The Effect of Four P & K Rates on the Drawdown and Buildup of Soil Test Levels

Illinois Fertilizer Conference Proceedings
January 24-26, 1994

Home 1994 Index Search

The Effect of Four P & K Rates on the Drawdown and Buildup of Soil Test Levels

Michael J. Mainz and Glenn A. Raines¹

spacer

Introduction

Crop yield response to added phosphate or potash fertilizer has been minimal most years since these studies were started in 1980 and 1982 for the Orr and Northwest Research Centers, respectively. Soil tests values have been inconsistent several years with known response curves. This revised study was started in 1990 to a) monitor drawdown and buildup of pH, P₁, P₂ and K soil test levels at the 0-8, 8-16 and 16-24 inch depths, when 4 rates of phosphorus (P) and potassium (K) are applied annually to dark prairie and moderately dark forest soils that have received varying rates of P or K biannually and annually, for 8 and 10 years at the NW and Orr Research Centers respectively, and then annually since 1990; b) correlate soil test changes at the 3 depths with fertilizer applications and removal in the grain; c) monitor movement of P & K in the 0-24 inch soil profile; and d) monitor the effects of P & K rates on grain nutrient analysis, grain yields and crop growth for corn, soybeans and wheat (wheat at Orr only).

Materials and Methods

Separate P and K rate studies were started in 1980 and 1982 at the Orr and Northwest Agricultural Research Centers, respectively. Annual applications of 0, 60, 120 and 180 lbs P₂O₅ and 0; 100, 200 and 300 lbs K₂O/acre were made to the respective plots in each study. These rates were compared on corn, soybeans and wheat each year at the Orr Center (Perry). Because of high initial soil test levels at the NW Center (Monmouth) P₂O₅ and K₂O rates of 0, 40, 80 and 120 were applied every other year rather than annually in an attempt to lower the soil test levels. Starting in 1990 all of these large plots were subdivided into 4 subplots for the split or "New" rates. For example: every "Old" 0, 40, 80 and 120 lb/acre plot now receive 0, 40, 80 and 120 lb P₂O₅ or K₂O/acre annually. Instead of 4 treatments there are now 16 combinations of the old and new P or K rates.

Other changes made in 1990 include: a) additional soil sampling depths of 8-16 and 16-24 inches are now taken annually and analyzed for pH, Pl, P2 and K, b) leaf tissue samples were taken at early pollination for corn and soybeans and analyzed for 11 elements (these samples were discontinued in 1993), c) grain samples were collected and analyzed for 11 elements, and d) surface soil samples are now analyzed for organic matter (OM) and cation exchange capacity (CEC).

Uniform nitrogen (N) applications are made annually to the wheat and corn. Ag lime and P or K are applied to maintain soil tests at recommended levels. Herbicides and/or mechanical tillage are used to control weeds. Insecticides are applied when needed to control pests. Hybrid/variety selection is maintained in 3 year cycles. Tillage practices were changed at Monmouth in 1985 and again in 1990. Prior to 1985 and starting again in 1990, corn stalks were chisel plowed, generally in the fall, soybean ground was left untouched until prepared for planting with secondary tillage implements such as the disk or DynaDrive. Following harvest at Perry all ground was chisel plowed with the exception of when the moldboard plow was used in 3 year cycles following wheat.

Results and Discussion

Plant responses to previous soil tests and added phosphorus or potassium fertilizer rates were very small, if any, in 1993. Rather than listing all 16 fertilizer rate combinations and their effect on plant parameters only the response to the old and new P or K rates are given in the tables below. Very few significant interactions between the old and new P or K rates were observed, none of which were for grain yield. Contrast analysis was run on all the data and those showing a significant Linear (Lin), Quadratic (Quad) or Cubic response are given in the tables. Most of the parameters were not significantly (N.S.) different at the 0.05 confidence level. The "Soil Tests" listed in Tables 1, 2, 3, and 4 are the average of either the old or new rates for the 0-8 inch sample taken after harvest in 1992. Actual soil test values have a range from low to high for each of the values listed. Figures 13, 14, 15, and 16 give the actual range of soil test values found in 1992.

Soil and grain samples collected in 1993 have not been analyzed and that data will not be available until later this winter. Leaf tissue samples were collected but will not be analyzed until sufficient funds become available to do so.

Grain yields were converted to a "Relative Yield" basis with 100% equal to the highest yield in any given year so that all years can be compared. These "Relative" yields were then compared to the previous years soil tests, either P₁, or K for each respective field. Yield response at the 4 fertilizer rates are indicated in each graph. Phosphorus and potassium soil test levels are lower at Perry than Monmouth by 20-30 lbs/a and 100-150 lbs/a, respectively. Optimum soil test levels at Monmouth appear to be in the 40-501b/a (P,) and 250-350 lb/a (K) range (Figures 1&2, 5&6). Optimum levels are lower at Perry and generally in the 20-40 lb/a (P₁) and 200-3001b/a (K) range (Figures 3, 4, 7, 8, 9, and 10).

Both P and K soil tests are significantly affected by sampling date and soil moisture level. Generally all soil samples have been taken in the fall after harvest at both locations. But, there have been some years when it was necessary to take them in the spring. Prior to making the split (New Rate) fertilizer applications in 1990 soil samples were taken in 8 inch increments to a 24 inch depth. At Monmouth we sampled the 0-8 inch depth the previous fall and the 8-16 and 16-24 inch depths the following spring. Figures 11&12 show the difference in soil test results caused by fall vs. spring sampling. Spring sampling appears to have increased Pl tests by 30-50 lbs/a and K tests by 100-150 lbs/a.

Peaks in the Monmouth data (Figures 13&15) during 1986 & 88 correspond to when ridges were made and an above normal summer precipitation in 1986. Similar increases in precipitation appear to have increased soil test levels at Perry, especially during the years of 1984-86 and 1991. Several of the peaks and valleys in the annual soil test values at Perry appear to be related to precipitation during May-September. Additional study is needed to prove this theory.

Net fertilizer addition rates are shown in Figures 17, 18, 19, and 20 for representative fields in the P & K rate studies at both locations. Removal rates were calculated using grain nutrient analysis data from 1990-92 and book values for previous years. Results were graphed using rate averages for each year of the study or total for the length of each respective study. The totals gave a better fit to the line than did yearly averages. Estimates of fertilizer needed to raise soil tests one pound per acre are 10.5 and 17.8 for P₂O₅ and 5.7 and 7.5 for K₂O for Monmouth and Perry, respectively. These values will be calculated rather than estimated at the completion of the study.

Summary

Soil testing is not an exact science. There are many factors affecting the results, many of which we have no control over nor totally understand. Yield response to added phosphorus or potassium fertilizer varies with soil fertility levels, crop, year and other factors. After 12 and 14 years of none to high P or K applications, soil tests now range from very low to very high at Perry and medium to high at Monmouth. Predictions on removal and the subsequent affect on soil test is proving difficult to calculate and make predictions with. Movement of P or K through the soil profile into lower depths appears to be occurring, but at very slow rates. We hope to continue these studies as long as funding is available and this type of research is needed. Follow the fertilizer recommendations given in the Illinois Agronomy Handbook.