SSSAJ Grow Your Career with SSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 1 January 2007
Published in Soil Sci Soc Am J 71:125-132 (2007)
DOI: 10.2136/sssaj2006.0060
© 2007 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
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 ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Murashkina, M. A.
Right arrow Articles by Pettygrove, G. S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Murashkina, M. A.
Right arrow Articles by Pettygrove, G. S.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Murashkina, M. A.
Right arrow Articles by Pettygrove, G. S.
Related Collections
Right arrow Soil Analysis
Right arrow Soil Chemistry
Right arrow Soil Fertility and Productivity

SOIL CHEMISTRY

Potassium Fixation in San Joaquin Valley Soils Derived from Granitic and Nongranitic Alluvium

M. A. Murashkina*, R. J. Southard and G. S. Pettygrove

Department of Land, Air and Water Resources, Soils and Biogeochemistry Graduate Group, One Shields Ave, Univ. of California, Davis, CA 95616

* Corresponding author (mmurashkina{at}ucdavis.edu).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 
Potassium fixation influences the effectiveness of fertilization in soil–plant systems. A rapid method for measuring K fixation could help clarify relationships with other soil properties, especially mineralogy. Our objectives were to compare some existing measurement techniques for soil K fixation and availability, develop an alternative K fixation test, and evaluate the utility of soil texture and parent material for predicting K fixation in soils derived from granitic Sierra Nevada (SN) and nongranitic Coast Range (CR) alluvium. Potassium pools were estimated by 1 mol L–1 NH4OAc and sodium tetraphenylboron (TPB) extractions. Our 1-h fixation method correlated well (R2 = 0.95, P = 0.001) with a 7-d procedure, so the 1-h method was used for subsequent work. The SN soils fixed up to 740 mg K kg–1; CR soilsf ixed up to 263 mg K kg–1. There was no significant relationship between K fixation and soil clay or silt content for either parent material. The TPB test had a stronger correlation with NH4OAc-extractable K in SN soils (R2 = 0.77, P = 0.001) than in CR (R2 = 0.49, P = 0.001). Plant-available nonexchangeable K (PANK = TPB minus NH4OAc) did not correlate with K fixation potential for pooled data from all pedons (R2 < 0.11), and had negative correlation (R2 from 0.97 to 0.99, P = 0.01) for individual pedons. The PANK probably represents K that has already been fixed and satisfies some of the K fixation capacity. The 1-h test is a reliable, rapid method for predicting K fixation potential. Together, the TPB and NH4OAc tests could be useful for identifying K already fixed by soils, thereby reducing K fixation potential.


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 
The production of cotton (Gossypium spp.) is a significant contributor to California agriculture, one of California's dominant industries. In 2004, cotton was California's second highest valued field crop, totaling US$808 million (California Agricultural Statistics, 2006). Irrigated cotton has been grown on as much as 500 000 ha each year in the San Joaquin Valley (SJV) of California. It is estimated, however, that K deficiency reduces cotton yield on about one-fifth of the cotton acreage in the SJV (Cassman et al., 1990). Cotton's K uptake requirement is intense during the boll-fill stage, a time when the cotton root system is more dependent on subsurface layers, below about 15-cm depth. Potassium deficiency during this critical stage of development affects both yield and quality of cotton fiber (Bennett et al., 1965; Kerby and Adams, 1985).

Late-season K deficiency in cotton in California and response to K fertilizer was first reported by Stromberg (1960). The deficiencies are widespread in the SJV, apparently occurring on many soils derived from Sierra Nevada granitic alluvium, which contains significant amounts of K-fixing minerals such as vermiculite, hydrous biotite, and biotite mica at different weathering stages (Page et al., 1967; Shaviv et al., 1985). In these soils under field conditions, long-term fixation of added K continues for at least 3 yr and K inputs in excess of 1500 kg K ha–1 are sometimes required to achieve maximum cotton (Acala type: Gossypium hirsutum L.) yields (Cassman et al., 1989, 1992). Such micaceous soils that are intensively cropped in high-yield environments may not only require a higher-than-normal rate of K fertilizer application to overcome deficiency, but may also need a special soil test that reflects the K fixation potential.

The NH4OAc (1 mol L–1) method, which extracts both soluble and exchangeable K, is the most common test used to develop K fertilizer recommendations (Soil Survey Staff, 2004). It has been shown, however, that this method is inadequate for soils that have micaceous or vermiculitic mineralogy, which can release some of nonexchangeable K when solution and exchangeable K pools are depleted (McLean, 1976; Cassman et al., 1990; Eckert and Watson, 1996).

Some commercial laboratories use a simplified K fixation test that was developed based on work by Cassman et al. (1990). The Cassman K fixation test involves incubating a soil sample with 18 mmol K kg–1 soil for 7 d with daily shaking for 45 min. This test is simple, but a 7-d incubation period and the daily shaking limit its usefulness for routine laboratory operations.

Current recommendations for management of K in cotton production in California (Miller et al., 1997) include the recommendation for a critical value or threshold of 110 mg kg–1 for soil NH4OAc-extractable K in the 15- to 45-cm depth. At this K level and above, the likelihood of a positive lint response to added K is <10% . Soil levels <80 mg kg–1, or if K fixation is >60%, indicate a high possibility of K fertilizer response. No recommendation for which K fixation soil test procedure to use is specifically provided by Miller et al. (1997), but they referenced Cassman et al. (1990).

Another method for measuring plant-available K in midwestern U.S. soils was published by Cox et al. (1999). A 5-min incubation with sodium tetraphenylboron (TPB) allows the BPh4– ion to combine with K+ in solution, causing precipitation of potassium tetraphenylboron (KBPh4). Solution Na+ acts as an exchanger for exchangeable and interlayer K+. This test extracted 1.5 to 6 times more K than did NH4OAc, and the amount of K extracted was shown to be closely correlated with plant uptake of K in greenhouse studies (Cox et al., 1999).

Although numerous methods for measuring K fixation exist, there is still an absence of a simple and fast test that would help farmers predict K fixation. Our objective was to assess the relationships among the Cassman et al. (1990) K fixation method, the TPB method, and NH4OAc-extractable K, and to evaluate the effect of parent material and soil texture on soils collected from cotton-producing areas of the SJV. We sought to develop a modified procedure more suitable for routine determination of K fixation potential that might benefit cotton growers and agricultural laboratories.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 
Soil Sampling
To investigate the effect of parent material and particle size distribution on K fixation, we used soils from two parent materials (SN granitic and CR nongranitic alluvium) that covered a range of soil textures. Study sites for this research were chosen to represent two categories: (i) soils derived from granitic, SN parent material, the east side of the SJV; and (ii) soils derived from CR nongranitic parent material, the west side of the SJV. We expected SN soils to have higher K fixation potential than CR soils due to greater amounts of micaceous minerals in the granitic alluvium.

We produced preliminary maps of the SJV based on soil family-level characteristics from the Soil Survey Geographic (SSURGO) database of the National Cooperative Soil Survey. The maps displayed soil information overlain by land use information from the California Department of Food and Agriculture that showed where cotton is grown. We hypothesized that coarse-loamy and fine-loamy families were likely to fix K. The maps were used to locate sampling sites. Most of the sites were either in cotton production at the time of sampling or in a cotton rotation. A reconnaissance survey of 48 agricultural fields that received no K fertilizer since fall of 2000 was conducted in 2001 to examine the status of the extractable K. During the survey, soil grab samples from the 15- to 45-cm depth, the depth of the most efficient K uptake by cotton roots (Gulick et al., 1989; Miller et al., 1997), were collected from Fresno, Kings, and Kern counties. At each site, three grab samples were collected by auger; samples were separated from each other by 10 crop rows. Soil texture was estimated by feel to a depth of 100 cm for one of the three subsamples. In the lab, after air drying and grinding samples to pass a 2-mm sieve, 1 mol L–1 NH4OAc-extractable K was measured by atomic emission spectrometry (data not shown). These results were used to determine locations for full profile sampling.

Many soils in the SJV have >110 mg kg–1 NH4OAc-extractable K, above the threshold for K-deficient soils (Miller et al., 1997), especially in the surface horizons. Surface horizons with NH4OAc-extractable K concentrations >200 mg kg–1 were excluded from data analysis due to the greater likelihood of recent K fertilizer and manure applications. Soil samples were collected by horizon to a maximum depth of 150 cm from 14 on-farm locations: seven pedons from SN, and seven from CR alluvium (Fig. 1 , Table 1). The soils included eight Entisols, three Aridisols, two Mollisols, and one Vertisol (Huntington, 1971; Arroues and Anderson, 1986; Chang, 1988; Arroues, 2004). Soil pits for profile descriptions were dug in furrows after hand leveling cotton beds to provide a consistent reference point for depth.


Figure 1
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 1. Map of the study area in the San Joaquin Valley, California, showing location of the sampling sites. Numbers indicate pedon identification codes.

 

View this table:
[in this window]
[in a new window]

  		Table 1. Geographical location and soil classification.

 
Laboratory Analyses
In the lab, samples were air dried at room temperature, manually ground, and passed through a 2-mm sieve.

To prepare samples for particle size fractionation, carbonates were dissolved using a 1 mol L–1 NaOAc solution buffered at pH 5.0, and organic matter was removed with NaClO (pH 9.5). Particle size distribution was determined by pipette and sieving (Soil Survey Staff, 2004).

Ammonium Acetate Extractable Potassium and Cation Exchange Capacity
Potassium was extracted by 1 mol L–1 NH4OAc (pH 7) with a mechanical vacuum extractor (Soil Survey Staff, 2004). Soil samples (2.5–3 g) were saturated and extracted overnight with 1 mol L–1 NH4OAc (pH 7), and K was determined by flame emission spectrometry. The same samples were used to determine cation exchange capacity by washing with ethanol to remove excess NH4+ and displacing NH4+ with Na+. The displaced NH4+ then was determined by a flow injection analyzer (Lachat QuikChem 8000, Lachat Instruments, Loveland, CO).

Potassium Fixation Potential
To test the Cassman et al. (1990) K fixation method, and to develop a less time- and labor-consuming procedure, we used seven of the pedons, six of which (203, 210, 215, 219, 224, and 225) represent soils derived from granitic Sierra Nevada alluvium, and presumably fix K, and one pedon (208) derived from Coast Range alluvium. Pedons were selected based on preliminary data to represent a range of soil textures and NH4OAc-extractable K.

Following the procedure of Cassman et al. (1990), 3 g of soil were equilibrated with 30 mL of 2 mmol KCl (780 mg K kg–1 soil) and one drop of toluene for 7 d with daily shaking for 45 min. After shaking and centrifuging on the seventh day, the K concentration of a 5-mL aliquot of the supernatant (solution-phase K) was determined. Ten milliliters of 3.5 mol L–1 NH4Cl was then added to each tube to make a 1 mol–1 NH4Cl solution. After 30-min shaking followed by 45-min centrifuging at 16200 x g (Sorvall RC-5B refrigerated superspeed centrifuge, Thermo Electron Corp., Asheville, NC), solution K was measured by flame emission using a PerkinElmer AAnalyst 800 atomic absorption spectrophotometer (PerkinElmer, Wellesley, MA). Added K that was not recovered in the solution phase or in the 1 mol L–1 NH4Cl extract was considered to be fixed.

To determine the effect of incubation duration on K fixation, primarily to see if shorter incubation times gave comparable results, we modified the Cassman procedure. Three replicates of each soil sample were shaken for 1 h with 2 mmol KCl solutions at a 1:10 soil/solution ratio. Ten milliliters of 4 mol L–1 NH4Cl was then added to each sample to yield a 1 mol L–1 NH4Cl solution. After 30-min shaking and 15-min centrifuging at 650 x g, K was measured in the solution. The amount of K+ not displaced by NH4+ was considered fixed and was calculated as the difference between the initial K added and the extracted K.

Sodium Tetraphenylboron-Extractable Potassium
"Plant-available" K, which includes soluble, exchangeable K and a portion of fixed K, was extracted with NaTPB (Cox et al., 1996, 1999). Soil samples (three replicates) of 1 g soil were weighed into 50-mL Erlenmeyer flasks, and 3 mL of extracting solution (0.2 mol L–1 NaTPB + 1.7 mol L–1 NaCl + 0.01 mol L–1 ethylenediaminetetraacetic acid [EDTA]) were added. Soils were allowed to stand at room temperature for 5 min, and then 25 mL of quenching solution (0.5 mol L–1 NH4Cl + 0.11 mol L–1 CuCl2) was added to the solutions for recovery of precipitated K. The flasks were heated to about 150°C on a hot plate and the mixture boiled gently for 30 to 45 min until the precipitate dissolved completely (Cox et al., 1999). After the precipitate was completely dissolved, deionized water was added to bring the volume to 50 mL. Three drops of 6 mol L–1 HCl were added to prevent precipitation of Cu and breakdown products of TPB. Samples were shaken by hand and then filtered using Whatman no.1 filters. Standards were prepared to match the matrix of the samples. Solutions were analyzed for K with a PerkinElmer AAnalyst 800 atomic absorption spectrophotometer using flame emission.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
SUMMARY
 REFERENCES
 
Ammonium Acetate Extractable Potassium
Selected physical and chemical properties of the studied soils are shown in Tables 2 and 3. In pedons from both parent materials, NH4OAc-extractable K contents were highest in the Ap horizons and decreased with depth. Soils from SN alluvium had a range of 59 to 170 mg kg–1 NH4OAc-extractable K in the Ap horizons, and 25 to 136 mg kg–1 in the subsurface layers. These values are similar to data reported by Cassman et al. (1990) for five soil series of the SJV, where the range of extractable K (1 mol L–1 NH4Cl) in the upper 20 cm was 54 to 251 mg kg–1. In a later study (Cassman et al., 1992) of Grangeville soil (coarse-loamy, mixed, superactive, thermic Fluvaquentic Haploxeroll), dominated by vermiculite, hydrobiotite, and biotite mica, values of extractable K (1 mol L–1 NH4Cl) were reported to be 56 to 134 mg kg–1 for the 0- to 20-cm depth, and 37 to 90 mg kg–1 for the 20- to 40-cm depth.


View this table:
[in this window]
[in a new window]

  		Table 2. Selected soil physical and chemical properties for soils derived from Sierra Nevada alluvium.

 

View this table:
[in this window]
[in a new window]

  		Table 3. Selected soil physical and chemical properties for soils derived from Coast Range alluvium.

 
Three pedons derived from CR alluvium had >200 mg kg–1 of NH4OAc-extractable K in the surface horizons: the Ap1 and Ap2 horizons of Pedon 200, and the Ap1 horizon of Pedons 201 and 207 (Table 3). We did not include these horizons in our data analysis, correlations, or graphs. In the other CR pedons, NH4OAc-extractable K ranged from 108 to 198 mg kg–1 in Ap and Ad horizons, and from 46 to 134 mg kg–1 in the subsurface horizons.

Potassium Fixation Potential
The 1-h incubation method produced K fixation results similar to the 7-d incubation test developed by Cassman et al. (1990) (R 2 = 0.95, P < 0.0001; Fig. 2 ). Our results support work by others showing that K fixation is rapid. Potassium fixation was completed in 24 h for additions up to 25 mmol K kg –1 (Stanford and Pierre, 1946; Sparks et al., 1980). Stanford and Pierre (1946) reported that fixation for the Webster soil series (fine-loamy, mixed, superactive, mesic Typic Endoaquolls) in the Midwest was 90% complete in 10 min. In their study, soils fixed no more K after 8 d than after 24 h. Olk (1993) reported that two-thirds of the K added, or about 8.2 mmol kg–1 soil, was fixed rapidly during the first 2 to 5 d, followed by much slower fixation lasting for at least 100 d. Olk concluded that rapid fixation is abiotic, while much of the fixation during the slow phases was possibly linked to microbial activity (Olk, 1993).


Figure 2
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 2. Comparison of K fixation potentials estimated by 7-d (Cassman et al., 1990) and 1-h incubation methods. Standard deviation bars (±1 SD) are shown.

 
The 1-h test recovered about 73% (slope of 0.73) of the values from the Cassman 7-d procedure (Fig. 2), which could be attributed to the shorter incubation time with KCl. Sparks (1987) and Sparks et al. (1980) showed that K exchange is a diffusion-controlled process. Wedge zones in interstratified vermiculite–mica are shown to have high selectivity for K (Rich, 1964; Rich and Black, 1964) but require lower diffusion rates (Sawhney, 1966). When comparing results between the two procedures, we recommend using the regression equation provided in Fig. 2.

Our modified procedure is less time- and labor-consuming than the Cassman method. After 7 d of incubation, our soils were highly dispersed, so that solutions had to be centrifuged for at least 45 min at 16200 x g. Centrifugation after the 1-h procedure required 2000 rpm for only 15 min.

Soils developed from granitic SN alluvium fixed up to 740 mg K kg–1 (1-h test), while CR soils fixed up to 263 mg K kg–1 (Tables 2 and 3). In all pedons, K fixation was lowest in the surface horizons and increased with depth. In two SN profiles, K fixation was negative in the surface horizons, meaning that more K was extracted than was added, presumably from cation exchange sites.

All CR pedons had negative K fixation values throughout the upper part of the profiles. Most of the CR pedons, however, fixed some K in C horizons. There was no evidence of vermiculite in the clay or silt fractions by x-ray diffraction. Smectites and chlorite are dominant minerals in the clay fractions, and chlorite is dominant in the silt, although some mica is also present in most pedons (data not shown; Murashkina, 2006). The degree of K fixation is strongly influenced by the charge density on the layer silicates (Sparks and Huang, 1985), and K fixation in these CR soils might be due to high-charge smectites (Weir, 1965).

Among the pedons derived from granitic SN alluvium, K fixation occurred within the whole range of NH4OAc-extractable K (27–170 mg kg–1) with the exception of three surface samples (Fig. 3A ). In CR soils, K fixation occurred when NH4OAc-exctractable K was in the range of 48 to 120 mg kg–1 (Fig. 3B). In general, there was a significant negative correlation between K fixation and NH4OAc-extractable K for both parent materials. The trend was more significant for CR soils (R2 = 0.46, P = 0.0005) than for SN soils (R2 = 0.21, P = 0.02). Correlation for SN soils improved (R2 = 0.47, P = 0.0006, slope = –3.12) after Pedon 19 was excluded from the population. Pedon 19 had a similar trend to other SN profiles, but had a different magnitude (slope = –4.02, R2 = 0.96, P = 0.019), presumably due to mineralogical differences (Fig. 3A). Addition of cation exchange capacity as a second independent variable into the regression analysis for predicting K fixation slightly improved correlations (R2 = 0.40, P = 0.004 for all SN soils, and R2 = 0.59, P = 0.0002 for all CR soils). Other variables, such as clay or silt content, had no significant influence (P > 0.12) on correlation with K fixation when included in multiple regression analysis with NH4OAc-extractable K and cation exchange capacity.


Figure 3
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 3. Relationships between K fixation potential (1-h test) and NH4OAc-extractable K in soils derived from (A) Sierra Nevada and (B) Coast Range alluvium. Standard deviation bars (±1 SD) are shown. The relationship for Sierra Nevada soils excluding Pedon 19 is y = 535.9 – 3.12x (R2 = 0.47 at P = 0.001), and for Pedon 19 is y = 236.8 – 4.02x (R2 = 0.96 at P = 0.05).

 
Soils from both SN and CR alluvium represented a range of soil textures. Among SN soils, Pedon 19, with sandy loam and loamy sand textures and NH4OAc-extractable K values of 37 to 72 mg kg–1, fixed the least K (–48 to 97 mg kg–1). Pedon 224 fixed the most K, with a clay content ranging from 17 to 41% and K fixation from 384 to 740 mg kg–1 (about 49–95% of added K was fixed; Table 2). The most fixation occurred in the C1 horizon with a sandy loam texture (17.4% clay) and 48 mg kg–1 NH4OAc-extractable K.

For soils developed from CR alluvium, Pedon 209 released the most K (no fixation: –65.5 to –354 mg kg–1; Table 3). Textures for this profile ranged from sandy loam to loam and silt loam. Among CR soils, Pedons 200 and 201, with sandy loam textures throughout the profile, fixed the most K in the subsurface horizons (152–263 mg K kg–1), which is about 19 to 33% of the K added.

Potassium fixation did not correlate with particle size distribution in either parent material. In SN soils (n = 27), regressions between K fixation and clay, silt, and sand contents revealed R2 values of 0.003, 0.037, and 0.016, respectively (P > 0.3). Similarly, in CR soils (n = 24), R2 values were 0.001 (K fixation vs. clay), 0.059 (K fixation vs. silt), and 0.036 (K fixation vs. sand), with P > 0.3. This lack of correlation could be explained by variations in mineral composition and distribution between soils within each parent material (Murashkina, 2006). In our other study we found that, in soils derived from SN granitic alluvium, vermiculite and hydrobiotite, major K-fixing minerals, dominated the silt and very fine sand fractions but not the clay fraction, while soils from CR alluvium were dominated by chlorite and smectite (Murashkina, 2006).

Sodium Tetraphenylboron-Extractable Potassium
Sodium tetraphenylboron-extracted K showed a stronger correlation with NH4OAc-extractable K for SN soils (R2 = 0.77, P < 0.0001, Fig. 4A ) than for CR soils (R2 = 0.49, P < 0.001, Fig. 4B). The slope of 2.19 for SN soils indicates that TPB extracted more than twice as much K as 1 mol L–1 NH4OAc. For CR soils, TPB extracted about 50% more K than NH4OAc (slope of 1.53). We speculate that the large difference between TPB- and NH4OAc-extractable K in SN soils is from nonexchangeable K fixed in vermiculite and from biotite in the early stage of alteration to vermiculite (Murashkina, 2006).


Figure 4
View larger version (21K):
[in this window]
[in a new window]

  		Fig. 4. Relationships between NH4OAc-extractable K and sodium tetraphenylboron-extractable K (TPB) in soils derived from (A) granitic Sierra Nevada and (B) nongranitic Coast Range alluvium. Standard deviation bars (±1 SD) are shown.

 
Figure 5 summarizes depth-weighted average (DWA) values of K pools, calculated on a horizon basis. Soils developed on SN granitic alluvium had less NH4OAc-extractable K than CR soils (78 vs. 97 mg kg–1), and SN soils fixed almost seven times as much K as soils derived from CR (278 and 40 mg kg–1, respectively).


Figure 5
View larger version (20K):
[in this window]
[in a new window]

  		Fig. 5. Comparison of K pools estimated by different methods in soils derived from Sierra Nevada and Coast Range alluvium. Potassium pools, calculated as a depth-weighted average on a horizon basis, include NH4OAc-extractable K (NH4OAc-K), K fixation by 1-h test (K fix), tetraphenylboron-extracted K (TPB-K), and plant-available nonexchangeable K (PANK).

 
The DWA of TPB-extracted K for CR soils (311 mg kg–1) was much higher than for SN soils (190 mg kg–1; Fig. 5). It is not obvious, though, why granitic SN soils yield so much less TPB-extractable K than CR soils. It has been shown that TPB extracts more K from vermiculite than from mica minerals (Scott and Reed, 1962). Wentworth and Rossi (1972) demonstrated that extractability of K by TPB in some layer-silicate minerals was in the order vermiculite > illite > biotite > phlogopite > muscovite. Cox and Joern (1997) observed expansion of mica within 1 min to 1 h of treatment with TPB and suggested that the pattern of x-ray diffraction peak broadening corresponded to edge weathering of micas. Song and Huang (1988) showed that TPB could also extract K from feldspars. Their sequence of K release by TPB was biotite > muscovite > microcline = orthoclase (Song and Huang, 1988). We speculate that TPB in CR soils may be extracting K from mica, perhaps glauconite, an Fe-rich, marine, dioctahedral mica that is often interstratified with chlorite layers (Fanning et al., 1989), or from a chlorite–vermiculite intergrade that is not apparent from x-ray diffraction.

Plant-available nonexchangeable K (PANK) that was extracted by the TPB method (defined as TPB-extractable K minus NH4OAc-extractable K) accounted for 35 to 80% of TPB-extractable K for soils derived from either SN or CR parent material (Tables 2 and 3). The contribution of PANK to TPB-extractable K increased with depth in most of the pedons, mainly due to a decrease in NH4OAc-extractable K with depth. The PANK values ranged from 33 to 296 mg kg–1 in SN soils (DWA of 110 mg kg–1) and from 98 to 351 mg kg–1 in CR soils (DWA of 218 mg kg–1; Tables 2 and 3, Fig. 5). These low values of PANK and TPB-extractable K suggest that these SN soils have a large amount of unsatisfied K fixation potential but not much fixed K. The PANK and K fixation potential were not significantly correlated when all pedons of each parent material were taken into consideration as one population (Fig. 6 , Table 4). Lack of statistically significant correlation between PANK and K fixation potential is possibly related to the differences among the pedons in soil mineralogy and particle size distribution. When data were analyzed on an individual pedon basis, five of the seven SN soils (except for Pedons 19 and 225) exhibited strong negative correlation between PANK and K fixation potential (R2 of 0.85–0.99). Although the trend was similar for all SN soils, the magnitude of the correlation varied, with slopes ranging from –0.23 to –0.43 (mean –0.32 ± 0.08, Pedon 19 excluded; Table 4, Fig. 6A). This relationship suggests that soils wherein K has already been fixed and that yield high PANK values should exhibit low K fixation potential measured by the 1-h test in the lab. Conversely, soils that have high potential to fix K should have low PANK values.


Figure 6
View larger version (26K):
[in this window]
[in a new window]

  		Fig. 6. Relationships between K-fixation potential (1-h test) and plant-available nonexchangeable K (PANK) for soils derived from (A) Sierra Nevada and (B) Coast Range alluvium.

 

View this table:
[in this window]
[in a new window]

  		Table 4. Linear regression parameters for plant-available nonexchangeable K (PANK) vs. K fixation potential for studied pedons.

 
Analyzed on a pedon basis, CR soils (Pedons 200 and 207, where n = 2, were excluded as discussed above, see Table 3) showed a negative correlation between PANK and K fixation potential with a wider range of R2 values (0.35–0.96) and slopes (–0.17 to –0.97, mean 0.46 ± 0.36) than for SN profiles (Table 4, Fig. 6 B). Overall, SN soils with greater K fixation potential had lower values of PANK than CR soils.


   SUMMARY
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
REFERENCES
 
Our 1-h incubation K fixation method produced results very similar to those of the 7-d incubation and saved much time and labor. We found a significant negative correlation between K fixation potential and NH4OAc-extractable K for soils from both parent materials. The TPB test extracted 50 to 100% more K than NH4OAc did, and had a stronger statistical relationship with the NH4OAc test for extracting K in soils from granitic SN alluvium than for nongranitic CR soils. The lack of correlation between PANK and K fixation potential for all pedons from each parent material, but strong negative correlation for most SN profiles and some CR pedons, considered individually, suggests that SN soils we studied have high potential to fix K, but have less fixed K than the CR soils. The CR soils have lower potential to fix K, presumably because much K fixation potential has already been satisfied.

Soils derived from granitic SN alluvium fixed more K than CR soils. Although all of the soils derived from CR alluvium had negative K fixation, that is, K release, in the upper part of the profiles, most of them fixed K in the deeper horizons. We found no simple, clear relationship between K fixation and particle size distribution in these soils, suggesting that more detailed information about soil mineral composition and the distribution of soil minerals in the size fractions is required to predict K fixation.


   ACKNOWLEDGMENTS
 
This work was supported by the California Department of Food and Agriculture's Fertilizer Research and Education Program (CDFA FREP), the California State Support Committee of Cotton, Inc., and Hatch Project CA-D*-LAW-4525-H. We thank Craig Rasmussen for his expertise in GIS, Cooperative Extension county advisors Bruce Roberts, Brian Marsh, Dan Munk, and Steve Wright for providing help locating sampling sites, and Jiayou Deng, Donald G. McGahan, and Julie Baker for assistance in the field.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 SUMMARY
 REFERENCES
 
Abbreviations: CR, Coast Range; DWA, depth-weighted average; PANK, plant-available nonexchangeable potassium; SJV, San Joaquin Valley; SN, Sierra Nevada; TPB, tetraphenylboron.

Received for publication February 13, 2006.


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





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 ISI Web of Science (1)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Murashkina, M. A.
Right arrow Articles by Pettygrove, G. S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Murashkina, M. A.
Right arrow Articles by Pettygrove, G. S.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Murashkina, M. A.
Right arrow Articles by Pettygrove, G. S.
Related Collections
Right arrow Soil Analysis
Right arrow Soil Chemistry
Right arrow Soil Fertility and Productivity

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