Successful peanut production requires adequate Ca in the pegging zone. Soil testing for Ca based on pegging zone samples provides scientific basis for Ca supplementation for peanut, but concerns regarding inaccuracy of certain soil testing methods have been brought up by previous studies; therefore, reevaluation of routine soil testing methods for Ca on pegging zone samples is necessary. The objective of this study was to evaluate relationships and predictability of four major soil tests including Mehlich 1 (M1), Mehlich 3 (M3), 1 N neutral ammonium acetate (AA), and 0.01 M sodium nitrate (SN) for Ca, K, and Mg for peanut production in Coastal Plain soils common in the major peanut producing regions of the Southeast. Results showed that the four tests extracted varied amount of Ca, K, and Mg from soil. Soil treatments that included an addition of gypsum or lime reduced availability of soil K and Mg. Correlations were strong among soil K and Mg determined using M1 and M3, but the relationship was weak between M1 and SN. Application of gypsum shortly before soil sampling had minimal effect on correlations among the four tests; however, application of lime led to weaker correlations. In unamended or gypsum-treated soils, M1 is an adequate index of soil available Ca for peanut. The M1 test may overestimate soil available Ca if soil is recently limed; therefore, any of the other three soil tests may be used in such cases. The AA test for Ca had the best relationships to peanut yield, grade, and seed Ca concentration, followed by M1.
Sufficient Ca in the pegging zone (upper 7 to 10 cm of soil) is crucially important for peanut pod development and therefore quality peanut (
A summary of routine soil testing methods for Ca including Mehlich 1 (M1), Mehlich 3 (M3), 1 N neutral ammonium acetate (AA), and 0.01 M sodium nitrate (SN).
Some concerns have been raised about the accuracy of soil testing Ca for supplemental Ca recommendations for peanut production. Studies have shown that M1 may overestimate available Ca to developing pods in lime-amended soils, due to the dissolution of undissolved lime particles by the highly acidic M1 solution (
Soil Ca is not the only criteria that should be used to provide accurate peanut fertility recommendations. The Ca/K ratio also should be considered when determining whether a soil is limited by Ca. When the Ca/K ratio is <3:1, the capability of peanut to absorb Ca may be limited (
The objective of this study is to compare relationships among the common soil Ca tests, M1, M3, AA, and SN for pegging zone soils of the southeastern United States, determine effects of recent field applications of gypsum and lime on soil test Ca, Mg, and K, and evaluate the ability of soil Ca tests for correct evaluation of Ca availability for peanut.
Soil samples were taken from Ca supplementation field trials conducted at the Wiregrass Research and Extension Center (WREC) in Headland, AL (31.36°; N, 85.32°; W), and the Gulf Coast Research and Extension Center (GCREC) in Fairhope, AL (30.55°; N, 87.87°; W), in 2012 and 2013. Study sites at each location were not the same in 2012 and 2013 to avoid carryover effects of gypsum or lime from the previous season. Soil series include Dothan sandy loam (fine-loamy, kaolinitic, thermic Plinthic Kandiudults) and Lucy loamy sand (loamy, kaolinitic, thermic Arenic Kandiudults) at WREC, and Malbis fine sandy loam (fine-loamy, siliceous, subactive, thermic Plinthic Paleudults) at GCREC. The field trials were organized as randomized complete block designs with 4 replications under irrigated and non-irrigated conditions. Soil samples were taken from the following treatments: untreated control, 1120 kg/ha gypsum (251 kg Ca/ha) applied at early bloom, and 1120 kg/ha lime (381 kg Ca/ha as calcitic lime at WREC and 437 kg Ca/ha as dolomitic lime at GCREC) applied at planting. Lime was applied prior to planting due to its lower solubility, which requires more time to dissolve in soil than gypsum. Pegging zone soil samples (top 7 cm of soil) were collected from each plot immediately prior to planting and any Ca amendment and at mid-bloom after application of lime and gypsum. Mid-bloom is ∼70 days after lime application and ∼30 days after gypsum application. All pegging zone soil samples were composite samples of 14 individual samples collected from each plot with a hand-held soil probe, dried at 60 C for 48 h, ground using a mortar and pestle, and sieved to pass a 2-mm screen.
Soil Ca, Mg, and K were determined using four different soil testing methods (
Peanut yield from treatment plots was determined at harvest after peanuts were dried to approx. 10% moisture. In order to combine yield data from multiple site-years, relative yields, which represents the yield of the plot compared to the mean yield of the untreated control treatment for that site-year, were calculated for each experimental plot. Thus, relative yields could exceed 100%. Peanut grade represents the percentage of SMK in a 250 g sample. The SMKs were then dried at 60 C for 48 h, finely ground in a coffee grinder (Hamilton Beach Inc., Picton, Canada), and microwave digested using a Mars Xpress microwave (CEM Corp., Matthews, NC) using modified EPA-3051 procedure (
The recovery of Ca from various sources of gypsum and lime in M1, M3, AA, and SN was also analyzed. Three types of gypsum including Agri-Cal, which is flue gas desulfurized (FGD) gypsum (AGRI-B Plantation LLC, Albany, GA); USG® 500, which is naturally mined gypsum (United States Gypsum Company, Chicago, IL); and PCS Wetbulk, which is phosphogypsum (PotashCorp, White Springs, FL) were used along with two types of agricultural lime including calcitic lime (Tri-State Lime LLC, Arlington, GA) and dolomitic lime (Farmer's Favorite Fertilizer Company, Moultrie, GA). A large solution/solid ratio (> 1000/1, v/w) was used to assure that the solubility of gypsum and lime was not limiting Ca recovery. In detail, 200 mg gypsum material was measured and shaken with 250 mL extractant solution for 2 h, while 4 mg lime with 500 mL extractant solution. The suspension was allowed to settle for 1 h and the supernatant was filtered using Whatman No. 42 filter paper. Concentration of Ca in filtrate was analyzed by ICP-AES. Total Ca in each amendment was analyzed similarly to seed Ca using the EPA-3051 procedure without the hydrogen peroxide pre-treatment. Calcium recovered by each extracting solution represents a percentage of total Ca in each Ca amendment. Four analytical replications were performed for each amendment.
Mixed model methodology as implemented in SAS PROC GLIMMIX were used to analyze the soil Ca, K, and Mg data obtained at mid-bloom (
There was no significant location effect on soil Ca (P = 0.32), K (P = 0.68), or Mg (P = 0.22), therefore, data were combined for both locations to determine effects of soil tests and Ca treatment on soil Ca, K, and Mg. Extractable Ca, Mg, and K by the M1, M3, AA, and SN methods were notably different (P < 0.001), and no significant Ca treatment × soil test interaction on Ca (P = 0.60), K (P = 0.90), or Mg (P = 0.70) was observed; therefore, data were combined for all Ca treatments to evaluate effect of soil tests on soil extractable Ca, Mg, and K (
Soil Ca, K, Mg, Ca:K ratio, Ca:Mg ratio, and activity ratio of Ca determined using Mehlich 1 (M1), Mehlich 3 (M3), 1 N neutral ammonium acetate (AA), and 0.01 M sodium nitrate (SN) methods. Data were combined for all Ca treatments due to lack of Ca treatment × soil test interaction.
Significant variations of soil test Ca, K, and Mg have been reported due to diverse soils used by different studies. For example, using soils from central Florida,
Due to the varied trends for Ca, K, and Mg by different soil tests, the Ca/K ratios for each test were drastically different, ranging from 2.2 by the SN test to 9.5 by the M1 test (
In order to more thoroughly evaluate the contribution of Ca supplementation to correlations among soil extractable Ca, the four soil tests were evaluated directly on various Ca sources including FGD gypsum, mined gypsum, phosphogypsum, calcitic lime and dolomitic lime. Recovery of Ca from FGD, which is the gypsum used in the field study, and two other types of gypsum did not differ using M1, AA, and SN methods (
Recovery of Ca from different Ca amendments using Mehlich 1 (M1), Mehlich 3 (M3), 1 N neutral ammonium acetate (AA), and 0.01 M sodium nitrate (SN) extracting solutions. Calcium recovered in each extracting solution was calculated as a percentage of total Ca in the Ca amendment.
Recovery of Ca from the two agricultural limes using the M1 method was 25, 50, and 200% greater than with M3, AA, and SN (
Application of Ca supplements showed inconsistent influence on extractable Ca (P = 0.11), Mg (P < 0.01), and K (P < 0.01). Due to lack of Ca treatment × soil test interaction (P > 0.5), data were combined for all soil tests to evaluate the effects of Ca supplements. Calcium supplementation in this study did not significantly elevate soil Ca level (
Soil Ca, Mg, and K in untreated, gypsum-treated, and lime-treated soils. Data were combined for all soil testing methods due to lack of Ca treatment × soil testing method interaction. Means for each element followed by different letters indicate significant differences at 𝜶 = 0.05.
Addition of gypsum and lime notably decreased soil Mg and K extraction from the pegging zone relative to the untreated control (
Even though there was no Ca treatment × soil test interaction, some of the relationships among Ca extracted by M1, M3, AA, and SN differ. Regression lines were fitted separately by locations first. Slopes of regression lines at different locations were not notably different (P = 0.49); therefore, data were combined for both locations and the regression lines among different soil tests was fitted to determine effects of Ca amendments on the correlations among the soil tests. Soils taken at mid-bloom were used. The correlations between M1 and M3 (r2 > 0.9), M1 and AA (r2 > 0.9), M1 and SN (r2 > 0.8) were strong in untreated and gypsum-treated soils (
Relationships between soil Ca extractable by Mehlich 1 (M1) and (A-C) Mehlich 3 (M3), (D-F) 1 N neutral ammonium acetate (AA), and (G-I) 0.01 M sodium nitrate (SN) methods in untreated (A, D, G), gypsum-treated (B, E, H), and lime-treated (C, F, I) soils sampled at mid-bloom, which was 30 d after gypsum application and 60-70 d after lime application.
The slopes of regression lines for soil test Ca in lime-treated soils between M1 and M3 (P < 0.001), M1 and AA (P = 0.001), and M1 and SN (P < 0.001) were significantly higher than in untreated and gypsum-lime soils (
Soils taken prior to planting were defined as pre-gypsum and pre-lime, while those taken at mid-bloom after gypsum and lime application were defined as post-gypsum and post-lime. Using this method of comparison, variation among plots can be reduced, and evidence of differences in extraction would be reflected by different slopes. Slopes of regression lines between M1 and M3 (P = 0.81), M1 and AA (P = 0.66), and M1 and SN (P = 0.88) in pre- and post-gypsum soils did not differ significantly (
Relationships between soil Ca extractable by Mehlich 1 (M1) and (A) Mehlich 3 (M3), (B) 1 N neutral ammonium acetate (AA), and (C) 0.01 M sodium nitrate (SN) methods in pre- and post-gypsum soils. Pre- and post-gypsum data points represent soil samples taken before and 30 d after gypsum application.
Relationships between soil Ca extractable by Mehlich 1 (M1) and (A) Mehlich 3 (M3), (B) 1 N neutral ammonium acetate (AA), and (C) 0.01 M sodium nitrate (SN) methods in pre- and post-lime soils. Pre- and post-lime data points represent soil samples taken before and 60-70 d after lime application.
Relationships between the soil test methods for K were evaluated to generate approximate conversion equations. Data were combined for all locations and Ca treatments due to a lack of location effect and Ca treatment × soil test interaction. There were strong correlations for K extracted by M1, M3, and AA (r2 > 0.7); however, M1 and SN were poorly correlated (r2 = 0.37,
Relationships between soil K extractable by Mehlich 1 (M1) and (A) Mehlich 3 (M3), (B) 1 N neutral ammonium acetate (AA), and (C) 0.01 M sodium nitrate (SN) methods.
Soils from GCREC were 5 to 10 times greater in Mg than those from WREC, which is likely due to the longstanding use of dolomitic lime at GCREC and calcitic lime at WREC. As a result of these differences, the relationships between M1 and other methods differed by location (P = 0.018), and regression lines were fitted separately (
Relationships between soil Mg extractable by Mehlich 1 (M1) and (A) Mehlich 3 (M3), (B) 1 N neutral ammonium acetate (AA), and (C) 0.01 M sodium nitrate (SN) methods.
Correlations using the boundary line method among peanut relative yield and soil extractable Ca by (A) Mehlich 1 (M1), (B) Mehlich 3 (M3), (C) 1 N neutral ammonium acetate (AA), and (D) 0.01 M sodium nitrate (SN) generated by boundary line method.
Relationships between soil test Ca and relative yield and grade of peanut in the current study were weak. For relative yield of peanut, r2 using the boundary line method was 0.49, 0.46, 0.70, and 0.28 for M1, M3, AA, and SN, respectively (
Correlations using the boundary line method among peanut grade represented as percentage of sound mature kernels (SMK) and soil extractable Ca by (A) Mehlich 1 (M1), (B) Mehlich 3 (M3), (C) 1N neutral ammonium acetate (AA), and (D) 0.01 M sodium nitrate (SN) generated by boundary line method.
Several factors could account for the poor relationships between soil tests and peanut yield and grade observed in the current study as well as previous studies. In the current study, the background Ca level was ≥ the established critical value (150 mg/kg) for peanut production in Alabama. Therefore, Ca may not be a limiting factor for peanut production in those soils, and a yield or grade response to increased soil Ca may not occur. Also, there could be an effect of peanut cultivar. Each of these evaluations was performed with different peanut cultivars that may have different Ca uptake patterns or requirements.
Strong correlations (r2 > 0.6) were found between seed Ca concentration and soil Ca using the boundary line method with the four different soil tests (
Correlations using the boundary line method among peanut seed Ca concentration and soil extractable Ca by (A) Mehlich 1 (M1), (B) Mehlich 3 (M3), (C) 1 N neutral ammonium acetate (AA), and (C) 0.01 M sodium nitrate (SN) generated by boundary line method.
Addition of gypsum shortly (e.g., within 30 d) before soil sampling had minimal effect on correlations among soil test Ca. In addition, strong correlations for soil Ca exist with M1, M3, AA, and SN in gypsum-treated soils; therefore, M1 is a satisfactory index of soil available Ca for peanut if soil is unamended or gypsum is used for Ca supplementation.
The relationships are also strong among M1 and other methods for lime-treated soils. The slope changes because the M1 test extracts more Ca from recently applied lime than other tests. It is likely that the M1 test overestimates soil Ca in lime-treated soils, if undissolved lime is present in the soil sample. In such cases, reduced peanut yield and substandard grade may occur due to inadequate Ca, even though M1 extractable Ca is above the critical level for Ca supplementation. Therefore, it is important to know whether a soil has been recently limed prior to converting one soil test value to another. If soil is recently (e.g., within 6 mo) limed, care should be taken when evaluating soil Ca extracted by M1. An alternative way is to use any of the other three soil tests for soil Ca and convert to M1-Ca to get recommendations for Ca supplementation.
The soil Ca extracted by AA showed the best correlation with peanut yield, grade, and seed Ca concentration among the four evaluated tests. The M1 test is the second best, and the SN test was the worst. In practice, M1 test should continue serving as routine soil test in order to improve the efficiency of soil testing, since M1 extracts nutrients other than Ca (e.g., P, Mg, and K) at the same time.
Thanks to continuous support of National Peanut Foundation and Alabama Peanut Producers Association.
Graduate Research Assistant, Dept. Crop, Soil and Environmental Sciences, Auburn University, Auburn, AL 36849
Associate Professor, Dept. of Soil and Crop Sciences, Texas A&M University, College Station, TX 77843
Extension Specialist, Dept. Crop, Soil and Environmental Sciences, Auburn Unviersity, Wiregrass Research and Extension Center, Headland, AL 36345