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 (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wang, J. J.
Right arrow Articles by Bell, P. F.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Wang, J. J.
Right arrow Articles by Bell, P. F.
Agricola
Right arrow Articles by Wang, J. J.
Right arrow Articles by Bell, P. F.
Related Collections
Right arrow Sorption/Exchange
Right arrow Nutrient Management
Right arrow Soil Chemistry
Published in Soil Sci. Soc. Am. J. 68:654-661 (2004).
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

DIVISION S-8—NUTRIENT MANAGEMENT & SOIL & PLANT ANALYSIS

Potassium Buffering Characteristics of Three Soils Low in Exchangeable Potassium

Jim J. Wang*, Dustin L. Harrell and Paul F. Bell

Agronomy Dep., 313 Sturgis Hall, Louisiana State Univ. Agricultural Center, Baton Rouge, LA. 70803

* Corresponding author (jjwang{at}agctr.lsu.edu).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
Plant availability of soil K is controlled by dynamic interactions among its different pools. Misunderstanding of these dynamics leads to mismanagement of soil fertility. This study was conducted to evaluate buffering characteristics of low exchangeable-K soils that showed different sugarcane (Saccharum spp. L.) responses to K amendment. Three silt loams, Crowley (fine, smectitic, hyperthermic Typic Albaqualfs), Dundee (fine-silty, mixed, active, thermic Typic Endoaqualfs), and Norwood (fine-silty, mixed, superactive, hyperthermic Fluventic Eutrudepts), were evaluated by using a modified quantity/intensity (Q/I) approach, which allows partitioning of K changes in the soil–solution system into exchangeable and nonexchangeable pools. Total potential buffering capacity (PBCt) was found to correlate significantly (r = 0.97, P < 0.01) with the buffering capacity due to nonexchangeable K (PBCn) rather with that due to exchangeable K (PBCe). Impact factor (ß), a measure of effect of added K on nonexchangeable K, was inversely correlated with a soil's conversion magnitude ({alpha}) of converting added K to exchangeable K (r = 0.95, P < 0.01). Of the three soils, Dundee exhibited much smaller ß values than Crowley and Norwood but the soil converted much of added K to exchangeable K (60–65%) throughout its surface and subsurface soils. Both Crowley and Norwood possessed higher PBCt as well as higher PBCn and PBCe than Dundee, but Crowley required relatively lower critical exchangeable K (EKr) and solution K (CKr) levels below which release of nonexchangeable K was initiated. The overall results indicated that the partitioned Q/I approach could be used to explicitly evaluate short-term K dynamics in soil–solution systems. The nonexchangeable K buffering characteristics along with the differences between critical levels (EKr or CKr and EK0 or CK0) have important implications in assessing the likelihood of nonexchangeable K release and could be calibrated for soil fertility management.

Abbreviations: AR, activity ratio • CEC, cation-exchange capacity • CKi, initial solution K • CK0, equilibrium solution K • CR, concentration ratio • CR0, equilibrium concentration ratio • EKo, equilibrium exchangeable K • EKr, critical exchangeable K • KG, Gapon selectivity coefficient • LPBC, linear potential buffering capacity • MHA, mobile humic acids • PBCe, potential buffering capacity due to exchangeable K • PBCn, potential buffering capacity due to nonexchangeable K • PBCt, total potential buffering capacity • Q/I, quantity/intensity • {alpha}, slope of linear regression between final exchangeable K (EKf) and change in solution K ({Delta}K) • {Phi}, initial constraint • ß, slope of linear regression between change in nonexchangeable K ({Delta}Non-Exch K) and initial constraint


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
AVAILABILITY OF K TO PLANTS depends on its intensity, capacity, and renewal rate in soils. Intensity is the K concentration in soil solution, capacity is the total amount of K in soil solids available to go into the solution, and renewal rate is a kinetic factor describing the K transfer rate from capacity to intensity (Barber, 1984). Most soil K tests focus on measurements of K intensity, using 1 M neutral ammonium acetate (NH4OAc) or similar extractants, to characterize solution and exchangeable K (Mehlich, 1984; McLean and Watson, 1985), although strong extractants such as sodium tetraphenylboron (NaBPh4) and nitric acid have been found to correlate even better with plant K uptake in some cases (Richards and Bates, 1988; Cox et al., 1999).

It is well known that when fertilizer K is applied to soil, some of the readily dissolved K can be fixed by soil clays and become nonexchangeable, thus nonavailable to immediate uptake by plants (Scott and Smith, 1987). It has been generally accepted that nonexchangeable K can contribute to plant growth, especially in soils that are low in soil test (predominately exchangeable) K (Bertsch and Thomas, 1985; Cox et al., 1999). Several attempts have been proposed to characterize the capacity of total available K by extracting a fraction of nonexchangeable K using strong extractions (Scott and Welch, 1961; Richards and Bates, 1988). The dynamics of this fraction of K are, however, not fully understood simply because not all nonexchangeable K will become plant available. Factors affecting K fixation into or release from the nonexchangeable pool include soil mineralogy, equilibrium among different pools of soil K, drying–wetting cycles, and the amount and nature soil organic matter (Rich, 1968; Olk and Cassman, 1995). For these reasons, recent work has suggested that exchangeable K, nonexchangeable K and clay mineralogy be considered together when making fertilizer recommendations (Ghosh and Singh, 2001).

Various attempts have been made to characterize the relationship between intensity and capacity of soil K or soil K buffering characteristics (Evangelou et al., 1994; Nair, 1996). The incorporation of soil buffer properties into a soil-testing program enhances accuracy of soil tests in predicting K availability (Nair et al., 1997). Soil cation-exchange capacity (CEC) has traditionally been used to index soil buffering characteristics (McLean, 1976; Vitosh et al., 1995). However, this practice has been questioned because indexing soil buffering ability based on CEC solely may not reflect soils that are dominated by K fixation and release from the nonexchangeable pools (Cox et al., 1999). Another way to characterize soil K buffering ability is to use linear potential buffering capacity (LPBC), the slope of the linear portion of a Q/I plot between the amount of K absorbed and the K activity ratio (AR) (Beckett, 1964). Soil K buffering has also been characterized by various ratios of exchangeable K to solution K (Mengel and Busch, 1982; Nair, 1996). These attempts, however, focused primarily on soil K buffering characteristics of the exchangeable pool.

Potassium fertilization to enhance plant-available K in soils is a common practice in sugarcane production. This is because, unlike some grain crops, sugarcane acquires K in a steadily increasing fashion during the growth process (Filho, 1985). This persistent K requirement suggests that soil buffering is critical to sugarcane growth, especially in soils that are low in K intensity. The objective of this study was to evaluate buffering properties of exchangeable and nonexchangeable pools of soil K in three selected cane-producing soils that were low in NH4OAc-extractable K levels, using a modified Q/I isotherm procedure.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
Soil Samples and Properties
Three soils that were low in NH4OAc-test K were chosen for this study. Soils were: a Gulf coastal prairie Crowley silt loam, a subtropical Mississippi valley alluvium Dundee silt loam, and a Red River valley alluvium Norwood silt loam. The Dundee soil was collected from a field that had been cropped to sugarcane for decades, while the Crowley and Norwood soils were collected from fields cropped to sugarcane for six continuous years. Soil samples were taken from depths of 0 to 15, 15 to 30, and 30 to 45 cm at each site. Soil samples were passed through a 2-mm sieve and air-dried. Selected properties of these soils are presented in Table 1. Soil pH was determined using a pH meter in deionized water with a 1:1 soil/solution ratio. Cation-exchange capacity was determined by repeated saturations using 1 M NH4OAc followed by washing, distillation, and titration (Soil Survey Staff, 1996). Organic matter was determined by using the Walkley and Black procedure (Nelson and Sommers, 1982) and particle size by pipette method (Gee and Bauder, 1986). Clay mineralogy was characterized by X-ray diffraction. In addition, nonexchangeable K, or slowly released K, was determined by the NaBPh4 method described by Cox et al. (1999).


View this table:
[in this window]
[in a new window]
  		Table 1. Physical and chemical properties of selected soils from long-term sugarcane (saccharin spp.) fields.{dagger}

 
Potassium Adsorption and Desorption Experiments
Potassium Q/I isotherms were conducted according to a procedure used by Beckett (1964) with some modifications described here. Soil samples of 2.5 g were placed in 30-mL centrifuge tubes (preweighed) containing 25 mL of 0.01 M CaCl2 solutions with KNO3 concentrations of 0.0, 0.5, 1.0, 2.0, 3.5, and 5.0 mM. To obtain enough data points on K release, 0.62 and 1.25 g soil samples were also mixed with 25 mL of 0.01 M CaCl2 solutions containing no KNO3. The prepared soil suspensions were shaken for 30 min, allowed to equilibrate for 18 h, and centrifuged. After collecting supernatant solutions for K, Ca, and Mg analyses, the weight of each centrifuge tube with the remaining soil was recorded. Then, 25 mL of NH4OAc was added to each centrifuge tube, the suspensions were shaken again for 30 min, and then centrifuged to collect supernatants for K analysis. Two replicates were used for collecting each data point, and concentrations in supernatant solutions were determined by inductively coupled plasma (ICP) analysis. All isotherm experiments were conducted at 295.2 ± 0.2°K.

The final exchangeable K (EKf) for each equilibrium point was calculated based on NH4OAC extraction at end of contact of CaCl2 with correction of interstitial solutions remaining in centrifuge tubes from weight differences. Other parameters were obtained as follows:

[1]
where {Delta}K is the change of K in solution, CKi and CKf are the initial (K concentration added) and final equilibrium concentrations of K in solution, whereas v and w are the solution volume and soil mass, respectively. Positive {Delta}K values indicate K adsorption by the soil solid phase whereas negative values indicate K release from the soil phase into solution.

Potassium concentration ratio (CR) was used to describe the intensity of K in the presence of Ca and Mg as

[2]
where Caf and Mgf are concentrations of Ca and Mg in final equilibrium solutions, respectively (Wang et al., 1988). Originally, the AR was used by Beckett (1964) to express K intensity. However, the 0.01 M CaCl2 background electrolyte solution commonly used in such an experiment dominates ionic strength of the soil solution and subsequently yields a constant activity coefficient ratio (Wang et al., 1988). This suggests that CR differ from AR by a constant factor. For this reason, CR was adopted instead of AR to express K intensity in this study.

Traditionally, change in solution K ({Delta}K) has been attributed to the change in the amount of K adsorbed by or released from the soil solid phase. No specific partition of {Delta}K has been performed in the Q/I analysis. In this study, {Delta}K was partitioned into changes in K due to the exchangeable pool ({Delta}Exch K) and the nonexchangeable pool ({Delta}Non-Exch K). Calculations of {Delta}Non-Exch K and {Delta}Exch K were:

[3]
and

[4]
where EK0 was the exchangeable K corresponding to {Delta}K = 0, which was estimated from the linear regression equation of EKf vs. {Delta}K. This estimate of {Delta}Non-Exch K had been used by Schneider (1997) to describe soil K fixation. Least squares regression equations, based on a quadratic model ({Delta}K = {alpha}1 + {alpha}2CR + {alpha}3CR2), were used to describe the Q/I relationships between {Delta}K and CR (Wang et al., 1988).

Changes in the amount of K adsorption or release by soil solids depend not only on the nature of soil solids, but also on the initial disequilibrium of soil solution K applied to the soil. The initial disequilibrium of soil solution applied to soil can be described by {Phi}, an initial constraint (Schneider, 1997), which is calculated:

[5]
where CK0 is the initial concentration corresponding to {Delta}K = 0, and CKi, v, and w have been defined previously.

Regression analysis and other statistical studies were conducted using SAS statistical software package version 8.2 (SAS Institute, Cary, NC).


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSION
 REFERENCES
 
Partitioned Q/I Relationships and Buffering Capacities
Traditional Q/I curves have been termed "immediate" relations of the labile K pool, although K fixation was often suspected to happen when longer soil–solution contacts were used (Beckett, 1964). We attempted to partition the Q/I curve so it allowed us to explicitly assess K changes associated with slowly available or nonexchangeable K. The amount of total K adsorbed or released ({Delta}K) during the isotherm experiment was partitioned into changes due to exchangeable K and nonexchangeable K. The change due to exchangeable K was measured by re-extracting the amount of K at the end of soil-CaCl2 contact with 1 M NH4OAc and by correcting for remaining intersititial solution K. The change due to nonexchangeable K was calculated by difference between total amount of K adsorbed and the amount of K re-extracted by NH4OAc. This change in nonexchangeable K is considered short-term fixed K (Schneider, 1997). Results of partitioned Q/I curves for the three soils at the 0- to 15-cm depth are presented in Fig. 1 .



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 1. Partitioned quantity/intensity (Q/I) relationships for surface (0–15 cm) soils of Crowley, Dundee, and Norwood. The quadratic equation corresponds to each data set legend. Both replicates of each data point are plotted.

 
Crowley and Norwood soils showed strong total adsorption of K as solution CR increased (Fig. 1). The Dundee soil released K when CR was ≤0.01 (mol L–1)1/2 as compared with ≤0.005 (mol L–1)1/2 for the other two surface soils. For Crowley and Dundee soils, the quantity of {Delta}K due to nonexchangeable pools was less than that due to exchange once CR was >0.02 (mol L–1)1/2. The Norwood soil, in contrast, released more K from its nonexchangeable fraction than from exchangeable sites, probably because of its very large nonexchangeable pool as indicated by NaBPh4–extractable K (Table 1).

Two important parameters, equilibrium concentration ratio (CR0), a measure of K intensity in the soil-CaCl2 system, and equilibrium potential buffering capacity (PBC), an indicator of soil's buffering ability, were estimated from least squares regression equations that described the partitioned Q/I curves. The CR0 value corresponds to the system solution CR at which {Delta}K = 0 (no K adsorption and desorption), whereas the PBC value is the slope of the Q/I curve at CR0, derived from the second derivative of respective quadratic equation (Wang et al., 1988). Partitioned Q/I curves for subsurface samples of these soils were also constructed (not shown) and estimated CR0 and PBC values are presented in Table 2. The Dundee soil had a higher CR0 than the Crowley and Norwood soils, and CR0 values decreased along with soil depth. The later generally corresponded to the changes in NH4OAc-extractable K in the profile (Tables 1, 2). These CR0 values were greater than the critical CR range of 0.0004 to 0.0008 (mol L–1)1/2 or equivalent to the critical AR range of 0.0005 to 0.001 (mol L–1)1/2 found for many soils and crops (Beckett, 1972), as estimated by the conversion equation AR = 1.218CR + 2 x 10–5, r = 0.999 (Wang et al., 1988). This suggests that the three soils have a relatively sufficient K intensity, though low in NH4–extractable K (≤100 mg kg–1) (Peevy, 1972). In an unpublished field study, we noted that application of K fertilizers at recommended rate of 112 kg ha–1 for these soils increased sugarcane yield in the Crowley soil, but not in the Dundee and Norwood soils. Other factors rather than NH4–extractable K or CR0 alone may better indicate the K fertility status of these soils.


View this table:
[in this window]
[in a new window]
  		Table 2. Parameters derived to describe K dynamics in soil–solution system (mean ± SE) for three low soil-test K soils.{dagger}

 
The three soils exhibited different capacities for buffering K changes in soil–solution system (Table 2). Total potential buffering capacity increased as sampling depth increased in all three soils. The PBCt of the Crowley and Norwood soils were twice that of the Dundee at the 0- to 15-cm depth, and higher at lower depths. Soil PBCe and PBCn were similar in surface soils (0–15 cm) but significantly different in the subsurface (30–45 cm) depths. Higher PBC in subsurface soils than surface soils have been reported for Iowa soils (Wang and Scott, 2001).

The phenomenon of higher PBC in subsurface soils is interesting. Beckett and Nafady (1968) pointed out that the buffering capacity of a soil depends primarily on the surface area available for ion exchange and, to a lower degree, on the character and charge density of soil surfaces. Since major exchange surfaces in soils include inorganic clays and organic humic substances, it is expected that surface soils would exhibit higher PBC values due to the fact that surface soils often receive more crop residues. Previous studies showed, however, that PBC was generally not significantly affected by soil organic matter (Evangelou et al., 1986; Sailakshmiswari et al., 1986). Although an increase in soil organic matter may increase certain high K+ affinity sites that are not controlled by K-(Ca + Mg) exchange but possibly by K-NH4 or K-H exchange (Evangelou and Blevins, 1988), the overall affinity of organic matter itself for K+ was generally smaller than that of clays (Evangelou et al., 1986). Recent research has indicated that organic matter may possess a lower surface area then previously thought, and the interaction between organic matter and inorganic clays tends to reduce the surface area of the latter (Pennell et al., 1995). This implies that it is not the available total area of exchange surface that controls the PBC, as originally suggested by Beckett and Nafady (1967)( 1968). Rather, it is the available total area of high K-affinity surfaces that determines the PBC in the soil. On the other hand, Olk and Cassman (1995) showed that the nature of organic matter, e.g., mobile humic acids (MHA) vs. calcificated humate, could also have a different influence on K fixation. High MHA content can prevent solution K from fixation in vermiculitic soils (Cassman et al., 1992; Olk and Cassman, 1995).

In this study, we found from regression analysis that only 30 to 36% of the variation in PBCt was explained by clay content, confirming that the PBC may not be simply dependent of soil clay content alone but possibly its interaction with organic matter. The notion of higher total PBC corresponding to lower organic matter (greater clay surface exposed) may explain our observation of generally increasing buffering capacity in deeper profiles of these soils as well as that in the Norwood soil, in which its clay contents remain unchanged throughout three depths (Fig. 1, Tables 1, 2). We did not perform mineralogical analyses on samples below 15 cm. A significant change in clay mineralogy in the lower depths of these soils may also contribute to PBC increases.

Most PBC values reported in the literature refer to PBC obtained from the linear portion of the Q/I curve at high CR or AR regions. This linear potential buffering capacity (LPBC) of a soil is based on the assumption that the linear portion of Q/I curve conforms to the Gapon exchange equation (Beckett, 1964). Note that quadratic curves of the Q/I relations due to exchangeable sites as shown in Fig. 1 are similar to a linear relationship, particularly for the Crowley and Dundee soils. A similar situation was also observed for deeper samples of soils (not shown). These experimental results confirm that the partitioned PBCe indeed characterizes K buffering properties associated with exchangeable sites that are nonspecific and are governed by the Gapon exchange equation.

The Norwood soil demonstrated a curved Q/I relationship over the entire experimental range (Fig. 1) and does not exhibit a linear portion common in a typical Q/I plot, indicating that K is interacting with more specific sites of soil surfaces (Beckett and Nafady, 1967, 1968). Besides the impact of low organic matter, the large percentage of apparent specific sites in the Norwood soil may result from its high pH, low exchangeable K saturation, and presence of a large percentage of smectite and clay mica (Table 1). Since soils exhibiting high K-affinity sites are high in K fixation capacity (Rich, 1968; Poonia and Niederbudde, 1986; Aide et al., 1999), clay mineralogical differences and factors affecting nonexchangeable K dynamics may be the key in controlling PBCn differences in surface and subsurface soils.

It has been shown that for soils that are primarily characterized by LPBC, the equivalent of PBCe as in our approach, the buffering ability of a soil is controlled by the product of the CEC and Gapon selectivity coefficient (KG) for K-Ca exchange (Evangelou et al., 1994):

In such a case, soil-buffering capacity can be predicted by CEC as long as KG is constant. This would suggest that K fertilization may be linearly adjusted, based on differences in soil CEC, to reflect buffering capacity in mineralogically similar soils (relatively similar KG). In fact, this practice has been used in certain places (McLean, 1976; Vitosh et al., 1995). Our results, however, indicate that this method of predicting soil buffering capacity could not accurately account for soils that are predominately controlled by specific K sites or fixation, and release from the nonexchangeable K pools. In that case, PBCn may have to be determined to accurately assess K fertility status. In a separate regression analysis, we found that PBCt was significantly (P < 0.01) linearly dependent on PBCn (R2 = 0.94) but not on PBCe (R2 = 0.16). The results clearly demonstrated that the overall buffering power of these low-test K soils was dominated by K dynamics of nonexchangeable or fixed K pools. Use of CEC to predict soil-buffering capacity may be suitable only where the overall buffering ability of a soil is overwhelmingly controlled by similar and simple exchange reactions, which are not the case for these sugarcane-producing soils. This is further evident by the low linear regression between PBCe and CEC (R2 = 0.097). With field calibration, PBCn and PBCe could be used to understand and characterize K behavior in representative soil groups.

Conversion of Added Potassium to Exchangeable and Nonexchangeable Pools
There was a linear relationship between final exchangeable K and {Delta}K (Fig. 2) . Due to the fact that the {Delta}K (the change in solution K) when positive corresponds to K increase in soil solid phase, the slope (defined as {alpha}) of the linear regression between final exchangeable K and {Delta}K can be used to measure the magnitude of conversion from solution K to exchangeable K by each soil. The {alpha} values for all soils are presented in Table 2. Surface soils of Crowley, Dundee, and Norwood exhibited {alpha} values of 0.44 to 0.60, suggesting that 44 to 60% of {Delta}K converted to exchangeable K. Subsurface soils (30–45 cm) showed a much larger variability in conversion rate among the three soils with {alpha} values of 0.17, 0.62, and 0.41 for Crowley, Dundee, and Norwood, respectively. Crowley had the most difference in {alpha} (0.17–0.59) between its surface and subsurface samples, whereas Dundee had similar surface and subsurface characteristics in exchangeable K turnover (60–65%) (Table 2). Regression analysis indicated that 60% of the variability in {alpha} could be explained by clay content in surface and subsurface samples of the three soils. In general, the higher the clay content is, the lower the conversion rate is. The {alpha} value was found to correlate significantly with PBCe (r = 0.80, P < 0.01).



View larger version (28K):
[in this window]
[in a new window]
  		Fig. 2. Final exchangeable K as a function of total {Delta}K for surface (0–15 cm) and subsurface (30–45 cm). The quadratic equation corresponds to each data set legend. Both replicates of each data point are plotted.

 
A linear relationship was also observed between added K, expressed as initial constraint ({Phi}), and the change ({Delta}) in nonexchangeable K (Fig. 3) . The slope, defined as ß, between the change in the nonexchangeable K and the initial constraint measures the impact of the nonexchangeable pools on K dynamics in the soil–solution system (Schneider, 1997). The larger the ß, the greater the portion of added K converted to nonexchangeable K (fixed) at positive {Phi}, or the more fixed K released at negative {Phi}. The range of ß values between surface and subsurface soils was 0.12 to 0.23 for Norwood, 0.10 to 0.37 for Crowley, and 0.05 to 0.08 for Dundee (Table 2). The largest ß was 0.37 for the Crowley subsurface soil (30–45 cm), suggesting that its K fixation capacity was 37%. The ß value correlated significantly (P < 0.01) with PBCn (r = 0.87) and PBCt (r = 0.79), but not with PBCe (r = 0.026). As expected, the ß value was negatively correlated with {alpha} value (r = 0.95, P < 0.01), indicating an antagonistic relation between the conversion rates of exchangeable K and nonexchangeable K in the soil–solution system.



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 3. Change in nonexchangeable K as a function of initial constraint for surface (0–15 cm) and subsurface (15–30, 30–45 cm) soils. The quadratic equation corresponds to each data set legend. Both replicates of each data point are plotted.

 
Regression analysis showed that 70% of the variability in ß could be explained by soil clay content. This implies that ß may be a different indicator from PBCn since PBCn was less linearly dependent on clay contents (R2 = 0.30). A close examination showed that a larger quantity of Ca was adsorbed by the Crowley soil as opposed to Ca adsorbed by the Dundee and Norwood soils (on average, 6.0% vs. 0.1 and 1.0% of added Ca (0.01 M) was adsorbed, respectively). Since PBCn is based on the plot of {Delta}K as a function of CR, which references to soil solution Ca and Mg, and ß is based on the change in nonexchangeable K vs. initial K disequilibrium ({Phi}) only, we suspect that relative Ca status may contribute to the difference between PBCn and ß in relation to soil clay contents. It has been shown that K selectivity of smectite in the presence of Ca is different from that of mica due to the difference in interlayer expansion of minerals as well as the difference in size of exchange cations (Rich, 1968; Evangelou and Lumbanraja, 2002). This difference may contribute to the effect of clay content on PBCn and ß values. These results suggest that when fertilizer K is applied to a soil characterized with a large {alpha} and a small ß, much of the fertilizer K is expected to be held at exchangeable sites and would stay available without being fixed during the growing season. This could be a favorable situation for crops that require large amounts of persistent available K during the growing season such as sugarcane.

Threshold of Nonexchangeable Potassium Release
The characteristics of nonexchangeable K pools were further evaluated by CKr and EKr using a method described by Schneider (1997). Data points showing total net nonexchangeable K release were regressed against CKf and EKf and extrapolated for this purpose. According to Schneider (1997), CKr and EKr are critical values at which {Delta}Non-Exch K = 0, below which the release of nonexchangeable K was initiated. In addition, the CKi at which {Delta}K = 0 (defined as CK0) and the final exchangeable K at which {Delta}K = 0 (defined as EK0) were estimated from the linear regression of CKi vs. {Delta}K and EKf vs. {Delta}K, respectively. The results of these estimates are also presented in Table 2.

The CKr and EKr values were similar to those of CK0 and EK0, except for the Crowley surface soil, which showed a larger difference. Statistical analyses indicated that CKr was significantly (P < 0.01) correlated with CK0 (r = 0.94) and CR0 (r = 0.95) in these soils. The high correlations imply that soil properties affecting these parameters are similar. On the other hand, EKr and EK0 were also correlated (r = 0.85). Schneider (1997) reported a very high correlation between EK0 and EKr for surface samples of a silt loam soil (r > 0.99). The correlation found in this study, however, covers soils of different mineralogies and sampling depths.

The order of EKr values for surface soils was Dundee ≥ Norwood > Crowley (Table 2). For subsurface samples, Norwood showed higher EKr values than Crowley and Dundee soils. The CKr values, critical K solution concentrations, of Dundee soils were greater than those of Crowley and Norwood throughout the profile. This may imply that a smaller depletion of solution K would be needed by Dundee than by Crowley and Norwood soils for initiation of nonexchangeable K release. On the other hand, the nonexchangeable pools of the Dundee soil was less significantly affected by initial K addition and concentration ratio than those of the Crowley and Norwood soils as indicated by ß and PBCn (Table 2 and Fig. 1). This suggests that, as high EKr or CKr values may be an indicator of easiness of nonexchangeable K release (threshold easy to reach), the extent of the release (capacity) would be still controlled by PBCn. The correlation between EKr and nonexchangeable K release was significant (r = 0.77, P < 0.05). The EKr was even more significantly correlated with total K release (total desorption) during the experiment (r = 0.91, P < 0.01). In comparing EKr and CKr values, we speculate that in the Crowley soil, a crop would take up K largely from the added K because the added K may easily prevent exchangeable K or solution K from reaching EKr or CKr for initiation of nonexchangeable K release. This is because the EKr and CKr values of the Crowley soil are very low and largely different from its EK0 and CK0. This may explain the fact that Crowley soil responded to K fertilization, whereas Dundee and Norwood did not.


   CONCLUSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
REFERENCES
 
Our results showed that these sugarcane-producing soils with similar low levels of NH4AOc-extractable K possessed different characteristics of K dynamics in the soil–solution system. While these soils possessed high levels of NaBPh4–extractable K throughout the profile, K buffering abilities of these soils differed. Partitioning of traditional Q/I curves into K changes due to exchangeable and nonexchangeable pools of K allowed us to explicitly evaluate short-term K dynamics in the soil–solution system. Results demonstrated that buffering capacity due to nonexchangeable K is more likely to control the overall buffering capacity of a low test-K soil. The K total buffering capacity increased with soil depth, corresponding to decreases in soil organic matter. A significantly negative correlation between {alpha} and ß suggests that soil ability of converting to nonexchangeable K (fixation) from solution K inversely and linearly relate to that of converting to exchangeable K. Crowley and Norwood soils exhibited higher K buffering capacity, especially PBCn, than the Dundee soil at all three depths. The Crowley soil, however, required relatively low critical exchangeable K (EKr) and solution K (CKr), which implies that a large depletion of exchangeable K and solution K was needed to initiate nonexchangeable K release. The results suggest that a combined assessment of nonexchangeable K buffering capacity and release threshold would be useful for grouping soils that test low in exchangeable K for fertility management.


   ACKNOWLEDGMENTS
 
This work was supported in part by the American Sugar Cane League and the Louisiana Soybean and Grain Research and Promotion Board.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 
Contribution of Louisiana Agric. Exp. Stn. Journal No. 02-09-0434 and is published with the approval of the Director.

Received for publication March 25, 2003.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 Soil Sci.Home page
W. Nilawonk, T. Attanandana, A. Phonphoem, R. Yost, and X. Shuai
Potassium Release in Representative Maize-Producing Soils of Thailand
Soil Sci. Soc. Am. J., May 1, 2008; 72(3): 791 - 797.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
J. J. Wang and D. L. Harrell
Effect of Ammonium, Potassium, and Sodium Cations and Phosphate, Nitrate, and Chloride Anions on Zinc Sorption and Lability in Selected Acid and Calcareous Soils
Soil Sci. Soc. Am. J., June 2, 2005; 69(4): 1036 - 1046.
[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 (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wang, J. J.
Right arrow Articles by Bell, P. F.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Wang, J. J.
Right arrow Articles by Bell, P. F.
Agricola
Right arrow Articles by Wang, J. J.
Right arrow Articles by Bell, P. F.
Related Collections
Right arrow Sorption/Exchange
Right arrow Nutrient Management
Right arrow Soil Chemistry

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