Nutrient and pH Stratification
With Conventional and No-Till Management

Our work addressed the following management questions for North Carolina conditions (Piedmont, Coastal Plain, Tidewater regions):
1) What are reasonable and/or commonly recommended soil sampling depths?
2) Does soil sampling depth affect soil test results?
3) How sensitive are lime and fertilizer application rate recommendations to sampling depth?

Interest in these then leads to the following more basic questions:

4) Does tillage or no-till management change how lime and fertilizer are distributed in the soil profile?
5) Do soil pH and nutrient concentration differ predictably among soil layers?
6) With no-till management, does the surface soil layer become more acidic? Less acidic? Why?

Why the Interest in Soil Sampling Depth?

Many analytical laboratories now recommend shallower soil sampling depths for fields with no-till management than with conventional tillage since nutrients and plant roots are generally more concentrated near the surface. Surface soil properties, such as bulk density, organic matter, pH, extractable nutrients, and plant root distribution change with the adoption of no-till methods. Nevertheless, the impacts on soil test results and fertilizer recommendations are not well understood. Soil pH stratification patterns are less consistent than those reported for nutrients and roots with no-till management. Soil surface acidification can occur, presumably due to near-surface nitrification of ammoniacal N fertilizers and decomposition of crop residues. In contrast, the soil surface can become less acidic than the underlying soil, presumably due to surface liming. Thus, variation in soil profile pH patterns is probably related to liming practices. Recommendations made by soil testing laboratories need to be applicable on a regional basis across a wide range of soils, climates, and management situations. The objective of our study was to determine the effect of soil sampling depth on soil test results for a broad range of soils on commercial and large-scale research fields managed with conventional tillage and no-till for varying lengths of time. This study differs from many others in that it investigates replicate fields from a broad spectrum of soils and crop management practices, rather than focusing on uniformly managed research plots. Thus, it better represents the nature of samples likely to be submitted to a soil test laboratory.

Our Methods

Fields from several geologic regions and with different tillage histories were selected from across North Carolina. All fields were to be planted with corn in 1996. There were 45 commercial fields and 14 research plots representing a great deal of soil diversity: 22 soil series. Tillage histories in this study were conventional tillage (CT), no-till for less than 3 years (NT<3), no-till for 3 to 6 years (NT 3-6), and no-till for more than 6 years (NT>6). Conventional tillage refers to a variety of tillage intensities, including disking, chisel plowing, and subsoiling, but in most cases does not involve moldboard plowing. Fields were sampled between January and April 1996, prior to lime and fertilizer applications. From each field, samples were collected from 0-8 inches deep, 0-4 inches and 4-8 inches, and from 0-2 inches. A composite sample from each depth was submitted for analysis following procedures of the Agronomic Division, North Carolina Department of Agriculture (Mehlich-3 extractant, results converted to index units). Data were analyzed separately for each tillage history category. For statistical testing, we used Wilcoxon's signed ranks test to determine if soil test results for a given parameter differed significantly among depth intervals for each tillage category. We used this rather than an analysis of variance since soil test values varied so widely among fields. We grouped data across regions to obtain a minimum sample size for statistical testing. No statistical comparison of fertility status across tillage categories was possible with our sampling design since field selections within each region were based on the tillage history and next crop (corn), without regard for soil type or crop management decisions.

Our Answers to the Questions Posed

1) What are reasonable and/or commonly recommended soil sampling depths?

For a routine soil sample, NCDA currently recommends sampling field crops to the plow layer depth, usually 6-8 inches. Samples should be collected to a depth of 4 inches for established pasture, turf, and minimum/no-tillage systems. For problem diagnosis, a separate subsoil sample (8-16 inches) is also requested. [Tucker et al., Soil Testing Services, Agronomic Division, N.C. Dept. of Agriculture, 1995].

A national publication reports that a survey of several states suggests a sampling depth of 2-4 inches for conservation tillage fields. [James and Wells, p. 37 in Soil Sample Collection and Handling; Chapter 3 in Soil Testing and Plant Analysis, 3rd Edition, R.L. Westerman (ed.). 1990.]

2) Does soil sampling depth affect soil test results?

Table 1. Differences between soil test results of shallow (0-4") and deeper (0-8") soil depth layers. Values shown are the differences: means of 0-4" samples minus means of 0-8" samples. A "+" indicates higher values in the shallower surface layer sample.

AssayCTNT < 3NT 3-6NT > 6
pHNS+0.1*NSNS
P-IndexNS+ 20 ***+ 27 **+ 28 **
K-IndexNS+ 24 ***+ 14 ***+ 15 **
Mg%NSNSNS+ 1 *
Zn-IndexNS+ 27 **+ 58 **+ 63 **
Cu-IndexNSNSNS+ 22 *
S-IndexNS- 8 *- 2 *NS
Statistical significance: p < *0.05, ** p < 0.01, and *** p < 0.001.

No significant differences in soil test results were detected in the conventionally tilled fields unless the comparison was made with an even shallower sampling depth (0-2 inches, data not shown). In all no-till duration categories, soil test P, K, and Zn concentrations were higher with the shallow sampling depth. In the long-term no-till category (NT > 6), soil test Mg and Cu indexes were also higher with the shallow sampling depth. Our results are consistent with previous reports describing higher nutrient concentrations in the surface soil with no-till management.

3) How sensitive are lime and fertilizer application rate recommendations to sampling depth?

Table 2. Differences between lime and fertilizer application rate recommendations based on shallow (0-4") and deeper (0-8") soil depth layers. Values shown are the differences: means of 0-4" samples minus means of 0-8" samples. A "-" indicates lower rates recommended based on the shallower surface layer sample.

MaterialCTNT < 3NT 3-6NT > 6
     
Lime (T/A)NSNSNSNS
P2O3 (lb/A)NSNS - 12 *NS
K2O (lb/ANS- 5 * - 12 **- 6 *
Statistical significance: p<0.05, ** p < 0.01, and *** p < 0.001.

Since P and K concentrations are above sufficiency levels in many North Carolina fields, many sites received no fertilizer recommendation regardless of sampling depth. If P or K concentrations were below the sufficiency level, higher nutrient concentrations near the surface led to reduced fertilizer rate recommendations when based on the shallower sampling depth. Since the overall pH effects were relatively small, the only significant relationship between sampling depth and lime recommendation noted was for the NT<3 sites when comparing 0-2 inch and 0-8 inch samples (data not shown). On average, less lime was recommended if rates were based on the shallower sampling depth.

4) Does tillage or no-till management change how lime and fertilizer are distributed in the soil profile?

Table 3. Differences between soil test results of surface (0-4") and underlying (4-8") soil depth layers. Values shown are the differences: means of 0-4" samples minus means of 4-8" samples. A "+" indicates higher values in the shallower surface layer sample.

AssayCTNT < 3NT 3-6NT > 6
     
pHNS+ 0.2**+ 0.3**NS
AcidityNSNS- 0.2*NS
P-Index+ 30**+ 31** + 52**+ 44**
K-Index+ 29** + 30***+ 23**+ 31***
Ca%NS+ 4+ + 5*;NS
Mg%NSNS+ 2*+ 2*
Mn-IndexNS+ 23***+ 33* NS
Zn-Index+ 36**+ 47**+ 125**+ 117***
Cu-IndexNS+ 15*+ 31*+ 42**
S-IndexNS-7+ -7*NS
 
Statistical significance: NS (not significant), + (p < 0.1),* (p < 0.05), ** (p < 0.01), and *** (p < 0.001).

Comparison of the surface (0-4 inch) and underlying (4-8 inch) soil depths indicate that some degree of chemical stratification occurred for all tillage categories. Note that P, K, and Zn concentrations were greater (positive difference) for all tillage categories. Nutrient and pH stratification probably occurs in conventionally tilled fields since moldboard plows and rotary tillers are not used frequently, and since fertilizer banding is relatively common in North Carolina. Stratification in no-till fields is likely to be enhanced due to the surface application of materials and the lack of soil mixing.

Stratification of pH, Ca, Mn, Cu, and S was also noted for fields with NT < 3 years. For NT 3-6, in addition to the previously mentioned parameters, Mg concentrations were also significantly greater in the surface layer. The CT and NT > 6 fields did not exhibit consistent variation with respect to stratification of pH, Ca, Mn, or S. Thus, either these nutrients were uniformly distributed in these soils or variation was inconsistent across the tillage category.

Except for S, elements were more concentrated near the surface if stratification was detected. The S assay measures extractable sulfate (SO4-2), a highly leachable anion. Thus, S concentrations tend to be greater in lower soil layers. Although the P assay also measures extractable anions, phosphates are less mobile in soils due to complexation and precipitation reactions.

Data from the 0-2 inch sampling depths (data not shown) followed soil pH, P, Ca, and Mg stratification trends detected with the other sampling increments. For example, mean P concentrations were even greater in the 0-2 inch increment than in the 0-4 inch increment.

5) Do soil pH and nutrient concentration differ predictably among soil layers?

We use our statistical tests to evaluate how predictable these differences are. In general, we observed some stratification even in conventionally tilled fields. This is probably because soil mixing is not complete, especially when disks and chisel plows rather than moldboard plows are used. With the adoption of no-till methods, stratification becomes even more pronounced. Interestingly, stratification in pH, Ca, Mn, and S were more likely for fields in the early stages of no-till than in longer-term no-till.

6) With no-till management, does the surface soil layer become more acidic? Less acidic? Why?

Although surface soil pH was consistently higher than pH in underlying layers in the short-term no-till sites, the magnitude of these differences was often very small (0.2-0.3 pH units). Stratification was inconsistent in the long-term no-till plots (NT > 6). Thorough liming prior to the onset of no-till management has been promoted in North Carolina, and may have countered surface acidification in these soils (G. Naderman, unpublished). The less consistent soil pH profile in older no-till fields may be due to the variation in liming history among fields. In our study, some long-term (NT > 6) fields had received lime within as recently as 1 year, while others had not received any lime for 10 years prior to our sampling. Earlier detection and correction of surface acidification is one reason for shallower sampling in no-till. For these North Carolina fields with higher surface soil pH, shallower soil sampling could actually delay lime application since it fails to detect subsoil acidity.

Conclusions

Some chemical stratification occurred for all tillage categories, this became more pronounced with the adoption of no-till management. Soil sampling is a valuable tool for 1) specific lime and fertilizer recommendations, and 2) monitoring long-term soil changes. In consideration of specific lime and fertilizer recommendations: switching from a 0-8 inch to a 0-4 inch sampling depth in no-till situations yields higher soil P, K, Zn, and Cu levels, and thus reduces recommended fertilizer rates. In consideration of monitoring long-term soil changes: sampling no-till fields requires even more attention to depth than for conventionally tilled fields.

By: C.R. Crozier*, G.C. Naderman, M.R. Tucker, and R.E. Sugg

Published in: Communications in Soil Science and Plant Analysis 30:65-74, 1999.


Return to Soil Acidity & Liming - Part II