Illinois
Fertilizer Conference Proceedings
January 26-28, 1998
| Home | 1998 Index | Search |
| Home | 1998 Index | Search |
E.C. Varsa, S.A. Ebelhar, S.K. Chong, and S.J. Indorante 1
Variable Rate Technology (VRT) and associated application systems provide a means of assuring fertilizer applications are made only in amounts and locations where they are needed (Wollenhaupt, et al., 1993). Agronomically, variable-rate systems provide a means of targeted application of fertilizer based upon detailed soil tests and other related databases. Economically, variable-rate systems allow fertilizer dollars to be spent on areas within a field where they will provide a response, and to be saved where a response is unlikely. Environmentally, variable-rate systems help to prevent over-application of fertilizer where it could result in environmental problems. Research into successfully implementing VRT is in its early states (Francis, et al., 1996; Schefcik, et al., 1996), and some economic benefits have been reported.
More efficient crop production through the use of VRT has received widespread attention in the fertilizer industry and farmer/producer publications. Advances in computer software technology and improvements in fertilizer application equipment that reliably delivers what is desired when it is desired is now available. However, considerable uncertainty abounds as to just how to fully utilize VRT in the management of fertilizers across farm fields.
A goal of these studies is to help fertilizer dealers and farmers better understand the interaction of soil chemical properties (soil tests and crop responses) with soil physical properties. In southern Illinois, yields of corn and soybeans are correlated to the available water supplying power of the soil. In drought years, yield patterns can be closely tied to the presence of naturally occurring claypans that restrict root development for nutrients and water. Kitchen et al. (1996) reported that the depth to claypan, as assessed by electromagnetic conductivity (EM) readings, closely correlated with crop yield response especially in dry years.
Two southern Illinois farm fields have been identified which have distinct variations in terms of morphological properties but have somewhat uniformly low levels of soil test potassium. Soil sites within each field (called squares) with a wide range in depth to claypan will be used as sites for crop response studies to increasing rates of K application.
The objectives of this research were to:
Two commercial farm fields, one in Jefferson County, IL and the other in Pope County, IL, were identified in April 1997 as sites for this research. The Jefferson County field is about 24 acres in area and the soils are predominantly Wynoose and Bluford silt loams. The Pope County field is about 18 acres in area and the soil is predominantly Grantsburg silt loam. The previous crop for the Jefferson County field was corn (east portion) and red clover (harvested for hay in June 1997 in west portion). The previous crop at the Pope County field was no-till corn.
The fields were gridded into about 0.45 acre parcels using GPS navigational systems. Center points of each grid, about 140 ft spaced, were used as loci for initial soil sampling for basic soil fertility assessment. Five subsamples of soil were collected and composited to form a sample taken from within a 10 foot radius of each grid center point. A total of 54 grid samples were collected at the Jefferson County field site and 45 grid samples were collected at Pope County. These grid center points were also used as loci for the collection of elevation data (topography), depth to claypan measurements, soil bulk density assessments, and electromagnetic induction (EM) readings.
Based upon the initial soil tests for K and soil topography/soil depth variations within each field, 10 grid cells (called squares) were identified at Jefferson County (Figure 1) and nine cells were identified at Pope County (Figure 2) for intensive variable K rate evaluations in 1997. Each grid cell (square) was subdivided into 16 plot areas in which additional soil samples for K analysis were collected and which formed the individual plots for K rate evaluations. Potassium rates of 0, 40, 80, and 120 lb K20/acre as muriate of potash (0-0-62) were applied in a Latin square design (4 K rates x 4 reps). Individual plot sizes were 15 ft by 30 ft. The K rate treatments were broadcast-applied as a liquid formulation immediately following corn planting at Jefferson County and soybean planting at Pope County.
Ear leaf samples of corn and trifoliate leaf samples of soybean were collected from each plot to assess K concentrations in the plant tissue as a function of site and K treatment effects. For harvest, each individual plot had the center 2 rows by 20 feet harvested for yield and moisture correction. After individual plots were harvested the whole field was later harvested with a combine equipped with a yield monitor and GPS guidance system for whole-field yield variability assessment.
Field maps of elevation above mean sea level (site topography) for the Jefferson County and Pope County locations are given in Figure 1 and Figure 2. As one can observe, the Jefferson County field (Figure 1) is rather flat with most of the field having a slope of one to three percent. Northwest portions of the field are more strongly sloping (three to six percent). The flatter areas are dominated by Wynoose soils (with strongly developed claypans) whereas the more strongly sloping portions of the field are Bluford soils. The Pope County field (Figure 2) is steeply sloping on the north and south ends (five to ten percent) with gentler slopes through the mid portions. The dominant soil throughout the field is a Grantsburg series.
Soil test levels for K in the selected grids (squares) used for the potassium rate studies for the Jefferson County and Pope County fields are shown in Table 1 and Table 2, respectively. The 10 squares at the Jefferson County site (Table 1) averaged 141 lb K/acre with a range of 117 to 169 lb K/acre. These K test values were from the initial soil samples collected in April at grid centers to obtain baseline fertility levels. Later, in June, soil samples were taken from the K0 plots (no K2O applied; 4 plots per grid square), and for that set of samples, the average K test was 117 lb K/acre and the range was from 81 to 178 lb K/acre. Therefore, the average soil test decreased from the April to the June samplings but in both cases indicated that the field was well below the suggested optimum soil test value of 260 lb K/acre for corn or soybeans. A response to applied K would therefore be expected. The soil pH for the field ranged from 6.9 to 7.6 (data not shown).
Equally low soil test K levels were observed at the Pope County location (Table 2). Initial soil samples for K to establish baseline fertility levels (taken in April) revealed an average K test of 147 lb K/acre for the nine grid squares. Samples taken two months later from K0 plots across the nine grids (n = 36) showed that the average K test decreased to 130 lb K/acre and ranged from 87 to 261 lb K/acre for the 36 samples. A recommendation for K fertilization would be warranted for this field to achieve optimum production of corn or soybeans. The soil pH for the field ranged from 6.0 to 6.9 (data not shown).
Increased levels of ear leaf K were observed with increasing rates of applied K in all four squares (19, 20, 21, and 29) where corn followed corn at the Jefferson County location but only four of six squares responded to increasing K fertilization where corn followed the removal of red clover (Table 3). Also the average ear leaf K levels were considerably lower in the squares (41, 42, 43, 44, 47 and 48) where corn followed red clover. It was apparent in all squares that enhanced leaf K resulted from K fertilization and the tissue K increase was about 0.2 percent for the 120 lb K2O/acre rate compared to the control.
Factors other than applied K (probably drought) were most responsible for final grain yield at Jefferson County even though plant uptake of K was increased by K fertilization (Table 4). A trend, though not significant, toward a decrease in yield was observed in six of ten squares and a significant decrease was observed in a seventh square. A trend toward an increase in yield was observed in three squares with one being significant. Correlations of yield and ear leaf K with soil test K were mostly non-significant (Table 1), suggesting that soil characteristics and site factors relating to drought were probably more responsible for corn yield variations in 1997 than K fertility.
At Pope County, soybeans responded more positively to applied K in terms of increased leaf K levels and yield (Table 5 and Table 6). Averaged over all nine squares, leaf K composition increased linearly from 2.37 percent for the control to 2.60 percent for the 120 lb K2O/acre rate. At the same time a significant linear yield increase was observed with added K (about 1.5 bu/acre for high rate of K). The largest yield increase with added K was observed in squares 69 and 70 which had the greatest topsoil depth to claypan and to the fragipan. It is obvious that when soils are responsive to K having a large volume of soil non-restrictive to roots is important to the optimization of yield. The seasonal growing conditions for soybeans were nearly ideal at this location and, as a result, yields averaged from 51 to 63 bu/acre across cells and K rates.
Two field locations were identified for variable rate K studies on corn and soybean, one in Jefferson County and the other in Pope County, IL. Both fields had soil types and topography typical for the soils of the area. Soil test K levels were "low" at both sites, averaging 141 lb K/acre in Jefferson County and 147 lb K/acre in Pope County. Ten grid cells (squares) with differing K levels and depths to claypan were used as sites for K rate studies at Jefferson County and nine squares were utilized at Pope County. Potassium rates evaluated were 0, 40, 80, and 120 lb K2O/acre.
Corn yields obtained in 1997 at Jefferson County were strongly affected by drought factors such that there were mostly non-significant correlations between yield and soil test K level and between yield and added fertilizer K. Even though drought conditions prevailed, ear leaf K composition was increased in most of the grid squares with added K but this increased leaf K did not translate to increased yield. Data analysis is yet incomplete to determine what factors were most strongly correlated to yield at this drought-stressed site in 1997. At Pope County soybean leaf K composition and yield responded linearly to applied fertilizer potassium in most all cells. The greatest responsiveness to added K was observed in those squares that had the greatest depth to the claypan or fragipan.
The authors wish to express their deep appreciation to Mr. Matt McCauley and Mr. David Weber of the Natural Resources Conservation Service for their assistance in taking EM, depth to claypan, and elevation (topography) measurements.
Table 3. Potassium rate effects on corn ear-leaf K concentrations, Jefferson County site, 1997.
Table 4. Potassium rate effects on corn grain yields, Jefferson County site, 1997.
Table 5. Potassium rate effects on soybean trifoliate-leaf K concentrations, Pope County site, 1997.
Table 6. Potassium rate effects on soybean grain yields, Pope County, 1997.
1E.C. Varsa is an Associate Professor, Plant, Soil and General Agriculture Dept., SIUC; S.A. Ebelhar is an Agronomist, Dixon Springs Agricultural Center, University of Illinois, Simpson, IL; S.K. Chong is a Professor, Plant, Soil and General Agriculture Dept., SIUC; S.J. Indorante is a Soil Scientist, Natural Resources Conservation Service, Carbondale, IL.
