Abstract

Soils in northeastern Oklahoma are generally acid where winter wheat is grown.  Experiment 802 was established in the fall of 1977 to evaluate one-time fixed rates of phosphorus fertilizer in continuous winter wheat.  All one-time preplant P rates, excluding the 1200 P₂O₅ rate, became inadequate within 7 years (60-600 lb P₂O₅/ac) on this highly acid soil.  Soil test P levels were 100% sufficient for the 600 and 1200 lb P₂O₅ rates in 1995, however, the 600 lb P₂O₅ rate no longer resulted in grain yields near that achieved by the 1200 lb P₂O₅ treatment.  The one-time application of 1200 lb P₂O₅/ac in 1977 has resulted in near maximum grain yields for 20 years.  From 1988 to 1997, yields were depressed due to the effects of decreased soil pH.  Following 20 years of continuous wheat, all plots were limed in 1998 and 1999 (1 ton 71% ECCE).  In the last two years (1999-2000) following the lime application, the one-time 1200 lb P₂O₅ (applied in the fall of 1977) has resulted in average grain yields in excess of 45 bu/ac.  Although frequency of application was not evaluated in this study, higher rates over less frequent intervals may prove to be advantageous for highly acid soils.  Based on high soil test P levels for the 600 lb P₂O₅/ac rate in 1995 (32.6 ppm or a P soil test index of 65.2 which is 100 % sufficient), and observed decreased yields in recent years, these results also suggest the need for re-calibration of soil test P levels at low pH.

Materials and Methods

Experiment 802 was established in the fall of 1977 under conventional tillage on a Taloka silt loam (fine-mixed, thermic Mollic Albaqualf).  Wheat has been planted for 19 continuous years in 10 inch rows at seeding rates of 90 pounds per acre (rates increased when planting was delayed).  Fertilization, planting and management dates are listed in Table 1. The experimental design in this continuing experiment employs  a randomized complete block with four replications.  Fertilizer treatments used in this experiment along with average grain yields for selected periods are reported in Table 2.  Individual plots at this site are 17 feet wide and 50 feet long.  The center 10 feet are harvested with a conventional combine the entire 50 feet in length.  In addition to wheat grain yield measured every year, periodic soil and grain samples were taken for further chemical analyses.  Results from surface (0-6 inches) soil samples collected in 1995 are reported in Table 3.

Results

From 1978 to 1986, wheat grain yields were maximized from the one-time application of 60 lb P₂O₅/ac applied in the fall of 1977.  From 1987 to 1997, the maximum yield was achieved from applied P at a rate of 1200 lb P₂O₅/ac in the fall of 1977.   When observing relative wheat grain yields plotted by treatment over time, it is evident that all P treatments, excluding the 1200 P₂O₅ rate, became inadequate soon after 1985, with little difference between the 240 and 600 lb P₂O₅ rates (Figure 1).  Soil test P levels were 100 % sufficient for the 600 and 1200 lb P₂O₅ rates in 1995, however, the 600 lb P2O5 rate no longer resulted in grain yields near that achieved by the 1200 lb P₂O₅ treatment.  The one-time application of 1200 lb P₂O₅/ac in 1977 has resulted in near maximum grain yields for 20 years.  Although frequency of application was not evaluated in this study, higher rates over less frequent intervals may prove to be advantageous for highly acid soils.  Based on high soil test P levels for the 600 lb P₂O₅/ac rate in 1995 (32.6 ppm or a P soil test index of 65.2 which is 100 % sufficient), and observed decreased yields in recent years, these results also suggest the need for re-calibration of soil test P levels at low pH.

Table 1.  Treatment  applications, and experimental management for continuous winter wheat Experiment 802, Haskell, OK, 1978-2000.

L – lime applied at a rate of 2240 kg/ha (1 ton/acre) in July 1998 (71% ECCE)
L – lime applied at a rate of 2240 kg/ha (1 ton/acre) in July 1999 (71% ECCE)

Table 2.  Treatment structure and wheat grain yields of long-term wheat Experiment 802, Haskell, OK, 1978-2000.

N applied as 34-0-0, P as 0-46-0, and K as 0-0-60. SED - standard error of the difference between two equally replicated means. CV - coefficient of variation.

Table 3.  Treatment structure of long-term winter wheat Experiment 802, and surface (0-6 inches) soil test analyses from samples collected in 1995, Haskell, OK.

N applied as 34-0-0, P as 0-46-0, and K as 0-0-60. SED - standard error of the difference between two equally replicated means.  CV - coefficient of variation.

Figure 1.  Relative wheat grain yields, by treatment and year, Experiment 802, Haskell, OK, 1978-2000.