|
||
 | |
|
|
IDENTIFYING VALUE-ADDED TRAITS IN EXPANDED CROPPING SYSTEMS: AN UPDATE ON EXPLORATORY TRIALSJosh Posner (1), Ron Doetch (2), Janet Hedtcke (3), Jon Baldock (4) INTRODUCTIONResearch from the Wisconsin Integrated Cropping Systems Trial indicates that “expanded rotations” (longer than just corn-soybean) are important for the environmental sustainability of Wisconsin agriculture. Agronomic work is on-going for the addition of small grains, hay, and canning crops to the common corn-soybean rotation. Many of these options are quite productive and result in lower input use. However, due to the lower frequency of “program” crops (e.g. corn and soybeans) they are not as profitable as simpler rotations. As a result, it is crucial to identify ways to increase the value of the crops grown in these expanded low input systems. Our hypothesis is that expanded rotations with their increased use of organic sources of nitrogen, reduced or eliminated use of herbicides and altered soil flora and fauna, may be associated with grain traits such as altered amino acid profiles, higher protein, different starch fractions, or higher levels of nutriceuticals that would increase their nutritional value. Apparently a number of conventional processing sectors of the market (e.g. baby food, brewers, pet food, poultry, canning industry) are already willing to pay premiums for crops with special grain characteristics. In 2002, corn with protein greater than 7.8% received $0.08/point/bu premium and with soybeans there was a $0.01/pt/bu premium for protein+oil > 65% from poultry producers (Doetch, pers. Comm.). In addition, the poultry and pet food industries are looking for specific amino acids like methionine, cystein and lysine that are typically low in corn. As another example, ethanol plants have recognized that all corns are not equal in product yield. Improved values of “fermentable” starch greatly impacts their bottom line. One Minnesota plant currently has a list of corn varieties that will receive a $.05/bushel premium. On site testing equipment will soon be available to test for fermentable starch. Brewers are interested in increasing the total starch fraction in corn they purchase. A typical plant will earn approximately $1 million per point increase in total starch. This report focuses on exploratory work, initiated in 2000, on the effect of cropping system on grain quality. MATERIALS & METHODSCorn and soybean grain was harvested from a number of different production plots and research trials. All samples (whole kernels, dried to 60° F) were tested for grain quality (protein, oil, starch and kernel density) using a Foss Infratec 1229 Near Infrared Transmittance (NIT) instrument at Quality Traders Inc (QTI). The Infratec 1229 compares reflectance on 10 small samples that total approximately 1 pound. Over 5,000 wet chemistry lab test were used in writing the protein, oil and starch calibrations resulting in standard error of predictions of 0.2%, 0.3%, and 0.5%, respectively. All values are expressed at 0% moisture. Newer calibration for specific amino acids and fermentable starch exist, but unfortunately are not yet sufficiently well developed for rapid “load-by-load” testing at grain elevators. RESULTSTRIAL #1. ON-FARM SCREENING OF QTI CORN HYBRID 9664. At the elevator in Sunrich MN, grain from producers growing QTI corn hybrid 9664 was collected for NIT analysis. These samples came from a range of corn-based rotations and different management systems (fertilizer rates, herbicides, tillage). In 2000, 20 environments (farm fields) were sampled, and in 2001, 33 environments were sampled. (1) all numbers at 0% moisture As can be seen in Table 1, the grain quality traits for hybrid QTI 9664 were actually quite stable across these 53 random “environments”. TRIAL #2. SOYBEAN ROTATION AND MANAGEMENT TRIAL Soybeans from the 2001 season were collected from Joe Lauer’s long-term factorial rotation trial. Three rotations were sampled: 1) soybeans following 5 years of corn; 2) soybeans alternated with corn every year; and, 3) the 5th year of continuous soybeans; the two tillage treatments were no-till and conventional; and the two row spacing were 7.5 and 30”. The variety of soybean was ‘Asgrow 2301’. As can be seen in Table 2, there were statistical differences for main effect of rotations and tillage treatments. Highest protein levels were found in continuous soybeans and highest oil in first year soybeans following corn. Conventional tillage resulted in higher protein levels and no-till higher oil levels. However, the change in the grain quality traits was so small as to have no economic significance. There was no effect of row spacing and there were no interactions. (1) Corrected to 13% moisture. The NIT calibration on this Infratec 1229 was set up to indicate corn component value at 0.0% moisture while indicating soybean values at 13% moisture. This is typical problem area as nutritionist normally use “as-is” values and grain trade is generally charted at to 0.0% moisture to determine premiums. TRIAL #3. TEN CORN HYBRIDS MANAGEMENT TRIAL. Ten QTI hybrids were evaluated for yield and output characteristics under varying agronomic practices on land adjacent to the WICST trials at the Lakeland Agricultural Complex. The ten hybrids were planted following corn or soybeans to evaluate the effect of previous crop on yield and output characteristics. Superimposed on this factor were two N rates (120 and 160 N) and three planting populations (33,000, 30,000, and 27,000 pl/a). The six row plots were conventionally tilled, nitrogen was side dressed as UAN and accent + buctril (2/3 oz and 1 pt/a respectively) were applied at V6. Other fertilizer included 100 pounds of 9-23-30 applied as starter. Two rows of 25 feet were harvested with a plot combine and samples analyzed by QTI for grain quality. As can be seen in Table 3, there was no “main effect” of previous crop (row 2), N-rate (row 5), or population (row 9) on grain characteristics, nor were there any significant interactions between hybrids and these management components (rows, 4, 7, 11). Not surprisingly, there was a statistical difference, between hybrids for % protein and % oil (row 3). Table 4 represents the main effect of hybrid averaged across previous crop, N rate, and population. As can be seen, QTI 9664 (row 4) was the highest performer for oil and protein. TRIAL #4. WISCONSIN INTEGRATED CROPPING SYSTEMS TRIAL: The corn phases of the core rotation trail were harvested and analyzed for grain quality traits in 2001 and 2002 at Lakeland, and only in 2002 at Arlington Lakeland (LAC): Results from the Lakeland site are presented in Table 5. In both years, corn yields were very erratic due to heavy spring rains and very droughty summers creating difficult conditions for weed control (delayed rotary hoeing in the organic systems, delayed application of herbicides in the conventional systems). In 2001, all the corn plots were planted to the same hybrid (Cargill 4111) and there were significant grain quality differences between cropping systems. Both forage rotations had significantly higher corn yield (due to lower weed pressure), and higher protein levels (due to a richer N-environment) and grain density compared to the grain systems. In 2002, the conventional systems were planted to hybrid Agrigold A6382 (RM=105) while the organic systems were planted to shorter cycle Growmark FS3969 (RM=95). Among the conventional systems, the poor yields of continuous corn (CS1) probably explain the high protein levels (it was fertilized for 160 bu/a yield). With the organic systems, our hypothesis is that N-deficiency explains the low yield and protein levels in the grain system (CS3) vs. the forage system (CS5) with its alfalfa and manure credits. The other traits generally were not affected by cropping system. * = p< 0.10, ** = p<0.05, *** = p<0.01 Arlington Research Station (ARS). The WICST plots at ARS were also used for this exploratory study in 2002. The same two hybrids were used as at Lakeland: A6382 on the conventional systems (CS1, CS2, and CS4); and FS3969 was used for the organic systems (CS3 and CS5). Five subsamples, each from 10 adjacent plants, were collected in each plot at harvest to study the variability of grain characteristics within plots as well as between systems. The variance components for experimental error and sampling error were determined with the Restricted Maximum Likelihood method as implemented in JMP 4.04. As can be seen in Table 6, A6382 had similar grain characteristics, regardless of which of the three cropping systems it was grown in. The organically managed systems did have a significant influence on %protein, % starch and density for hybrid FS3969. The forage system was higher in these aforementioned traits likely due to the accumulated legume credits and manure application the previous fall. The low protein in the organic grain rotation suggests a N-deficiency during grain fill, similar to the situation at Lakeland in 2002. * = p< 0.10, ** = p<0.05, *** = p<0.01 The results of the subsampling effort are summarized in Tables 7a and 7b. Table 7a shows that the variability among subsamples was larger than that across blocks (also called replications in these trials) and the variance components were relatively consistent between the two hybrids. The dominance of the sampling error is even more clear when each component is expressed as a percent of the total random variation (Table 7b). The meaning of the dominating effect of the sampling error is that the variability within these large plots is greater than the variability between blocks (replications). Consequently, more progress will be gained in reducing the error and detecting significant differences by increasing the number of subsamples than by increasing the number of blocks. While more subsamples would help detect differences, once such variability is quantified, the usual procedure is to make one well mixed composite from them so there is only one sample to analyze. TRIAL #5 ALTERNATIVE CORN PRODUCTION ENVIRONMENTS AT ARS. In addition to the WICST plots, ‘Agrigold A6382’ was planted on two other fields with conventional tillage at Arlington in 2002. One field was in corn following corn and was divided in half: one side was manured (30 ton/a) and the other received anhydrous ammonia fertilizer (120#N/ac). In a second field was corn following corn and was split into two pieces: one piece had deliberately low fertility (STP=17, STK=59) and the other had typically high fertility (STP=55, STK=212). Recommended rates of 160 # N/acre were applied to both halves. Five samples were collected from each strip near harvest and the data was analyzed as independent two-treatment experiments with a t-test. * = p< 0.10, ** = p<0.05, *** = p<0.01 As can be seen in Table 8, there were no differences between manured and nonmanured corn except for the 300-weight which was significantly higher under the manured treatment. Not surprisingly, high N fertilization in a low yielding situation (very low soil P and K) resulted in elevated protein levels. CONCLUSIONResults from this exploratory work suggest that macro grain quality factors like % protein, % oil and % starch in corn and perhaps soybeans are not particularly sensitive to cropping system. Relatively small differences in these factors however, were observed under very varied cropping systems. Future work will focus on selecting hybrids or varieties with known marketable characteristics (e.g. methionine levels) and see if this specific trait is enhanced by alternative cropping system
|
|
||
|
Home | Who we are | Research | Publications | Links | Site map | Contact us University of Wisconsin-Madison Site hosted by UW-Madison Center for Integrated Agricultural Systems |
|||||