no information on soil was available. In another liming study, Long et al. (1997) showed that soil Mn concentrations decreased in the upper soil horizons (0–5 cm) 3 yr after liming (22.4 Mg ha-1). Moore et al. (2000) reported significant foliar Mn decreases caused by the lime treatments with the same experimental design as our study. This suggests that Mn bioavailability was reduced following liming despite the increase in soil exchangeable Mn. The suspected reduction in bioavailability was probably caused by soil pH increases observed in the limed humus, which should likely affect Mn solubility (Guyette et al., 1992). Bohn et al. (1979) mentioned that reduced Mn solubility and availability is common in acidic soils after liming. Interestingly, the reduction in the xylem Mn concentration observed in limed trees implies the existence of a mechanism capable of expulsing Mn out of the tree. This situation is in conformity with the finding of Momoshima and Bondietti (1990) for red spruce, who showed that the sapwood can act as a chromatographic exchange column. In this view, previously adsorbed Mn within the wood complex probably desorbed when the solution passing through the column was impoverished in Mn, leading to a new equilibrium for Mn between the adsorbed and liquid phase. The desorbed Mn likely circulated from the bottom to the top of the tree and it is suggested that it was probably expulsed through foliar exudation.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Bilodeau Gauthier, D. Houle, C. Gagnon, B. Cote, and C. Messier Extractability of Elements in Sugar Maple Xylem along a Gradient of Soil Acidity J. Environ. Qual., May 1, 2008; 37(3): 871 - 879. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Vadose Zone Journal | Journal of Plant Registrations | ||||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal |
